Selective reduction of nitroaromatics to azoxy compounds on supported Ag-Cu alloy nanoparticles through visible light irradiation

Article abstract of DOI:10.1039/c5gc01726b

The selective hydrogenation of aromatic nitrocompounds to their corresponding azoxy compounds is challenging in organic synthesis, which are typically performed under harsh reaction conditions. The core issue involved in this reduction is to finely control the product selectivity. Herein, we report an efficient photocatalytic process using supported silver-copper alloy nanoparticles (Ag-Cu alloy NPs) to selectively transform nitrobenzene to azoxybenzene by visible light irradiation under green mild reaction conditions. Ag-Cu alloy NPs can absorb visible light, causing excited hot-electrons due to the localized surface plasmon resonance (LSPR) effect, and the excited electrons can activate the reactant molecule adsorbed on the NP surface to induce a reaction. The photocatalytic performance was affected by the Ag-Cu ratio, and the catalyst with an Ag-Cu molar ratio of 4-1 exhibited the optimal performance. Tuning the wavelength of incident light manipulated the product selectivity between azoxybenzene and aniline. Compared to pure Ag NPs, the alloying of Cu was found to be responsible for the product selectivity shifting from azobenzene to azoxybenzene. The reaction pathway was investigated to explain the selectivity difference and thus a tentative reaction pathway was proposed.

Full text of DOI:10.1039/c5gc01726b

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: Z. Liu, Y. Huang,
Q. Xiao and H. Zhu, Green Chem., 2015, DOI: 10.1039/C5GC01726B.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/greenchem

