860
H. Xiao et al. / Electrochimica Acta 297 (2019) 856e863
Raman spectra of graphene support and corresponding Pd-
loading samples were shown in Fig. 5. As for graphitic materials,
ꢁ
1
ꢁ1
three characteristic peaks at about 1350 cm , 1580 cm
2
and
ꢁ1
650 cm assigned to D, G and 2D bands respectively are often
investigated [31,32]. The D band is assigned to edges, other defects
and disordered carbon, while G band is attributed to the zone
2
center E2g mode corresponding to the ordered sp bonded carbon
atom [31,32]. Therefore, the I
and of G(I ) band is used to measure the disorder degree. The I
increases with the disorder degree for graphitic materials [1]. As for
these three supports, their I /I ratios follow the order: I /I (HG-
0) > I /I (HG-300) > I /I (HG-800), in accordance with their
/I
D G
ratio of the intensities of the D (I
D
)
G
/I
D G
D
G
D G
6
D
G
D G
hydrogenation degree inferred from FTIR results. It's concluded that
the hydrogenation in preparation process mainly contributed to the
defects on HG-60 agreeing with previous research [33]. That's to
say, like oxidation of graphene producing defects in the synthesis
process of graphene oxide, hydrogenation also leads to defects
D G
sites. Compared to HG-300 and HG-800, the I /I ratios in corre-
sponding Pd-loading catalysts (Pd/HG-300 and Pd/HG-800) in-
crease. Such an enhancement was also found in GO decorated with
metal nanoparticles, suggesting formation of strong chemical
interaction between the metal nanoparticles and graphene [34]. It
was further verified by the Pd 3d XPS results. Pd 3d BE value of Pd/
HG-60 (Fig. S3) is a little higher than that of Pd/HG-300 and Pd/HG-
8
00 reflecting stronger Pd-graphene support interactions. How-
ever, when compared Pd/HG-60 with HG-60, the I /I ratio
D
G
decreased, that's because the decrease of dehydrogenation degree
with introduction of Pd leads to drastic reduction of defects
deduced from FTIR results. In addition, the hydrogenation degree of
these three Pd-loading catalysts are very low, and their I
D
/I
G
ratios
(Pd/
follow the order: I /I (Pd/HG-60)> I /I (Pd/HG-300) > I
D
G
D
G
D
/I
G
2
HG-800), indicating smaller average size of the sp domains and
higher dispersion of Pd nanoparticles on Pd/HG-60 than that on Pd/
HG-300 and Pd/HG-800. In other words, Pd/HG-60 should show
the super catalytic activity for ethanol oxidation because of its
higher dispersion of Pd nanoparticles to Pd/HG-300 and Pd/HG-
8
00.
C1s spectra of graphene supports and corresponding Pd-loading
Fig. 7. Cyclic voltammograms of Pd/HG-60, Pd/HG-300, Pd/HG-800, commercial Pd/C
and Pd/GO in 1 M KOH (a) and 1 M KOHþ1 M ethanol (b) respectively at room
temperature.
samples were exhibited in Fig. 6. The spectrum of graphitic mate-
rials is usually fitted to five different peaks centered at 284.5, 285.5,
2
3
2
86.6, 287.2 and 288.7 eV, assigned to sp C (C]C), sp C(CeC/
CeH), CeOH, C]O and eCOOe groups, respectively [35]. But, as for
Table S1 for electrochemical activity comparison. Based on the
ethanol oxidation current peak values obtained from Fig. 7b and
Table S1, it's concluded that the rate of oxidation of ethanol is
3
HG-60 and HG-300, both their sp C peaks shift to about 285.1 eV
3
probably due to relatively high hydrogenation of graphene. The sp
2
3
ꢁ1
C/sp C ratios for three graphene samples follows the order: sp C/
higher on Pd/HG-60 (I z 3442 mA mg Pd) than that on Pd/HG-300
sp C (HG-60) > sp3 C/sp C (HG-300) > sp C/sp C (HG-800).
2
2
3
2
ꢁ1
ꢁ1
(I z 3197 mA mg Pd) or Pd/HG-800 (I z 2505 mA mg Pd) or
ꢁ
1
Compared to these graphene supports, as for the corresponding Pd-
commercial Pd/C (I z 532 mA mg Pd) under similar experimental
conditions, suggesting the Pd/HG-60 showed the best ethanol
oxidation activity among these samples. The test results verify the
conclusion inferred from TEM, Raman and FTIR results that small
size and high dispersion of Pd nanoparticles more easily formed on
hydrogenated graphene support contributes to high electro-
catalytic performance for ethanol oxidation. It's seen that EOR ac-
tivity for these catalysts increased with increase of the
hydrogenation. For verification, the EOR of Pd supported on non-
hydrogenated graphene (graphene oxide (GO)) was evaluated for
comparison. As seen from the Fig. 7b, the rate of oxidation of
ethanol is lowest on Pd/GO compared with other catalysts, sug-
gesting the crucial rule of hydrogenated graphene substrate for the
improvement of EOR activity.
3
2
loading catalysts, their sp C/sp C ratios all decrease. The high ratio
3
2
of sp C/sp C means the efficient hydrogenation of the graphene
support [36]. Thus, the above results suggest HG-60 show the
highest hydrogenation degree among these supports, and the hy-
drogenation degree decreases with the introduction of Pd, agreeing
with FTIR results.
Fig. 7a and b show CVs of Pd/HG-60, Pd/HG-300, Pd/HG-800,
commercial Pd/C and Pd/GO electrodes in the potential range
from 0.224 to 1.424 V in 1 M KOH and 1 M KOHþ1 M ethanol
respectively. As seen from the Fig. 7a, four CV curves determined in
1
M KOH are almost similar and don't show obvious characteristic
current peaks of ethanol oxidation. As for Pd-loading catalysts,
their CV curves (Fig. 7b) measured in 1 M KOHþ1 M ethanol also
appear to be similar, but each curve exhibits two characteristic
current peaks: one in the forward scan is under anodic condition
while the other in the reverse scan is under cathodic condition.
Moreover, values of the forward oxidation current peak and its
corresponding peak potential for ethanol oxidation are given in
Fig. 8 presents the potentiodynamic polarization plots obtained
ꢁ1
at a scan rate of 0.5 mV s for the Pd loading catalysts. The po-
larization currents of the ethanol oxidation on these catalysts are in
the order as follows: Pd/HG-60 > Pd/HG-300 > Pd/HG-800 > com-
mercial Pd/C at the anodic peak potential. The corresponding Tafel