Hydrogenation–dechlorination of 2-chloro-4,6-dinitroresorcinol over Pd/C catalysts

Article abstract of DOI:10.1007/s11164-018-3477-y

Pd catalysts supported on active carbon (Pd/C) prepared by impregnation–deposition were used in catalytic hydrogenation and hydrodechlorination (HDC) of 2-chloro-4,6-dinitroresorcinol (CDNR). Characterization by X-ray diffraction, X-ray photoelectron spectroscopy (XPS), scanning electron microscopy, and transmission electron microscopy revealed Pd species in the form of ~3.0-nm Pd nanoparticles composed of Pd⁰ and Pdⁿ⁺ dispersed highly on active carbon. Pd/C catalysts with high Pd⁰/(Pd⁰?+?Pdⁿ⁺) exhibited high active for hydrogenation of CDNR, while the yield of the target product 4,6-diaminoresorcinol (DAR) largely depended on the solvent media. The correlations of the dielectric constants of the solvents with the activity and selectivity were obtained. Alkanol or acetic acid aqueous solutions were favorable for HDC. No 2-chloro-4,6-diaminoresorcinol product was detected when the reaction was carried out using 80?% ethanol/water media. XPS, energy-dispersive spectroscopy, and high-resolution transmission electron microscopy revealed that strong Cl adsorption caused the decrease in activity and selectivity for DAR of the Pd/C catalysts.

Full text of DOI:10.1007/s11164-018-3477-y

Res Chem Intermed
https://doi.org/10.1007/s11164-018-3477-y
Hydrogenation–dechlorination of 2-chloro-4,6-
dinitroresorcinol over Pd/C catalysts
1
1
1
1
•
Xinzheng Li
Xingyi Wang
•
Feng Qin
•
Qiguang Dai
•
Shijie Shao
1
Received: 1 March 2018 / Accepted: 11 May 2018
Ó Springer Science+Business Media B.V., part of Springer Nature 2018
Abstract Pd catalysts supported on active carbon (Pd/C) prepared by impregna-
tion–deposition were used in catalytic hydrogenation and hydrodechlorination
(HDC) of 2-chloro-4,6-dinitroresorcinol (CDNR). Characterization by X-ray
diffraction, X-ray photoelectron spectroscopy (XPS), scanning electron microscopy,
and transmission electron microscopy revealed Pd species in the form of *3.0-nm
Pd nanoparticles composed of Pd and Pd dispersed highly on active carbon. Pd/C
0
n?
0
0
n?
catalysts with high Pd /(Pd ? Pd ) exhibited high active for hydrogenation of
CDNR, while the yield of the target product 4,6-diaminoresorcinol (DAR) largely
depended on the solvent media. The correlations of the dielectric constants of the
solvents with the activity and selectivity were obtained. Alkanol or acetic acid
aqueous solutions were favorable for HDC. No 2-chloro-4,6-diaminoresorcinol
product was detected when the reaction was carried out using 80 % ethanol/water
media. XPS, energy-dispersive spectroscopy, and high-resolution transmission
electron microscopy revealed that strong Cl adsorption caused the decrease in
activity and selectivity for DAR of the Pd/C catalysts.
Keywords Hydrogenation Á Hydrodechlorination Á Pd/C catalysts Á Solvent
effect Á 4,6-Diaminoresorcinol
&
Xingyi Wang
wangxy@ecust.edu.cn
1
Research Institute of Industrial Catalysis, East China University of Science and Technology,
Shanghai 200237, China
123

X. Li et al.
Introduction
,6-Diaminoresorcinol (DAR) is known to be an indispensable monomer used in
4
preparation of polybenzoxazoles (PBO) that can be processed into special fibers
with both high tensile strength and thermal stability, enabling wide applications in
the military and aerospace fields [1]. There are few available studies on synthesis of
DAR [2–5], with one-pot catalytic hydrogenation and hydrodechlorination (HDC)
of 2-chloro-4,6-dinitroresorcinol (CDNR) being one of most cost-effective routes.
CDNR is readily obtained from 1,2,3-trichlorobenzene through nitrosation and
hydration. An important issue to be solved regarding this technique is complete
hydrodechlorination of 2-chloro-4,6-diaminoresorcinol (2-Cl-DAR), an intermedi-
ate during formation of DAR. Otherwise, the quality of PBO will be significantly
lowered, because 2-Cl-DAR reacts with dissociated hydrochloric acid to form
hydrochloride salt, which is difficult to remove in purification steps.
During liquid-phase hydrogenation, the selectivity for products may be related to
the properties of the solvent media [6–11], which provide a chemical environment
for catalyst dispersion and hydrogen dissolution. Solvent polarity, dielectric
constant, acid/base properties, and solvation ability are considered to be the factors
that significantly influence both the reaction rate and product distribution [12].
Recent studies have focused on HDC of halogenated organic compounds in water
[
[
13–20], organic liquids [21–24], or water–organic liquid coexisting solvents
25–28] under mild conditions, among which alcohols were considered to be
superior solvents, due to their excellent solubility of hydrogen gas and substrates
and their participation as hydrogen donors [29]. Methanol and 2-propanol were used
as hydrogen donors to dechlorinate chloroanisole with Pd catalysts in an inert
atmosphere [30]. For HDC of 4-chloroanisole over Pd-Al-MCM-41 catalyst, the
order of preference of solvents was methanol [ ethanol [ 2-propanol [31]. Water/
alkanols or glycols/organic solvents have also been used to investigate the
hydrogenation and hydrodechlorination of CDNR [32]. Ethanol was dehydro-
genated to supply hydrogen for hydrogenolysis of the C–Cl bond of CDNR [33].
Additionally, acetic acid was considered to be an important medium for
?
hydrogenation of CDNR [1]. Under acidic condition, the interaction between H
and amino groups of aniline masks the electron pair on N and blocks nucleophilic
attack of the amino group on the carbonylic ion, stopping N-alkylation and
inhibiting deamination [34]. Study of the effect of solvents on CDNR hydrogenation
is still of significance for design of highly effective techniques for synthesis of
DAR. To the best of the authors’ knowledge, the striking influence of addition of
water into alkanol or acetic acid in HDC of CDNR is not yet well understood. In this
work, hydrogenation and HDC of CDNR over Pd/C catalysts in organic solvent
media with or without addition of water were investigated in terms of the effects of
solvents, initial temperature, desorption of Cl produced during the reaction, and
kinetic results.
1
23

