Preparation and reactivity of UV light-reduced Pd/α-Fe2O3 catalyst towards the hydrogenation of o-chloronitrobenzene

Article abstract of DOI:10.1016/j.apcata.2016.04.007

A type of Pd/α-Fe₂O₃ catalyst was synthesized by a convenient UV light-induced reduction in the presence of Fe³⁺ ions. The synthesized Pd/α-Fe₂O₃ particles were characterized by X-ray diffraction (XRD), low-resolution transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HR-TEM), X-ray photoelectron spectrum (XPS), atomic force microscope (AFM), scan electron microscopy with energy dispersion spectra (SEM-EDS), FT-IR and BET. Related characterizations confirm that UV light-induced reduction is a feasible, simply and environment-friendly method for the preparation of supported catalysts containing active metal. Furthermore, hydrogenation of o-chloronitrobenzene (o-CNB) to o-chloroaniline (o-CAN) by the selected Pd(0.5%)/α-Fe₂O₃ was investigated systematically. The α-Fe₂O₃-supported Pd catalyst was found to exhibit good catalytic activity and selectivity for the o-CNB reduction under mild conditions. The hydrogenation of o-CNB over the resulting Pd-based catalyst gave a 100% o-CNB conversion and 91.4% selectivity of o-CAN under the reaction conditions of 50 °C, 2 h, 1 MPa hydrogen pressure, n_o-CNB/n_Pd = 6000:1. Moreover, the resulting Pd-based catalyst remained higher hydrogenation reactivity after three runs even if a sharply decrease in activity was observed at the fourth run. The obtained results reveal the as-prepared Pd/α-Fe₂O₃ is an efficient catalyst for the hydrogenation of o-CNB even at a high level of substrate concentration.

Full text of DOI:10.1016/j.apcata.2016.04.007

Accepted Manuscript
Title: Preparation and Reactivity of UV Light-Reduced
Pd/␣-Fe O Catalyst Towards the Hydrogenation of
2
3
o-Chloronitrobenzene
Author: Weidong Jiang Bin Xu Zhen Xiang Xiaoqiang Liu
Fuan Liu
PII:
S0926-860X(16)30177-6
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2016.04.007
APCATA 15835
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
5-2-2016
9-4-2016
11-4-2016
Please cite this article as: Weidong Jiang, Bin Xu, Zhen Xiang, Xiaoqiang Liu, Fuan
Liu, Preparation and Reactivity of UV Light-Reduced Pd/rmalpha-Fe2O3 Catalyst
Towards the Hydrogenation of o-Chloronitrobenzene, Applied Catalysis A, General
http://dx.doi.org/10.1016/j.apcata.2016.04.007
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Preparation and Reactivity of UV Light-Reduced Pd/α-Fe
2
O
3
Catalyst Towards the Hydrogenation of o-Chloronitrobenzene
Weidong Jiang a, b*, Bin Xu a, b, Zhen Xiang a, b, Xiaoqiang Liu a, b, Fuan Liu b
(1
School of Chemical and Pharmaceutical Engineering, Sichuan University of Science & Engineering, Zigong
6
43000, China;
2
Key Laboratory of Green Catalysis of Sichuan Institute of High Education, Zigong 643000, China)
1

Graphical abstract
The present work reported the convenient preparation of Pd-based catalyst over α-Fe
2
3
O as
support via a UV light-induced reduction process. The resulting Pd/α-Fe
2
3
O catalyst exhibited
good performance for the hydrogenation reduction of o-chloronitrobenzene (o-CNB), yielding
completely o-CNB conversion and 91.4% selectivity of o-CAN at high substrate/catalyst ratio
(
6000:1) under mild conditions.
2 3
Pd/α-Fe O
1
1
20
00
8
6
4
2
0
0
0
0
0
o-CNB Conversion%
AN Selectivity%
o-CAN Selectivity%
0
2000
4000
6000
8000
10000
Molar Ratios of o-CNB to Pd/α-Fe
2 3
O
2

HIGHLIGHTS


Pd/α-Fe
Pd/α-Fe
2
2
O
3
3
catalyst was prepared by a convenient UV light-induced reduction.
O
maintained good reactivity towards o-CNB hydrogenation at high o-CNB/Pd
ratios up to 7000.
Pd/α-Fe -promoted o-CNB hydrogenation gave 100% o-CNB conversion and 91.4%
o-CAN selectivity under mild conditions.

2
O
3
Abstract: A type of Pd/α-Fe
presence of Fe3+ ions. The synthesized Pd/α-Fe
low-resolution transmission electron microscopy (TEM), high-resolution transmission electron microscopy
HR-TEM), X-ray photoelectron spectrum (XPS), atomic force microscope (AFM), scan electron microscopy
2
O
3
catalyst was synthesized by a convenient UV light-induced reduction in the
2
O
3
particles were characterized by X-ray diffraction (XRD),
(
with energy dispersion spectra (SEM-EDS), FT-IR and BET. Related characterizations confirm that UV
light-induced reduction is a feasible, simply and environment-friendly method for the preparation of supported
catalysts containing active metal. Furthermore, hydrogenation of o-chloronitrobenzene (o-CNB) to
2 3
o-chloroaniline (o-CAN) by the selected Pd(0.5%)/α-Fe O was investigated systematically. The
α-Fe -supported Pd catalyst was found to exhibit good catalytic activity and selectivity for the o-CNB
2
O
3
reduction under mild conditions. The hydrogenation of o-CNB over the resulting Pd-based catalyst gave a 100%
0
o-CNB conversion and 91.4% selectivity of o-CAN under the reaction conditions of 50 C, 2 hrs, 1MPa
hydrogen pressure, no-CNB/nPd=6000:1. Moreover, the resulting Pd-based catalyst remained higher hydrogenation
reactivity after three runs even if a sharply decrease in activity was observed at the fourth run. The obtained
2 3
results reveal the as-prepared Pd/α-Fe O is an efficient catalyst for the hydrogenation of o-CNB even at a high
level of substrate concentration.
2 3
Keywords: catalytic hydrogenation, o-chloronitrobenzene, UV light-induced reduction, Pd/α-Fe O catalyst
3

