Diruthenium complex catalyzed reduction of nitroarenes-investigation of reaction pathway

Article abstract of DOI:10.1016/j.mcat.2019.01.005

A diruthenium complex, [(L)Ru₂(η⁶-C₆H₆)₂Cl₂](PF₆)₂ (1) (L = 5-phenyl-2,8-di-2-pyridinylanthyridine), was prepared and characterized. This diruthenium complex 1 was found to be an efficient catalyst for the reduction of aromatic nitro compounds leading to the corresponding aniline derivatives with the use of hydrazine as the reducing agent at 80 °C in an ethanol solution. Catalytic activity of 1 towards various possible intermediates leading to anilines was investigated to understand the reaction pathway. These studies indicate that this reduction proceeds via a direct route as evidenced by hydroxylamines being observed as the major intermediate followed by the appearance of aniline under the catalytic conditions. Thus, the reaction pathway of this catalytic system is discussed.

Full text of DOI:10.1016/j.mcat.2019.01.005

Molecular Catalysis 466 (2019) 46–51
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Diruthenium complex catalyzed reduction of nitroarenes-investigation of
reaction pathway
⁎
Shih-Chieh Aaron Lin, Yi-Hung Liu, Shie-Ming Peng, Shiuh-Tzung Liu
Department of Chemistry, National Taiwan University, Taipei, Taiwan
A R T I C L E I N F O
A B S T R A C T
6
Keywords:
Diruthenium
Reduction
Nitroarene
Hydrazine
Aniline
2 6 6 2 2 6 2
A diruthenium complex, [(L)Ru (η -C H ) Cl ](PF ) (1) (L = 5-phenyl-2,8-di-2-pyridinylanthyridine), was
prepared and characterized. This diruthenium complex 1 was found to be an efficient catalyst for the reduction
of aromatic nitro compounds leading to the corresponding aniline derivatives with the use of hydrazine as the
reducing agent at 80 °C in an ethanol solution. Catalytic activity of 1 towards various possible intermediates
leading to anilines was investigated to understand the reaction pathway. These studies indicate that this re-
duction proceeds via a direct route as evidenced by hydroxylamines being observed as the major intermediate
followed by the appearance of aniline under the catalytic conditions. Thus, the reaction pathway of this catalytic
system is discussed.
1. Introduction
of binuclear complexes as catalysts in chemical transformation has re-
ceived much attention [4]. Here, we report the preparation of a dir-
Anilines are important chemicals as the starting materials for or-
uthenium complex containing an anthyridine-based ligand L and the
catalytic activity of the complex on reduction of nitroarenes. Further-
more, the reaction pathway of this catalytic system is investigated.
ganic synthesis, polymers, pharmaceuticals, and other aspects [1].
Generally, the catalytic reduction of nitroarenes is considered to be the
most convenient way to prepare anilines and numerous catalysts ac-
companied with various reductants have been successfully employed.
Despite these achievements, chemists are also interested in the reduc-
tion pathway operated under a specific catalytic condition. Reduction
of nitroarenes generally comprises multi-step pathways and two routes
are often discussed in the homogeneous catalysis (Scheme 1) [2]. The
first step involves the reduction of nitro functionality into nitroso
group, which is subsequently reduced to the hydroxyaminobenzene.
This intermediate can undergo two competing routes to provide the
final product: the direct reduction of hydroxyaminobenzene leading to
aniline (Scheme 1, route A), and the condensation of nitroso- and hy-
droxyamino-benzene to generate azoxybenzene, which is further re-
duced to azobenzene, diphenylhydrazine and finally to aniline (Scheme
2. Experimental section
2.1. General information
Chemicals and solvents were of analytical grade and were used
without further purification. Nuclear magnetic resonance spectra were
recorded on a 400 MHz spectrometer. Chemical shifts are given in parts
1
13
per million relative to Me
thenium complex [(bipy)Ru(η -C
4
Si for H and C NMR. Ligand L and ru-
6
6 6
H )Cl]Cl (2) [bipy = 2,2′-bipyridine]
were prepared according to the reported methods [5].
6
2.2. Preparation of Complex 1, [(L)Ru
2
(η -C
6
H
6
)
2
Cl
2
](PF
6
)
2
1
, route B).
Pathway operating in a catalytic system really depends on the
nature of catalyst and reaction conditions [2]. In a previous study, we
found that reduction of nitroarenes catalyzed by a dipalladium complex
This complex was prepared according to Scheme 2. A mixture of L
(70 mg, 0.17 mmol), [Ru(η -C H )Cl ] (90 mg, 0.18 mmol) and KPF
6 6 2 2 6
(60 mg, 0.34 mmol) in a 25 ml round-bottom flask was flashed with
nitrogen. Anhydrous acetonitrile (5 mL) was syringed into the mixture
and the resulting solution was heated under refluxing condition for
18 h. Upon cooling, the volume of the reaction mixture was reduced
down to 1 ml and diethylether was added to precipitate out the desired
6
(
Fig. 1) undergoes the condensation pathway (route B) presumably due
to the synergistic effect caused by the dinuclear catalyst to facilitate the
condensation [3]. It is believed that metal ions in close proximity are
expected to “cooperate” in promoting reactions. In this context, the use
⁎
Corresponding author.
