M. Mirza-Aghayan et al. / Catalysis Communications 69 (2015) 97–103
101
displays the TGA curves of graphite, graphite oxide, and Pd NPs/rGO
nanocomposite samples. Graphite material shows a weight loss of near-
ly 1% in the experimental temperature range (Fig. 2a) [31]. In contrast,
the weight loss of graphite oxide occurs in three successive steps. The
first one is a steady weight loss (7%), attributed to the vaporization of
adsorbed water molecules and occurs at around 100 °C. It is followed
by a rapid loss (25%) due to the decomposition of the labile oxygen-
containing functional groups (hydroxyl, epoxy, carbonyl, and carboxyl
groups) in the temperature range of 100–208 °C. Finally, a weight loss
(19%) due to the combustion of the carbon skeleton is observed in the
temperature range of 208–900 °C (Fig. 2b) [32]. Pd NPs/rGO exhibited
overall less than 10% weight loss in the same temperature range
(Fig. 2c). This result indicates the effective reduction of graphite oxide
to rGO by removing most of the carboxyl, epoxy, carbonyl, and hydroxyl
groups.
Fig. 3 depicts the SEM images and the corresponding EDX spectra of
graphite flakes, graphite oxide and Pd NPs/rGO nanocomposite.
As clearly seen in the SEM images, the structure of graphite (Fig. 3a)
was completely changed to flower-like structure with appearance of
graphitic sheets after the oxidation process (Fig. 3b). EDX analysis
shows that the initial graphite comprises mainly carbon (97.29 at.%)
with a small amount of oxygen (Fig. 3d). After extensive oxidation of
graphite to graphite oxide, EDX spectrum reveals the formation of a
new structure with higher oxygen content of 17.91 at.% with C/O ratio
of 3.24 (Fig. 3e). It should be noted that the presence of sulfur
(4.64 at.%) is most likely due to impurities from H2SO4 used in this pro-
cedure. The SEM image of the Pd NPs/rGO nanocomposite exhibits a
high density of nanoparticles in form of aggregates (Fig. 3c). EDX anal-
ysis of Pd NPs/rGO nanocomposite indicates that the palladium loading
on the rGO is about 4.29 at.% (Fig. 3f). The FTIR spectrum of graphite
oxide exhibits an intensive broad peak of O\\H stretching vibrations
at about 3393 cm−1, and several bands at 1719, 1574, 1215 and
1044 cm−1 assigned to C_O, C_C, (C\\O) epoxy and (C\\O) alkoxy
groups, respectively (Fig. 4a) [33]. The comparison of FT-IR spectra of
graphite oxide and Pd NPs/rGO clearly shows that chemical changes oc-
curred during Pd NPs deposition using ultrasonication. The strong peaks
at 1719 and 1044 cm−1 disappear and a weak peak at 1217 cm−1 still
remains in the FTIR spectrum of Pd NPs/rGO. The FTIR spectrum is dom-
inated by a broad band at ~1636 cm−1 due to C_C stretching modes,
suggesting partial restoration of the aromatic network during the chem-
ical process (Fig. 4b) [33].
Fig. 5. TEM image of Pd-NP/rGO.
average diameter (Fig. 5). The TEM image of the Pd NPs/rGO also shows
nanoparticles in form of aggregates.
3.2. Catalytic performances of Pd NPs/rGO
In connection with our interest in heterogeneous catalysis, we have
investigated rGO as a supporting matrix for Pd NPs and Et3SiH as a hy-
dride source in the chemoselective reductive dehydroxylation of ben-
zylic alcohols. In a control experiment, we have studied the efficiency
of graphite oxide (50 mg) for the hydrogenolysis of benzyl alcohol
(1 mmol) in ethanol (5 mL). The resulting mixture was stirred for the
time indicated in Table 1 prior to GC–MS analysis. Only the starting ma-
terial was recovered after 5 h (entry 1, Table 1), suggesting that graphite
oxide alone is not active for this chemical transformation. In another
control experiment, we have tested the efficacy of triethylsilane
(2 mmol) for the reductive dehydroxylation of benzyl alcohol in anhy-
drous ethanol in an inert atmosphere. Again Et3SiH alone was not effec-
tive for the reduction of benzyl alcohol and toluene was obtained only in
8% yield after 5 h (entry 2, Table 1). Finally, we have investigated graph-
ite oxide and Et3SiH system for this chemical process. After 5 h (entry 3,
Table 1), only the starting material was recovered. From this set of con-
trol experiments, it is clear that graphite oxide, Et3SiH or graphite oxide/
Et3SiH is not effective for the reductive dehydroxylation of benzyl
alcohol.
TEM analysis of the synthesized Pd NPs/rGO material clearly reveals
highly exfoliated rGO sheets covered by Pd nanoparticles of 37 22 nm
In contrast when the reaction of benzyl alcohol (1 mmol) and Et3SiH
(2 mmol) in dry ethanol was performed in the presence of Pd NPs/rGO
(50 mg), an exothermic reaction occurred during the first 5 min. GC/MS
analysis of the crude product indicated that toluene was obtained in 99%
yield after 5 min (entry 4, Table 1). It should be noted that similar results
were obtained using only 10 mg of Pd NPs/rGO under otherwise identi-
cal experimental conditions (entry 5, Table 1). Decreasing the catalyst
amount to 1 mg led to a decrease of the reaction yield to 65% even
Table 1
Different conditions for the hydrogenolysis of benzyl alcohol.
Entry
Catalyst
(mg)
50
Et3SiH (mmol)
Time (min)
Yielda (%)
1
2
3
4
5
6
Graphite oxide
–
Graphite oxide
Pd NPs/rGO
Pd NPs/rGO
Pd NPs/rGO
–
2
2
2
2
2
300
150
300
5
5
5
–
8
–
99
99
65
50
50
10
1
a
Fig. 4. FT-IR spectra of (a) graphite oxide and (b) Pd NPs/rGO.
Determined by GC–MS.