1
. Introduction
The hydrogenation of nitro compounds is commonly conducted to prepare aromatic amines that
are important intermediates for dyes, drugs, and agrochemicals [1, 2]. Thereupon, many research
[
3]
groups have engaged in the synthetic methods and process of aromatic amines . Especially, the
hydrogenation of corresponding nitroaromatics proved to be an easier and convenient route for the
production of aromatic amines [4-13]. Hence, many chemists and researchers devoted to the
development and improvement of this technology.
For one catalytic reaction two main aspects, including developing efficient catalysts and
optimizing reaction conditions, are highly attracted scientist’s attention. Noble metal catalysts are
well known for their high catalytic activities in hydrogenation reactions [5, 7, 14-19]. Among these
catalysts, palladium-based catalysts were widely used in the reduction of various compounds due
to its high activity and selectivity [20-26]. To improve the stability and recyclability of metal
catalysts, moreover, almost all active metal components applied in previous works were loaded on
the varied supports, such as activated carbon [27], metallic or non-metallic oxidates [15, 28-30]
,
polymer [31, 32], grapheme [33], etc. At present, chemists focused on the design and preparation of
different supports with dazzling morphologies.
In the catalytic hydrogenation reactions, it was widely accepted that zero-valent metal atoms on
the carriers are the real catalysts. This is the reason for a reduction process of cationic metals in the
preparation of immobilized metal catalysts. Nowadays, chemical reductants [34, 35] were applied in
the reduction of metal cations to zero-valent ones. Compressed hydrogen as classic reducing agent
is used to reduce metal ions loaded on supports [36], and whereas it is potential flammable and
explosive owing to its intrinsic characters. Meanwhile, in the general preparation of supported
metal catalysts, chemical reductants are widely applied in the reduction of metal ions. Howbeit,
these reductants have their limitations concerning expensive price, environmental hazard and
operated dangerous.
Interestingly, Enrico Borgarello et. al reported a facile UV light-induced reduction method [37]
for the reduction of metal ions. Nevertheless, this strategy was seldom used in the preparation of
various supported catalysts for the hydrogenation reaction so far. In the present study, we prepared
a kind of immobilized Pd catalyst over stable α-Fe
2
3
O using UV light-induced reduction method
as a facile and environment-friendly one in comparison with chemical reduction. Furthermore, we
evaluated the hydrogenation reactivity of the resulting Pd-base catalyst, systematically. Impact of
varied reaction conditions on the hydrogenation was screened for the aim of achieving much
4

higher o-CAN yield.
2
2
. Experimental
.1 Material and Methods
All reagents were of analytical grade and used as received. PdCl
2
·2H O was a product of
2
Kunming Institute of Precious Metals (China). Fe(NO
3
)
3
·9H O was purchased from Tianjin
2
Fuchen Chemical Reagents Factory (China). Standard samples of o-CNB, o-CAN and aniline were
purchased from Sigma-Aldrich Co.
Preparation of α-Fe
Fe(NO ·9H O solution (0.1mol/L) was prepared in deionized water. Ammonia solution was
slowly dropped into the Fe(NO ·9H O solution using constant-flow pump with stirring, during
2
3
O . The α-Fe
2
3
O carriers were obtained according to following process.
3
)
3
2
3
)
3
2
which the pH value of the mixture was adjusted to 8 ~9. Afterwards, the mixed solution was
heated to remove the residual ammonia. The mixture was cooled down to room temperature, and
filtrated under vacuum along with a washing process with anydrous ethanol. The yielded solidwas
0
dried under vacuum at 80 C for 4 hrs. Furthermore, the dried solid was calcinated in a muffle
0
furnace at 250 C for 2 hrs. Ultimately, the desired brown α-Fe
2
3
O powder sample was obtained via
grinding operation.
Preparation of Pd/α-Fe
O
2 3
using facial UV-light reduction route. α-Fe
2
O
3
powder was added to
the prepared solution of PdCl
2
·2H O (theroretic Pd loading content 0.5%). The mixture was stirred
2
thoroughly under ultraviolet radiation [37] (30W) with a constant wavelength of 365nm for 24 hrs.
The mixture was filtered and washed with deionized water. The solids obtained were dried in
0
vaccum at 80 C for 4 hrs. And then, grinding the dried solids produced the desired
α-Fe
2
3
O -supported Pd catalyst.
2
.2 Structural and Morphologic Characterizations
Low resolution TEM (JEOL JEM 1200EX working at 100 kV) and high-resolution TEM (FEI
Tecnai G2 F20 S-Twin working at 200 kV) were utilized to characterize the morphology of the
Pd/α-Fe catalyst. BET areas were recorded on a SSA-4200 Specific Surface Area and Porosity
Analyzer (Beijing Builder Electronic Technology Co., Ltd.) with a 30% v/v N /H flow using pure
(99.9%) as an internal standard. The crystalline phase of the as-prepared Pd catalyst was
2
O
3
2
2
N
2
recorded on a Bruker D2-PHASER X-ray diffractometer (Germany Bruker AXE) with
nickel-filtered Cu Kα radiation (λ= 1.5417 Å, 40kV, 25mA) in the 2θ zone of 10-90º, and with a
scan speed of 0.06º/sec. Atom Force Microscope (AFM) images were obtained on an E-SWEEP
5

AFM (Hitachi High-Technologies Co., Japan). Scanning Electron Microscope with an EDS
detector (Germany Bruker AXE) was measured on a VEGA 3SBU instrument from Tescan Co.,
Czech Republic. The loading content of Pd metal on the resulting Pd-base catalyst was determined
on Spectro Arcos ICP-OES (Germany Spectro Analysis Co., Ltd).
2.3 Catalytic Tests
The catalytic hydrogenation of o-CNB was preformed in a 100ml GYF-60-2 stainless-steel
autoclave (China Weihai Autocontrolled Autoclave Ltd.) with a thermocouple and a mechanical
stirrer. Reaction reagents and catalyst were added into the stainless-steel reactor before each run.
The hydrogenation reaction of o-CNB started after the reaction system was heated to the desired
temperature.
2
.4 Analytical Methods for the Products
The reaction mixtures were filtrated before the product analysis. The analysis of the obtained
organic phase was performed using a SC-3000B-204 gas chromatography (Chongqing Chuanyi
Analyzer Co., LTD, China) equipped with a flame ionization detector (FID) and an SE-54
0
capillary column (30m×Ф0.53mm). The oven temperature was tuned to 150 C, while injector
0
0
temperature and detector temperature were respectively 260 C and 220 C. The column
0
0
temperature was progressively increased from 100 C (3min) to 220 C (6min) with a heating rate
0
of 35 C/min. Data processing was performed with an HW-2000 chromatography
workstation (Nanjing Qianpu Software, China). The major reactant and products were identified
by comparison with corresponding authentic samples.
3
3
Results and Discussions
.1 Varied Characterizations of the As-prepared Pd/α-Fe
2
O Catalyst
3
To elucidate the structure-activity relation of the synthesized Pd catalyst, it is highly important
that the structure characterizations by some classic physical methods, such as XRD, SEM, element
mapping, TEM, XPS, AFM, etc. As shown in Fig. 1, it was found that the crystalline phase of
Pd/α-Fe
peak of nano α-Fe
1.89°, and 75.34° are assigned to the scattering from (012), (104), (110), (113), (024), (116),
2
O
3
catalyst is almost the same as that of the used support, α-Fe
2
O3. The X-ray diffraction
2
O
3
at 2θ=24.16°, 33.11°, 35.62°, 40.86°, 49.46°, 54.11°, 57.60°,62.34°, 64.02°,
7
(
018), (214), (300), (101), and (220) planes and well match with the JCPDS card no.89-596. No
6

crystal phase of Pd was detected by XRD analysis, which is mainly due to two reasons. The one is
that the particle size of the Pd(0) species were too small to be detected as well the metallic Pd(0)
exhibited amorphous-like or less ordered structural features. Another one is that high dispersion of
metallic Pd (0) on the surface of supports leads to no characteristic peaks of metal species that can
be used for the observation in the XRD spectra. Therefore, average crystallite sizes of metallic Pd
species cannot be estimated on the basis of the XRD patterns using the Scherrer equation.
The SEM images of α-Fe
Pd/α-Fe powders display spherical morphography (Fig. 2B) which is highly accordance with
those of α-Fe powders (Fig. 2A). From the SEM images, the powder sizes distribute in a range
of 0.5 to 5.0 μm. And, the size distribution of most of Pd/α-Fe powders are in the range from
.0 to 1.5 μm. Moreover, TEM images (Figs. 2C and 2D) denote that those Pd/α-Fe powders
2
O
3
powder and Pd/α-Fe
2
O
3
catalyst were shown in Fig. 2. The
2
O
3
2
O
3
2
O
3
1
2
O
3
are aggregates of nano α-Fe
2
O
3
crystallites on which catalytic Pd particles are loaded. One can
crystal particle possesses rice-like morphography with almost
observe that each one α-Fe
2
O
3
uniform size of 10nm×8nm×8nm. Fig. 2D also shows the Pd particals were successfully loaded on
the surface of α-Fe . The crystallite size of Pd partical is found to be about 6nm×4nm×4nm,
which can be also confirmed by the high-resolution TEM (Fig. 2E) of Pd catalyst over α-Fe
2
O
3
2
3
O .
The interplanar spacing of the {200} plane of metallic Pd(0) shown in Fig. 2E is determined as
aproximate 0.19 nm, which is consistent with the reported figure [38]. Furthermore, the
selected area electron diffraction pattern (SAED) (Fig. 2F) of Pd/α-Fe
2
3
O catalyst displays a
concentric circling diffraction character, indicating the polycrystallic microstructure of the
resulting supported Pd catalyst. This fact is highly in agreement with the results of HR-TEM
shown in Figs. 2C and 2D.
The X-ray photoelectron spectrum (XPS) was obtained to analyze the valence state of metal
element and the surface chemical composition of the Pd/α-Fe
2
3
O catalyst. Fig. 3A displays the
total XPS spectra of Pd3d, Fe2p, O1s and C1s, respectively. The spectrum of C1s can be neglected
because of the adsorbed carbon dioxide in air. The XPS binding energies (BE) for Pd (0) 3d are at
3
35.0 eV (3d5/2) and 340.0 eV (3d3/2) as reported in literature [39, 40]. As shown in Fig. 2B, the BE
of Pd3d5/2 level in Pd/α-Fe is about 337.0 eV, which is 2.0 eV higher than that (= 335.0 eV) of
2
O
3
the standard zero-valent state of Pd. The peak at 342 eV is assigned to the Pd3d3/2, which also
gives an increased binding energy by approximate 2 eV. The increase in BE of Pd(0) arises from
the electron transformation from palladium to α-Fe
2
3
O due to the strong interaction between Pd
7

and α-Fe
the binding energy of Fe 2p3/2 is approximate 711eV for Fe3+ [41]. The weak satellite peak situated
at about 719 eV is a characteristic peak of Fe3+ in γ-Fe
, suggesting a partial formation of
γ-Fe during the preparation of Fe powders. The XPS spectra of O1s shown in Fig. 3D are
2
O
3
. Fig. 3C shows the X-ray photoelectron spectra signals of the Fe 2p regions, in which
2
O
3
2
O
3
2
O
3
wide and asymmetric, indicating that two kinds of oxygen species are present inside the
as-prepared catalyst. Namely, the binding energy of the crystal lattice oxygen (OLat2-) and the
oxygen-deficient regions (ODef2-) were 529.5 and 531.5 eV, respectively. OLat is attributable to the
2-
2
-
Fe–O bonding, whereas ODef is closely related to the catalytic efficiency of Pd/α-Fe
2
O catalyst.
3
Further, the area-selected SEM-EDS mapping was performed to determine the Pd, Fe, O
element distribution inside the Pd/α-Fe catalyst (Fig. 4). Fig. 4D shows the element mapping of
the carbon in carbon conductive adhesive for the sample preparation. Figures 4E and 4F present
the element mappings of α-Fe particles. Moreover, the element mapping of Pd in Fig. 4C
suggest a uniform distribution and efficient loading of Pd element on the surface of α-Fe
2
O
3
2
O
3
2
3
O .
In addition, the AFM section analysis can be utilized to obtain the size of the nanostructures. Fig.
shows the AFM height and topographic phase images of the Pd/α-Fe particles. The phase
5
2
O
3
image in Fig. 5C indicates that those sample phases are almost uniform and dispersed ball-like
particles exhibiting diameters which range from 25 nm to 42 nm. As for the HRTEM image (Fig.
2
D), the size of each α-Fe
images in Fig. 2B reveal that most of Pd(0.5%)/α-Fe
from 1.0 to 1.5 μm. As a consequence, it is reasonable that these ball-like particles shown in the
AFM phase image are indeed the aggregates from α-Fe crystals. The distinguishable
information on the particle sizes determined by the three characterizations, i.e., HRTEM, SEM and
2
O
3
crystal was found to be about 10nm×8nm×8nm. Moreover, the SEM
2
O
3
particles possess avarage sizes ranged
2
O
3
AFM, probably results from the re-aggregation of those dispersed α-Fe
ultrasonic processing of the original ethanol solution of the Pd(0.5%)/α-Fe
2
O
3
crystals during the
particles.
2
O
3
3
.2 Hydrogenation Performance of o-CNB over Pd/α-Fe
.2.1 Effect of the Pd Loadings on the Hydrogenation of o-CNB
In this work, the effect of the Pd loadings on the reduction of o-CNB under selected operation
2
O
3
Catalyst
3
conditions was evaluated. Fig. 6 shows the variations of the o-CAN selectivity and the conversion
of o-CNB upon the loading content of Pd. It was found that the hydrogenation of o-CNB yields a
8

4.22% conversion of o-CNB and 95.8% o-CAN selectivity in the absence of Pd. Further, the
conversion of o-CNB increased with the loading of Pd, obtaining 100% conversion when the Pd
loading content is above 0.5%. However, the o-CAN selectivity displayed a decreased trend
increasing Pd content companying with an increasing selectivity of AN. This fact probably results
from the high activity of Pd resulting in the concomitant dechlorination of
o-CNB, which produces more AN [42]. Hence, the subsequent o-CNB hydrogenation runs were
performed with the Pd/α-Fe
2
3
O catalyst at a 0.5% Pd loading level.
3
.2.2 Effect of the Ratios of o-CNB to Pd(0.5%)/α-Fe
2
O
3
As for a catalytic reaction, it is highly desired to obtain good conversion and selectivity with
high substrate/catalyst molar ratios. In the two decades, many groups have studied the
hydrogenation of o-CNB with various catalysts, showing o-CNB conversion and o-CAN
selectivity. However, previous catalytic hydrogenation runs of o-CNB were carried out at low
substrate/catalyst ratios [43, 44]. In our case, the effect of o-CNB/Pd molar ratios (1000 to 10000)
was evaluated for obtaining optimal ratio. The as-prepared Pd(0.5%)/α-Fe
2
3
O has been found to be
obviously effective in the o-CNB hydrogenation with high o-CNB/Pd molar ratios (up to 6000),
which is comparable to the hydrogenation of o-CNB promoted by other supported catalysts [42, 45,
4
6]
including commercial Pd/C [47, 48]
.
Fig. 7 indicates that in the ratio range of o-CNB/Pd molar ratios (1000 to 6000) the o-CNB
hydrogenation achieved 100% conversion and approximate 90% o-CAN selectivity. With the
further increase in molar ratio, a type of inhibition effect was observed for higher substrate
(
o-CNB) concentration. For instant, the o-CNB conversion with the highest o-CNB/Pd molar ratio
10000:1) is only 0.4 time lower than that with another ratio (6000:1). Based on the general
(
catalytic hydrogenation mechanism, we think that one mainly reason probably result in the
declining conversion of o-CNB at high ratios of o-CNB to Pd. That is, substrate molecules with a
high concentration competitively inhibit the absorption of hydrogen molecules on the surface of
catalyst [49]. This negitive behavior weakened the generation of more active H atoms, greatly
impeding the hydrogenation reduction of o-CNB. This trend is highly consistent with
those previous works [50, 51]. Moreover, the o-CAN selectivity kept a higher value ranging from 89
to 95%, following by a decreased trend of o-CAN selectivity when more than about an o-CNB/Pd
molar ratios of 6000. This result is provided an evidence from the dechlorination of o-CNB,
9