Please do not adjust margins
Page 1 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C5GC01726B
Journal Name
ARTICLE
Selective reduction of nitroaromatics to azoxy compounds
on supported Ag-Cu alloy nanoparticles through visible
light irradiation
Received 00th January 20xx,
Accepted 00th January 20xx
Zhe Liu, Yiming Huang, Qi Xiao* and Huaiyong Zhu
DOI: 10.1039/x0xx00000x
The selective hydrogenation of aromatic nitrocompounds to their corresponding azoxy compounds is challenging in
organic synthesis, which were typically performed under harsh reaction conditions. The core issue involved in this
reduction is to finely control the product selectivity. Herein, we report an efficient photocatalytic process using supported
silver-copper alloy nanoparticles (Ag-Cu alloy NPs) to selectively transform nitrobenzene to azoxybenzene by visible light
irradiation under green mild reaction conditions. Ag-Cu alloy NPs can absorb visible light causing excited hot-electrons due
to the localized surface plasmon resonance (LSPR) effect, and the excited electrons can activate the reactant molecule
adsorbed on the NP surface to induce the reaction. The photocatalytic performance was affected by the ratio of Ag-Cu,
and the catalyst with Ag-Cu molar ratio of 4-1 exhibited the optimal performance. Tuning the wavelength of incident light
remarkably manipulate the product selectivity between azoxybenzene and aniline. Compared with pure Ag NPs, the
alloying of Cu was found responsible for the product selectivity shifting from azobenzene to azoxybenzene. The reaction
pathway was investigated to explain the selectivity difference and thus a tentative reaction pathway was proposed.
www.rsc.org/
through irradiation of visible light—the major component of
abundant solar spectrum. In previous work, we have found
that Au NPs supported on zirconium oxide (ZrO2) effectively
Introduction
The selective reduction of nitroaromatics is one of the most
significant reactions in organic synthesis,1-3 because it
produces important industrial intermediates such as amines,4
azo and azoxy compounds.5 Among them, azo and azoxy
compounds attract particular interest for their widely
application in the production of dyes, agrochemical, and
pharmaceuticals.6-7 Therefore, a variety of reduction processes
have been developed to catalyse this reaction selectively to
obtain the desired products.8-9 Generally, the stoichiometric
amount of reducing agents such as pressured hydrogen as well
as transition metal catalysts are demanded, which may result
in undesired by–products and the uncontrollability of product
selectivity.10-11 Grirrane et al. demonstrated that supported
gold nanoparticle (Au NP) catalyst is capable of catalysing the
reduction of nitro aromatics to the corresponding azo
compounds through a two-step process.12 However, high
reaction temperature, pressured hydrogen and oxygen gas
were required, the harsh reaction conditions not only increase
the energy consumption but also the safety risk. From the
point of view of green chemistry, it would be a promising
process to drive the reduction under mild reaction conditions
catalyse the reductive coupling of nitrobenzene to azo
compound under ambient pressure at 40°C by visible light
irradiation,13 this is mainly due to the reason that Au NPs can
strongly absorb visible light through the localized surface
plasmon resonance (LSPR) effect.14 The LSPR effect is the
collective oscillation of conduction electrons in the metal NPs
induced by the electromagnetic field of the incident light.15
Through this process, the light excited electrons of NPs can
gain the energy from light and activate reactant molecules
adsorbed on metal surface and trigger chemical reactions.16
Apart from Au NPs, silver (Ag) and copper (Cu) NPs also
exhibits strong light absorption in the visible range.17 It has
been reported that supported Ag NPs can be used as efficient
photocatalyst under ultraviolet (UV) and visible light
irradiation.18-19 Very recently, Cu NPs supported on graphene
were also been found to exhibit high photocatalytic activity for
the reductive coupling of nitroaromatics to aromatic azo-
compounds under irradiation of solar spectrum.20 However,
the main products in both of the Au and Cu NPs catalysed
reaction systems were azo compounds, and azoxy compounds
appear as the intermediates during these processes which is
not easily controllable. We thus envision the design of a high-
performance catalyst for the selective reduction of nitro
compounds to azoxy compounds under much milder and
greener reaction conditions driven by visible light irradiation.
In the present study, we reported the selective reduction
of nitroaromatics to azoxy compounds on Ag-Cu alloy NPs
School of Chemistry, Physics and Mechanical Engineering, Science and
Engineering Faculty, Queensland University of Technology, Brisbane, QLD 4001,
Australia.
E-mail: q2.xiao@qut.edu.au
Fax: +61 7 3138 1804; Tel: +61 7 3138 1581.
† Electronic Supplementary Information (ESI) available: See
DOI: 10.1039/x0xx00000x
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 2 of 22
ARTICLE
Journal Name
under the irradiation of visible light. We successfully fabricated due to the CuNPs cannot be stabilized on the ZrO2 surface, and
View Article Online
DOI: 10.1039/C5GC01726B
Ag-Cu alloy NPs with air stability by adjusting the Ag-Cu ratio. the copper oxides on the surface strongly supressed the
The Ag-Cu alloy NPs can efficiently drive the reduction of nitro catalytic activity. Interestingly, when using Ag-Cu alloy NPs as
aromatics and selectively yield azoxy compounds at moderate the catalyst, the main product selectivity was switched from
reaction temperature with the irradiation of visible light. From azobenzene to azoxybenzene. The catalytic performance of
a green chemistry point of view, it would be a highly attractive alloy NPs was enhanced with higher Ag ratio, and the catalyst
and challenging goal to develop active, easily separable and with Ag-Cu molar ratio of 4-1 exhibited the optimal
reusable catalyst systems that can perform such desirable performance (entry 3). Thus, the Ag-Cu (4-1) alloy @ZrO2 was
syntheses of aromatic azoxy compounds under more selected as the optimal catalyst for the following experiments.
controlled, simplified, and greener conditions.
Further increasing the amount of Ag or Cu may form large
amount of metal oxide species on the surface which
significantly supressed the catalytic performance. The different
Ag-Cu ratio did not affect the reaction selectivity. The general
applicability of Ag-Cu (4-1) alloy @ZrO2 for the photocatalytic
reduction of nitroaromatics with various functional groups was
also investigated, the product selectivity can be maintained
mainly at azoxy compounds and no azo product was detected
(Table S1, ESI). These results confirmed that alloying Cu with
Ag to form Ag-Cu alloy NPs can change the product selectivity
for the reduction of nitroaromatic effectively at moderate
conditions under visible light irradiation. Compared with some
literature reported reaction conditions using Cu, Ag or Au
catalysts for the reduction of nitrobenzene (Table 2), it can be
seen that Ag-Cu alloy NPs can efficiently drive this reaction to
obtain good azoxybenzene yield under mild conditions through
visible light irradiation. Thus, to the best of our knowledge, this
is the first example of selective reduction of nitro aromatics to
yield azoxy compounds at moderate reaction temperature
driven by “green” solar energy.
Results and discussion
Photocatalytic performance
Ag-Cu alloy NPs supported on ZrO2 (Ag-Cu@ZrO2) catalysts
with different Ag:Cu ratio were prepared in the present study.
The total metal amount (Ag+Cu) of the catalysts was
maintained at 3 wt%. The photocatalytic reductions of
nitrobenzene over Ag-Cu alloy NPs with different ratios under
visible light irradiation were tested where isopropyl alcohol
was acting as both the solvent and reducing agent. The
reactions isolated from light while other reaction condition
remained unchanged were carried out for comparison. The
results are summarized in Table 1. Monometallic Ag@ZrO2
catalyst exhibited good catalytic activity with 90% conversion,
while the control experiment in the dark only gave 25%
conversion (entry 1). However, the main product was
azobenzene (83%), and no azoxybenzene was detected.
Cu@ZrO2 catalyst gave poor activity for this reaction with only
38% conversion under irradiation (entry 2), this is probably
Table 1. Activity test and catalyst screening for reduction of nitrobenzene.
-
O
+
N
N
N
N
hv
+
NH
+
NO
2
2
catalyst
Conv. (%) (dark)
Select. (%)
Azoxy
TON
TOF (h-1)
Entry
Ag:Cu[a]
Aniline
Azo
1
2
3
4
5
Ag@ZrO2
Cu@ZrO2
4:1
90 (25)
38 (4)
96 (16)
74 (2)
57 (2)
17
12
14
18
16
0
83
12
0
32
8
2.0
0.5
2.0
1.3
0.9
76
86
82
83
32
21
14
1:1
0
1:4
1
Reaction condition: photocatalyst 50 mg, reactant 0.5 mmol, 3 mL isopropyl alcohol (IPA) as solvent, 0.15 mmol KOH as base, 1 atm argon atmosphere,
reaction temperature 60°C, reaction time 16 h, and the light intensity was 0.8 W/cm2. The conversions and selectivity were analysed by gas chromatography
(GC). [a] Molar ratio. TON=[amount of nitrobenzene (mol) × conversion]/[amount of Ag (mol)+amount of Cu (mol)], TOF=TON/reaction time (h).
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 22
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
Table 2. Comparison of the reaction conditions and achieved conversion of metal catalysts reported in literatures for the
View Article Online
DOI: 10.1039/C5GC01726B
reduction of nitrobenzene.
Yield (%)
Entry
Catalyst
Reaction conditions
Aniline
Azobenzene
Azoxybenzene
1
2
3
4
5
6
7
8
9
Cu phthalocyanines 21
70°C, 2h
120°C, 11h
93
86
84
-
-
-
-
-
22
Cu/HCOONH4
18
Ag/TiO2
1h
-
-
0.85 wt% Au/MgAl HT 23
50°C, 3.5 h, 2 MPa H2
40°C, 5h, visible light
90°C, 5h, visible light
65°C, 8h
98
99
96
-
-
13
Au/ZrO2
-
-
Cu/graphene 20
-
-
24
5.0 % Pd/Al2O3
-
44
72
75
Pd nanoparticle 25
80°C, 24h
-
-
Ag-Cu alloy@ZrO2
60°C, 16h, visible light
-
-
sample was tested by X-ray photoelectron spectroscopy (XPS)
and results are shown in Figure 1. For comparison, the
monometallic Cu sample was also tested. It can be seen from
Figure 1a that the XPS spectra of Cu in Ag-Cu (4-1)@ZrO2 and
Cu@ZrO2 show much difference. First, two characteristic
shake-up peaks at 942.7 eV and 962.2 eV can be observed in
Cu@ZrO2 sample (blue curve), which suggested that Cu(0) was
oxidized to Cu(II) in the monometallic Cu sample. Thus, no
further catalytic test was conducted with Cu@ZrO2 sample due
to it exists mainly as CuO. In the contrast, the shake-up peak
disappeared in the Ag-Cu alloy sample (red curve), illustrating
that Cu species existed in metallic state in alloy NPs when the
Ag-Cu molar ratio was adjusted to 4-1.26 On the other hand, as
shown in Figure 1b, the binding energy peaks of Ag in alloy
sample can be located at 368.2 eV and 374.2 eV attributed to
the zero valence of Ag, thus the metallic state of Ag in
bimetallic alloy can be confirmed.27 These results suggested
that we may use alloying effect to maintain the metal
nanoparctiles’ stability.
The representative transmission electron microscopy (TEM)
images of the catalyst are shown in Figure 2. The Ag-Cu alloy
NPs are well distributed on the ZrO2 crystal surface (Figure 2a)
with a mean particle size of 7 nm (Figure 2b). The high
resolution (HR)-TEM image indicates that the crystal face of
alloy nanoparticle is predominantly (111) face and the space of
lattice fringe is 0.23 nm (Figure 2c). Figure 2d is a cross-
sectional compositional line profile analysis of the energy
dispersion X-ray (EDX) spectrum from TEM for a typical Ag-Cu
(4-1) alloy NP, showing that the NP consists of both Ag and Cu
elements distributed uniformly, which suggests that the two
metals exist as binary alloy NPs in this sample. The
composition of Ag and Cu elements in the alloy catalysts was
also analysed from the EDX spectrum (Figure 2e). The Ag/Cu
Figure 1. (a) XPS profile of Cu species in Ag-Cu (4-1)@ZrO2 (red
curve) and in Cu@ZrO2 samples (blue curve). (b) XPS profile of
Ag species in Ag-Cu (4-1)@ZrO2 sample.
Catalyst characterisation
Monometallic Cu NPs are unstable in the air, thus alloying Ag
with Cu should be an effective approach to maintain its
stabilization. To confirm the valence state of Cu species, the
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 4 of 22
ARTICLE
Journal Name
ratio (4/1) is well matched with the initial experimental design. The ZrO2 support exhibits negligible absorption VoiefwlAigrtihclte Ownilitnhe
DOI: 10.1039/C5GC01726B
This was further supported by the detailed data summarized wavelengths above 400 nm, thus the support itself does not
from both TEM and scanning electron microscopy (SEM) show any light absorption in the visible range.28 As a result, the
analysis (Table S1, Figure S1 and S2, ESI). The appearance of absorption of the photocatalyst in the visible light range is due
both metals, Ag and Cu, in the elemental mapping from TEM to the formation of Ag-Cu alloy NPs, indicating that the alloy
also confirmed the alloy character of the bimetallic Ag-Cu NPs may have the potential to utilise the irradiation energy
catalyst (Figure 2f).
delivered by the wider range of solar spectrum.
Figure 3. The normalized DR-UV-vis spectra of the Ag-Cu (4-1)
@ZrO2 catalyst, as well as its comparison with monometallic
Ag@ZrO2 and ZrO2 support.
The X-ray diffraction (XRD) patterns of the Ag@ZrO2 and
Ag-Cu alloy@ZrO2 photocatalysts are shown in Figure 4. It can
be seen that the diffraction peaks of all the samples can be
indexed to a monoclinic structure of the ZrO2 crystal.29 The
reflection peaks of Ag and alloy cannot be identified due to the
low metal content (3 wt%). As a result, the loading of alloy NPs
have no impact on the crystal structure of ZrO2 supporting
material.
Figure 2. (a) TEM image of Ag-Cu (4-1)@ZrO2 sample. (b)
Particle size distribution of Ag-Cu (4-1)@ZrO2 sample. (c) HR-
TEM image of an Ag-Cu (4-1)@ZrO2 particle. (d) The line profile
analysis of Ag-Cu NP corresponding to the particle indicated by
the red arrow in inset. (e) EDX spectrum of Ag-Cu (4-1)@ZrO2
samples. (f) EDX mapping of Zr, Ag and Cu elements.
Figure 4. The XRD patterns of the photocatalysts.
One of the important features of the photocatalyst is its
light absorption property, which represents the light
absorption ability of the catalyst. The diffuse reflectance UV-
visible (DR-UV-vis) spectra of supported Ag-Cu alloy NPs are
shown in Figure 3. The light absorption peak of the Ag−Cu (4-1)
NPs is located at 408 nm while the light absorption peak of
pure Ag NPs is 392 nm which is attributed to the LSPR effect.
Impact of wavelength and light intensity
To further understand the contribution of light and verify the
photocatalytic mechanism, the dependence of catalytic activity
on wavelength and light irradiance (intensity) was investigated.
Action spectrum presents the mapping of the wavelength-
dependent photocatalytic rate and the light extinction
spectrum, which is a useful way to determe which wavelength
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 22
Green Chemistry
Please do not adjust margins
Journal Name
of light can drive reaction effectively.30-31 We conducted the
ARTICLE
In addition to the reaction rate, we also foundVtihewatArttihclee Olinglihnet
reduction of nitrobenzene over Ag-Cu (4-1) alloy photocatalyst wavelength can remarkablely affect theDpOroI: d10u.1c0t3s9e/Cle5cGtCiv0i1t7y26aBs
as model reaction. Five LED lamps with wavelengths of 400 ± 5, shown in Table 3. When shining by the light with the shortest
470 ± 5, 530 ± 5, 590 ± 5, and 620 ± 5 nm were used as light wavelength (365 nm), the highest selectivity to aniline was
source. The reaction rates were calculated and presented in obtained and the selectivity decreased with the increasing of
apparent quantum yield (AQY).32-33 The AQY for the reduction wavelength. Such result illustrated that we can manipulate the
of nitrobenzene were compared with the light absoroption selectivity of nitrobenzene reduction by the wavelength of
spectrum of Ag-Cu alloy NPs. It can be seen from Figure 5 that incident light. The possible reason is that transfer of the
the trend of AQY well matches with the light absorption curve energetic electrons to transform nitrobenzene to aniline
of the Ag-Cu alloy photocatalyst. The most significant require greater energy than that of azoxybenzene, light with
enhancement of reaction rate was obseved at 400 nm, where shorter wavelengths can excite electron of metal to higher
LSPR peak of Ag-Cu alloy NPs located. The action spectrum energy level and have greater chance to overcome the energy
suggested that the enhancement of the catalytic performance barrier to yield aniline.
was mainly due to the light absorption of the alloy NPs. The
Besides wavelength dependence, the impact of light
metal electrons gain sufficient energy from the LSPR effect intensity is another important feature of the photocatalytic
under irradiation which can be then transfered into the reactions.34 The dependence of catalytic activity on light
reactant molecules absorbed on the metal surface to induce irradiation intensity was investigated and the results is shown
the reaction. Here, we can confirm that it is the alloy NPs that in Figure 6. The conversion in the dark was regarded as the
act as an antenna that harvest visible light enhancing the contribution of thermal effect, we conducted the reaction
catalytic reaction activity.
under different irradiance (from 0.5 to 1.0 W/cm2) while all the
other reaction conditions were kept identical. We can see that
the conversion increased with the incresing of light irridiance.
The contributions of irradiation to the conversion efficiency
were calculated by subtracting the extent of conversion
achieved in the dark from the overall extent of the irradiated
system.35 To be more clear, the contribution of irradiation
accounts for only 66% for the reaction when the irradiation
was 0.5 W/cm2, while it increased to 83% with the irradiation
of 1.0 W/cm2. It shows that the contribution of irradiation
effect is increasing with the raise of light intensities, this is due
to the reason that with higher light intensity, larger amount of
light energetic electrons can be excited thus give higher
reaction rate.
Figure 5. Photocatalytic action spectrum for the reduction of
nitrobenzene using Ag-Cu(4-1)@ZrO2 photocatalyst. The light
absorption spectrum (left axis) is the DR−UV/vis spectrum of
the supported Ag-Cu(4-1)@ZrO2 catalyst.
Table 3. Performance of Ag-Cu (4-1)@ZrO2 photocatalyst in
nitrobenzene reduction with irradiation of different
wavelengths.
Conv. (%)
Select. (%)
Wavelength (nm)
AQY (%)
Aniline Azoxy
365
400
470
530
590
620
2.8
4.2
3.2
2.4
1.8
1.6
35
60
55
46
38
35
90
34
25
19
16
11
0
58
75
81
84
89
Figure 6. Dependence of the catalytic activity of Ag-Cu (4-
1)@ZrO2 for reduction of nitrobenzene on the intensity of light
irradiation. The numbers with percentages show the
contribution of the light irradiation effect. A photometer was
used to measure the light intensity; the other experimental
conditions were kept the same.
Reaction condition: photocatalyst 50 mg, reactant 0.5 mmol, 3 mL IPA as
solvent, 0.15 mmol KOH as base, 1 atm argon atmosphere, reaction
temperature 70°C, reaction time 16 h, and the light intensity was 0.2 W/cm2.
The conversions and selectivity were analysed by GC.
The relationship of reaction activity with the light
irradiance and wavelength confirmed that the light excited
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 6 of 22
ARTICLE
Journal Name
energetic eletrons play the key role in photocatalysed temperature, we did not obtain any siginificantVicewhaArntigclienOgnlionef
reactions. The population of energetic electrons can be product selectivity. It can be concluded DthOaI:t10te.1m03p9/eCr5aGtuCr0e17h26aBs
enhanced by applying higher light irradiance and turning the limited influence on selectivity of nitrobenzene redcution
