46
M.N. Pahalagedara et al. / Journal of Catalysis 336 (2016) 41–48
was comparatively low (75%) with the azoxybenzene selectivity of
85% and aniline selectivity of 15%. Conversion (100%) and selectiv-
ity toward the desired azoxybenzene product were observed when
the ratio was 1:1.5. With a further increase up to 1:2.5, although
the conversion was 100%, azoxybenzene selectivity decreased sig-
nificantly to 70%, due to the formation of aniline as the major side
product (Fig. S2b).
very small conversion (1%). In both reactions, aniline was the only
product detected from GC (Table 2, entries 10 and 11).
To illustrate the general applicability of the Ni/G nanocompos-
ite, the scope of the reaction was extended to reductive coupling
of a range of nitroarenes (Table 3). Under the optimized conditions,
a variety of structurally different nitroarenes were selectively
transformed to their corresponding azoxy products with a high
conversion and selectivity regardless of the presence of electron
donor or acceptor groups in the phenyl ring.
The reusability was tested by recycling the spent catalyst in
consecutive runs. The catalyst was recovered from the reaction
mixture by magnetic separation (Scheme 1) and washed thor-
oughly with acetone and dried at 120 °C for 12 h. Over a period
of four reaction cycles under the same conditions, no significant
loss of activity was observed (93% conversion, 100% selectivity
after the 4th reuse). XRD patterns of the catalyst before and after
the reaction are given in Fig. S4 and the catalyst did not show
any obvious change.
In order to calculate the rate constant for the catalytic reduction
of nitrobenzene, ln (Ct/C0) was plotted against time for the Ni nano-
material and the Ni/G nanocomposite (Fig. S3). C0 is the initial con-
centration of nitrobenzene and Ct is the concentration of
nitrobenzene at time t. The linear fit with a coefficient of determi-
nation very close to unity supports the pseudo-first-order kinetics.
The reaction rate constants k, calculated using the rate equation ln
(Ct/C0) = kt, are 1.2 Â 10À2 and 4.5 Â 10À2 minÀ1 for Ni nanomate-
rial and Ni/G nanocomposite catalysts, respectively. Incorporation
of graphene has increased the reaction rate for the reduction of
nitrobenzene to azoxybenzene, nearly 4-fold compared to the cat-
alyst without graphene. When increasing the nitrobenzene:hy-
drazine molar ratio from 1:1.5 to 1:2.5, the conversion was not
changed suggesting that the reaction is zeroth order with respect
to the reductant. Blank experiments were performed with gra-
phene and in the absence of the catalyst in order to confirm the
catalytic nature of the present catalyst. Graphene showed a 5% con-
version after 2 h. Reaction carried out without the catalyst gave a
4. Discussion
This study reports the synthesis of magnetic Ni/G nanocompos-
ites that act as a promising catalyst in the selective reduction of
nitrobenzene to azoxybenzene at room temperature and under
atmospheric pressure. Here, the composite is synthesized via the
reduction of Ni2+ and GO in a single step where hydrazine monohy-
drate is used as the reducing agent [19,20]. Oxygen containing
groups in GO, such as –OH and –COOH facilitate the homogeneous
dispersion of Ni2+ on its surface and both Ni2+ and GO are simulta-
neously reduced by hydrazine. Therefore, this can be considered as
an in situ reduction growth process. Raman spectroscopy is widely
used to characterize crystal structure, disorder and defects in
graphene-based materials. For example, the reduction process of
GO can manifest itself in Raman spectra by the changes in relative
intensity of two main peaks: D and G. We use this information to
verify the reduction process. Fig. 3 shows the Raman spectra of
GO and Ni/G nanocomposite which contains reduced GO. Both
spectra show the existence of the G band, related to the C–C vibra-
tions of carbon with a sp2 orbital structure, and the D band (disor-
der band), related to the disorder-induced vibration of the C–C
Table 3
Catalytic activity of the Ni/G nanocomposite for the formation of azoxybenzene via
selective reduction of different substituted nitroarenes.a
Entry
1
Substrate
Conversion (%)
96
Selectivity (%)
90
2
100
>99
bond. The decrease in the intensity ratio between
D band
(1350 cmÀ1) and G band (1600 cmÀ1) in Raman spectroscopy con-
firms the successful reduction of GO by hydrazine and is in agree-
ment with previous reports [21,22]. Due to its remarkable physical
properties, graphene has been used as an ideal substrate to anchor
inorganic nanoparticles for a wide range of applications [23–27].
According to SEM images, the in situ reduction growth process
has produced Ni nanomaterials with an urchin-like surface mor-
phology. TEM studies further confirm that incorporation of gra-
phene does not radically change their shape or surface
morphology. Enhancement of the catalytic activity by graphene
incorporation can be explained in two ways. As a catalyst support,
there is inhibition of the aggregation of active Ni nanoclusters by
dispersing them on the surface and providing a desirable chemical
interface between the catalyst and the reaction media. On the
3
4
100
88
>99
95
other hand, graphene can adsorb nitrobenzene via
p–p stacking
interactions, and again improves contact between the reactant
and the catalyst surface and facilitates transport of electrons from
graphene to the Ni nanomaterial [28]. These factors lead to the
accumulation of a high concentration of nitroaromatic compounds
over the active Ni surface, being subject to efficient reduction to
the corresponding azoxy compound [29–31]. The higher activity
of Ni/G nanocomposite can be attributed to its accessible surfaces
toward the substrate molecules due to their greater surface/vol-
ume ratio compared to that of Ni nanomaterial. Therefore, the
5
85
90
a
Reaction conditions: solvent (ethanol) = 5 mL, substrate = 15 mmol, nitroarene:
hydrazine (molar ratio) = 1:1.5, 0.025 g catalyst, room temperature, reaction
time = 2 h. Conversion and selectivity determined and confirmed by GC–MS.