Efficient synthesis of p-chlorobenzaldehyde through liquid-phase oxidation of p-chlorotoluene using manganese-containing ZSM-5 as catalyst

Article abstract of DOI:10.1039/c5ra16206h

Manganese-containing MFI-type Mn-ZSM-5 zeolites were prepared and characterized by XRD, UV-vis DRS, SEM, XPS, N₂ adsorption-desorption, NH₃-TPD and ICP-AES techniques. The zeolites show high catalytic activity and selectivity in the heterogeneous oxidation of p-chlorotoluene to p-chlorobenzaldehyde. The effects of catalyst concentration, water addition, HBr amount, as well as reaction time and temperature on product yield were investigated. Under the optimized conditions (catalyst 20 mg, p-chlorotoluene 1 mL, solvent (acetic acid) 10 mL, HBr (40 wt%) 30 mg, H₂O 3 g, oxygen flow rate 50 mL min^-1, time 8 h, temperature 100 °C), the Mn-ZSM-5 (Si/Mn = 48, Mn 1.7 wt%) catalyst shows p-chlorotoluene conversion of 93.8% and p-chlorobenzaldehyde selectivity of 90.5%. The excellent catalytic activity can be attributed to the distribution of Mn species and the mild acid sites.

Full text of DOI:10.1039/c5ra16206h

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Efficient synthesis of p-chlorobenzaldehyde
through liquid-phase oxidation of p-chlorotoluene
Cite this: RSC Adv., 2015, 5, 74162
using manganese-containing ZSM-5 as catalyst
Wei-Fang Zhou, Lang Chen, Jun Xie, Chak-Tong Au and Shuang-Feng Yin*a
a
a
a
ab
Manganese-containing MFI-type Mn–ZSM-5 zeolites were prepared and characterized by XRD, UV-vis
2 3
DRS, SEM, XPS, N adsorption–desorption, NH -TPD and ICP-AES techniques. The zeolites show high
catalytic activity and selectivity in the heterogeneous oxidation of p-chlorotoluene to
p-chlorobenzaldehyde. The effects of catalyst concentration, water addition, HBr amount, as well as
reaction time and temperature on product yield were investigated. Under the optimized conditions
Received 12th August 2015
Accepted 20th August 2015
(
2
catalyst 20 mg, p-chlorotoluene 1 mL, solvent (acetic acid) 10 mL, HBr (40 wt%) 30 mg, H O 3 g,
ꢀ1
ꢁ
oxygen flow rate 50 mL min , time 8 h, temperature 100 C), the Mn–ZSM-5 (Si/Mn ¼ 48, Mn 1.7 wt%)
catalyst shows p-chlorotoluene conversion of 93.8% and p-chlorobenzaldehyde selectivity of 90.5%. The
excellent catalytic activity can be attributed to the distribution of Mn species and the mild acid sites.
DOI: 10.1039/c5ra16206h
www.rsc.org/advances
With micropores close to the molecular diameter of many
desired products, zeolites are used as shape-selective cata-
As an important intermediate in the production of dyes, phar- lysts. It is known that the Mobil Fih (MFI)-type ZSM-5
1
. Introduction
7,8
maceutical drugs, optical brighteners and agricultural chem- zeolites are unique because their three-dimensional micropo-
1
icals, p-chlorobenzaldehyde (PCB) is manufactured on an rous channels are accessible only for molecules that are within
9,10
industrial scale by side-chain chlorination of p-chlorotoluene 0.6 nm in diameter.
There are reports on the selective
(
PCT) followed by acid hydrolysis using soluble acetates of adsorption of benzene and p-substituted benzenes over that of
2
transition metals as catalysts. The process, however, suffers o- and m-substituted aromatic products on nano-sized ZSM-5
1
1–13
from low conversion and poor selectivity, and with the use of catalysts.
As reported, ZSM-5 zeolites modied with
3
14–18
soluble catalysts, it is hard to be environmentally-benign. Thus, manganese show attractive properties and reactivity.
Excel-
research efforts were employed to investigate the use of solid lent catalytic performance is ascribed to the well-isolated Mn
4
catalysts.
species in the zeolite framework or those anchored on the
17
To be commercially viable, the process for oxidation of surface. In addition, oxygen can be easily adsorbed on the
aromatic hydrocarbons by molecular oxygen or hydrogen defective sites and strongly bound to the manganese centres
5
16
peroxide has to be highly selective. In the last decades, there that are active for selective oxidation. Previously, we synthe-
were formulations for PCT oxidation using heterogeneous sized a series of manganese oxide octahedral molecular sieves
3,4,6
catalysts.
