2
Chen et al. Sci China Chem
spectroscopy with rotating disk electrochemistry to in-
vestigate the oxygen species intermediates in ORR process
of iron porphyrin compounds [21,22]. However, these
techniques focused on the extended metal surface or the
well-defined compounds, and were not suitable for the
complex ternary Fe/N/C catalysts. Inspired by the extensive
studies of active oxygen species in biological system, the
methods of the electron spin resonance (ESR) spectroscopy
and fluorescent probes are mainly employed to the detection
of the active oxygen species [23–25]. For the ESR spectro-
scopy, several groups have applied it into the detection of
•OH in ORR, but this method can only provide the qualita-
tive conclusions, and is difficult to apply to quantitatively
analyse •OH production due to low sensitivity [12]. Fluor-
escent probe method is hydroxyl radical-sensitive and a
powerful tool to estimate the amount of •OH through the
fluorescence intensity. Therefore, to pick up a suitable
fluorescent probe for the detection of •OH in Fe/N/C ORR is
prerequisite. Previous work proposed that •OH is common
intermediate for ORR and OER in the bio-inspired copper
catalyst by using the coumarin as a fluorescent probe in
ethanol-diluted alkaline solution [26]. Note that coumarin
would decompose in alkaline solution, and the existence of
•OH is qualitatively conclusive. Therefore, more detailed
experimental evidence is required to further understand the
nature of •OH generated from ORR on the Fe/N/C catalysts.
Among the M/N/C catalysts, rare fluorescent probes have
been employed into the detection of the •OH, and the re-
lationship between the •OH and H2O2 remains unclear. As
coumarin is effective in acid solution and its detection limit
for •OH is 5 nM [25,27,28], in this study, we use coumarin as
the fluorescent probe to distinguish the role of •OH in the Fe/
N/C ORR. As depicted in Scheme 1, during the Fe/N/C
ORR, non-fluorescent coumarin will react with the •OH to
generate the intense fluorescent 7-hydroxyl coumarin. Ac-
cording to the fluorescent intensity of 7-hydroxyl coumarin,
the •OH can be quantitatively analysed. We found that as
potential decreased from 0.70 to 0.60 V, the amount of •OH
generated from the ORR process on Fe/N/C catalyst in-
creased, reached a maximum at 0.65 V, and then declined.
This variation of •OH amount is greatly different from that of
H2O2, and the latter increased monotonically with decreasing
potentials. Our result indicates that •OH is not mainly gen-
erated by the H2O2 decomposition as traditional viewpoint.
The synthesis of Fe/N/C catalyst was similar to that re-
ported in our previous work [29]. To avoid the micropore
adsorption of coumarin which can block the mass transfer
channels for ORR, we used the external-surface graphene
oxide as the carbon source. The Fe/N/C catalyst was pre-
pared by heat treatment of the graphene oxide at 950 °C in an
NH3 (10%)+Ar (90%) atmosphere for 1 h with the heat
treatment of FeCl3 in next zone at 310 °C as the Fe source.
The transmission electron microscopy (TEM) showed that
Scheme 1 Schematic illustration of the conversion from non-fluorescent
coumarin to fluorescent 7-hydroxyl-coumarin by •OH generated in ORR on
Fe/N/C catalyst (color online).
the as-prepared Fe/N/C catalyst remained the layered struc-
The ORR test was carried out in O2-saturated 0.1 M H2SO4
solution with a glassy carbon RRDE at 900 r/min. Figure 1
(a) shows the stability test of the Fe/N/C catalyst at 0.65 V
with and without 0.5 mM coumarin. With the addition of the
coumarin, cathodic current decreased more quickly than that
in the blank solution. Meanwhile, the electrolyte with
0.5 mM coumarin showed a fluorescence emission at 472 nm
and the fluorescence signal grew continuously with the
electrolyzing time varying from 0 to 6 h as shown in Figure 1
(b). The appearance of the fluorescence emission at 472 nm
was due to the formation of the 7-hydroxyl coumarin [26,30].
This result indicates that •OH was involved in the Fe/N/C
ORR process.
In order to confirm that the fluorescence was generated
from •OH via the ORR on the Fe/N/C catalyst, we carried out
a series of control experiments as shown in Figure 2. Firstly,
to exclude the influence of electrooxidation, the potential
was cycled between 0.2 to 1.0 V for 6 h in Ar-saturated
0.1 M H2SO4 solution with 0.5 mM coumarin. The fluores-
cence intensity at 472 nm was extremely low (66 vs. 1,083).
This result is consistent with the previous reports because the
oxidation potential of coumarin is above 2.1 V (vs. NHE)
[28,31]. Secondly, during the Fe/N/C ORR process, the
counter electrode occurs oxidation reaction. We performed
the stability at 0.65 V for 6 h in 0.1 M H2SO4 containing
0.5 mM coumarin with a Nafion 211 proton exchange
membrane to separate the work electrode and counter elec-
trode. There was no significant fluorescence emission that
was observed in the solution in counter electrode compart-
ment. Thirdly, the O2 oxidation effect was also ruled out by
treating the Fe/N/C catalyst solely with the O2-saturated
0.1 M H2SO4 containing 0.5 mM coumarin without applying
potential, and the fluorescent intensity was also very low
(79 vs. 1,083). Lastly, we checked the influence of H2O2.
35 μM H2O2 was mixed with 0.5 mM coumarin and Fe/N/C
catalyst in 0.1 M H2SO4, and the fluorescence intensity was
also quite low (88 vs. 1,083). So, the chemical oxidation of
coumarin by H2O2 in Fe/N/C ORR process can be excluded