wavelength of irradiation.
compared with that of light wavelength in this reaction.
Impact of temperature
Table 4. Performance of Ag-Cu (4-1)@ZrO2 photocatalyst in
nitrobenzene reduction at different temperatures.
Another important feature of the photocatalytic process on
metal NP catalysts is that the photocatalytic activity of the NPs
can be increased by elevating the reaction temperature.36-37 As
shown in Figure 7, higher reaction yield was achieved with the
temperature raising both under irradiation and in the dark.
However, the contribution of light irradiation reduced with the
rising of reaction temperature. For example, when at 50°C, the
difference between the light reaction and dark reaction was
68% accounting for 97% of the total conversion, which
decreased to 61% when the temperature was increased to
70°C. Under thermal heating conditions, the thermal activated
electons of metal NPs can activate adsborbed moleculars and
trigger the chemical reactions as well. At high reaction
temperature, larger population of thermal activated electrons
were generated with greater chance to surmount the
activation barrier, thus it is easier for the activation of the
reactions. Meanwhile, the vibrational states of the adsorbed
molecules were excited by high reaction temperature and
therefore the activation of substrate molecules is much
easier.35 This theory can also explain that the contribution
decreases as the reaction temperature was raised. At low
temperature, the thermal energy is insufficient to activate
metal electrons to overcome the activation barrier therefore
light excited electrons dominates the catalytic activity. In
Conv. (%)
Select. (%)
Aniline
Temp (°C)
Azoxy
73
40
50
60
70
80
40
70
27
21
14
11
13
78
86
85
87
96
100
100
Reaction condition: photocatalyst 50 mg, reactant 0.5 mmol, 3 mL isopropyl
alcohol (IPA) as solvent, 0.15 mmol KOH as base, 1 atm argon atmosphere,
reaction time 16 h, and the light intensity was 0.8 W/cm2. The conversions
and selectivity were analysed by GC.
Additionally, the apparent activation energy of the reaction
can be calculated by the reaction rate at different
temperatures using Arrhenius equation. We conducted the
reduction of nitrobenzene at diferent temperatures: 30°C,
40°C, 50°C, 60°C, 70°C and 80°C both under irradiation and in
the dark (thermal process). The apparent activation energy of
light reaction and dark reaction are shown in Figure 8, it can be
seen that the apparent activation energy for reduction of
nitrobenzene in the dark is 104.4 kJ mol-1 while only 20.4 kJ
mol-1 for the reaction under irradiation. The visible light
irradiation reduced the activation energy of the reaction by 80
kJ mol-1, which accounts for 80.4 % of the apparent activation
energy in the thermal reaction. The reduced apparent
activation energy in irradiation indicates the contribution of
light irradiation.38
contrast, high reaction temperature can generates
a
remarkable amount of excited metal electrons capable of
crossing activation barrier thus the contribution of light energy
is suppressed.
Figure 7. Dependence of catalytic activity on different reaction
temperatures for reduciton of nitrobenzene under a thermal
heating process in dark (blue triangles) and the light irradiation
process (red circles). The light intensity was 0.8 W/cm2.
Figure
8. Apparent activation energies of nitrobenzene
reduction calculated for the photoreaction and the reaction in
the dark.
The effect of temperature on the product selectivity was
also investigated. As shown in Table 4, with the raising of
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 22
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
Proposed mechanism
widely accepted mechanism for the reduction of
nitrobenzene was proposed by Haber in 1898 that nitro group
stepwise reduced to aniline through nitroso- and
hydroxylamine, additionlay hydroxylamine is an highly active
intermediate causing the yield of azoxy and azo compounds,39
the further reduction of azo compoDuOnId: 1s0.1r0e3s9u/Clt5sGCin017t2h6Be
hydroazo compounds which is eventually transformed into
aniline.40-41
View Article Online
A
Table 5. Photocatalytic performance of Ag-Cu (4-1)@ZrO2 using various intermediates as reactant.
Conv. (%)
Select. (%)
Azo
Entry
Substrate
Aniline
14
Azoxy
86
NO
2
1
96
0
1
9
2[a]
3[a]
100
100
8
91
61
NO
NHOH
29
4[a]
5
100
0
6
-
9
-
85
-
+
+
NO
NHOH
-
O
N
N
N
N
6
trace
-
-
-
Reaction condition: photocatalyst 50 mg, reactant 0.5 mmol, 3 mL IPA as solvent, 1 atm argon atmosphere, environment temperature 60°C, reaction time 16
h and the light intensity was 0.8 W/cm2. The conversions and selectivity were analysed by gas chromatography (GC). [a] reaction time 5 h.
In the Ag-Cu alloy NPs catalysed reaction system, trace
aniline was obatined when using azoxybenzene and
azobenzene as substrate (entries 5 and 6, Table 5.) indicating
the azoxybenzene through azobenzene to aniline route is
unfavorable. Therefore, it can be concluded the formation of
aniline and azo-compound are parallel reactions. Compared
with nitrobenzene, the reduction of nitrosobenzene and
hydroxylamine is much faster. It is worth noticing that
hydroxylamine gave greater yield to aniline than nitrobenzene
and nitrosobenzene indicating its responsible to the direct
production of aniline (entries 2 and 3, Table 5). When two
intermediates were used togther as the substrate, the similar
selectivity to that of nitrobenzene was observed (entry 4,
Table 5.). Hence, we proposed the reaction mechanism as
shown in Scheme 1, Ag-Cu alloy NPs can catalyse the reduction
of nitrobenzene through the coupling of nitrosobenzene and
Scheme 1. Possible reaction pathways for the reduction of
nitrobenzene with Ag-Cu alloy NPs in the irradiation of visible light.
hydroxylamine, instead of futher reducing to azobenzene or
In this study, the fact that only azoxybenzene was obtained
aniline. The Ag-Cu alloy NPs cannot transfer azoxybenzene into
when using Ag-Cu Alloy NPs catalyst indicates a different
azobenzene possiblely due to the limited reductive ability.
reaction pathway regarding to the alloying of Cu and Ag (entry
Whereas monometallic Ag NPs could favour the further
1, Table 5.). First of all, the reduction of nitrobenzene over Ag-
reduction of azoxybenzene to form azobenzene under the
Cu alloy NPs produce azoxybenzene with minor aniline without
other intermediates being detected. To further understand
same conditions. (Table S2 and Scheme S1, ESI)
reaction route with Ag-Cu alloy NPs catalyst, several
intermediates such as nitrosobenzene, hydroxylamine and
Recycle of photocatalyst
The reusability is a significant character of heterogeneous
catalyst.42 Therefore we performed the recycle experiments
for the catalyst. The catalyst was reused for 10 times to test
azobenzene were used as substrates for the redcution to gain
some insight into the reaction pathways.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 8 of 22
ARTICLE
Journal Name
the activity and the other reaction conditions were kept
identical. After each reaction cycle, the catalyst was separated
by centrifugation, washed with ethanol twice, and then dried
for subsequent reactions. As shown in Figure 9, the catalyst
was recycled for 10 times with less than 20% reduction of the
catalytic activity. Noticeably, the selectivity to azoxybenzene
can be still maintained above 70%. The observed slightly loss in
catalytic activity was possibly due to the leaching of the NPs or
the loss of NPs during washing after each catalytic run. To
further confirm the heterogeneous nature of the alloy catalyst,
we tested the amount of Ag and Cu in the reaction
supernatant liquid by Inductively coupled plasma-atomic
emission spectroscopic (ICP-AES) analysis, the amount of
leaching was detectable (<10 ppm), but there was very little
leaching of Ag during the reaction and the amount of the
leached Cu decreased apparently with the recycle times (Table
S4, ESI). We further conducted the hot filtration test,32 as
shown in Figure S3, the filtrate was incompetent to provide
further conversion (only 8% increase), while the undisturbed
control reaction showed excellent conversion (78% increase),
which indicated that the filtrate solution did not contain a
meaningful catalytically active species. Thus, overall it is
evident that the reaction was predominately catalysed by the
supported heterogeneous alloy NPs. The TEM image indicates
that the Ag-Cu alloy NPs still distributed evenly after several
times reactions, and the mean particle size remained 7 nm
(Figure 10). These results confirmed that the Ag-Cu alloy NPs
are stable and resuable photocatalyst for the reduction of
nitrobenzene.
View Article Online
DOI: 10.1039/C5GC01726B
Figure 10. (a) TEM image of Ag-Cu (4-1)@ZrO2 sample after 10
catalytic cycles. (b) Particle size distribution of Ag-Cu (4-
1)@ZrO2 sample after 10 catalytic cycles.
Conclusion
In Summary, a green photocatalysis process for selective
reduction of nitro compounds to azoxy compounds, driven by
light irradiation without intensive heating and pressured
reagent, can be achieved using Ag-Cu alloy NP catalyst
prepared simply by alloying small amount of Cu into Ag NPs.
The alloying of Cu into Ag NPs can maintain the azoxy
compounds as main product, which is not easily controllable
when using pure Ag NPs as catalyst. The catalyst with Ag-Cu
ratio 4:1 exhibited the best photocatalytic performance and
maintained stability in air. The selectivity to azoxybenzene and
aniline can be controlled by turning the wavelength of
irradiation. Meanwhile, the conversion of reactions can be
enhanced with the rising of light irradiance and reaction
temperature. The solid catalyst can be recycled and reused for
several runs without significantly losing activity. The catalytic
system provides us with a new direction for the application of
alloy NP photocatalyst and also contribute to understanding
the development of photocatalytic systems for more complex
organic reactions driven by visible light under “green” mild