For example, Bautista et al. reported 19% PCT that show excellent catalytic performance towards the oxidation
19–21
conversion and 70% PCB selectivity with aqueous H O over of PCT for the formation of PCB.
vanadium oxide catalysts supported on TiO –sepiolite. On the investigation to the use of supports that have manganese active
Lately we extended our
2
4
2
2
other hand, Yoo reported 28% PCT conversion and 61.6% PCB sites, and explored the relationship between PCB selectivity and
selectivity over Fe/Mo/borosilicate molecular sieves in the the distribution of active sites.
6
presence of O
2
. In addition, Wang et al. achieved PCB yield of
In this article, we report for the rst time the liquid-phase
over cobalt-doped mesoporous catalytic oxidation of PCT (O as oxidant) to PCB in a reux
4
8.5% with aqueous H
2
O
2
2
3
titania. Nevertheless, those approaches are unsatisfactory system using hydrogen bromide as reaction initiator (Scheme 1)
either in conversion or in selectivity. It is therefore necessary to over manganese-containing ZSM-5. They are Mn–ZSM-5 (Si/Mn
design solid catalysts active enough for the oxidation reaction. ¼ 48, Mn 1.7 wt%) prepared by a hydrothermal method and Mn/
ZSM-5 prepared by an impregnation method. The former shows
better activity (93.8%) and selectivity (90.5%) than the latter. As
revealed by XRD, UV-vis DRS, SEM, XPS, N2 adsorption–
a
Hunan University, College of Chemistry and Chemical Engineering, Changsha,
desorption, NH -TPD and ICP-AES analysis, the catalytic
3
Hunan, China. E-mail: sf_yin@hnu.edu.cn
b
performance is signicantly inuenced by specic surface area,
distribution of active sites and mild acid sites.
Hong Kong Baptist University, Chemistry, Waterloo Road, Kowloon Tong, Kowloon,
Hong Kong, China
74162 | RSC Adv., 2015, 5, 74162–74169
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
scanning microscope with a Schottky emitter at an accelerating
voltage of 5 kV and a beam current of 1 mA. X-ray photoelectron
spectroscopy (XPS) analyses were conducted with a K-Alpha
1063 (Thermo Fisher Scientic, USA) spectrometer using Al
Ka radiation (1486.6 eV). The C 1s signal of contaminant carbon
(
BE ¼ 284.6 eV) was taken as a reference for BE calibration. The
specic surface areas and pore size distributions were
measured on a Tristar 3000 instrument by the Brunauer–
Emmett–Teller (BET) and Saito–Foley (SF) method, respectively.
The element analysis was carried out using ICP-AES (IRIS1000).
2
.4. Catalyst evaluation
All reactions were carried out in a three-necked round-bottom
ask equipped with a reux condenser, and the experimental
Scheme 1 Liquid-phase catalytic oxidation of p-chlorotoluene with
22
2
O over Mn–ZSM-5.

21
setups are similar to those of our previous work. Oxygen was
introduced at atmospheric pressure into the solution at a
desired ow rate, and the ask with a magnetic stirrer was
placed in an isothermal paraffin oil bath. Aer the reaction, the
2. Experimental
2.1. Materials
The chemical reagents were of analytical grade and were used catalyst was removed by centrifugation. The products were
without further purication. No impurities were found in dissolved in acetonitrile and qualitatively analyzed by GC/MS
p-chlorotoluene by GC analysis. Pure gaseous O was used as the (6890N/5973N) using an Agilent HP-5MS capillary column
2
oxygen source.
(30 m ꢂ 0.45 mm ꢂ 0.8 mm). For quantitative determination,
the collected liquid products were analyzed on an Agilent 7820A
GC with FID and an Agilent AB-FFAP capillary column (30 m ꢂ
2.2. Catalysts preparation
0
.25 mm ꢂ 0.25 mm). The internal standard method was
All silica ZSM-5 zeolite was prepared by hydrothermal synthesis
using tetraethylorthosilicate (TEOS, Aldrich, 98%) as a silicon
source and TPAOH (25 wt%) aqueous solution as a structure-
directing agent. Typically, 12.5 g of TEOS, 12.5 g distilled
water, and 12.4 g aqueous TPAOH (25 wt%) were sequentially
added and adequately mixed. The well-mixed solution (solution
A) was stirred for 3 h. Then the synthesis was carried out in an
employed with biphenyl being the standard.
