W.D. Wang et al.
Molecular Catalysis xxx (xxxx) xxx
(Fig. 1h).
Table 1
Raman spectra mainly showed that D (1350 cm–1) and G (1587
Preparation of benzonitrile from benzaldehyde and ammonia by catalysts:
comparison of catalytic activity under different conditions.a.
cm–1) peaks for a series of Fe2O3@NC catalysts varied with the tem-
perature (Fig. 2a). The relative intensity ratio of ID/IG represents the
density of defects in the catalysts. The ID/IG value exerted a certain
relationship with the carbonization temperature. The ID/IG firstly
increased and then decreased. Moreover, with increase in carbonization
temperature to 700 ◦C, the ID/IG value was the maximum, and then it
decreased with increased carbonization temperature. The higher the
carbonization temperature, the higher the graphitization degree of the
prepared materials. However, due to the effect of Fe, additional defects
were observed in the vicinity of Fe2O3 NPs. Therefore, this change is
caused by the effect of changes from carbon matrix to carbon nanotubes
with increase in carbonization temperature. XRD patterns of different
Fe2O3@NC materials shows a wide diffraction peak at 26◦, corre-
sponding to the presence of the C(002) crystal plane in all samples
(Fig. 2b). Especially, Fe2O3@NC-900 prepared at a high carbonization
temperature of 900 ◦C exhibited strong peak strength. By comparing the
XRD patterns of different Fe2O3@NC samples, a new peak at 44.7◦ which
attributed to Fe(110) crystal plane was observed in Fe2O3@NC-900
pattern [34]. In addition, it was difficult to observe the characteristic
peak for the Fe(110) crystal plane in Fe2O3@NC-500, Fe2O3@NC-600,
Fe2O3@NC-700, and Fe2O3@NC-800, because Fe NPs were hardly
formed at such carbonization temperatures.
Entry
Catalyst
T
Ammonia
(μL)
Solvent
Yield
(%)
(◦C)
1
2
3
4
5
Fe2O3@NC-500
(30 mg)
60
60
60
60
60
800
800
800
800
800
EtOH
EtOH
EtOH
EtOH
EtOH
0.15
Fe2O3@NC-600
(30 mg)
0.42
Fe2O3@NC-700
(30 mg)
32.00
69.25
59.79
Fe2O3@NC-800
(30 mg)
Fe2O3@NC-900
(30 mg)
6
7
NCNS (50 mg)
Fe2O3@NC-800
(10 mg)
60
60
800
800
EtOH
EtOH
trace
13.26
8
Fe2O3@NC-800
(50 mg)
60
60
60
60
60
50
70
60
60
800
800
800
800
800
800
800
600
700
EtOH
DMSO
DMF
99.45
78.61
49.37
29.93
27.25
96.39
69.65
46.82
63.64
9
Fe2O3@NC-800
(50 mg)
10
11
12
13
14
15
16
Fe2O3@NC-800
(50 mg)
Fe2O3@NC-800
(50 mg)
1, 2-
To investigate the specific surface area, pore diameter and pore
volume of Fe2O3@NC-800, nitrogen adsorption–desorption isotherms
were recorded (Fig. S2a). A typical type IV nitrogen adsorption isotherm
was observed for Fe2O3@NC-800, and hysteresis was observed in the
region of the medium and high pressures, near P/P0 = 0.4, the relative
pressure of closed, indicative of a mainly mesoporous aperture distri-
bution for Fe2O3@NC-800 [35]. The specific surface area of
Dichloroethane
n-Hexane
Fe2O3@NC-800
(50 mg)
Fe2O3@NC-800
(50 mg)
EtOH
EtOH
EtOH
EtOH
Fe2O3@NC-800
(50 mg)
Fe2O3@NC-800
(50 mg)
Fe2O3@NC-800 was 418.8 m2
g
–1. Such a high specific surface area
Fe2O3@NC-800
(50 mg)
plays an active role in catalysis. Fig. S2b shows the aperture distribution
curve of Fe2O3@NC-800. Consistent with the conclusion from the ni-
trogen adsorption–desorption isotherm analysis, the types of pores in
Fe2O3@NC-800 are mainly mesoporous: at 4.2 nm and 20–50 nm.
Mesopores are more favorable for the diffusion of the substrate within an
abundant pore size distribution.
a
Reaction condition: benzaldehyde (1 mmol), solvent (5 mL), ammonia (28
%–30 %) and at O2 atmosphere for 24 h.