Hydrogenation–dechlorination of 2-chloro…
Experimental
Preparation of catalysts and starting materials
Activated carbon (17 g) was placed into 300 ml nitric acid aqueous solution
(
10 wt%), agitated at 80 °C for 2 h, then washed with distilled water until the pH
was 7, and finally dried under vacuum at 60 °C for 12 h. A given amount of PdCl
2
(
Aldrich) was dissolved in HCl aqueous solution, into which was added 17 g treated
carbon followed by stirring at 80 °C for 1.5 h. Pd precursor was reduced by addition
of 37 wt% formaldehyde aqueous solution, maintaining the pH value of the
suspension at 8.5 using sodium bicarbonate aqueous solution. The filter was washed
with deionized water until no chloride ions could be detected with AgNO solution,
3
and dried at 80 °C overnight. The Pd loading of the obtained catalysts was
determined by inductively coupled plasma (ICP) to be 4.6, 8.8, and 14.4 wt%.
Catalyst characterization
Powder X-ray diffraction (XRD) patterns of samples were recorded on a Rigaku
D/Max-rC powder diffractometer using Cu Ka radiation (40 kV and 100 mA).
Diffractograms were recorded in the 2h range of 10–80° with step size of 0.01° and
step time of 10 s. Nitrogen adsorption–desorption isotherms were measured at
-
196 °C on an ASAP 2400 system in static measurement mode. Samples were
outgassed at 160 °C for 4 h before measurement. Specific surface area was
calculated using the Brunauer–Emmett–Teller (BET) model. The chemical com-
position of the prepared catalysts was determined by ICP-atomic emission
spectroscopy (AES) using a Vanan 710 instrument. The residual amount of surface
chloride species was determined by energy-dispersive spectrometry (EDS, Falion).
XPS measurements were carried out on a VG ESCALAB MK II spectrometer with
Mg Ka (1253.6 eV) radiation as excitation source. Charging of samples was
corrected by setting the binding energy of adventitious carbon (C 1 ) at 284.6 eV.
s
Transmission electron microscopy (TEM) images of powder samples were recorded
on a JEM-1400 instrument operated at 120 kV. High-resolution transmission
electron microscopy (HRTEM) images were taken on a JEM-2100 microscope
operated at 200 kV.
Hydrogenation and hydrodechlorination of CDNR
The reaction was carried out in a 100-ml autoclave with mechanical stirrer and
hydrogen inlet above the liquid level. A typical reaction mixture contained
0
.025 mol CDNR, 0.025 mol NH Ac, 56 ml solvent, and 0.25 g Pd/C catalyst.
4
After filling with cold suspension, the reactor vessel was flushed with hydrogen six
times to remove air completely before being stirred vigorously. The initial
temperature was set at 60 °C, the hydrogen pressure at 2.0 MPa, and the stirring rate
at 900 rpm. H consumption was determined using an HLMFM flow counter. The
2
reaction was considered to be finished when H consumption was not observed
2
123

X. Li et al.
within 5 min. At that time, the reaction liquor at room temperature was added into
4
0 ml HCl aqueous solution (2.0 mol/L) containing 1 g SnCl Á2H O (as antioxi-
2
2
dant), in which DAR and by-products, such as 2-Cl-DAR and 2-aminoresorcinol
AR), reacted with HCl to form salts. The composition of salts was analyzed
qualitatively and quantitatively by high-performance liquid chromatography
HPLC)-mass spectrometry (MS) (Q Exactive Plus LCMS). According to the
(
(
results of HPLC, yield data of DAR and 2-Cl-DAR could be obtained. The cycle test
was carried out with direct addition of catalyst filtrated from warm liquor into fresh
reaction liquor without treatment.
Results and discussion
CDNR can be hydrogenated to DAR via two reaction pathways (Fig. 1), viz.
hydrogenation of nitro groups followed by dechlorination or dechlorination
followed by hydrogenation of nitro groups. HCl produced during the reaction is
neutralized by a base additive.
Characterization of Pd/C catalysts
The Pd content of the as-synthesized Pt/C catalysts was determined by ICP to be
4.6–14.4 wt%, close to the nominal value (5–15 wt%) (Table 1), indicating that the
Pd(II) precursor could be effectively deposited on active carbon. The narrow pore
size distribution obtained from the N adsorption isotherm branch (not shown) based
2
on the Barrett–Joyner–Halenda (BJH) model shows that the pore structure was
mainly composed of 3.2-nm pores. The BET surface area was estimated to be
2
90–995 m g , and the pore volume as 0.137–0.142 cm g . The slight decrease
-1
3
-1
9
HO
OH
O
OH
O N
NO2
2
H N
2
NH2
Cl
HO
H N
OH
HO
HO
O N
OH
OH
OH
NO2
2
NH2
H N
2
2
Cl
HO
HO
OH
H N
2
NH2
Fig. 1 Proposed reaction pathways for hydrogenation of CDNR
1
23