3
. Results and discussion
23
3.1. Catalyst characterization
The XRD patterns of the M–Z (x), M/Z (2.0 wt%) and regenerated
M–Z (48)-1 aer the rst cycle of reusability test are shown in
Fig. 1. They show typical MFI-type structure and the peaks
match those of ZSM-5. There were no detections of peaks
assignable to metal or metal oxide, suggesting that the Mn
entities are either well dispersed as amorphous species or
ꢁ
autoclave with a Teon liner at 180 C for 2 days. The obtained
gel was ltered out and washed exhaustively with distilled
ꢁ
water, dried overnight at 120 C and then calcined in air at
ꢁ
5
50 C for 5 h. The Mn–ZSM-5 zeolite was synthesized by adding
18
manganese(III)–acetylacetonate (Sigma Aldrich) to solution A.
The prepared catalysts are denoted hereinaer as M–Z (x),
where x means the Si/Mn molar ratio.
Mn/ZSM-5 was prepared by impregnation. An appropriate
amount of manganese(III)–acetylacetonate precursor was dis-
solved in deionized water and mixed with 2 g ZSM-5 at 80 C
aggregated into minicrystals that are too small (<4 nm) to be
24,25
detected by XRD.
Comparing the XRD patterns with that of
ZSM-5 zeolite, one can see that there are diffraction peaks of
M–Z (x) that are slightly shied to a smaller angle. This is likely
ꢁ
under magnetic stirring for 4 h. The as-synthesized sample was
ꢁ
ꢁ
dried overnight at 120 C and calcined in air at 550 C for 5 h.
The nal product is denoted hereinaer as M/Z (y), where y
stands for the weight percentage of Mn in terms of the whole
catalyst.
2.3. Catalyst characterization
The powder X-ray diffraction (XRD) experiment was conducted
on a Bruker D8 Advance diffractometer using Cu Ka radiation.
ꢁ
ꢀ1
The data were recorded at a scan rate of 0.02 (2) s in the 10–
ꢁ
80
range. UV-vis diffuse reectance spectra (UV-vis DRS) of
samples were obtained over a UV-vis spectrophotometer (Cary
00) using BaSO as reference. Scanning electron microscopy
1
4
Fig. 1 XRD patterns of ZSM-5, M–Z (x), M/Z (2.0 wt%) and regenerated
(
SEM) images were taken on a Hitachi S-4800 emission M–Z (48)-1 after the first cycle of reusability test.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 74162–74169 | 74163

View Article Online
RSC Advances
Paper
to be due to isomorphous substitution of Si by Mn(II or III) that
ꢀ
are larger in size (0.5–1.0 A). Furthermore, there is no obvious
difference between the patterns of fresh and regenerated cata-
lysts aer the rst cycle of reusability test. It is hence deduced
that there is no long-range change in catalyst structure during
the reaction.
We examined the ZSM-5 and Mn-modied ZSM-5 samples by
SEM (Fig. 2). They are all in the form of intergrown hexagonal
platelets with particle size of 200–300 nm. The surface of M–Z
(48) is smooth whereas that of M/Z (2.0 wt%) is decorated with
small particles. To determine the surface atomic concentration
and to characterize the oxidation states of surface Mn, XPS tests
were carried out. The actual percentages of Mn were obtained by
ICP analysis (Fig. 3). The atomic weight percentage of Mn in the
M–Z (48) and M/Z (2.0 wt%) samples is 1.7 and 1.5 wt%
respectively. The results are slightly less than the respective
nominal values. Such discrepancy could be owing to the loss of Fig. 3 XPS analysis of M–Z (48) and M/Z (2.0 wt%).
Mn species during the centrifugation and/or stirring processes.
As shown in Table 1, M/Z (2.0 wt%) and M–Z (48) have similar
Mn loading, while the surface Si/Mn of M–Z (48) is only half that Table 1 Element composition of M–Z (48) and M/Z (2.0 wt%)
of the M/Z (2.0 wt%), indicating Mn accumulates on the surface
of M/Z (2.0 wt%). As seen from Fig. 3, M–Z (48) and M/Z
Sample
XPS Si/Mn
Mn concentration by ICP (wt%)
(
2.0 wt%) both show a variety of Mn oxidation states that is
M–Z (48)
M/Z (2.0 wt%)
54.9
21.5
1.7
1.5
20,26
essential for electron transfer in a material.