conditions.
Experimental section
Catalysts preparation
3 wt% Ag-Cu alloy NPs on ZrO2 (Ag-Cu alloy @ZrO2) with
different ratios were prepared by the impregnation−reduction
method. For example, Ag-Cu alloy NPs of molar ratio 4:1 was
prepared as follow, AgNO3 (82.3 mg) and Cu(NO3)2∙6H2O (29.3
mg) was dissolved into 60 mL deionized water, to the above
solution ZrO2 nanopowder (2.0 g) was dispersed followed by
adding 20 mL lysine aqueous (0.01 M) solution. The mixture
was kept under magnetic stirring at room temperature for 20
min. To this suspension, 10 mL of freshly prepared aqueous
NaBH4 (0.35 M) solution was added dropwise. The mixture was
aged for 24 h, and then the precipitate was separated by
centrifugation, washed with water (three times) and ethanol
(once), and dried at 60°C in vacuum for 24 h. The collected
powder was used directly as catalyst. Other alloy catalysts with
different Ag-Cu ratio, as well as NPs of pure Ag on the ZrO2
support, were prepared via a similar method but using
different amount of quantities.
Figure 9. The reusability of the Ag-Cu alloy NPs for reduction of
nitrobenzene under visible light irradiation. The selectivity is to
azoxybenzene. Reaction conditions: photocatalyst 50 mg,
reactant 0.5 mmol, 3 mL isopropyl alcohol (IPA) as solvent,
0.15 mmol KOH as base, 1 atm argon atmosphere, reaction
temperature 60°C, reaction time 16 h and the light intensity
was 0.8 W/cm2. The conversions and selectivity were analysed
by gas chromatography (GC).
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 22
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
identical to those of typical reaction procedures. The apparent
View Article Online
Catalysts characterisation
quantum yield (AQY) was calculated DaOsI: 1f0o.1ll0o3w9/:C5aGpCp0a17re26nBt
quantum yield = [(Mlight–Mdark)/Np] × 100%, where Mlight and
Mdark are the molecules of products formed under irradiation
and dark conditions respectively, M = mole number of the
reactant × conversion × 6.02 × 1023 (Alvarado constant). Np is
the number of photons involved in the reaction. Np = Etotal/E1,
Etotal (the total energy involved in the reaction irradiation) =
intensity × light spot area × reaction time, E1 (the energy of
one photon) = h × c / L (h is Planck constant, c is light speed, L
is wavelength of the LED light).
The morphology study of catalysts was carried out on a JEOL
2100 transmission electron microscopy (TEM) equipped with a
Gatan Orius SC1000 CCD camera, with an accelerating voltage
of 200 kV and nickel grids were used as the supporting film.
Scanning electron microscope (SEM) imaging, elemental
mapping and EDX were performed using a ZEISS Sigma SEM at
accelerating voltages of 20 kV. Diffuse reflectance UV-visible
spectra of the sample catalysts were examined by a Varian
Cary 5000 spectrometer with BaSO4 as a reference. X-ray
photoelectron spectroscopy (XPS) data was acquired using a
Kratos Axis ULTRA X-ray Photoelectron Spectrometer
Acknowledgements
incorporating
a 165 mm hemispherical electron energy
We gratefully acknowledge financial support from the
Australian Research Council (ARC DP110104990 and
DP150102110). We also thank Arixin Bo and Mitchell De Bruyn
for the support of SEM and ICP analysis.
analyser. The incident radiation was Monochromatic Al Kα X-
rays (1486.6 eV) at 225W (15 kV, 15 ma). Narrow high-
resolution scans were run with 0.05 ev steps and 250 ms dwell
time. Base pressure in the analysis chamber was 1.0x10-9 torr
and during sample analysis 1.0x10-8 torr. X-ray diffraction (XRD)
patterns of the sample powders were collected using a Philips
PANalytical X’pert Pro diffractometer. Cu Kα radiation (λ=
1.5418 Å) and a fixed power source (40 kV and 40 mA) were
Notes and references
1
2
3
4
A. M. Tafesh and J. Weiguny, Chem. Rev., 1996, 96, 2035-
2052.
L. He, L.-C. Wang, H. Sun, J. Ni, Y. Cao, H.-Y. He and K.-N. Fan,
Angew. Chem. Int. Ed., 2009, 48, 9538-9541.
R. Mantha, K. E. Taylor, N. Biswas and J. K. Bewtra, Environ.
Sci. Tech., 2001, 35, 3231-3236.
A.-J. Wang, H.-Y. Cheng, B. Liang, N.-Q. Ren, D. Cui, N. Lin, B.
H. Kim and K. Rabaey, Environ. Sci. Tech., 2011, 45, 10186-
10193.
F. Figueras and B. Coq, J. Mol. Catal. A: Chem., 2001, 173,
223-230.
J. M. Joseph, H. Destaillats, H.-M. Hung and M. R. Hoffmann,
J. Phys. Chem. A, 2000, 104, 301-307.
S. S. Acharyya, S. Ghosh and R. Bal, ACS Sustain. Chem. Eng.,
used.
Inductively
coupled
plasma-atomic
emission
spectroscopic (ICP-AES) analysis was performed using a Perkin
Elmer 8300DV ICP fitted with an ESI SC-4DX auto-sampler and
PrepFAST
2 sample handling unit for online internal
standardisation and auto-dilution of samples and calibration
standards. Nitric acid, purified by sub-boiling distillation was
used for the preparation of all standards and blank solutions
used throughout the analysis.
5
6
7
8
9
Activity test
In a typical activity test, a 20 mL Pyrex glass tube was used as
the reaction vessel. The vessel containing reactants and
catalyst was irradiated with visible light using a halogen lamp
(from Nelson, wavelength in the range of 400−750 nm) under
magnetic stirring; the irradiance was measured to be 0.8
W/cm2. The reaction temperature was carefully controlled
with an air conditioner attached to the reaction chamber. The
reactions in the dark were conducted using an oil bath placed
above a magnetic stirrer to maintain the reaction temperature
the same to corresponding reactions with light illuminations;
reaction tubes were wrapped with aluminium foil to avoid
exposure to light. At given irradiation time intervals, 0.5 mL
aliquots were collected, the catalyst particulates was removed
by a Millipore filter (pore size 0.45 μm). The liquid−phase
products were analysed by gas chromatography (GC)
technique (Agilent 6890) with a HP−5 column to monitor the
change in the concentrations of reactants and products. An
Agilent HP5973 mass spectrometer was used to identify the
product.
2014, 2, 584-589.
A. Agrawal and P. G. Tratnyek, Environ. Sci. Tech., 1996, 30
,
153-160.
Y. Moglie, C. Vitale and G. Radivoy, Tetrahedron Lett., 2008,
49, 1828-1831.
10 K. Ohe, S. Uemura, N. Sugita, H. Masuda and T. Taga, J. Org.
Chem., 1989, 54, 4169-4174.
11 A. Roucoux, J. Schulz and H. Patin, Chem. Rev., 2002, 102
,
3757-3778.
12 A. Grirrane, A. Corma and H. García, Science, 2008, 322
,
1661-1664.
13 H. Zhu, X. Ke, X. Yang, S. Sarina and H. Liu, Angew. Chem. Int.
Ed., 2010, 49, 9657-9661.
14 S. Sarina, H. Zhu, E. Jaatinen, Q. Xiao, H. Liu, J. Jia, C. Chen
and J. Zhao, J. Am. Chem. Soc., 2013, 135, 5793-5801.
15 H. Zhu, X. Chen, Z. Zheng, X. Ke, E. Jaatinen, J. Zhao, C. Guo,
T. Xie and D. Wang, Chem. Commun., 2009, 48, 7524-7526.
16 S. Linic, U. Aslam, C. Boerigter and M. Morabito, Nat. Mater.,
2015, 14, 567-576.
17 J. C. Scaiano, K. G. Stamplecoskie and G. L. Hallett-Tapley,
Chem. Commun., 2012, 48, 4798-4808.
18 H. Tada, T. Ishida, A. Takao and S. Ito, Langmuir, 2004, 20
,
7898-7900.
In action spectrum experiments, light emission diode (LED)
lamps (Tongyifang, Shenzhen, China) with wavelengths 360±5
nm, 400±5 nm, 470±5 nm, 530±5 nm, 590±5 nm and 620±5 nm
were used as light source. The reaction temperature was
controlled with oil bath with all the other reaction conditions
19 X. Chen, Z. Zheng, X. Ke, E. Jaatinen, T. Xie, D. Wang, C. Guo,
J. Zhao and H. Zhu, Green. Chem., 2010, 12, 414-419.
20 X. Guo, C. Hao, G. Jin, H.-Y. Zhu and X.-Y. Guo, Angew. Chem.
Int. Ed., 2014, 53, 1973-1977.
21 U. Sharma, P. Kumar, N. Kumar, V. Kumar and B. Singh, Adv.
Synth. Catal., 2010, 352, 1834-1840.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 10 of 22
ARTICLE
Journal Name
33 Q. Xiao, S. Sarina, A. Bo, J. Jia, H. Liu, D. P. Arnold, Y. Huang,
22 A. Saha and B. Ranu, J. Org. Chem., 2008, 73, 6867-6870.
23 X. Liu, H.-Q. Li, S. Ye, Y.-M. Liu, H.-Y. He and Y. Cao, Angew.
Chem. Int. Ed., 2014, 53, 7624-7628.
24 C. Morales-Guio, I. Yuranov and L. Kiwi-Minsker, Top. Catal.,
2014, 57, 1526-1532.
25 L. Hu, X. Cao, L. Shi, F. Qi, Z. Guo, J. Lu and H. Gu, Org. Lett.,
2011, 13, 5640-5643.
26 J. P. Espinós, J. Morales, A. Barranco, A. Caballero, J. P.
View Article Online
H. Wu and H. Zhu, ACS Catal., 2014, 4,D1O7I2:510-1.1703394/.C5GC01726B
34 Q. Xiao, E. Jaatinen and H. Zhu, Chem. Asian. J., 2014, 9,
3046-3064.
35 Q. Xiao, Z. Liu, A. Bo, S. Zavahir, S. Sarina, S. Bottle, J. D.
Riches and H. Zhu, J. Am. Chem. Soc., 2015, 137, 1956-1966.
36 S. Linic, P. Christopher and D. B. Ingram, Nat. Mater., 2011, 10,
911-921.
Holgado and A. R. González-Elipe, J. Phys. Chem. B, 2002, 37 P. Christopher, H. Xin, A. Marimuthu and S. Linic, Nat.
106, 6921-6929.
Mater., 2012, 11, 1044-1050.
38 S. Sarina, E. R. Waclawik and H. Zhu, Green. Chem., 2013, 15
1814-1833.
27 M. Liu and W. Chen, Nanoscale, 2013,
5
, 12558-12564.
,
28 A. Marimuthu, J. Zhang and S. Linic, Science, 2013, 339
,
1590-1593.
39 K. Möbus, D. Wolf, H. Benischke, U. Dittmeier, K. Simon, U.
Packruhn, R. Jantke, S. Weidlich, C. Weber and B. Chen, Top.
Catal., 2010, 53, 1126-1131.
29 G. Ranga Rao and H. R. Sahu, J. Chem. Sci., 2001, 113, 651-
658.
30 A. Tanaka, S. Sakaguchi, K. Hashimoto and H. Kominami, ACS 40 A. Corma, P. Concepción and P. Serna, Angew. Chem. Int. Ed.,
Catal., 2013,
3
, 79-85.
2007, 46, 7266-7269.
31 E. Kowalska, R. Abe and B. Ohtani, Chem. Commun., 2009,
2
,
41 H.-U. Blaser, Science, 2006, 313, 312-313.
42 Y. Zhang, Q. Xiao, Y. Bao, Y. Zhang, S. Bottle, S. Sarina, B.
Zhaorigetu and H. Zhu, J. Phys. Chem. C, 2014, 118, 19062-
19069.
241-243.
32 Q. Xiao, S. Sarina, E. Jaatinen, J. Jia, D. P. Arnold, H. Liu and H.
Zhu, Green. Chem., 2014, 16, 4272-4285.
10 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 11 of 22
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
Table of contents
View Article Online
DOI: 10.1039/C5GC01726B
Photocatalytic reduction of nitroaromatics on Ag-Cu alloy NPs can maintain selectivity to azoxy compounds
whereas Ag NPs mainly obtain azo compounds.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 11
Please do not adjust margins