E-mail address: stliu@ntu.edu.tw (S.-T. Liu).
https://doi.org/10.1016/j.mcat.2019.01.005
Received 26 November 2018; Received in revised form 3 January 2019; Accepted 4 January 2019
2468-8231/ © 2019 Elsevier B.V. All rights reserved.

S.-C.A. Lin et al.
Molecular Catalysis 466 (2019) 46–51
2.5. Crystallography
Crystal suitable for X-ray determination was obtained for 1·(DMF)2.
Cell parameters were determined either by a Siemens SMART CCD
diffractometer. The structure was solved using the SHELXS-97 program
[
6] and refined using the SHELXL-97 program [7] by full-matrix least-
squares on F2 values. Crystal data of 1(DMF)
43Cl Ru , Mw = 1276.84, Orthorhombic, space group
P2(1)2(1)2(1); a = 14.6196(3) Å, b = 14.9766(3) Å, c = 21.6427(4) Å,
2
:
C
45
H
F
2 12
N
7
O
P
2 2
2
3
α
=
90°,
β
=
90°,
γ
=
90°; V = 4738.71(16) Å ; Z = 4;
F(000) = 2552; Crystal size:
.20 × 0.15 × 0.10 mm ; reflections collected: 25,765; independent
−
3
ρ
calcd. = 1.790 Mg m
;
Scheme 1. Pathways for reduction of nitrobenzene.
3
0
reflections: 10,133 [R(int) = 0.0541]; θ range 2.71–27.50°; goodness-
of-fit on 1.012; Final indices [I > 2sigma(I)]R1 = 0.0461,
wR2 = 0.0862; R indices (all data) R1 = 0.0674, wR2 = 0.0977. CCDC
,864,735. These data can be obtained free of charge via www.ccdc.
F
2
R
1
cam.ac.uk/data_request/cif, or by emailing data_request@ccdc.cam.ac.
uk, or by contacting The Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336,033.
Fig. 1. Structures of dipalladium complex and ligand L.
3. Results and discussion
3.1. Preparation and characterization of complex 1
The ligand L used for preparation of the diruthenium complex was
prepared by the previously reported method [5a]. Complexation of L
6
6 6 2 2
with an equimolar amount of [Ru(η -C H )Cl ] in the presence po-
tassium hexafluorophosphate in a refluxing acetonitrile solution gave
Scheme 2. Preparation of complex 1.
the desired di-ruthenium complex as dark purple solids in 88% yield
1
(
Scheme 2). H NMR spectrum of 1 reveals that the anthyridine protons
complex. The crude product was washed with water and then dried
under vacuum to give 1 as dark purple solid (192.2 mg, 88%). H NMR
of the ligand L are split into two set of doublets, clearly suggesting the
1
symmetrical nature of the proposed structure. It is noticed that the
6
(
2
400 MHz, CD
3
CN) δ = 9.57 (d, J = 5.2 Hz, 2 H), 8.68 (d, J = 8.1 Hz,
signal for the η -benzene ring protons shows as a singlet at δ 6.19 with
H), 8.62 (d, J = 9.0 Hz, 2 H), 8.51 (d, J = 9.0 Hz, 2 H), 8.36 (t, J
3
the integration of 12H at room temperature. ESI-HRMS of 1 in CH CN
=
7.8 Hz, 2 H), 7.94 (t, J = 6.5 Hz, 2 H), 7.86 (m, 3 H), 7.66 (m, 2 H),
matrix shows a peak at m/z = 420.4943, which is matched with the
1
3
2+
6.19 (s, 12 H); C NMR (100 MHz, CD
3
CN) δ = 165.2, 158.5, 157.7,
6
theoretical value of 420.4932 of [1 – 2(PF )] , supporting a dinuclear
156.8, 155.9, 142.2, 141.3, 133.7, 132.2, 131.8, 130.5, 130.4, 128.3,
nature of the complex.
3
−1
−1
1
24.4, 121.7, 88.7 ppm. UV–vis: λmax (ε, 10 M cm ) : 505 (1.7),
In addition to the spectroscopic characterization, the detail struc-
tural formulation of 1 was confirmed by a single-crystal X-ray study.
ORTEP plot (Fig. 2) of cationic portion of 1 shows a pair of ruthenium
atoms surrounded by L. Each metal ion is in an octahedral environment,
as expected, with the coordination sphere formed by the bipyridine
415 (8.2), 377 (7.8), 294 (10.8), 252 (12.6), 227 (12.7) nm; HRMS (ESI-
2
+
TOF): m/z [M-2·(PF
6
)]
2 5 2
calcd. for C39H29Cl N Ru , 420.4943; found
4
20.4932.