producing more anilines as the main by-product [39]
.
3
.2.3 Effect of Reaction Temperature on the Hydrogenation of o-CNB
Reaction temperature is an important factor that influences the hydrogenation reactivity by
controlling reaction route and by changing of catalyst microstructure, or providing extra energy for
overcoming the energy barrier. In this case, we have investigated the impact of three different
temperatures on the hydrogenation of o-CNB promoted by Pd(0.5%)/α-Fe
2
3
O (Fig. 9). The results
in Fig. 9 show that the prepared Pd catalyst was active in the hydrogenation of o-CNB, exhibiting
an increased conversion as a function of temperature. Nevertheless, no significant differences in
o-CAN selectivity were observed in the given temperature range. As follows from the data in Fig.
9
, it was found that the hydrodehalogenation of o-CNB and o-CAN was not efficiently inhibited in
the o-CNB hydrogenation over the as-prepared Pd(0.5%)/α-Fe . This experimental fact
2
O
3
probably results from the intrinsic activity of the Pd-base catalyst and the temperature-dependent
hydrodehalogenation of o-CNB.
Conditions: P(H2)=1MPa, t=1h, no-CNB/nPd=7000:1, Vethanol =5ml.
3
.2.4 Dependence of the o-CNB Hydrogenation upon Reaction Time
Table 1 presents the variations of o-CNB hydrogenation with different reaction times. Under the
selected conditions, the conversion of o-CNB gradually increased with the increasing reaction
time covering a range of 0.5 to 2.0 hours. Based on the hydrogenation pathway, the substrate
conversion and product selectivity are correlated to the efficient adsorption on the active catalytic
sites [52]. Prolonging reaction time generally facilitates the interactions between reactants and
catalyst. Hence, the present work suggests an increasing tendency of o-CNB conversion.
Nevertheless, lengthening time brought about a slight lowering of the o-CAN selectivity. The
main cause probably originates from the further accumulation of more by-products (mainly AN)
and intermediates from some side reactions along with the increase of the reaction time. In
addition, no other by-product except AN appeared at shorter reaction time (0.5 to 1.5 hrs),
suggesting to some extent the Pd-based catalyst used here inhibits the formation of other
by-products even if it does not absolutely inhibit the hydrodehalogenation of o-CAN. By the way,
it is difficult to achieve a high chemoselectivity in ethanol as the solvent in which significant
accumulation of intermediates or dechlorination takes place.
3
.2.5 Dependence of Hydrogen Pressure on the Reaction Rates
As for the liquid phase hydrogenation reaction, hydrogen content in catalytic system is a vital
1
0

factor for the efficient hydrogenation of various compounds [53]. The observations in the present
work (Fig. 10) reveal that the conversion of o-CNB increased with the increasing hydrogen
pressure. The soluble hydrogen in solvent benefits the increase of absorbed hydrogen on the
surface of catalyst, improving the interaction with substrate molecules absorbed on the catalyst
surface. Therefore, the conversion of o-CNB gradually increases (up to 96.3% at 2.0MPa
hydrogen pressure) along with increasing hydrogen pressure. Meanwhile, the hydrogenation of
o-CNB kept relatively stable o-CAN selectivity with a region of 81~86 percent. In addition, it is
possible to observe that that one side reaction concerning the formation of aniline was reduced as
the hydrogen pressure increases. The main reason for this fact consists in the dominate
hydrogenation of nitro group over the reductive dechlorination [54] during the hydrogenation of
o-CNB over α-Fe
2
3
O -supported Pd catalyst at high hydrogen pressure.
3
.2.6 Recycling Tests
In this work, the stability and recyclability of Pd/α-Fe
2
O
3
for the o-CNB hydrogenation was
0
investigated with successive runs under the selected conditions (50 C, 1h, no-CNB/nPd=1000:1,
Vethanol=5ml). The solid catalyst was obtained from the reaction mixture by centrifugal separation
after each run. The left solid was thoroughly washed with ethanol, and dried in vacuum following
by its application in next hydrogenation run. The recycling test was conducted for four times in the
present work, totally.
As shown in Fig.11A, the catalyst is relatively stable and no obvious deactivation towards the
hydrogenation of o-CNB as the Pd-base catalyst was reused three times. Namely, the o-CNB
hydrogenation is maintained at a higher o-CNB conversion (96.34%) and o-CAN selectivity
(
94.44%). Further, the reactivity of the Pd/α-Fe
revealing approximately one sixth of the initial catalytic activity of the prepared Pd catalyst though
00% selectivity of o-CAN was obtained. This observation is probably ascribed to the Pd leaching
and the aggregation of Pd/α-Fe particles. The ICP determination demonstrates Pd leaching with
about 0.2% Pd residue, and distinguishable TEM images of original and reused Pd catalysts (Figs.
1B and 11C) indicates the observable aggregation of the Pd-based catalyst during the recycling
2
O
3
catalyst sharply decreases at the fourth run
1
2
O
3
1
tests. In addition to above-mentioned factors, the accumulated products (such as AN) of side
reactions can bind permanently to a sufficient number of active sites to reduce catalytic efficiency
over time, poisoning the Pd/α-Fe
2
O catalyst. In spite of the achieved higher conversion and
3
selectivity for the o-CNB hydrogenation after the third test, however, to some extent the
UV-light-induced reduced Pd-base catalyst possesses its intrinsic drawback mainly concerning its
11