Preliminary studies on the distribution of manganese
species on the catalysts were conducted by means of UV-vis DRS
(Fig. 4). It is generally accepted that manganese species in ZSM-
5
display a number of UV-vis bands: (i) 220–380 nm ascribable
2
ꢀ
3+
2ꢀ
2+
to O / Mn charge-transfer transition or O / Mn
charge-transfer transition, (ii) 380–600 nm attributable to iso-
lated or extra-framework Mn species, and (iii) above 600 nm
due to manganese oxide aggregates on the external crystal
3
+
Fig. 4 UV-vis patterns of ZSM-5, M–Z (x) and M/Z (2.0 wt%).
14,18,27–29
surface.
has no absorption in the 200–800 nm range. In the case of M–Z
x), there is absorbance at 265 nm together with a small feature
at 500 nm. The latter indicates the presence of Mn(III) sites of an
Without the modication of manganese, ZSM-5
(
14
elongated tetragonal or distorted octahedron system. It is clear
that with Si/Mn increases from 72 to 48, absorption at 500 nm
increases proportionally. The M/Z (2.0 wt%) sample only shows
an absorption band around 250 nm with a tail extending to 800
nm, plausibly due to the existence of Mn and Mn oxide clusters
on the external surface of zeolite.
Fig.
5 illustrates the nitrogen adsorption–desorption
Fig. 2 SEM images of ZSM-5 (a and b), M–Z (48) (c and d) and M/Z (2.0
wt%) (e and f).
isotherms and micropore (SF method) size distribution of the
74164 | RSC Adv., 2015, 5, 74162–74169
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Fig. 5
2
N adsorption–desorption isotherm and pore diameter distri-
bution of ZSM-5, M–Z (48) and M/Z (2.0 wt%).
3
Fig. 6 NH -TPD profiles of as-prepared ZSM-5, M–Z (72), M–Z (60),
M–Z (48), M–Z (36), M/Z (2.0 wt%).
Table 2 Textural properties of Mn-containing ZSM-5 materials
BET surface
area (m g
Microporous
volume (cm g
Total pore
volume (cm g )
2
ꢀ1
3
ꢀ1
3
ꢀ1
Sample
)
)
ZSM-5
373.0
375.4
373.8
396.2
393.9
0.14
0.14
0.14
0.14
0.12
0.12
0.3943
0.3360
0.3384
0.3349
0.2953
0.2969
M–Z (72)
M–Z (60)
M–Z (48)
M–Z (36)
M/Z (2.0 wt%) 334.5
catalysts. The textural parameters are summarized in Table 2.
All samples show a type-I isotherm typical of microporous
materials, and the isotherms exhibit a hysteresis loop at P/P
0
of
30
0.1–1.0, indicating the presence of slit-shaped pores. The pore
3
Fig. 7 Acidity distribution of catalysts measured by NH -TPD.
size distribution curves suggest structures of uniform
molecular-scale pores. Meanwhile, the total pore volume of M–Z
(x) is smaller than that of ZSM-5, indicating that there is pore
ꢁ
around 140 C increases with the decline of Si/Mn ratio. The
weak acidic sites on M–Z (x) or M/Z is related to the amount of
external or extra-framework manganese ions that engender sites
blockage and narrowing as a result of Mn deposition at the pore
31
entrance and/or on the channel walls. However, in the cases of
M–Z (x ¼ 72, 60, 48), the total pore volume and micro-pore
volume are rather similar. It is deduced that using the hydro-
thermal synthesis method, there is homogeneous substitution
of Mn to Si at Si/Mn molar ratio of 72, 60 and 48. With Mn
dispersed in the framework throughout the materials, there is
little change in pore volume. Nonetheless, the pores of M–Z (36)
and M/Z (2.0 wt%) are severely blocked and narrowed as a result
of excess Mn species that are scattered on the catalyst surface
and inside the channels.
of weak acidity in the form of (Mn)
Al OH groups but much weaker in strength. The amount of
strong acidic sites decreases in the sequence of M–Z (48) > M–Z
60) > M–Z (72) > M–Z (36). Sample M–Z (48) has the largest
n
OH, similar to those of
32
n
(
amount of weak acidic sites mainly because of the abundance of
Mn ions in the framework, giving sites such as Mn–OH–Si and
33
Mn/(OH)–Si with Br ¨o nsted acidity. Overall, the total amounts
of accessible acid sites in all the samples are relatively low.