.3. Catalytic-reduction of nitroarenes
mixture of nitro compound (0.6 mmol) and complex
6
moiety of L, a η -benzene and a chloride. It is noticed that two benzene
2
rings are opposite along the anthyridine rings, avoiding the steric in-
teraction. Selected bond distances and bond angles are collected in
A
1
−
3
(
1.5 × 10
mmol) in ethanol (2 mL) was placed in a reaction tube.
The reaction vessel was flushed with nitrogen gas and a solution of
·H O (1.2 mmol) in ethanol was added. The mixture was stirred at
0 °C for a period of time as indicated in the tables. After the reaction,
N
2
H
4
2
8
ethanol was removed under reduced pressure. The residue was ana-
lyzed by NMR spectroscopy. For the purification, chromatography on
silica gels provided the desired compound in the pure form. The spec-
tral data of the organic products are essentially identical to those re-
ported ones.
2.4. Kinetic studies
−3
A mixture of nitro compound (0.6 mmol) and complex (1.5 × 10
mmol) in a reaction tube with a stirring bar. The reaction vessel was
flushed with nitrogen gas and a solution of N ·H O (1.2 mmol) was
added. The mixture was stirred at 80 °C for a period of time (from 3 to
h). At appropriate time intervals, 0.1 ml aliquots were removed using
2
H
4
2
7
a syringe and quickly passed through Celite to remove the metal com-
Fig. 2. ORTEP plot of cationic portion of 1 (30% probability level; label of some
aromatic carbons are omitted for clarity).
plexes with elution of diethylether. The filtrate was then concentrated
1
under reduced pressure and analyzed by H NMR spectroscopy.
47

S.-C.A. Lin et al.
Molecular Catalysis 466 (2019) 46–51
Table 1
In order to understand more about the pathway of the catalytic
process, we set up few experiments to study the nature of the catalyst
and possible intermediates operated in the catalysis. First, the kinetics
of the resultant reaction profile for the reduction of p-nitrotoulene
catalyzed by 1 was performed (Fig. 4). It appears that there is a very
short induction period for this catalytic system. p-Tolylhydroxylamine,
a key intermidate, is generated at the beginning of the catalysis, and it
remains as a constant concentration in the first two hours. After that,
this species decreases slowly and eventually converted into the product
at the end the reaction. The amount of p-toluidine cumulates gradually
from the beginning and reaches to a full conversion at the end. It is
noticed that p,p'-azoxytoluene appears after 1.5 h and remains as a very
low concentration along the reaction and diminishes finally. From these
observations, it seems that the reduction of p-nitrotoluene catalyzed by
selected bond distances (Å) and angles (deg).
Ru(1)-N(1)
Ru(1)-N(2)
Ru(1)-Cl(1)
N(2)-Ru(1)-N(1)
Cl(1)-Ru(1)-N(2)
Cl(1)-Ru(1)-N(1)
2.075(4)
2.137(4)
2.383(1)
77.5(2)
88.2(1)
83.6(1)
Ru(2)-N(5)
Ru(2)-N(4)
Ru(2)-Cl(2)
N(4)-Ru(2)-N(5)
Cl(2)-Ru(2)-N(4)
Cl(2)-Ru(2)-N(5)
2.068(4)
2.120(4)
2.390(1)
77.5(2)
88.1(1)
80.1(1)
1
proceeds via pathway A (Scheme 1). However, we are not able to
exclude the possibility of pathway B due to the appearance of trace p,p'-
azoxytoluene. Thus, we carried out a series of testing to find out the
reactivity of catalytic system toward various possible intermediates
under the optimized conditions.
Fig. 3. Side view of cationic portion of 1 (30% probability level).
The product distribution under the optimal catalytic conditions was
examined with p-nitrosotoluene as the substrate (Table 3, entry a).
Within 2 h, this substrate was fully converted into a mixture of p,p'-
azoxytoluene, toluidine, p,p’-azotoluene and 1,2-di-p-tolylhydrazine
with the major portion (61%) of azoxytoluene. For a longer period
(20 h), the final reduction product (toluidine) did not increase much,
whereas the amount of p,p’-azotoluene, which is the reduction product
of azoxytoluene, increased slightly. On starting with p,p'-azoxytoluene,
minor portion of reduction products were obtained, while more than
Table 1. All Ru-N bond lengths are quite similar, at 2.07–2.13 Å, which
are in agreement with those of the reported bipyridine ruthenium
complexes. Both angles of N(1)-Ru(1)-N(2) [77.5(2)°] and N(4)-Ru(2)-N
(
5) [77.5(2)°], deviated from 90°, are comparable to those of [(bipyr-
idine)RuCl(L)] presumably due to the geometrical constrain of donors.
The average Ru-C(benzene) distances (2.183(5) Å) are in the normal range
6
…
of η coordination modes. The distance of Ru(1) ·Ru(2) is 5.4170(5) Å,
indicating that two metal centers are far away from each other. No
significant discrepancies in other bond lengths and angles are noticed in
complex 1. An interesting observation is the ligand itself is not seated in
a plane of this crystal structure, as illustrated in Fig. 3. The chelating
ring with Ru(2) is twisted and the metal center is above the coordina-
tion plane. The torsional angles of N(4)-C(16)-C(17)-N(5) is -17.1°,
which is larger than that of N(1)-C(5)-C(6)-N(2) [- 4.5°]. Presumably,
this is due to the steric congestion caused by the coordinating ligands.
75% of substrate remained un-reacted (Table 3, entry b), indicating that
the activity of 1 on reduction of azoxybenzenes is poor. Meanwhile,
treatment of azotoluene under the catalytic conditions provided the
1,2-di-p-tolylhydrazine as the major reduction product accompanied
with trace amount of toluidine (Table 3, entry c). Similarly, the dir-
uthenium complex 1 showed a poor activity on the conversion of 1,2-di-
p-tolylhydrazine into the desired product toluidine (Table 3, entry d).
On the contrary, on using p-hydroxyaminotoluene as the substrate,
complex 1 granted a full conversion of the substrate to p-toluidine
3.2. Reduction of nitro-compounds
(
Table 3, entry e). Unlike nitrosotoluene as the substrate (entry a), the
It is well-documented that transition metal complexes are frequently
reduction of p-hydroxyaminotoluene under the optimal conditions did
not give any accumulation of p,p'-azoxytoluene, a condensation of ni-
trosotoluene with p-hydroxyaminotoluene, indicating that the hydro-
xylamine is directly reduced to the final product [10–13].
Some generalizations can be made from the above control reactions.
2 4
The catalytic process by complex 1 with N H in the reduction of ni-
troarene undergoes preferentially via the direct reduction of hydro-
xyaminoarene. The catalytic system has poor activity on reduction of
azoxy- and azo-arenes as compared to hydroxyaminoarenes, i.e. the
activity of complex 1 is quite poor on the reductive cleavage of NeN
bonds. Furthermore, once the nitrosoarene species appears in the re-
action medium, the formation of azoxy species cannot be avoided, i.e.
the rate of condensation of hydroxyaminoarene with nitrosoarene is
comparable to the reduction steps. To our surprise, the catalyst showed
a poor activity toward nitrosotoluene, which is generally known to be
the key intermediate of this reduction (Scheme 1).
used as catalysts for reduction of nitroarenes [8]. Accordingly, the ac-
tivity of the diruthenium complex 1 on reduction of nitroarenes with
hydrazine as the reductant was investigated [9]. For the optimization,
we chose the reduction of p-nitrotoluene as the model reaction with 1 as
the pre-catalyst. A control experiment clearly showed the essentiality of
a metal complex playing the role of catalysis (Table 2, entry 1). After
several trials, it appears that a full conversion was achieved by carrying
out the reaction using excess of hydrazine in the presence of 0.25 mol%
of catalyst in an ethanol solution at 80 °C (Table 2, entry 2). Various
solvents were screened and alcohols were found to be the best choice
(
Table 2, entries 2–8). Other solvents gave poor results. It is noticed that
N-hydroxy-p-toluidine and 4,4′-azoxytoluene were produced in an
acetonitrile solution (Table 2, entry 4). By lowering the ratio of hy-
drazine/nitrotoluene to 2:1, we were still able to obtain the full re-
duction product quantitatively even with a shorter period of time (7 h)
(
Table 2, entry 9). By comparison to the outcome of entry 11, the
catalyst loading of 0.25 mol% appeared to be the best choice. By run-
ning the reaction under O atmosphere, the product yield diminished by
For further understanding the reduction pathway whether involving
nitrosotoluene intermediate, we carried out another experiment by
adding nitroso compound into the catalytic reaction. After a standard
reaction of the reduction of nitrotoluene proceeded for 1 h, a quantity
of 30 mol% p-nitrosotoluene was syringed into the reaction mixture.
2
about 50%. Furthermore, the catalysis proceeded smoothly even with
the addition of mercury during the reaction, indicating that the cata-
lytic system is a homogeneous metal-catalyzed reaction. Two ruthe-
1
This reaction mixture was then monitored by H NMR. In this reaction
6
6
nium complexes, [Ru(η -benzene)Cl]
2), were also tested for comparison. However, these two complexes
showed poor activity in this catalytic reduction. (Table 2, entries
4–16). By carrying out the reaction in air, the aniline product was
2
and [(bipy)Ru(η -benzene)Cl]Cl
medium, a mixture of nitrotoluene, p-hydroxyaminotoluene, p,p'-azox-
ytoluene, p,p'-azotoluene and toluidine was identified, i.e. production of
azoxy species appeared immediately. For a longer period, both p,p'-
azoxytoluene and p,p'-azotoluene still remained, which is consistent
with the results of experiments in Table 3 (entries a–d). Clearly, the
appearance of nitroso intermediate in the catalytic process would
(
1
obtained in poor yields accompanied with other intermediates even
with the use of excess of hydrazine (Table 2, entries 17–18).
48

S.-C.A. Lin et al.