slightly fewer reusability. Thus, our next work will focus on the improvement of the reusability of
Pd/α-Fe catalyst.
2
O
3
4
. Conclusions
In summary, in the present work the hydrogenation of o-chloronitrobenzene (o-CNB) was
investigated over a UV-light reduced Pd-based catalyst loaded on α-Fe . The experimental
results demonstrate that the light-induced reduction successfully achieved the preparation of
α-Fe -supported Pd-based catalyst. Furthermore, the as-prepared Pd catalyst exhibited good
2
O
3
2
O
3
performance for the hydrogenation of o-CNB, including higher activity and o-CAN reactivity.
0
Under optimum conditions (50 C, 2hrs, P(H
2
)=1MPa, no-CNB/npd=6000:1, Vethanol=5ml), the
conversion of o-CNB and the selectivity of o-CAN reach values respectively up to 100% and
1.4%. Here, the UV-light induced reduction as one convenient method is anticipated to facilitate
9
the preparation of the supported catalysts towards the hydrogenation reactions. Additionally, the
further improvement of the stability and activity of supported catalysts based on the
UV-light-induced reduction methodology shall be our next project.
Acknowledgement
We gratefully thank financial supports from Key Subject of China Zigong Science &
Technology Burear (No. 2013H07), Key Project of Key Laboratory of Green Catalysis of Sichuan
Institutes of High Education (No. LYJ1404). Especially, we highly appreciate Mr. Jinlong Fan as
an employee of the Analysis and Testing Center (Sichuan University of Science & Engineering)
for his supports in those characterizations, i.e., SEM, AFM and EDS element mapping.
1
2

References
1
2
3
4
5
6
7
8
9
. H.-U. Blaser, H. Steiner, M. Studer. ChemCatChem, 2009, 1(2)210-221.
. B. Nair. Int. J. Toxicol., 2001, 20(Suppl. 3)23-50.
. A. Stanislaus, B. H. Cooper. Catal. Rev., 1994, 36(1)75-123.
. P. Serna, A. Corma. ACS Catal., 2015, 5 (12)7114-7121.
. Y. Y. Chen, C. Wang, H. Y. Liu, J. H. Qiu, X. H. Bao. Chem. Commun., 2005, 5298-5300.
. F. Cárdenas-Lizana, S. Gómez-Quero, N. Perret, M. A. Keane. Gold Bull., 2009, 42(2) 124-132.
. G. Y. Fan, W. J. Huang, C. Y. Wang. Nanoscale, 2013, 5, 6819-6825.
. J. Li, X. Y. Shi, Y. Y. Bi, J. F. Wei, Z. G. Chen. ACS Catal., 2011, 1 (6)657–664.
. R. V. Jagadeesh, T. Stemmler, A. -E. Surkus, H. Junge, K. Junge, M. Beller. Nature Protocols,
2
015, 10(4)548-557.
1
1
0. M. Pietrowski. Green Chem., 2011, 13(7)1633-1635.
1. X. D. Wang, N. Perret, J. J. Delgado, G. Blanco, X. W. Chen, C. M. Olmos, S. Bernal, M. A.
Keane. J. Phys. Chem. C, 2013, 117(2)994-1005.
1
1
2. Z. Z. Wei, J. Wang, S. J. Mao, D. F. Su, H. Y. Jin, Y. H. Wang, F. Xu, H. R. Li, Y. Wang.
ACS Catal., 2015, 5 (8)4783-4789.
3. C. Antonetti, M. Oubenali, A. M. R. Galletti, P. Serp, G. Vannucci. Appl. Catal. A: Gen., 2012,
4
21-422, 99-107.
14. F. Harraz, S. El-Hout, H. Killa, I. Ibrahim. J. Catal., 2012, 286, 184-192.
15. Y. Y. Fang, H. L. Xiao, N. Sui, M. H. Liu, W. W. Yu. RSC Adv., 2014, 4, 11788-11793.
16. F. Cárdenas-Lizana, Y. F. Hao, M. Crespo-Quesada, I. Yuranov, X. D. Wang, M. A. Keane, L.
Kiwi-Minsker. ACS Catal., 2013, 3 (6)1386-1396.
1
1
1
2
2
7. G. Vilé, N. Almora-Barrios, N. López, J.Pérez-Ramírez. ACS Catal., 2015, 5 (6)3767-3778.
8. P. Lara, K. Philippot. Catal. Sci. Technol., 2014, 4(8)2445-2465.
9. W. T. Zang, G. Z. Li, L.Wang, X. W. Zhang. Catal. Sci. Technol., 2015, 5(5)2532-2553.
0. S. Furukawa, Y. Yoshida, T. Komatsu. ACS Catal., 2014, 4(5)1441-1450.
1. K. Fulajtárova, T. Soták, M. Hronec, I. Vávra, E. Dobročka, M. Omastová. Appl. Catal. A:
Gen., 2015, 502, 78-85.
2
2
2
2. J. A. Johnson, J. J. Makis, K. A. Marvin, S. E. Rodenbusch, K. J. Stevenson. J. Phys. Chem.
C, 2013, 117(44)22644-22651.
3. A. M. R. Galletti, C. Antonetti, A. M. Venezia, G. Giambastiani. Appl. Catal. A: Gen., 2010,
3
86, 124-131.
4. A. M. R. Galletti, L.Toniolo, C. Antonetti, C. Evangelisti, C. Forte. Appl. Catal. A: Gen., 2012,
1
3