Hydrogenation–dechlorination of 2-chloro…
Table 1 Structure of Pd/C catalysts
a
b
c
0
Pd /(Pd ?
0
Catalyst
Pd content
(wt%)
Surface
area
Pore
radius
(nm)
Pore
volume
R (nm)
L (nm)
n?
Pd ) (%)
d
2
m g
–1
3
(cm g
–1
(
)
)
5
1
1
a
wt%Pd/C
4.6
8.8
998
995
990
3.2
3.2
3.2
0.142
0.139
0.137
2.6
3.1
3.2
1.9
2.4
3.0
65.9
70.4
89.1
0 wt%Pd/C
5 wt%Pd/C
14.4
Pd content measured by ICP
b
Size of Pd nanoparticles calculated by applying Scherrer equation to h111i
Estimated by TEM
c
d
XPS data
Fig. 2 XRD patterns of
1
5wt% Pd/C
0wt% Pd/C
5
1
wt%Pd/C, 10 wt%Pd/C, and
5 wt%Pd/C
1
5
wt% Pd/C
1
0
20
30
40
50
60
70
80
2
Theta (deg.)
in pore volume with Pd loading should resulted from blockage by Pd species. The
XRD patterns (Fig. 2) showed a broad peak at 23°, assignable to amorphous carbon,
and several peaks at 40.1°, 46.5°, and 68.0° due to reflections from h111i , h200i,
and h220i planes of Pd face-centered cubic lattice according to Joint Committee on
Powder Diffraction Standards (JCPDS) card no. 05-0681. The size of the Pd
nanoparticles was estimated to be 2.6–3.2 nm by applying the Scherrer equation to
h111i plane of the Pd cubic lattice. TEM images showed Pd nanoparticles with
amorphous shape and size of 1.9–3.0 nm, depending on the Pd loading. The
15 wt%Pd/C sample presented the largest size of Pd nanoparticles, which is related
to growth of Pd nanoparticles due to the increased Pd loading (Fig. 3). HRTEM
˚
Fig. 3b) images showed an interplane distance of 2.2 A, corresponding to the
(
123

X. Li et al.
4
3
3
2
2
1
1
0
5
0
5
0
5
0
(a) (b)
d=2.4nm
5
0
0
.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Particle Size(nm)
0
(
.22nm
111)
40 nm 5 nm
(
c) (d)
3
3
2
2
1
1
5
d=3.0nm
0
5
0
5
0
0
.22 nm
111)
5
0
(
0
1
2
3
4
5
6
7
Particle Size(nm)
40 nm
5 nm
(
e) (f)
d=4.5nm
3
3
2
2
1
1
5
0
5
0
5
0
0.22 nm
111)
(
5
0
0
1
2
3
4
5
6
7
8
9
10
Particle Size(nm)
40 nm
5 nm
Fig. 3 TEM images (a, c, e) and HRTEM images (b, d, f) of 10 wt%Pd/C (a, b), 15 wt%Pd/C (c, d), and
5 wt%Pd/C-s (e, f)
1
exposed h111i plane of the face-centered cubic Pd lattice. Pd species were identified
n?
n?
in the form of Pd (337.4–337.7 eV) and Pd (335.2–335.5 eV) by deconvolution
0
of the Pd 3d5/2 XPS spectrum (Fig. 4) [35]. The presence of Pd species is related
to the interaction between Pd species and groups containing oxygen on the carbon
0
surface. The Pd /(Pd ? Pd ) ratio increased with the Pd loading, probably due to
0
n?
decreased Pd defects on growth of the Pd nanoparticles.
1
23

Hydrogenation–dechlorination of 2-chloro…
Fig. 4 Pd 3d XPS spectra for
Pd/C catalysts
1
5wt%Pd/C
1
0wt%Pd/C
5wt%Pd/C
3
48
345
342
339
336
333
330
Binding energy (eV)
a
Table 2 Hydrogenation of CDNR over Pd/C catalysts
b
Entry Catalyst
H
2
Duration Conversion Rate Yield (%)
(%)
–
1
(s )
consumption (min)
mol)
DAR 2-Cl-DAR Other
(
1
2
3
4
5
a
5 wt%Pd/C
0.194
0.187
0.183
0.198
0.197
6.3
5.4
5.0
5.5
6.5
100
100
100
100
100
5.9
3.5
2.1
1.6
3.0
64.2
75.5
80.3
53.3
56.5
1.06
0.10
0.00
1.20
1.03
34.0
24.4
27.7
46.5
42.5
10 wt%Pd/C
15 wt%Pd/C
15 wt%Pd/C-s
c
d
10 wt%Pd/C
4
Reaction conditions: 0.025 mol CDNR; 0.025 NH Ac; 0.25 g catalyst; 56 ml 20 % water/ethanol
solution; 2.0 MPa hydrogen pressure; stirring at 900 rpm; initial temperature, 60 °C
b
LC results
c
Active carbon, from Sinopharm Chemical Reagent Limited Corporation
d
Commercial catalyst
Hydrogenation and hydrodechlorination of CDNR
Effect of Pd loading
Pd/C catalysts with various Pd loadings were investigated in hydrogenation of
CDNR, and the results are summarized in Table 2. The initial H consumption rate
2
within the first minute based on the number of H
-
transformed per mol of Pd per
mol
2
1
second was estimated to be 5.9, 3.5, and 2.1 s for 5 wt%Pd/C, 10 wt%Pd/C, and
5 wt%Pd/C, respectively. Obviously, the catalysts with small Pd nanoparticles
1
123