The acidic properties of M–Z (x) were studied to prove the
intrinsic acidity caused by the incorporation of Mn. The results 3.2. Catalytic performance
are shown in Fig. 6, and quantitative data of the amount of
ammonia desorbed from selected samples are reported in
Table 3 depicts the PCT conversion and PCB selectivity detected
over the catalysts. In a blank run (without a catalyst) or over
ZSM-5, there is no reactivity. However, the Mn-modied ZSM-5
catalysts perform well, conrming that catalytic activity is a
result of Mn presence in the ZSM-5 zeolite. There is an increase
of catalytic activity with increase of Mn loading up to Si/Mn ¼
Fig. 7. The M–Z (x) samples show two desorption peaks, one due
ꢁ
to weak acidic sites appears as a broad peak at 140 C with a
ꢁ
shoulder at 240 C, and the other due to strong acidic sites that
ꢁ
centre at about 500 C. The amount of ammonia desorption
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 74162–74169 | 74165

View Article Online
RSC Advances
Paper
a
b
Table 3 Results of p-chlorotoluene oxidation over different catalysts
external surface acidity is favourable for the partial oxidation
of PCT.
2
1
Entry
Catalyst
Conv. (%)
PCB
1
2
3
4
5
6
7
No catalyst
ZSM-5
0
0
0
0
3
.3. Effect of reaction conditions
3.3.1. Effect of catalyst amount. Fig. 8 describes the effect
M–Z (72)
M–Z (60)
M–Z (48)
M–Z (36)
M/Z (2.0 wt%)
49.7
71.5
93.8
91.8
80.3
86.9
87.9
90.5
82.0
76.9
of catalyst amount on PCT reaction over M–Z (48). When the
amount of catalyst is increased from 0 to 20 mg, there is a steep
increase in PCT conversion and a slight decrease in PCB selec-
tivity. Further increase of catalyst amount does not result in any
a
Reaction conditions: catalyst 20 mg, p-chlorotoluene 1 mL, solvent signicant changes. The results indicate that even with the
(
5
acetic acid) 10 mL, HBr (40 wt%) 30 mg, H
0 mL min , time 8 h, temperature 110 C. PCB: p-chlorobenzaldehyde.
2
O 3 g, oxygen ow rate
ꢀ1
ꢁ
b
enrichment of M–Z (48), only a small amount of the catalyst is
utilized. At the same time, the conversion of PCT is also limited
by the availability of dissolved oxygen. Based on the results,
the amount of catalyst for optimum production of PCB should
3
36
48. Further increase of Mn loading results in the decrease of be 20 mg.
PCT conversion and PCB selectivity, plausibly due to reduced
3.3.2. Effect of water addition. The effect of water addition
Mn incorporation and/or plugging of pores that are caused by on PCT conversion is displayed in Fig. 9. With the increase of
an excess amount of Mn. To investigate the effect of the water from 0 to 3 g, there is a signicant increase of PCT
distribution of active sites, the physical–chemical properties conversion and PCB selectivity. As shown in Scheme 1, there is
and catalytic activity of M/Z (2.0 wt%) and M–Z (48) of similar the rapid formation of p-chlorobenzylic bromide intermediate
Mn loading were studied.
and its decomposition for the progress of the reaction is
According to the BET results, the ZSM-5 zeolite shows high enhanced by water (step (1)). Furthermore, water could show
surface specic area and uniform micropores structure. The other positive effects such as acting as a deactivator for the
37
well dispersed Mn active sites on ZSM-5, as indicated in XRD production of aromatic acids. When water is in excess, the
analysis, should signicantly enhance the catalytic activity for concentration of active water decreases due to self-association,
21
PCT oxidation. As shown in Table 3, the PCT conversion over and there is decline in radical production and diminution of
19
M–Z (48) (93.8%) is signicantly higher than that over M/Z overall reaction rate. It is clear that a proper amount of water
2.0 wt%) (80.3%). With the insertion of Mn into the ZSM-5 can promote the oxidation of PCT whereas in excess the
framework, there is weakening of Mn–O bonds and enhanced oxidation reaction is inhibited. Taking PCB yield into consid-
(
34
amount of lattice defects. Hence, besides being higher in eration, the most suitable initial amount of water for the
2
ꢀ1
2
ꢀ1
specic surface area (396.2 m g compared to 334.5 m g ), present reaction system should be 3 g.
M–Z (48) is also a better oxygen carrier and donor.