Green Chemistry
Page 12 of 22
View Article Online
DOI: 10.1039/C5GC01726B
Table of contents
Photocatalytic reduction of nitroaromatics on Ag-Cu alloy NPs can maintain selectivity to azoxy
compounds whereas Ag NPs mainly obtain azo compounds.

Page 13 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C5GC01726B
Electronic Supplementary Information
Selective reduction of nitroaromatics to azoxycompounds on supported Ag-
Cu alloy nanoparticles through visible light irradiation
Zhe Liu, Yiming Huang, Qi Xiao* and Huaiyong Zhu
School of Chemistry, Physics and Mechanical Engineering, Science and Engineering Faculty, Queensland University of Technology,
Brisbane, QLD 4001, Australia. E-mail: q2.xiao@qut.edu.au; Fax: +61 7 3138 1804; Tel: +61 7 3138 1581.
LEGENDS
Table S1. The quantified composition of Ag and Cu for catalysts with different ratios
Table S2. Performance of Ag-Cu (4-1)@ZrO2 for reduction of nitrobenzene with different substituent groups
Table S3. Photocatalytic performance of pure Ag@ZrO2 using various intermediates as reactant
Table S4. The amount of Ag and Cu in the supernatant liquid after in the recycle of catalyst as detected by ICP-AES analysis
Scheme S1. Possible reaction pathways for the reduction of nitrobenzene with pure Ag NPs
Figure S1. Energy dispersive spectrometer (EDX) spectrum for Ag-Cu (1-1)@ZrO2 and Ag-Cu (1-4)@ZrO2 sample from TEM
Figure S2. Scanning electron microscopy (SEM) analysis of the catalysts
Figure S3. The results of the hot filtration test
Figure S4. Transmission electron microscopy (TEM) analysis of the Ag NPs