X. Li et al.
exhibited higher activity for CDNR hydrogenation. It took less than 6.3 min for the
CDNR conversion to reach 100 %. After that, the H consumption became very
2
slow, as the reaction approached completeness. The products were mainly
composed of DAR, 2-Cl-DAR, and AR, as confirmed by HPLC-MS. AR resulted
from deamination of DAR. The carbon balance reached 98 % or higher. No
azoxybenzene and azobenzene could be detected, which is not really consistent with
the reduction of m-dinitrobenzene over Pd/TiO , where the selectivity for
2
azoxybenzene and azobenzene was up to 17 % and that for m-phenylenediamine,
9 % [36]. The difference in the yields of DAR and 2-Cl-DAR should be related to
4
the properties of the catalyst. The yield of DAR was proportional to the Pd /
0
0
Pd ? Pd ) ratio of the Pd/C catalysts (Fig. 5a). As reported, Pd species is much
n?
0
(
more active for H dissociation than Pd species [37], especially for formation of
n?
2
n?
H atoms. On Pd sites, heterolytic dissociation of H into H and H occurs easily.
–
?
2
Moreover, H atoms can be highly active for hydrogenation, such as reduction of
?
nitro group and reductive substitution of chlorine. Meanwhile, H could be
-
favorable for electrophilic attack at N atom of NH groups, and H , for nucleophilic
2
attack at the corresponding carbon atom, leading to formation of AR through C–N
bond breaking. In fact, deamination of DAR was not observed under experimental
condition. It can be deduced that AR formed through reduction of CDNR in ‘‘one
n?
step’’ on Pd sites. The 15 wt%Pd/C sample presented the highest selectivity for
n?
DAR, due to its least Pd sites. Moreover, the theoretical H consumption for
2
conversion of 0.025 mol CDNR into DAR is 0.175 mol. Excess H consumption
2
resulted from deamination of DAR. For 5 wt%Pd/C, the theoretical H consumption
2
used for production of DAR and 2-Cl-DAR was 0.114 mol and for AR was
0
.08 mol, almost consistent with the yield of AR.
Reduction of nitro groups should alternate with HDC during liquid-phase
hydrogenation of CDNR. The HCl produced can be neutralized by base additive.
1
00
(
a)
(b)
1
1
0
0
0
0
.2
.0
.8
.6
.4
.2
9
8
7
6
5
0
0
0
0
0
0.0
5
0
60
70
80
90
50
60
70
80
n+
90
o
o
n+
o
o
Pd /Pd +Pd
Pd /Pd +Pd
0
Fig. 5 Effect of Pd /(Pd ? Pd ) ratio on yields of DAR (a) and 2-Cl-DAR (b)
0
n?
1
23

Hydrogenation–dechlorination of 2-chloro…
However, Cl atom near hydroxyl group on ring is removed through HDC with some
difficulty, especially for very low concentration of reactant. Figure 5b shows that
0
the yield of 2-Cl-DAR was also dependent on the Pd /(Pd ? Pd ) ratio of the
0
n?
0
catalysts. With increase in the Pd /(Pd ? Pd ) ratio from 65 to 75 %, 2-Cl-DAR
0
n?
decreased rapidly from 1.0 to 0.1 %, indicating that HDC became more effective
0
over the catalysts with high Pd /(Pd ? Pd ) ratio. Generally, Cl species tend to
0
n?
n?
adsorb strongly on active Pd sites. It can thus be expected that Pd sites will be
favorable for adsorption of Cl species. When hydrogenation approaches complete-
ness, HDC of a very low amount of 2-Cl-DAR should be more effective. Of course,
2
-Cl-DAR could be further converted to DAR given more time. As a result,
deamination would be promoted to a large extent. Because 2-Cl-DAR reacts with
dissociated hydrochloric acid to form hydrochloride salt, which is difficult to
remove in purification steps, the amount of 2-Cl-DAR must be controlled as low as
possible while meeting polymerization requirements. Decreased 2-Cl-DAR content
is required for desirable synthesis pathway. In later experiments, all data on various
aspects of liquid-phase hydrogenation of CDNR were obtained using 15 wt%Pd/C.
Effects of solvents
The effects of solvents on the reaction over 15 wt%Pd/C were investigated, and the
results are presented in Table 3. The reactions in different alkanols were more rapid
than those in other solvents. It took a long duration for H2 consumption in
cyclohexane and n-heptane. In nonpolar solvent systems, inorganic Cl species
tended to deposit on the catalyst surface due to the difficulty in transfer of inorganic
Cl species in nonpolar solvent, thus deactivation of the Pd catalyst is inevitable. The
surface of the Pd/C catalysts was covered by hydroxyl and carboxylic groups
(
confirmed by XPS) produced during the treatment of active carbon with nitric acid
aqueous solution, thus having affinity for polar solvents, while CDNR and DAR
with two hydroxyl groups have affinity for polar solvents such as alcohols and acetic
acid. Thus, the low rate and no DAR production can be ascribed to poor mass
transfer in nonpolar solvents. In contrast, the mass transfer in polar media became
smooth, with DAR yield reaching 76, 88, 80, and 73 % in acetic acid, isopropanol,
ethanol, and methanol, respectively. Liquid composition analyses showed that the
main by-products were 2-Cl-DAR and AR in almost all solvents. The yield of 2-Cl-
DAR lay in the range of 0–2 %. The carbon balance reached 98 % or higher. In
water, however, the duration for H consumption was much longer than in other
2
polar media, implying that another factor influences the reaction. The H2
consumption in water was 0.202 mol, higher than 0.175 mol (the theoretical value
needed for complete conversion of 0.025 mol CDNR to DAR), and the product was
composed almost entirely of AR, for which the theoretical H consumption is
2
0
.200 mol. Obviously, almost all the CDNR was completely converted into AR.
The difference in product distribution can be related to the properties of the
reaction media, such as solvent polarity, dielectric constant, and acid/base properties
12]. According to Table 3, with increase in the dielectric constant from 18.6 (2-
[
propanol) to 80.1 (water), the selectivity for AR increased, except for acetic acid.
The effect of water content in organic media on the reaction over 15 wt%Pd/C was
123