3.3.3. Effect of HBr amount. The amount of HBr is a crucial
With the increase of Mn active sites, there is signicant factor that affects the oxidation reaction (Fig. 10). In the absence
improvement of PCT conversion and PCB selectivity. It is of HBr, the auto-oxidation of PCT is not signicant. However,
deduced that the scattering of Mn active sites in the pores of when the amount of HBr (40 wt%) is increased from 10 to
the nanoscale zeolite framework, as in the case of M–Z (48), is 30 mg, there is a remarkable increase of PCT conversion from
benecial for the enhancement of PCB selectivity. With the 33.7% to 93.8%. The attack of bromine radicals toward methyl
2
2,35
mechanism suggested by Partenheimer,
bromide is hydrogen is a key step for reaction promotion. This facilitates
directly bonded to manganese for reaction initiation. Then the
oxidation of PCT in the catalyst channels leads to direct
formation of benzylic bromide. Therefore, compared to those
on the external surface, the active sites in the channels of a
catalyst are more benecial to PCB selectivity. However, in
comparison to M–Z (48) (conv.% ¼ 93.8, sel.% ¼ 90.5), there is
a decrease of PCB selectivity and yield over M–Z (36) (conv.% ¼
9
1.8, sel.% ¼ 82.0). Taking the NH -TPD results into account, a
3
higher content of Mn means a higher amount of both strong
and weak acid sites. However, a further increase of Mn content
gives rise to a decline of strong acid sites but an increase of
total acid sites. In other words, excess Mn species that are
scattered on the catalyst surface results in rise of the amount
of weak acid sites that enhance further oxidation of PCB to
carboxylic acid. In general, the high-silica MFI ZSM-5 samples
are extremely low in acidity and the amounts of total acidic
ꢀ1
Fig. 8 Effect of catalyst amount on PCT oxidation (reaction condi-
tions: substrate 1 mL, solvent 10 mL, HBr (40 wt%) 30 mg, H O 3 g,
oxygen flow rate 50 mL min , time 8 h, temperature 100 C).
sites on M–Z (x) are approximately 0.1 mmol g . The low
2
ꢀ1
ꢁ
74166 | RSC Adv., 2015, 5, 74162–74169
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Fig. 9 Effect of water addition on PCT oxidation (reaction conditions: Fig. 11 Effect of temperature on PCT oxidation (reaction conditions:
catalyst M–Z (48) 20 mg, substrate 1 mL, HBr (40 wt%) 30 mg, solvent catalyst M–Z (48) 20 mg, substrate 1 mL, HBr (40 wt%) 30 mg, H O 3 g,
2
ꢀ1
ꢁ
ꢀ1
1
0 mL, oxygen flow rate 50 mL min , time 8 h, temperature 100 C). solvent 10 mL, oxygen flow rate 50 mL min , time 8 h).
Fig. 10 Effect of HBr amount on PCT oxidation (reaction conditions: Fig. 12 Effect of reaction time on PCT oxidation (reaction conditions:
catalyst M–Z (48) 20 mg, substrate 1 mL, solvent 10 mL, H
O 3 g, catalyst M–Z (48), substrate 1 mL, solvent 10 mL, HBr (40 wt%) 30 mg,
2
ꢀ1
ꢁ
ꢀ1
ꢁ
2
H O 3 g, oxygen flow rate 50 mL min , temperature 100 C).
oxygen flow rate 50 mL min , time 8 h, temperature 100 C).
are depicted in Fig. 12. There is a signicant increase in PCT
conversion between 4 and 6 h. It is deduced that erce inter-
action of Mn active sites with bromide results in enhancement
of free radicals, and the reaction is greatly stimulated. Under
the adopted condition, PCT conversion increases to 96.4% in
2
electron transfer and radical formation. However, an excessive
amount of bromide results in an obvious decrease of conversion
38
and selectivity. It was pointed out by Hu that bromide in
excess results in the formation of inactive organic bromide as
well as metal–multibromides that are active for further oxida-
tion of aldehydes.
.3.4. Effect of reaction temperature. The inuence of
reaction temperature on the conversion of PCT was investigated
10 h, while there is a constant decrease in PCB selectivity. Based
39
on the results, we consider that 8 h is the reaction time for
optimal performance.
3
(
Fig. 11). It was found that the reaction is highly sensitive to the
ꢁ
change of temperature. Below 70 C, the conversion of substrate 3.4. Recycling of the catalyst
is insignicant. At a higher temperature, there is a large
increase of PCT conversion together with a slight decline in PCB
selectivity, and PCB yield (85.0%) reaches maximum at 100 C.
The slight decrease of selectivity is attributed to the further
oxidation of PCB. Besides, considering the boiling point of
substrate in the reaction system, the reaction temperature
should be controlled at 100 C.
Recyclability of the catalysts for liquid-phase oxidation of PCT
was tested. Aer each cycle, the catalyst was ltered out and
ꢁ
washed several times with acetone or anhydrous alcohol before
ꢁ
being dried at 120 C overnight. Listed in Table 4 are the results.