Green Chemistry
Page 14 of 22
View Article Online
DOI: 10.1039/C5GC01726B
Table S1. The quantified composition of Ag and Cu for catalysts with different ratios
Mole ratio of Ag to Cu
Weight ratio of Ag to Cu
Entry
Experimental design
TEM[a]
4.3-1.0
1.4-1.0
1.0-4.4
Experimental design
2.6-0.4
SEM[b]
1
2
3
4-1
1-1
1-4
2.65-0.35
1.97-1.03
1.05-1.95
1.9-1.1
0.9-2.1
[a] Determined from EDX spectrum and element mapping from TEM analysis in Figure 2e and Figure S2.
[b] Determined from EDX spectrum and elemental mapping from SEM analysis in Figure S3.

Page 15 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C5GC01726B
Table S2. Performance of Ag-Cu (4-1)@ZrO2 for reduction of nitrobenzene with different substituent groups:
-O
N+
R
hv
R
NH2 +
R
R
NO2
N
catalyst
Conv. (%) (dark)
Select. (%)
Entry
R
Aniline
Azoxy
1
2
3
4
Cl
Br
80 (46)
70 (25)
50 (3)
11
29
29
14
89
71
71
86
MeO
H
96 (16)
Reaction condition: photocatalyst 50 mg, reactant 0.5 mmol, 3 mL isopropyl alcohol (IPA) as solvent, 0.15 mmol KOH as base, 1 atm argon atmosphere,
reaction temperature 60°C, reaction time 16 h and the light intensity was 0.8 W/cm2. The conversions and selectivity were analysed by gas chromatography
(GC).