Molecular Catalysis 466 (2019) 46–51
Table 2
Optimization for the reduction of p-nitrotoluene.a
entry
cat. (mol%)
2
N H
4
solvent
conv.b
yield (%)b
(
mmol)
I
II
III
1
2
3
4
5
6
7
8
–
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.2
0.6
1.2
1.2
1.8
1.2
EtOH
EtOH
MeOH
0
0
0
0
trace
0
0
0
trace
0
trace
7
0
0
trace
0
0
0
6
0
0
trace
0
0
0
6
trace
0
0
0
1 (0.25 mol%)
1 (0.25 mol%)
1 (0.25 mol%)
1 (0.25 mol%)
1 (0.25 mol%)
1 (0.25 mol%)
1 (0.25 mol%)
1 (0.25 mol%)
1 (0.25 mol%)
1 (0.13 mol%)
1 (0.25 mol%)
1 (0.25 mol%)
100%
100%
46%
9%
50%
12%
32%
100%
53%
97%
52%
100%
6%
100
99
35
7
48
6
22
99
48
83
49
98
trace
CH
3
CN
DCE
toluene
dioxane
H
2
O
c
9
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
1
1
1
1
1
0c
1
2
3
4
c,d
e
[Ru(C
0.25 mol%)
2 (0.5 mol%)
6
H
6
)Cl]
2
(
c
c
f
1
1
1
1
5
6
7
8
1.2
1.2
1.2
3.0
EtOH
EtOH
EtOH
EtOH
6%
5%
56%
96%
trace
trace
18%
52%
0
0
6%
9%
trace
trace
30%
35%
2 (0.25 mol%)
1 (0.25 mol%)
1 (0.25 mol%)
f
a
Reaction conditions: p-nitrotoluene (0.6 mmol), N
2
H
4
·H
Determined by H NMR with 1,3,5-trimethoxybenzene as the internal standard.
h.
(1 atm).
2
O and catalyst in solvent (1 mL) at 80 °C for 10 h, unless noted.
b
c
d
e
f
1
7
O
2
Addition of mercury (0.24 g, 1.2 mmol) after the reaction proceeded.
In air.
azoxytoluene (56%) and toluidine (38%) (Scheme 3B). Additionally,
when p-hydroxyaminotoluene was treated with hydrazine in the ab-
sence of 1, only minor portion of the substrate was transformed into
p,p'-azoxytoluene (13%) and toluidine (6%) (Scheme 3C). Apparently,
hydrazine is able to react with hydroxylamine and nitroso compounds
directly without the assistance of metal catalysts, which may rationalize
the presence of trace p,p'-azoxytoluene during the reduction of ni-
troarenes under this catalytic study, i.e. the species is generated from
the direct reaction of p-hydroxyaminotoluene with hydrazine.
At this point, we believe that a direct route from nitrobenzene to
phenylhydroxylamine contributes to the production of aniline under
the complex 1 catalyzed condition, i.e. a pathway without going
through the nitosoarene intermediates (Scheme 4). Such a reaction
route has been proposed in few heterogeneous catalysis, but not
homogeneous systems. To the best of our knowledge, no direct forma-
tion of hydroxylamines from nitroarenes catalyzed by a metal complex
homogeneously has been described [10–12].
Fig. 4. Reaction profile of reduction of p-nitrotoulene (0.6 mmol) catalyzed by
(0.25 mol%) in EtOH at 80 °C with N ·H O (1.2 mmol) as reductant. p-
1
2
H
4
2
nitrotoluene (▲); p-toluidine (●); p-hydroxyaminotoluene
( ); p,p'-azox-
ytoluene (*).
readily undergo the condensation with p-hydroxyaminotoluene to yield
p,p'-azoxytoluene, thus inhibiting the reduction leading to anilines.
These observations clearly indicate that the reduction pathway by
catalyst 1 does not go through nitroso species. In other word, under this
catalytic condition, nitroarenes is directly reduced to hydro-
xyaminoarenes, which undergoes a further reduction step to the final
product [12].
In order to rationalize the appearance of trace amount of p,p'-
azoxytoluene during the reduction of nitrotoluene (Fig. 4), control ex-
periments using nitrotoluene and related intermediates as substrates
without complex 1 were performed, and results are listed in Scheme 3.
3.3. Mononuclear versus dinuclear ruthenium complexes
For comparison, the catalytic activities of the mono-nuclear com-
6
plex [(bipy)RuCl(η -C H )]Cl (2) under the above optimized reduction
6 6
conditions was examined. Table 4 summarizes results of a series of
reactions catalyzed by 2. It appears that the catalytic activity of 2 in
reduction of nitroarene is much less active than that of 1, even for 24 h
only 14% of conversion with 12% of amine product (Table 4, entry a).
On starting with p-nitrosotoluene (entry b), p,p'-azoxytoluene and to-
luidine were the observed products. However, it is noticed that the ratio
of these two products is similar to that of the control reaction men-
tioned above (Scheme 3B), revealing that complex 2 might not have
any catalytic activity in this reaction. As for p-hydroxyaminotoluene,
complex 2 did show catalytic activity on its reduction. However, the
2 4
It appears that the nitro compound cannot be reduced by N H in
ethanol at 80 °C (Scheme 3A), implying the necessity of catalyst.
However, when nitrosotoluene was treated with hydrazine, even in the
absence of complex 1, the substrate was fully converted into p,p'-
49

S.-C.A. Lin et al.
Molecular Catalysis 466 (2019) 46–51
Table 3
Catalytic activity of 1 toward various intermediates.a
entry
substrateb
time
conv.
a
b
2 h
100%
100%
22%
61%
49%
———
———
trace
17%
11%
10%
trace
trace
trace
trace
23%
27%
9%
2
0 h
7 h
2
4 h
25%
14%
c
7 h
7 h
59%
10%
———
———
———
55%
trace
6%
d
trace
———
e
7h
100%
0%
0%
0%
100%
a
Substrate (0.6 mmol), N
Ar = p-MeC H -.
6 4
H
2 4
·H
2
O (1.2 mmol) and 1 (0.25 mol%) in EtOH at 80 °C.
b
1
can be viewed as a linkage of two molecules of 2. However, the
catalytic activity of 1 shows a higher efficiency relative to its mono-
nuclear analogues, suggesting a possible cooperative effect between
two metal centers within 1, which is absent in a mononuclear system.
Thus, the cooperativity index (α) for a polymetallic catalyst with n
centers (Eq. (1)), previously proposed by Jones and James [14], was
used in this work to estimate the degree of cooperativity of complex 1.
Since the complex investigated here is a dinuclear species, the equation
is simplified as shown in Eq. (2). In this equation, A
activity of 1; A is the predicted total activity of the dinuclear complex,
which is summation of the measured activity of two molecules of
complex 2; and Aave = A /2. Base on the conversions of the reaction
o
stands for the
p
p
catalyzed by 1 and 2 (Table 2, entries 9 and 16), the cooperativity index
(α) for 1 is estimated to be 19, which is far larger than unity. This result
implies that complex 1 readily benefits from the cooperative effect
during the reaction course of this reduction. It would be interesting to
know how the cooperative effect operates. However, we have at-
tempted to reveal the possible intermediates of metal complexes, but
not succeed, even by monitoring a sample with NMR or mass spectro-
meters.
2 4
Scheme 3. Reductions of nitrotoluene by N H
without metal catalyst 1a.
Scheme 4. Reduction pathway of nitroarenes catalyzed by 1.
Ao:activity of the polymetallic catalyst with n centers
Ap:predicted total activity of the polymetallic catalyst
Ao − Ap
α=
α=
Aave
production of toluidine only reached 60% even for 24 h period (Table 4,
entry c). Not surprisingly, a poor activity of 2 toward azoxytoluene was
observed (Table 4, entry d). All of these investigations summarize that
the catalytic activity of the dinuclear complex 1 is superior to that of
the mono-nuclear complex 2.
Diruthenium complex 1 appears to exert an excellent catalytic ac-
tivity with a distinct pathway in this reduction and shows a higher ef-
ficiency relative to its mononuclear analogues. Structure-wise, complex
Aave: A /n
p
(1)
(2)
A
(complex 1) − 2x [A(complex 2)
]
[
2x(A(complex 2))]/2
We were expected to isolate some catalytic intermediates coordinating
to metal ions due to the suitable distance between Ru ions in 1, but in
vain. Nevertheless, the experimental data do show a possible synergistic
Table 4
a
Catalytic activity of 2 towards p-nitrotoluene and various intermediates.
entry
substrateb
time
conv.
a
Ar-NO
Ar-NO
2
7 h
6%
< 3%
ND
ND
———
———
NDc
< 3%
55%
< 5%
18%
ND
ND
ND
ND
8%
< 5%
12%
38%
30%
64%
2
4 h
14%
100%
32%
90%
b
c
2 h
7 h
2
4 h
d
24 h
20%
———
———
6%
12%
a
2 4 2
Substrate (0.6 mmol), N H ·H O (1.2 mmol) and 2 (0.5 mol%) in EtOH at 80 °C.
b
c
Ar = p-MeC
6 4
H -.
ND = not detected.
50

S.-C.A. Lin et al.
Molecular Catalysis 466 (2019) 46–51
Table 5
pathway of this reduction catalyzed by 1 involves the direct reduction
of nitro compound to form hydroxylaminearene, which is subsequently
reduced to the final product, without going through nitrosoarene as the
intermedaite. Such a direct reduction is quite rare in a homogeneous
system. In addition, various substrates with other reducible function-
ality such as halo functionality were unaffected during the reduction,
but carbonyl functionality may lead to the formation hydrazones during
the reductions. However, the mechanism in detail on the direct for-
mation of the hydroxyaminoarenes by the diruthenium complex re-
mains unclear in current stage.
Reduction of substituted nitroarenes catalyzed by 1.a
entry substrate
N
2
H
4
·H
2
O
time product (yield)
1
2
3
4
5
6
7
p-MeC
p-ClC
p-BrC
p-IC
6
H
4
NO
NO
NO
NO
NO
2
1.2 mmol
1.2 mmol
1.2 mmol
1.2 mmol
1.2 mmol
2.4 mmol
1.8 mmol
7 h
p-MeC
6
H
4
NH
2
(95%)
6
H
4
2
14 h p-ClC
6
H
4
NH
2
(93%)
(94%)
6
H
4
2
7 h
10 h
7 h
p-BrC
p-IC
6
H
4
NH
2
6
H
4
2
6
H
4
2
NH (95%)
C
6
H
5
2
C
6
H
4
NH
2
(91%)
NH
C(=N-NH
73%)
p-HC(=N-NH
30 h p-MeOOCC
30 h p-HOOCC
30 h p-H NC
30 h p-HOC
7 h p-H NCOC
b
b
p-CH
p-CH
3
OC
6
H
4
NO
NO
2
14 h p-MeOC
36 h p-CH
6
H
4
2
(92%)
)C
3
COC
6
H
4
2
3
2
6
H
4
NH
2
(
b
b
8
9
1
1
1
1
p-HCOC
p-MeOOCC
p-HOOCC
p-H NC
p-HOC
p-H NCOC
6
H
4
NO
NO
NO
NO
NO
2
2.4 mmol
2.4 mmol
3.6 mmol
3.6 mmol
3.6 mmol
1.2 mmol
24 h
2
4
)C
NH
6
H
4
NH
2
(72%)
Acknowledgments
6
H
4
2
6
H
2
(42%)
b
b
b
0
1
2
6
H
4
2
6
H
4
NH
NH (61%)
NH
2
(96%)
c
2
6
H
4
2
2
6
H
4
2
We thank the Ministry of Science and Technology, Taiwan for fi-
nancial support (MOST103-2113-M-002-002-MY3).
c
6
H
4
2
6
H
4
2
(40%)
NH
(59%)d
3.
2
6
H
4
NO
2
2
6
H
4
2
a
Appendix A. Supplementary data
Reaction conditions: nitroarene (0.6 mmol), complex 1 (0.25 mmol%) and
·H O (1.2 mmol) in ethanol (1 mL) at 80 °C; isolated yield.
Complex 1 (0.5 mmol%).
No isolation; NMR yields.
2 6 4
Accompanied with p- H NCOC H N(OH)H (40%).
N
2
H
4
2
b
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2019.01.005.
c
d
References
interaction to promote the reduction.
[
[
1] A.M. Tafesh, J. Weiguny, A review of the selective catalytic reduction of aromatic nitro
compounds into aromatic amines, isocyanates, carbamates, and ureas using CO, Chem.
Rev. 96 (1996) 2035–2052.
2] (a) R.V. Jagadeesh, G. Wienhöfer, F.A. Westerhaus, A.-E. Surkus, H. Junge, K. Junge,
M. Beller, A convenient and general ruthenium‐catalyzed transfer hydrogenation of nitro‐
and azobenzenes, Chem. Eur. J. 17 (2011) 14375–14379;
3
.4. Substrate scope of the reduction
With the above optimized conditions, a study to explore the re-
duction of various nitroarenes catalyzed by 1 was investigated
Table 5). Most aromatic nitro compounds were found to undergo the
(
b) S.P. Annen, H. Grützmacher, Nitrosobenzene as a hydrogen acceptor in rhodium
(
catalyzed dehydrogenation reactions of alcohols: synthesis of aldehydes and azox-
ybenzenes, Dalton Trans. 41 (2012) 14137–14145;
reduction to afford the respective anilines in good to excellent yields. In
particular, the halo-substituted nitrobenzenes were reduced to yield the
corresponding halo-anilines (Table 5, entries 2–4). In most cases, the
reduction of halo-substituted nitrobenzenes always accompanies with
the reduction of halogen, but this catalytic system offers an efficient
way in this regard. In order to reach good yields, it is required to use
excess of hydrazine for the reduction of acetyl or formyl substituted
nitrobenzene, because of the consumption of hydrazine in the forma-
tion of hydrazone functionality [15]. Reduction of methyl p-ni-
trobenzoate provided the desired product methyl p-aminobenzoate in
(
c) G. Wienhoefer, M. Baseda-Krueger, C. Ziebart, F.A. Westerhaus, W. Baumann,
R. Jackstell, K. Junge, M. Beller, Hydrogenation of nitroarenes using defined iron-phos-
phine catalysts, Chem. Commun. 49 (2013) 9089–9091;
(
d) S. Hohloch, L. Suntrup, B. Sarkar, Arene–ruthenium(II) and −iridium(III) complexes
with “click”-based pyridyl-triazoles, bis-triazoles, and chelating abnormal carbenes: ap-
plications in catalytic transfer hydrogenation of nitrobenzene, Organometallics 32 (2013)
7376–7385;
(
e) S. Chen, G. Lu, C. Cai, Iridium-catalyzed transfer hydrogenation of nitroarenes to
anilines, New J. Chem. 39 (2015) 5360–5365 and references therein.
[3] S.-T. Yang, P. Shen, B.-S. Liao, Y.-H. Liu, S.-M. Peng, S.-T. Liu, Catalytic reduction of
nitroarenes by dipalladium complexes: synergistic effect, Organometallics 36 (2017)
3
110–3116.
4] (a) Z. Bian, S. Das, M.H. Wai, P. Hongmanorom, S. Kawi, A review on bimetallic nickel-
based batalysts for CO reforming of methane, ChemPhysChem 18 (2017) 3117–3134;
b) Y.-B. Huang, J. Liang, X.-S. Wang, R. Cao, Multifunctional metal–organic framework
catalysts: synergistic catalysis and tandem reactions, Chem. Soc. Rev. 46 (2017) 126–157;
c) J. Fu, X. Huo, B. Li, W. Zhang, Cooperative bimetallic catalysis in asymmetric allylic
substitution, Org. Biomol. Chem. 15 (2017) 9747–9759;
d) J. Serrano-Plana, I. Garcia-Bosch, A. Company, M. Costas, Structural and reactivity
[
4
2% yield with unidentified products due to the reaction of hydrazine
2
(
with the ester functionality (Table 5, entry 9). However, the use of p-
nitrobenzoic acid as substrate gave p-aminobenzoic acid in 96% yield
(
(
Table 5, entry 10). Reductions of p-nitro-aniline and p-nitrophenol
(
proceeded similarly, but we had troublesome on the purification of the
products (Table 5, entries 11–12). From these results, it appears that the
developed procedure is applicable to nitroarenes containing both
electron withdrawing and electron donating substituents. Substrates
with electron withdrawing substituents require a longer time and excess
of hydrazine. However, we found that the reaction of p-
models for copper oxygenases: cooperative effects and novel reactivities, Acc. Chem. Res.
48 (2015) 2397–2406 and references therein.
5] (a) Y.-H. Lo, Y.-H. Liu, S.-M. Peng, S.-T. Liu, Ruthenium complexes with an anthyridine-
based ligand. Synthesis, characterization and catalytic activity, J. Chin. Chem. Soc. 60
[
(
2013) 839–845;
(b) D.A. Freedman, J.K. Evju, M.K. Pomije, K.R. Mann, Convenient synthesis of tris-
heteroleptic ruthenium(II) polypyridyl complexes, Inorg. Chem. 40 (2001) 5711–5715.
6] G.M. Sheldrick, SHELXS-97, Acta Crystallogr. Sect. A Found. Crystallogr. 46 (1990)
467–473.
7] G.M. Sheldrick, SHELXL-97, University of Göttingen, Göttingen, Germany, 1997.
8] N. Ono, The Nitro Group in Organic Synthesis, Wiley-VCH, New York, 2001.
9] Z. Zhao, H. Yang, Y. Li, X. Guo, Cobalt-modified molybdenum carbide as an efficient
catalyst for chemoselective reduction of aromatic nitro compounds, Green Chem. 16
[
H
2
NCOC
aniline in 59% accompanied with the hydroxyamine intermediate
Table 5, entry 13), indicating that the reduction of hydroxylamine is
6 4 2
H NO under the standard condition gave the corresponding
[
[
[
(
presumably to be the limiting step.
(
2014) 1274–1281.
However, when using 1-nitrocyclexene and 2-phenyl-1-nitroethane,
the corresponding reduced amine product was not obtained under the
catalytic conditions. These reactions provided a mixture of complex
substance. Apparently, this catalysis is only applicable to the aromatic
nitro compounds.
[
10] E.A. Gelder, S.D. Jackson, C.M. Lok, The hydrogenation of nitrobenzene to aniline: a new
mechanism, Chem. Commun. (2005) 522–524.
[11] A. Corma, P. Concepción, P. Serna, A different reaction pathway for the reduction of
aromatic nitro compounds on gold catalysts, Angew. Chem. Int. Ed. 46 (2007)
7266–7269.
[12] P.L. Gkizis, M. Stratakis, I.N. Lykakis, Catalytic activation of hydrazine hydrate by gold
nanoparticles: chemoselective reduction of nitro compounds into amines, Catal. Commun.
3
6 (2013) 48–51.
4
. Summary
[
[
[
13] S. Wu, G. Wen, X. Liu, B. Zhong, D.S. Su, Model molecules with oxygenated groups
catalyze the reduction of nitrobenzene: insight into carbocatalysis, ChemCatChem 6
(
2014) 1558–1561.
In this work, a diruthenium complex 1 hosted by an anthyridine-
14] N.D. Jones, B.R. James, Homo- and heterobimetallic precursor catalysts for the Heck
reaction, and a proposal for a general catalytic cooperativity index, Adv. Synth. Catal. 344
based ligand has been prepared and structurally characterized. Notably,
this complex appears to be an excellent pre-catalyst for reduction of
nitroarenes with the use of hydrazine as the reducing agent. From the
analysis of the activity of 1 towards various intermediates, the reaction
(
2002) 1126–1134.
15] U. Sharma, P. Kumar, N. Kumar, V. Kumar, B. Singh, Highly chemo- and regioselective
reduction of aromatic nitro compounds catalyzed by recyclable copper(II) as well as
cobalt(II) phthalocyanines, Adv. Synth. Catal. 352 (2002) 1834–1840.
51