X. Li et al.
a
Table 3 Solvent effects on hydrogenation of CDNR
b
c
Entry Solvent (vol
H
2
Dielectric
constant
Time
(min)
Conversion Yield (%)
(%)
%)
consumption
mol)
DAR 2-Cl-
DAR
AR
100
(
d
1
2
3
4
5
6
7
8
Water
0.202
0.188
0.182
0.181
0.184
0.080
80.1
33.0
24.3
18.6
6.2
44.5
1.8
100
100
100
100
100
61
ND
ND
Methanol
Ethanol
73
1.24
0.82
1.14
0.94
ND
26.5
2.6
80
20.6
10.5
23.0
40.2
41.5
33.2
Isopropanol
Acetic acid
n-Heptane
2.8
88
8.3
76
d
1.9
17.7
25.3
4.7
ND
ND
67
Cyclohexane 0.085
80 % 0.186
methanol
80 % ethanol 0.183
80 % 0.185
ethanol
1.2
58
ND
42.4
100
0.21
9
1
35.5
35.5
5.0
6.5
100
100
78
69
ND
27.2
32.0
0
0.73
e
1
1
1
1
2
3
50 % ethanol 0.186
30 % ethanol 0.188
52.2
63.4
30.7
9.1
16.5
5.3
100
100
100
56
44
86
0.10
0.24
0.28
44.2
55.7
13.9
80 %
isopropanol
0.179
0.188
0.186
0.187
1
1
1
a
4
5
6
80 % acetic
acid
21.0
43.2
57.9
15.8
18.2
20.5
100
100
100
72
47
38
0.17
0.17
0.30
28.7
52.3
61.6
50 % acetic
acid
30 % acetic
acid
Reaction conditions: 0.025 mol CDNR; 0.025 mol NH
4
Ac; 0.25 g 15 wt%Pd/C; initial reaction tem-
perature, 60 °C; 56 ml solvent; 2.0 MPa hydrogen pressure; stirring rate, 900 rpm
b
2
Duration until end of H uptake
c
Product yield
d
e
Not detected
4
Same as footnote ‘a’ except without NH Ac
investigated under typical reaction condition. For alkanol media containing 20 %
water, the 2-Cl-DAR content decreased greatly by 75–100 %, while the DAR yield
decreased to a small extent (Table 3). Increase in the dielectric constant of the
reaction media (Table 3) by addition of water seemed favorable for HDC of 2-Cl-
DAR. On the other hand, water promotes dissolution of NH Ac, which, in turn,
4
promotes removal of Cl species from the catalyst surface [38], thus the active sites
for HDC become more available. In other words, strong Cl absorption on active Pd
site can inhibit C–Cl bond breaking in 2-Cl-DAR. Further increase in the water
content makes the dielectric constant of the reaction medium increase (Fig. 6),
1
23

Hydrogenation–dechlorination of 2-chloro…
causing the yield of DAR to decrease gradually. A similar result was observed when
increasing the water content in acetic acid (Fig. 6).
The H consumption as a function of time during hydrogenation of CDNR over
2
1
5 wt%Pd/C under typical reaction conditions in the studied media was linear
Fig. 7), although the initial rate appeared to be quite different. This phenomenon
(
indicates that the rate is constant regardless of the concentration of reactant, which
could be described well by zero-order kinetics. This observation of zero-order
kinetics combined with the very quick reaction suggests that diffusion of reactants
or products is slow enough to be the rate-controlling step within the overall reaction
time. Transfer of DAR or CDNR is also slow, even if they are more soluble in pure
organic solvents. The main reason can be ascribed to very rapid hydrogenation of
CDNR. Moreover, it should be pointed out that adsorption of inorganic Cl species
produced during the reaction on Pd atoms is so strong that removal of Cl species
from the catalyst surface would be slow. Considering the difference in the rate of H2
consumption in reaction media with different dielectric constants, transfer of Cl in
the media can be readily affected by their compatibility.
Effect of initial reaction temperature
The effect of the initial reaction temperature on the reaction over 15 wt%Pd/C in
8
0 % ethanol/water was investigated (Fig. 8). The standard reaction procedure used
a feed composed of 0.025 mol CDNR, 0.025 mol NH Ac, 56 ml solvent, and 0.25 g
4
Pd/C catalyst filled into the reactor vessel, following by flushing with hydrogen six
times to remove air completely. The reaction liquor was heated at hydrogen pressure
of 2.0 MPa to given temperature without stirring. After that, the stirring rate was set
at 900 rpm, and H consumption occurred. On raising the initial temperature from
2
Acetic acid+H O
2
7
6
4
3
1
5
0
5
0
5
0
Ethanol+H O
2
1.4
Acetic acid+H O
2
1
1
0
0
0
0
0
.2
.0
.8
.6
.4
.2
.0
Ethanol+H O
2
0
10 20 30 40 50 60 70 80
H
O content / %
2
0
20
40
60
80
100
H O content (%)
2
Fig. 6 Effect of water content in organic solvents on yield of DAR and selectivity for 2-Cl-DAR (inset)
123

