Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Page 3 of 4
ChemComm
Please do not adjust margins
Journal Name
COMMUNICATION
disperse GO, the signal of 16O2 (m/z = 32) is strong but there is
no clear signal of 18O16O (m/z = 34) or 18O2 (m/z = 36) by
comparison. The integration of the peak areas for the three
signals revealed that the 16O2 accounts for about 98.8% of the
evolved oxygen. As shown in Fig 3b, the signals of 18O16O and
18O2 in the labeling experiment using heavy oxygen water can
be clearly observed compared to the above experiment using
unlabeled water. The contents of 16O2, 18O16O and 18O2 in the
labeling experiment were determined to be about 78.9, 18.7
and 2.4%, respectively. At this stage, it would be reckless to
reach a conclusion confirming the source of molecular oxygen
as adequate consideration should be given to some critical
issues, such as the oxygen exchange between water and GO.
Here, we first made the assumption that there is oxygen
exchange occurring between water and GO; the 18O content of
GO at the equilibrium state of oxygen exchange is α (with
regard to the overall oxygen atoms of GO). If GO is the only
oxygen source during photoreduction, based on above
assumption, the ratio of 16O2, 18O16O and 18O2 should be: (1-α)2
: 2α(1-α) : α2. Clearly, the actual ratio of the three oxygen
molecules (78.9% : 18.7% : 2.4%) is totally inconsistent with
above mathematical deduction. This finding leads us to
speculate that the evolved oxygen comes from GO and water
together. Thus, it is inferred that GO and water contribute
equally to oxygen evolution because the actual ratio of 18O16O
(18.7%) is closer to 20%. In other words, half of the atoms in
the evolved molecular oxygen come from GO, while the other
half is derived from water. Thus, we can conclude that the
occurrence of GO photoreduction in aqueous dispersion is
accompanied by water oxidation 24-28 as depicted by equation
1:
DOI: 10.1039/C8CC08427K
B
D
C
E
A
Ea=14.8
Ea=38.8
Ea=12.0
TS2
TS3
Ea=15.9
TS1
B
A
TS4
E
C
D
Fig 5. Carbon dioxide signal detected by mass spectrometry. m/z=44,
m/z=46 and m/z=48 are C16O2, C18O16O and C18O2 respectively. The zero
point of the X axis represents time of the detection starting. (a) Carbon
dioxide signal of GO in normal water under UV irradiation for 3 hours. (b)
Carbon dioxide signal of GO in 18O enriched water under UV irradiation for
3 hours.
oxide and carbonyl groups clearly decreased after
photoreduction. We compared the possibility of the reactions
of epoxide and carbonyl groups with water, as shown in Fig S4.
The results indicated that epoxide can directly react with
water molecules, while carbonyl and water molecules are
difficult to react. Accordingly, we propose a reasonable
pathway whereby two nearby homolateral epoxide groups and
one water molecule as the starting reactants participate in the
oxygen generation process during GO photoreduction. As
shown in Fig 4 and Fig S11, the deoxidization reaction consists
of four steps: First, the hydrogen atom of water is transferred
to the epoxide group to form a hydroxyl group and a hydroxyl
radical. Second, the hydroxyl radical attacks the second
epoxide group and causes the removal of the epoxide group to
produce the •OOH fragment. Then the hydrogen atom of
•OOH is transferred between the two oxygen atoms of •OOH.
In the last step, the hydrogen atom of •OOH is transferred to
the carbon atom in GO to form the C-H bond and at the same
time oxygen is produced. The overall reaction process
consumes two epoxide groups, one water molecule to
generate oxygen, C-H bonds and hydroxyl groups. The
activation energy of four steps is listed in Fig 4. The mechanism
is consistent with the results of the XPS analysis and isotope
labeling experiments. The contrast of the infrared and Raman
spectrum intensity before and after photoreaction (Fig S5)
reflects the increase of the C-H bond content, which is also
consistent with the mechanism we have proposed.
GO + H218O → rGO + 16O18O
(equation 1)
To gain more insight into the possible mechanism of oxygen
generation, we additionally performed DFT calculations.
According to the results of the XPS analysis, the content of ep-
a
b
Time (Hour)
Time (Hour)
In addition to oxygen, another important gaseous product in
the photoreduction of GO is carbon dioxide. Our study also
confirms the important role of water in carbon dioxide
production, as shown in Fig S6. In this experiment, solid
sample with low water content rather than GO aqueous
dispersion was put into the quartz reactor. The solid sample
was prepared by squeezing 50 mg freeze-dried GO into a
tablet (see the insert photo of Fig S6), while the wet tablet
sample contained 0.5 ml distilled water. Evidently, under UV
Fig.4 Minimum energy path for oxygen evolution reaction obtains by
DFT calculation. A is the configuration of initiating reactant. We
omitted the gray atoms to make the model clearer to see. B, C and D
are the configuration of intermediate. E is the configuration of
product. TS represent transition state of reaction. C, O and H atoms
are illustrated in green, red and white, respectively. The reaction
energy curve including activation energy of each step is given in the
bottom diagram.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins