Continuously photocatalytic production of H2O2 with high concentrations using 2-ethylanthraquinone as photocatalyst

Article abstract of DOI:10.1016/j.molcata.2016.04.009

We report a photocatalytic process for H₂O₂ synthesis by using 2-ethylanthraquinone (EAQ) as photocatalyst. H₂EAQ (or HEAQ^?) was in-situ produced by photocatalytic hydrogenation of EAQ with various solvents as hydrogen-donors (H-donors), and thus the pre-production of power-intensive H₂ was circumvented. The industrial hydrogenation of EAQ and the oxidation of H₂EAQ with O₂ here were combined into one pot for H₂O₂ photocatalytic production. EAQ molecule can be repeatedly used with high TON under both UV and visible light irradiation. When dual-phase reaction was adopted, the maximal H₂O₂ concentration of 574.0 mM in aqueous phase was achieved with mesitylene as H-donor. As the produced H₂O₂ concentration in the industrial AO process is estimated as ca. 400 mM in one cycle, our method with the dual-phase reaction greatly promotes the possibility for the practical production of H₂O₂ with photocatalytic method.

Full text of DOI:10.1016/j.molcata.2016.04.009

Journal of Molecular Catalysis A: Chemical 420 (2016) 66–72
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Continuously photocatalytic production of H2O2 with high
concentrations using 2-ethylanthraquinone as photocatalyst
Gangqiang Xu, Yan Liang, Feng Chen∗
Key Laboratory for Advanced Materials and Institute of Fine Chemicals, East China University of Science and Technology, 130 Meilong Road, Shanghai,
200237, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 March 2016
Received in revised form 11 April 2016
Accepted 12 April 2016
Available online 13 April 2016
We report a photocatalytic process for H2O2 synthesis by using 2-ethylanthraquinone (EAQ) as photo-
catalyst. H2EAQ (or HEAQ•) was in-situ produced by photocatalytic hydrogenation of EAQ with various
solvents as hydrogen-donors (H-donors), and thus the pre-production of power-intensive H2 was circum-
vented. The industrial hydrogenation of EAQ and the oxidation of H2EAQ with O2 here were combined
into one pot for H2O2 photocatalytic production. EAQ molecule can be repeatedly used with high TON
under both UV and visible light irradiation. When dual-phase reaction was adopted, the maximal H2O2
concentration of 574.0 mM in aqueous phase was achieved with mesitylene as H-donor. As the produced
H2O2 concentration in the industrial AO process is estimated as ca. 400 mM in one cycle, our method
with the dual-phase reaction greatly promotes the possibility for the practical production of H2O2 with
photocatalytic method.
Keywords:
Photocatalysis
Hydrogen peroxide
2-Ethylanthraquinone
Hydrogen transfer
Hydrogen-donor
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
Although new methods for H2O2 production are emerging in
endlessly, AO process is still presented as the most competi-
As a green oxidant, hydrogen peroxide (H2O2) is widely used
in industrial areas. More than 95% of the world’s production of
H2O2 so far is based on the anthraquinone oxidation (AO) pro-
cess [1]. The main advantage of the AO process is the very high
yield of H2O2 per cycle compared to other methods [2]. However,
the input of large quantities of H2 is indispensable, and risk of
explosion exists if a certain amount of O2 is mixed into the reac-
tor. Recently, novel strategies for the production of H2O2 by other
routes (electrochemical synthesis [3,4], direct synthesis [5,6], and
so on) are being explored. Alternatively, photocatalytic production
of H2O2 over semiconductor materials has been achieved, which
is expected to reduce the production cost of H2O2 by using solar
energy. Unfortunately, the concentrations of H2O2 produced with
semiconductor heterogeneous photocatalytic methods (Table S1)
are unimpressive, mM levels with UV irradiation [7–10] and ␮M
levels with visible light irradiation [11–16] in most cases. To date,
the highest concentrations of H2O2 produced through photocat-
alytic method just reach 40.2 mM under UV irradiation and 4.3 mM
under visible irradiation, respectively [10].
tive method for the industrial production of H2O2. Herein we
report a photocatalytic method for the synthesis of H2O2 with
2-ethylanthraquinone (EAQ), in which EAQ is utilized as a homo-
geneous photocatalyst. H2O2 is photocatalytically produced with
an hydrogen transfer process from the H-donor to the dissolved
oxygen molecule via the assistance of the excited EAQ.
In industrial AO process, the production of H2O2 includes
(i) the catalytic hydrogenation of EAQ with H2 to 2-
ethylanthrahydroquinone (H2EAQ) and (ii) the oxidation of
H2EAQ with O2 to give H2O2 and EAQ. The hydrogenation of EAQ
is separated strictly from the oxidation of H2EAQ [1]; otherwise,
the reaction of O2 with H2 would trigger an explosion. Briefly, the
solution containing H2EAQ is separated from the hydrogenation
catalyst (Nickel or supported Pd catalysts) and then oxidized with
air to produce equimolecular amount of H2O2 and regenerate
original EAQ as well [17]. In this work, the hydrogenation of
EAQ adopted a photocatalytic method. The observation of pho-
tocatalytic hydrogenation process was firstly operated under N2
atmosphere in the pre-deaerated solvent (CH3CH2OH) for safety
consideration, although no additional H2 was introduced into this
system. H2O2 was then obtained by bubbling air for ca. 5 min after
sampling.
∗
Corresponding author.
E-mail addresses: 442394042@qq.com (G. Xu), 623872330@qq.com (Y. Liang),
fengchen@ecust.edu.cn (F. Chen).
http://dx.doi.org/10.1016/j.molcata.2016.04.009
1381-1169/© 2016 Elsevier B.V. All rights reserved.