X. Li et al.
Fig. 7 Relation of H
consumption with reaction time;
a reaction condition: 0.025 mol
2
0.20
0.15
0.10
CDNR, 0.025 mol NH
ethanol/H O, 0.25 g 15 wt%Pd/
C, 2.0 MPa hydrogen pressure,
00 rpm stirring rate;
b condition similar to ‘a’ except
for free NH Ac; c reaction
condition similar to ‘b’ except
for 3.6 ml Et
4
Ac, 80%
2
9
4
3
N
8
8
8
8
8
0% CH OH-a
3
0% C H OH-a
2
5
0
0
.05
.00
0% C H OH-b
2
5
0% C H OH-c
2
5
0% C H OH-a
3
7
0
1
2
3
4
5
6
7
Time (min)
3
0 to 50 °C, the yield of DAR increased from 75 to 95 % with slight change in the
yield of 2-Cl-DAR. With further increase in the initial temperature to 60 °C or
higher, the DAR yield decreased below 76 %, accompanied by a significant
decrease in 2-Cl-DAR, indicating that high temperature favors HDC. At higher
temperature, deamination of amino groups of DAR becomes significant, leading to a
sharp decrease in DAR yield, e.g., 28 % at 70 °C. Considering both the suitable,
high yield of DAR and the lower yield of 2-Cl-DAR, it can be suggested that the
optimal initial temperature is 60 °C.
Fig. 8 Effect of initial reaction
temperature on yields of DAR
and 2-Cl-DAR (inset)
100
1
1
0wt% Pd/C
5wt% Pd/C
8
6
4
2
0
0
0
0
0
1
.6
2
-Cl-DARYield (%)
1.2
0
0
0
.8
.4
.0
3
0
40
50
60
70
o
Initial reaction temperature ( C)
1
23

Hydrogenation–dechlorination of 2-chloro…
Stability tests and catalyst deactivation
The stability of 15 wt%Pd/C in 80 % ethanol/water media was investigated. After
the reaction, the catalyst was directly separated from the reaction liquor then used
for the next reaction without any treatment. From run 1 to 5, the DAR yield
gradually decreased to 78,73, 68, 67, and 59 %, respectively, with increase of the
2
increased from 4.9 to 6.2 min. Characterization by XPS revealed that the Pd species
-Cl-DAR yield from 0 to 0.99 %. The corresponding H consumption duration
2
n?
in the used 15 wt%Pd/C catalysts were still in the forms of Pd (337.0–337.5 eV)
0
0
0
n?
and Pd (335.1–335.5 eV) (Fig. 9), while the Pd /(Pd ? Pd ) ratio decreased
from 89.1 % for the fresh catalyst to 74.5, 61.2, 56.9, and 47.4 % after run 2 to 5,
respectively (Fig. 9). EDS analyses confirmed that the corresponding content of Cl
species deposited on the catalyst surface increased to 0.90, 1.34, 1.39, and 2.77 %
(
n?
Fig. 10). The increase in the Pd content is probably related to oxidation of the
1
5 wt%Pd/C catalyst during its separation from the reaction liquor in air. On the
n?
other hand, Cl species deposited on the surface can lead to formation of Pd
through electron withdraw. Figure 10 shows the relation between the Cl content and
0
the Pd /(Pd ? Pd ) ratio. As mentioned above, Pd species are favorable for
0
n?
n?
?
heterolysis of H to produce H , which leads to denitrification. The corresponding
2
HDC activity can be reduced (Fig. 10, inset). In the other experiment, 15 wt%Pd/C
used for the first reaction was separated from the recrystallization step; the Cl
0
deposition increased up to 1.5 % with Pd /(Pd ? Pd ) ratio of 62.5 %. This result
0
n?
indicates that the increase in Cl deposition due to adsorption of Cl from the
n?
recrystallization solution promoted formation of Pd . For the 15 wt%Pd/C used for
Fig. 9 Pd 3d XPS spectra of
15 wt%Pd/C used for different
times in recycle test: a fresh,
b one time, c two times, and
d three times, and e 15 wt%Pd/C
after use in 80 % ethanol/water
following by treatment with HCl
aqueous solution or f 15 wt%Pd/
C used in 80 % ethanol/water
a
b
c
4
without NH Ac
d
e
f
345
340
335
330
Binding energy (eV)
123

X. Li et al.
Fig. 10 Effect of Cl deposition
on Pd /(Pd ? Pd ); inset:
100
0
0
n?
0
yield of DAR versus Pd /
0 n?
Pd ? Pd )
(
8
6
4
2
0
0
0
0
0
8
7
7
6
6
5
0
5
0
5
0
5
5
0
60 70 80 90
o o n+
Pd /Pd +Pd (%)
0
.0
0.5
1.0 1.5
2.0
2.5
3.0
Cl content (%)
the reaction in 80 % acetic acid/water media one time, the Cl deposition was
determined to be 2.5 % by EDS (Fig. 11), similar to the value obtained after the
reactions in 80 % ethanol/water media four times (Fig. 12). Here, the great decrease
in the 2-Cl-DAR content is related to low Cl deposition.
Figure 13 shows scanning electron microscopy (SEM) images of 15 wt%Pd/C
used for the reaction in 80 % ethanol/water media. It can be seen that the fresh
catalyst presented a smooth surface (Fig. 13), becoming rough after run 1, even with
washing by ethanol. However, with washing by ethanol then water, the rough
surface became smooth, indicating coverage by species soluble in either ethanol or
water, such as DAR and 2-Cl-DAR or their salts, as confirmed by the fact that a
Fig. 11 EDS spectrum of
5 wt%Pd/C used for the
10000
1
reaction in 80 % acetic acid/
water media
C
8
6
4
2
000
000
000
000
0
Pd
O
Cl
0
.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Energy (keV)
1
23