3
M–Z (48) shows an immediate decrease of activity aer the rst
cycle. However, aer successive recycling (cycles 2 to 4) of the
catalyst, there is a minor decrease in catalytic activity. Although
there was slight fading of colour aer each cycle, the XRD
pattern of a recovered catalyst (Fig. 1) is similar to that of the
ꢁ
3.3.5. Effect of reaction time. The effect of reaction time on
PCT reaction over Mn–ZSM-5 was investigated and the results
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 74162–74169 | 74167

View Article Online
RSC Advances
Table The recycling of M–Z (48) in the oxidation of p-
Paper
4
5 N. N. Tu ˇs ar, S. Jank and R. Gl ¨a ser, ChemCatChem, 2011, 3,
a
chlorotoluene
254–269.
6
7
J. S. Yoo, Appl. Catal., A, 1996, 135, 261–271.
C. Sprung and B. M. Weckhuysen, Chem.–Eur. J., 2014, 20,
3667–3677.
Recycle number
1 (fresh)
2
3
4
Conv. (%)
Sel. (%)
93.8
90.5
82.1
96.4
79.5
95.8
80.2
95.1
8 H. X. Tao, H. Yang, X. H. Liu, J. W. Ren, Y. Q. Wang and
G. Z. Lu, Chem. Eng. J., 2013, 225, 686–694.
9 H. L. Janardhan, G. V. Shanbhag and A. B. Halgeri, Appl.
Catal., A, 2014, 471, 12–18.
a
Reaction conditions: catalyst 20 mg, p-chlorotoluene 1 mL, solvent
O 3 g, oxygen ow rate
0 mL min , time 8 h, temperature 110 C.
(acetic acid) 10 mL, HBr (40 wt%) 30 mg, H
2
ꢀ
1
ꢁ
5
10 D. B. Shah, D. T. Hayhurst, G. Evanina and C. J. Guo, AIChE
J., 1988, 34, 1713–1717.
fresh one, indicating that there is no change in catalyst struc- 11 W. Tan, M. Liu, Y. Zhao, K. K. Hou, H. Y. Wu, A. F. Zhang,
ture. In other words, the deactivation of catalyst is likely to be
due to the leaching of a small amount of manganese from the
H. O. Liu, Y. R. Wang, C. S. Song and X. W. Guo,
Microporous Mesoporous Mater., 2014, 196, 18–30.
catalyst during reaction. It is observed that the leaching of 12 Y. T. Cheng, Z. Wang, C. J. Gilbert, W. Fan and G. W. Huber,
manganese decreases in further cycles. Therefore, catalytic
Angew. Chem., Int. Ed., 2012, 51, 11097–11100.
activity probably originates from the manganese ions in the 13 Q. Ouyang, S. F. Yin, L. Chen, X. P. Zhou and C. T. Au, AIChE
framework of M–Z (x). The results indicate that the catalyst can
J., 2013, 59, 532–540.
be recycled a number of times without showing signicant loss 14 D. Radu, P. Glatzel, A. Gloter, O. Stephan, B. M. Weckhuysen
of catalytic activity.
and F. M. F. de Groot, J. Phys. Chem. C, 2008, 112, 12409–
2416.
5 Y. Chen, G. Li, F. Yang and S. M. Zhang, Polym. Degrad. Stab.,
011, 96, 863–869.
Efficient synthesis of p-chlorobenzaldehyde by liquid-phase 16 G. Lv, F. Bin, C. L. Song, K. P. Wang and J. O. Song, Fuel,
oxidation of p-chlorotoluene with molecular oxygen was ach-
2013, 107, 217–224.
ieved over Mn–ZSM-5 (Si/Mn ¼ 48, Mn 1.7 wt%) under opti- 17 X. R. Lou, P. F. Liu, J. Li, Z. Li and K. He, Appl. Surf. Sci., 2014,
mized conditions (catalyst 20 mg, p-chlorotoluene 1 mL, solvent
307, 382–387.