4
47-448, 49-59.
5. A. M. R. Galletti, C. Antonetti, M. Bertoldo, F. Piccinelli. Appl. Catal. A: Gen., 2013, 468,
5-101.
2
2
2
9
6. C. Antonetti, L.Toniolo, G. Cavinato, C. Forte, C. Ghignoli, R. Ishak, F. Cavani, A. M. R.
Galletti. Appl. Catal. A: Gen., 2015, 496, 40-50.
7. P. Zhang, C. Yu, X. M. Fan, X. N. Wang, Z. Ling, Z. H. Wang, J. S. Qiu. Phys. Chem. Chem.
Phys., 2015, 17, 145-150.
28. W. C. Du, S. X. Xia, R. F. Nie, Z. Y. Hou. Ind. Eng. Chem. Res., 2014, 53(12)4589-4594.
29. H. M. Liu, H. B. Yu, C. R. Xiong, S. H Zhou. RSC Adv., 2015, 5, 20238-20247.
30. U. Jung, A. Elsen, Y. Y. Li, J. G. Smith, M. W. Small, E. A. Stach, A. I. Frenkel, R. G. Nuzzo.
ACS Catal., 2015, 5(3)1539-1551.
31. S. Bhattacharjee, M. L. Bruening. Langmuir, 2008, 24(6)2916-2920.
32. F. A. Harraz, S. E. El-Hout, H. M. Killa, I. A. Ibrahim. J. Catal., 2012, 286, 184-192.
33. H. Göksu, S. F. Ho, Ö. Metin, K. Korkmaz, A. M. Garcia, M. S. Gültekin, S. H. Sun. ACS
Catal., 2014, 4(6)1777-1782.
3
4. S. Y. Jung, S. J. Bae, W. J. Lee. Environ. Sci. Technol., 2014, 48(16)9651-9658.
5. A. M. Alshaibani ， Z. Yaakob ， A. M. Alsobaai ， M. Sahri. Brazil. J. Chem.
Eng., 2014, 31(1)69-78.
3
36. S. X. Xia, L. P. Zheng, R. F. Nie, P. Chen, H. Lou, Z. Y. Hou. Chin. J. Catal., 2013,
3
4(5)986-992.
7. E. Borgarello, N. Serpone, G. Emo, R. Harris, E. Pelizzetti, C. Minero. Inorg. Chem., 1986,
5(25)4499-4503.
3
2
3
8. S. H. Park, S. H. Kim, S. Y. Park, W. I. Lee, C.M. Lee. Sensors, 2014, 14(9) 15849-15860.
9. K. V. S. Ranganath, J. Kloesges, A. H. Schäfer, F. Glorius. Angew. Chem. Int. Ed., 2010,
3
4
9(42)7786-7789.
4
4
4
0. M. Brun, A. Berthet, J. C. Bertolini. J. Electron. Spectrosc., 1999, 104(1-3)55-60.
1. P. Li, E. Y. Jiang, H. L. Bai. J. Phys. D: Appl. Phys., 2011, 44(7)075003.
2. W. W. Lin, J. Zhao, H. Y. Cheng, X. R. Li, X. N. Li, F. Y. Zhao. J. Colloid Interf. Sci., 2014,
4
32, 200-206.
43. H. K. Kadam, S. G. Tilve. RSC Adv., 2015, 5, 83391-83407.
44. Q. Bai, D. Li, L. He, H. L. Xiao, N. Sui, M. H. Liu. Prog. Nat. Sci. –Mater. , 2015,
2
5(3)179-184.
4
5. D. Dutta, D. K. Dutta. Appl. Catal. A: Gen., 2014, 487, 158-164.
1
4

4
4
6. X. D. Wang, F. Cárdenas-Lizana, M. A. Keane. ACS Sustain. Chem.
Eng., 2014, 2(12)2781-2789.
7. X. S. Xu, X. Q. Li, H. Z. Gu, Z. B. Huang, X. H. Yan. Appl. Catal. A: Gen., 2012, 429-430,
1
7-23.
4
4
5
5
5
8. Y. L. Xie, N. Xiao, C. Yu, J. S. Qiu. Catal. Commun., 2012, 28, 69-72.
9. U. K. Singh, M. A. Vannice. Appl. Catal. Part A: Gen., 2001, 213(1)1-24.
0. S. Hu, Y. Chen. Ind. Eng. Chem. Res., 1997, 36(12)5153-5159.
1. U. K. Singh, M. A. Vannice. J. Mol. Catal. Part A: Chem., 2000, 163(1-2)233-250.
2. F. Devred, P. Dulgheru, N. Kruse. Elementary Steps in Heterogeneous Catalysis. Reference
Module in Chemistry, Molecular Sciences and Chemical Engineering, from Comprehensive
Inorganic Chemistry II (2 Ed.), 2013, 7, 7-38.
5
3. L. Guczi, Á. Moln á r, D. Teschner. Hydrogenation Reactions: Concepts and Practice.
Reference Module in Chemistry, Molecular Sciences and Chemical Engineering,
from Comprehensive Inorganic Chemistry II (2 Ed.), 2013, 7, 421-457.
4. M. Pietrowski. Curr. Org. Synth., 2012, 9(4)470-487.
5
1
5

2 3 2 3
Fig. 1. The XRD patterns of the α-Fe O (A) and Pd/α-Fe O catalyst (B).
1
6