ICP–OES were 1.73 %, 1.88 %, 4.22 %, 5.43 %, and 6.03 %.
Fig. 3a shows the wide-spectrum XPS of Fe2O3@NC-800 catalyst. The
result was consistent with the EDX results: C, N, O, and Fe were present
in Fe2O3@NC-800. The characteristic signal of Fe 2p was relatively
weak, possibly caused by Fe2O3 NPs being surrounded by N-doped
carbon. Fig. 3b shows the high-resolution C 1s spectrum of Fe2O3@NC-
2.2. Catalytic ammoxidation of benzaldehyde to benzonitrile
The catalytic properties of the prepared Fe2O3@NC materials were
examined by ammoxidation of benzaldehyde to benzonitrile. Table 1
summarizes the optimization of reaction conditions. Fe2O3@NC cata-
lysts prepared at different calcination temperatures exhibited different
catalytic properties under the same reaction conditions, and Fe2O3@NC-
800 exhibited the highest catalytic activity (entries 1–5). When N-doped
carbon nanosheets (NCNS) without iron was used as the catalyst, there
was no conversion of benzaldehyde (entry 6), indicating that the Fe2O3
should be the active sites for the catalytic ammoxidation of benzalde-
hyde to benzonitrile. With the reduction in the amount of the
Fe2O3@NC-800 catalyst to 10 mg, the conversion rate of benzaldehyde
was low at 24 h (entry 7). By using 50 mg of the Fe2O3@NC-800 catalyst,
almost all of the benzaldehyde was converted into benzonitrile at 24 h
(entry 8). Low-polarity solvents gave poor reaction results, and the yield
of benzonitrile accordingly decreased with the gradual decrease in the
solvent polarity (entries 8–12). Therefore, ethanol is the optimal solvent
herein. Reaction temperature certainly affected the reaction yield. The
yield of benzonitrile was the highest at 60 ◦C, and the yield decreased
with the decrease in the temperature to 50 ◦C (entry 13). However, with
increase in reaction temperature to 70 ◦C, the benzonitrile yield reduces
to about 70 %. The reason is that, in a high reaction temperature of
–
–
800. The C in Fe O @NC-800 mainly existed as C C (284.6 eV) and
–
2
3
–
–
–
CN (285.3 eV). The characteristic signal for O CO in the sample
–
was observed at 288.7 eV with weak strength, confirming that the
considerable pyrolysis of oxygen-containing functional groups occurs in
the catalyst during carbonization. Fig. 3c shows the high-resolution N 1s
spectrum of Fe2O3@NC-800. The two peaks at 398.1 and 400.7 eV
corresponded to pyridine N and graphite N, respectively [36]. Through
XPS analysis, the N content of Fe2O3@NC-800 was 12. 28 %, and the
corresponding contents of pyridine N and graphite N were 7.98 % and
4.3 %, respectively. Fig. 3d shows the high-resolution Fe 2p spectrum of
Fe2O3@NC-800. The Fe 2p3/2 signal with a combined energy of 710.7 eV
and the Fe 2p1/2 signal with a combined energy of 723.0 eV were
observed in the spectrum, indicating that Fe element is existing as Fe2O3
in Fe2O3@NC-800 sample [37,38]. It is worth mentioning that the sat-
ellite peak around 719.0 eV also suggests that the iron oxide is the main
Fe species in Fe2O3@NC-800 sample [39]. In addition, Fe-Nx was not
observed in Fe@NC-800 (whose characteristic signal was observed at
713.4 eV), providing supplementary evidence for the active component
in the catalytic reaction. Table S1 summarizes the contents of N and Fe
in Fe2O3@NC-800. As determined by elemental analysis, the contents of
N in Fe2O3@NC-500, Fe2O3@NC-600, Fe2O3@NC-700, Fe2O3@NC-800,
and Fe2O3@NC-900 were 27.20 %, 25.15 %, 15.12 %, 14.06 %, and 7.14
%, respectively, and the corresponding Fe contents estimated by
◦
above 70 C, the volatilization of NH3 from the reaction mixture will
lead the reduction of the NH3H2O concentration, which is not conducive
to the ammoxidation reaction. The ammonia content of ammonia water
led to the significant reduction in the yield of benzonitrile due to
4