Green Chemistry
Page 16 of 22
View Article Online
DOI: 10.1039/C5GC01726B
Table S3. Photocatalytic performance of pure Ag@ZrO2 using various intermediates as reactant
Conv. (%)
Select. (%)
Entry
Substrate
Aniline
17
Azo
83
Azoxy
0
NO
1
90
2
2[a]
3[a]
4[a]
5
100
100
4
6
4
92
23
NO
NHOH
+
71
NO
NHOH
-O
100
16
0
3
0
-
13
100
-
84
0
N+
N
N
N
6
-
Reaction condition: photocatalyst 50 mg, reactant 0.5 mmol, 3 mL IPA as solvent, 1 atm argon atmosphere, environment temperature 60°C, reaction time 16
h and the light intensity was 0.8 W/cm2. The conversions and selectivity were analysed by gas chromatography (GC). [a] reaction time 5 h.

Page 17 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C5GC01726B
Table S4. The amount of Ag and Cu in the supernatant liquid after in the recycle of catalyst as detected by ICP-AES analysis
Entry
Runs
Amount of Ag (ppm)
Amount of Cu (ppm)
1
2
3
4
1
2
0.067
0.587
0.017
0.008
8.004
4.569
1.238
0.181
5
10

Green Chemistry
Page 18 of 22
View Article Online
DOI: 10.1039/C5GC01726B
Scheme S1. Possible reaction pathways for the reduction of nitrobenzene with pure Ag NPs.
In the Ag NPs catalyst’s case, we observed the same result as Ag-Cu NPs catalyst that azobenzene cannot be further reduce to
aniline, it leads to same conculsion of formation of aniline and azocompound are parallel reactions (entry 6, Table S2). It was
demonstrated Ag NPs cannot efficiently convert azoxybenzene into azobenzene (entry 5, Table S2), this phenomenon implies
that in the reduction of nitrobenzene the great amount of azobenzene was not transfered from azoxybenzene but followes
another pathway. On the other hand, when both nitrosobenzene and hydroxylamine was introduced into the reaction, we
observed 84% yield of azoxybenzene (entry 4, Table S2). The reduction of nitrobenzene over Ag NPs catalyst cannot produce
both nitrosobenzene and hydroxylamine otherwise azoxybenzene instead of azobenzene would have been observed. We
seperately investigated the reduction of nitrosobenzene and hydroxylamine and found reduction of hydroxylamine gives similar
selectivity as the reduction of nitrobenzene only with faster reaction rate (entries 1 and 3, Table S2), however adverse selectivity
was observed with nitrosobenene (entry 2, Table S2). It illustrates the absence of nitrosobenzene as an intermediates in the
reduction of nitrobenzene because if it exists stablely in reaction system, azoxybenzene can be obtained as main product of
coupling. Therefore, the complete reaction route of nitrobenzene reduction over Ag NPs catalyst is illuminated in Scheme S1.

Page 19 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C5GC01726B
Figure S1. The EDX spectrum from TEM
(a) EDX spectrum from TEM for Ag-Cu (1-1)@ZrO2 samples. (b) EDX spectrum from TEM for Ag-Cu (1-4)@ZrO2 samples

Green Chemistry
Page 20 of 22
View Article Online
DOI: 10.1039/C5GC01726B
Figure S2. Scanning electron microscopy (SEM) analysis of the catalysts
(a) SEM image of Ag-Cu (4-1) @ZrO2 sample and the corresponding mapping of Zr, O, Cu and Ag elements. (b) EDX spectrum from SEM for
Ag-Cu (4-1) @ZrO2 sample. (c) EDX spectrum from SEM for Ag-Cu (1-1) @ZrO2 sample. (d) EDX spectrum from SEM for Ag-Cu (4-1) @ZrO2
sample.

Page 21 of 22
Green Chemistry
View Article Online
DOI: 10.1039/C5GC01726B
Figure S3. The results of the hot filtration test
The reaction was interrupted at 1.0 h, and the catalyst was removed by filtration, the solution was labelled as A, 0.5 mL sample was
collected for GC test. The rest solution without catalyst was re-irradiated as typical procedure, and after the reaction, sample was collected
for GC test (labelled as B). After moving catalyst, the conversion for B increased by 8 % (from 18 % from 26 %). The conversion of typical
reaction with catalyst (labelled C) was 96 %, which increased by 78 % from the first hour. It can be seen that the reaction did not proceed
too much for the solution B, which means that the NPs peeled off from catalyst has little effect to the reaction. Moreover, from the
reusability of Ag-Cu alloy NPs (Figure 9 in main text), we can also confirm that the catalytic activity doesn’t lose too much during the
reaction. Overall the main contribution of the activity results from the photocatalytic response of heterogeneous Ag-Cu alloy NPs.

Green Chemistry
Page 22 of 22
View Article Online
DOI: 10.1039/C5GC01726B
Figure S4. Transmission electron microscopy (TEM) analysis of the Ag NPs
(a) TEM image of the Ag@ZrO2 samples. (b) HR-TEM image of Ag particle indicated in part a (red square).