A
B
5μm
C
D
Pd
10 nm
E
F
0
.19nm
Pd{200}
0
.19nm
Pd{200}
0
.19m
5
1/nm
2
nm
(
B)
Fig. 2. SEM images of α-Fe
Patterns C and D display the TEM patterns of the Pd(0.5%)/α-Fe
100nm and 10nm), respectively. Images E and F respectively show the high-resolution TEM and
corresponding SAED pattern of Pd(0.5%)/α-Fe catalyst.
2
O
3
(A) and Pd/α-Fe
2
O
3
catalyst with 0.5% Pd theroretic loading (B).
2
O
3
catalyst at different scales
(
2
O
3
1
7

6
6
5
5
500
000
500
000
5
O1s
Pd(0)3d5/2
5μm
3
2
2
1
1
5
.0x10
Fe2p
(A)
5
.5x10
Pd(0)3d3/2
5
.0x10
5
.5x10
5
4500
4000
3500
.0x10
C1s
Pd3d
4
.0x10
0
.0
7
00 600 500 400 300 200 100
0
344 342 340 338 336 334 332
Binding Energy (eV)
Binding Energy (eV)
100000
Fe2p1/2
FF e 22 pp 33 // 22
O1s
(D)
(C)
5
0000
0000
0000
0000
0000
80000
60000
40000
20000
0
4
3
2
1
730
725
720
715
710
705
540 538 536 534 532 530 528
Binding Energy (eV)
Binding Energy (eV)
Fig. 3. X-ray photoelectron spectra of the Pd/α-Fe
2
3
O as catalyst for the hydrogenation of o-CNB.
1
8

A
B
C
Pd
Mapping
Operation
4μm
4μm
5μm
D
E
F
Carbon
Conductive
Adhesive
2 3
α-Fe O
4μm
4μm
4μm
Fig. 4. Scanning electronic micrograph image (A) and corresponding EDS elemental mappings (B-F)
of Pd(0.5%)/α-Fe catalyst in elements of (B) full mapping image, (C) Pd, (D) C (carbon
conductive adhesive), (E) Fe, (F) O, respectively.
2
O
3
1
9

A
B
C
[
μm]
[
μm]
[μm]
[
nm]
[deg]
[
μm]
Fig. 5. The AFM 3D view (A), 2D height (B) and topographic phase (C) images of
Pd(0.5%)/α-Fe particles at 5 μm×5 μm scale. The lower one shows a phase image of a selected
area at a smaller scale level (1.5μm×1.5μm) compared to the upper one.
2
O
3
2
0

1
1
20
00
8
6
4
2
0
0
0
0
0
o-CNB Conv.
AN Sel.
o-CAN Sel.
0
0.25 0.5 0.75
1
1.25 1.5 1.75
2
2.25
Content of Pd (%)
Fig. 6 Variety of the o-CNB hydrogenation as a function of the Pd loadings.
0
Conditions: T=50 C, t=1h, P(H
2
)=1MPa, no-CNB/nPd=1000:1, Vethanol =5ml.
2
1

1
1
20
00
8
6
4
2
0
0
0
0
0
o-CNB Conversion%
AN Selectivity%
o-CAN Selectivity%
0
2000
4000
6000
8000
10000
Molar Ratios of o-CNB to Pd/α-Fe
2 3
O
Fig. 7 Dependence of the hydrogenation reduction of o-CNB upon the molar ratios of
0
o-CNB to Pd(0.5%)/α-Fe
2
O
3
. Conditions: T=50 C, t=2h, P(H
2
)=1MPa, Vethanol =5ml.
2
2

1
00
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
o-CNB Conv.
AN Sel.
o-CAN Sel.
Others
2
5
35
45
55
65
75
85
Temperature (degree)
Fig. 9. Effect of temperature on the hydrogenation of o-CNB.
2
3

1
1
20
00
8
6
4
2
0
0
0
0
0
o-CNB Conv.
AN Sel.
o-CAN Sel.
Others
0
.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25
Hydrogen Pressure (MPa)
Fig. 10. Dependence of the o-CNB hydrogenation upon the hydrogen
0
pressure. Conditions: T=50 C, t=1h, no-CNB/nPd=7000:1, V ethanol =5ml.
2
4

1
1
1
40
20
00
(
A) (B)
o-CAN Conv.(%)
o-CAN Sel.(%)
AN Sel.(%)
8
6
4
2
0
0
0
0
0
50 nm
Recycling
(C)
50 nm
1
2
3
4
Recycling Times
2 3
Fig. 11. Recyclability survey (A) of the Pd/α-Fe O catalyst for the hydrogenation of o-CNB.
Images B and C shows the TEM images of the initial and reused Pd catalyst, respectively. The
0
insert of Fig. 11C is the reused Pd-based catalyst sample after three runs. Conditions: T=50 C,
t=1h, no-CNB/nPd=1000:1, Vethanol =5ml.
2
5

Scheme
others
UV
Cl
O
H
by-product
PdCl2
N
H N
2
H
O
Pd
2
alpha--Fe O3
Cl
aim product
H N
2
Scheme 1. Representation of the o-CNB hydrogenation by Pd/α-Fe
O
2 3
catalyst prepared by
UV light-induced route.
2
6

Table 1 Effect of reaction time on the catalytic hydrogenation of o-CNB
Reaction Conversion Selectivity (%)
Entry
time (h) (%)
o-CAN
AN
Others
1
2
3
4
0.5
1.0
1.5
2.0
41.0
72.8
84.7
93.9
84.7
15.3
17.8
16.5
15.4
-
82.2
-
80.5
3.0
6.1
78.5
0
Conditions: T=50 C, P（H2）=1MPa, no-CNB/nPd=7000:1, V ethanol =5ml
2
7