Hydrogenation–dechlorination of 2-chloro…
Fig. 12 Cl content and 2-Cl-
DAR yield in cycle reaction for
different times
3.0
2.5
Cl content
-Cl-DAR yield
2
2
2
1
1
0
0
.5
.0
.5
.0
.5
.0
2
1
1
0
.0
.5
.0
.5
0.0
1
2
3
4
5
Cycle-to-cycle number
1
1
2000
0000
C
1
0000
C
8
000
000
000
8
6
4
2
000
000
000
000
0
6
4
2000
O
Pd
Pd
Cl
O
N
0
0
.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Energy-keV
Energy-Kev
1
0000
6000
5000
C
C
8
000
000
000
000
0
4
3
2
000
000
000
6
4
2
Pd
Cl
O
1000
Pd
Cl
Sn
O
N
0
.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Energy-Kev
0
0
.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Energy-keV
Fig. 13 SEM images and EDS spectra of 15 wt%Pd/C catalyst treated under different conditions:
a fresh; b, c used for reaction in 80 % ethanol/water media followed by washing with ethanol, and
alternatively with ethanol and water; d used for the reaction in ethanol media followed by washing
alternatively with ethanol and water
123

X. Li et al.
significant amount of N element was detected by EDS. Addition of water into
organic solvent systems can prevent deposition of DAR salt on the catalyst surface
to some extent. The 15 wt%Pd/C used for reaction in pure organic solvent medium
one time presented almost no activity, which is related to DAR salt deposition.
HRTEM analyses (Fig. 3) showed 3.0-nm Pd nanoparticles uniformly distributed on
the fresh 15 wt%Pd/C, while the size of the Pd nanoparticles after the reaction in
8
0 % ethanol/water media increased to 5.2 nm (Fig. 14a), consistent with the XRD
analysis results (Fig. 15). Additionally, several peaks appeared in the XRD patterns,
probably due to deposition of crystals, such as DAR and 2-Cl-DAR salt (Fig. 15).
Without NH Ac, the aggregation became more severe (Fig. 14b), indicating that
4
(a)
(b)
5
nm
50 nm
Fig. 14 TEM images of 15 wt%Pd/C catalyst used for reactions in 80 % ethanol/water media for four
times (a) and 80 % ethanol/water without NH
4
Ac for one time (b)
Fig. 15 XRD pattern of
1
8
5 wt%Pd/C used for reaction in
0 % ethanol/water media
1
0
20
30
40
50
60
70
80
2
Theta (deg.)
1
23

Hydrogenation–dechlorination of 2-chloro…
neutralization of HCl by NH Ac is important for removal of Cl species from the
4
catalyst surface. Otherwise, growth of Pd nanoparticles can be promoted by Cl
species. ICP analyses showed that as little as 0.5 % of the whole Pd content was
leached in each run. Based on the results of all these analyses of the used catalyst, it
can be concluded that catalyst deactivation can mainly be ascribed to adsorption of
reactant and products, modification of Pd oxidation state, aggregation of Pd
nanoparticles, and Pd leaching.
Conclusions
Pd/C catalysts with Pd loading of 5–15 wt% prepared by impregnation–deposition
were characterized by XRD, N adsorption–desorption isotherms, XPS, SEM, and
2
0
TEM. The experimental results revealed Pd species, composed of main Pd and
n?
minor Pd , highly dispersed on active carbon with average size of Pd nanoparticles
in the range of 1.9–3.0 nm. In hydrogenation and HDC of CDNR, the Pd/C catalysts
presented high activity and selectivity for DAR in polar solvents. Yield of DAR up
to 73–88 % could be obtained. The by-products were composed of AR and trace
0
0
n?
2
-Cl-DAR. High Pd /(Pd ? Pd ) ratio was favorable for HDC of 2-Cl-DAR.
Moreover, addition of water into organic polar solvents could greatly promote HDC
of 2-Cl-DAR through removal of Cl species from the Pd/C catalyst surface. An
obvious effect of the initial temperature on the selectivity for DAR and 2-Cl-DAR
was observed. Considering both the suitable, high yield of DAR and lower yield of
2
-Cl-DAR, the optimal initial temperature should be 60 °C. No 2-Cl-DAR was
detected over 15 wt%Pd/C in 80 % ethanol/water media. XPS, EDS, and HRTEM
analyses showed that deactivation of the Pd/C catalysts was related to adsorption of
reactant and products, modification of Pd oxidation state, and aggregation of Pd
nanoparticles.
Acknowledgements The authors are grateful to Mrs. Xingxing Jiang for helpful discussions during the
revision process. We would like to acknowledge the National Key Research and Development Program of
China (no. 2016YFC0204300) and National Natural Science Foundation of China (nos. 21477036 and
21777043).
References
1. Z. Lysenko, N. Midland, High purity process for the preparation of 4,6-diamino-1,3-benzenediol[P].
US: 4766244, 1988
2
3
. S.N. Bassam, M. Midland, Synthesis of 4, 6-diaminoresorcinol. U.S. 5371291, 1994
. M.B. Horst, F. Helmut, Process for preparing 4,6-diaminoresorcinol dihydrochloride. U.S. 6093852,
2000
4
5
6
. G.P. Richard, Z. Lysenko, C.V. Paul, J. Org. Chem. 62, 8255 (1997)
. Z. Lysenko, G.P. Richard, Process for the preparation of diaminoresorcinol. U.S. 5399768, 1995
. S.K. Wilkinson, I. McManus, H. Daly, J.M. Thompson, C. Hardacre, N.S. Bonab, J.T. Dam, M.J.H.
Simmons, C. D’Agostino, J. McGregor, L.F. Gladden, E.H. Stitt, J. Catal. 330, 362 (2015)
. I.I. Kaminska, D. Lisovytskiy, S. Casale, A. Srebowata, S. Dzwigaj, Microporous Mesoporous Mater.
7
8
237, 56 (2017)
. B. Ren, M.H. Zhao, L.C. Dong, G.X. Li, Catal. Commun. 50, 92 (2014)
123