1
1
4
. Conclusions
2
(
acetic acid) 10 mL, HBr (40 wt%) 30 mg, H
2
O 3 g, oxygen ow 18 Y. T. Meng, H. C. Genuino, C. H. Kuo, H. Huang, S. Y. Chen,
ꢀ
1
ꢁ
rate 50 mL min , time 8 h, temperature 100 C). With 93.8%
p-chlorotoluene conversion and 90.5% p-chlorobenzaldehyde
L. C. Zhang, A. Rossi and S. L. Suib, J. Am. Chem. Soc., 2013,
135, 8594–8605.
selectivity, the highest product yield is 85.4%. It is shown that 19 Y. Q. Deng, T. Zhang, C. T. Au and S. F. Yin, Appl. Catal., A,
the Mn active sites resulted from manganese incorporation into
2013, 467, 117–123.
the ZSM-5 framework play an effective role in p-chlorotoluene 20 Y. Q. Deng, T. Zhang, C. T. Au and S. F. Yin, Catal. Commun.,
oxidation. The superior performance of the Mn–ZSM-5 catalyst
2014, 43, 126–130.
is attributed to its mild acidity and good distribution of 21 T. Zhang, Y. Q. Deng, W. F. Zhou, C. T. Au and S. F. Yin,
manganese species.
Chem. Eng. J., 2014, 240, 509–515.
2
2
2 W. Partenheimer, Adv. Synth. Catal., 2004, 346, 297–306.
3 E. M. Flanigen, J. M. Bennett, R. W. Grose, J. P. Cohen,
R. L. Patton and R. M. Kirchner, Nature, 1978, 271, 512–516.
Acknowledgements
The project was nancially supported by the National Natural 24 H. X. Yuan, Q. H. Xia, H. J. Zhan, X. H. Lu and K. X. Su, Appl.
Science Foundation of China (Grant No. 21476065, 21273067),
Catal., A, 2006, 304, 178–184.
the program for New Century Excellent Talents in Universities 25 H. H. Chen, H. P. Zhang and Y. Yan, Chem. Eng. J., 2014, 254,
NCET-10-0371), Program for Changjiang Scholars and Innova-
133–142.
tive Research Team in University (IRT1238), and the Funda- 26 L. Wang, H. P. Zhang, Y. Yan and X. Y. Zhang, RSC Adv.,
(
mental Research Funds for the Central Universities. C. T. Au
thanks the Hunan University for an adjunct professorship.
2015, 5, 29482–29490.
27 F. Milella, J. M. Gallardo-Amores, M. Baldic and G. Busca,
J. Mater. Chem., 1998, 8, 2525–2531.
2
8 S. Velu, N. Shah, T. M. Jyothi and S. Sivasanker, Microporous
Mesoporous Mater., 1999, 33, 61–75.
References
1
2
3
4
R. A. Sheldon and H. V. Bekkum, Fine Chemicals through 29 B. Qi, X. H. Lu, D. Zhou, Q. H. Xia, Z. R. Tang, S. Y. Fang,
Heterogeneous Catalysis, Wiley-VCH, Weinheim, 2001.
A. J. Hu, C. X. L u¨ , H. Y. Wang and B. D. Li, Catal. Commun.,
T. Pang and Y. L. Dong, J. Mol. Catal. A: Chem., 2010, 322,
73–79.
2
007, 8, 1279–1283.
30 B. J. Dou, G. Lv, C. Wang, Q. L. Hao and K. S. Hui, Chem. Eng.
J., 2015, 270, 549–556.
31 G. Q. Zhou, J. Li, Y. Q. Yu, X. G. Li, Y. J. Wang, W. Wang and
S. Komarneni, Appl. Catal., A, 2014, 487, 45–53.
J. Q. Wang, H. Fang, Y. Li, J. J. Li and Z. Y. Yan, J. Mol. Catal.
A: Chem., 2006, 250, 75–79.
F. M. Bautista, D. Luna, J. Luque, J. M. Marinas and
J. F. S ´a nchez-Royo, Appl. Catal., A, 2009, 352, 251–258.
74168 | RSC Adv., 2015, 5, 74162–74169
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
3
3
3
3
2 J. C. Xia, D. S. Mao, B. Zhang, Q. L. Chen and Y. Tang, Catal. 36 A. R. Li, S. W. Tang, P. H. Tan, C. J. Liu and B. Liang, J. Chem.
Lett., 2004, 98, 235–240. Eng. Data, 2007, 52, 2339–2344.
3 G. Centi, S. Perathoner and F. Trirb, J. Phys. Chem., 1992, 37 W. Partenheimer, Catal. Today, 1995, 23, 69–158.
9
6, 2617–2629.
4 R. H. Wang and J. H. Li, Environ. Sci. Technol., 2010, 44,
282–4287.
5 W. Partenheimer, J. Mol. Catal. A: Chem., 2001, 174, 29–33.
38 A. J. Hu, C. X. L u¨ , B. D. Li and T. Huo, Ind. Eng. Chem. Res.,
2006, 45, 5688–5692.
4
39 W. Partenheimer, J. Mol. Catal. A: Chem., 2003, 206, 105–119.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 74162–74169 | 74169