X. Li et al.
9
. I. McManus, H. Daly, J.M. Thompson, E. Connor, C. Hardacre, S.K. Wilkinson, N. Sedaie Bonab, J.
ten Dam, M.J.H. Simmons, E.H. Stitt, C. D’Agostino, J. McGregor, L.F. Gladden, J.J. Delgado, J.
Catal. 330, 344 (2015)
1
0. D.J. Segobia, A.F. Trasarti, C.R. Apestegu ´ı a, Catal. Commun. 62, 62 (2014)
11. L.G. Woolfolk, C. Geantet, L. Massin, D. Laurenti, J. De los Reyes, Appl. Catal. B Environ. 201, 331
(2017)
1
2. B.S. Akpa, C.D. Agostino, L.F. Gladden, K. Hindle, H. Manyar, J. McGregor, R. Li, M. Neurock, N.
Sinha, E.H. Stitt, D. Weber, J.A. Zeitler, D.W. Rooney, J. Catal. 289, 30 (2012)
3. Y.L. Ren, G.Y. Fan, C.Y. Wang, J. Hazard. Mater. 274, 32 (2014)
4. C.B. Molina, A.H. Pizarro, J.A. Casas, J.J. Rodriguez, Appl. Catal. B Environ. 148, 330 (2014)
5. J.A. Baeza, L. Calvo, J.J. Rodriguez, E. Carb o´ -Argibay, J. Rivas, M.A. Gilarranz, Appl. Catal.
B Environ. 168, 283 (2015)
1
1
1
16. M. Munoz, V. Kolb, A. Lamolda, Z.M. Pedro, A. Modrow, B.J.M. Etzold, J.J. Rodriguez, J.A. Casas,
Appl. Catal. B Environ. 218, 498 (2017)
17. Y.B. Huang, S.J. Liu, Z.J. Lin, W.J. Li, X.F. Li, R. Cao, J. Catal. 292, 111 (2012)
18. H.Y. Deng, G.Y. Fan, C.Y. Wang, L. Zhang, Catal. Commun. 46, 219 (2014)
19. N. Jadbabaei, T. Ye, D.M. Shuai, H.C. Zhang, Appl. Catal. B Environ. 205, 576 (2017)
20. O.V. Netskina, E.S. Tayban, A.P. Moiseenko, O.V. Komova, S.A. Mukha, V.I. Simagina, J. Hazard.
Mater. 147, 401 (2007)
21. W. Nishijima, Y. Ochi, T.Y. Tsai, Y. Nakano, M. Okada, Appl. Catal. B Environ. 51, 135 (2004)
2
2. M. Cobo, C.A. Gonz a´ lez, E.G. S a´ nchez, C. Montes, Catal. Today 172, 78 (2011)
23. V.I. Simagina, O.V. Netskina, E.S. Tayban, O.V. Komova, E.D. Grayfer, A.V. Ischenko, E.M.
Pazhetnov, Appl. Catal. A Gen. 379, 87 (2010)
24. O.V. Netskina, O.V. Komova, E.S. Tayban, G.V. Oderova, S.A. Mukha, G.G. Kuvshinov, V.I.
Simagina, Appl. Catal. A Gen. 467, 386 (2013)
25. T. Yoneda, T. Aoyama, T. Takido, K. Konuma, Appl. Catal. B Environ. 142, 344 (2013)
26. X.X. Ma, Y. Liu, S.J. Liu, C.H. Xia, Appl. Catal. B Environ. 144, 580 (2014)
27. X.X. Ma, Y. Liu, X.Q. Li, J.G. Xu, G.D. Gu, C.H. Xia, Appl. Catal. B Environ. 165, 351 (2015)
28. H. Sohn, G. Celik, S. Gunduz, S.L. Dean, E. Painting, P.L. Edmiston, U.S. Ozkan, Appl. Catal.
B Environ. 203, 641 (2017)
2
3
3
9. V.A. Yakovlev, V.I. Simagina, V.I. Likholobov, React. Kinet. Catal. Lett. 65, 177 (1998)
0. Y.J. Ukisu, T. Miyadera, React. Kinet. Catal. Lett. 89, 341 (2006)
1. T. Kawabata, I. Atake, Y. Ohishi, T. Shishido, Y. Tian, K. Takaki, K. Takehira, Appl. Catal.
B Environ. 66, 151 (2006)
3
3
2. T.K.T. Yin, C. Calif, Aqueous synthesis of 1-halo-4,6-dinitroresorcinol and 4,6-diamino- resorcinol.
U.S. 5001279, 1991
3. P.J. Mcgrath, M. Calif, Process for reducing aromatic halo-dinitro-diols in the presence of hydrogen
halide. U.S. 5360932, 1993
34. Z.H. Jin, D. Li, R. Ma, X.T. Peng, X.Y. Wang, G.Z. Lu, J. Ind. Eng. Chem. 15, 434 (2009)
35. M. Brun, A. Berthet, J.C. Bertolini, J. Electron Spectrosc. Relat. Phenom. 104, 55 (1999)
36. S.S. Liu, X. Liu, L. Yu, Y.M. Liu, H.Y. He, Y. Cao, Green Chem. 16, 4162 (2014)
3
7. M. Luisa, G. Sainero, L.S. Xos e´ , L.G.F. Jos e´ Adolfo, J. Catal. 209, 279 (2002)
38. Z.H. Jin, C. Yu, X.Y. Wang, D. Li, G.Z. Lu, Chem. Commun. 29, 4438 (2009)
1
23