G. Xu et al. / Journal of Molecular Catalysis A: Chemical 420 (2016) 66–72
67
2. Experimental
2.1. Determination of H2O2
The concentration of H2O2 was determined by iodomet-
ric method with sodium thiosulfate (Na2S2O3). Sample solution
(1.0 mL) was added into iodine flask, and then H2SO4 (0.5 mL,
0.5 M) and KI solutions (0.5 mL, 100.0 g L−1) were added. Two to
three drops of 3% ammonium molybdate ((NH4)6Mo7O24·4H2O)
was added as catalyst to catalyze the generation of I2. The solution
was diluted with H2O (several milliliters). After reaction of 10 min
in the dark, the generated I2 was titrated with Na2S2O3. When
the color turned yellowish, starch solution (1.0 mL, 10.0 g L−1) was
added for coloration, and then the titration went on until the blue
disappeared and no longer appeared within 1 min.
In order to verify the accuracy of the iodometric method, the
redox titration with KMnO4 (calibrated with Na2C2O4) was also
used to determine the concentration of H2O2. The measurement
error of the two methods is within 4%. Particularly, the measured
H2O2 concentrations with the KMnO4 redox titration were slightly
larger than those with the iodometric method, which should be due
to the additional KMnO4 consumption from the organic contami-
nants.
Fig. 1. The photocatalytic H2O2 production in ethanol with (a) deaerated condition,
(b) simply sealed condition, and aerating with an air flow of (c) 0.05 L min−1, (d)
0.1 L min−1, (e) 0.2 L min−1. Inset: zoomed curves of (a) and (b). The concentration
of EAQ: 2.0 g L−1
.
60% B to 82% B; 10–18 min, 82% B; 18–30 min, 60% B; flow rate:
1.0 mL min−1. DAD detector was used to trace the substances.
For the MS part, atmospheric pressure chemical ionization
(APCI) was selected as the ion source, and a high resolution TOF
detector was used. Ion source interface voltage: 4.5 kV, −3.5 kV;
Nebulizer gas: N2, 1.5 L min−1; Drying gas: N2, 10 L min−1; Oven
temperature: 40 ◦C; Interface temperature: 300 ◦C; CDL tempera-
ture: 250 ◦C; Heating block temperature: 250 ◦C; Detector voltage:
1.6 kV.
2.2. Photocatalytic synthesis of H2O2
The photocatalytic synthesis of H2O2 was carried out in either
mono-phase or dual-phase system. Mono-phase photocatalytic
reaction was performed in ethanol solution, while dual-phase pho-
tocatalytic reaction was performed with organic phase to dissolve
EAQ and aqueous phase to extract the generated H2O2.
2.2.1. Mono-phase photocatalytic reaction
2.4. The method for GC analysis
2-Ethyl-9,10-anthraquinone (0.20 g, Sinopharm Chemical
Reagent, 98%) was dissolved in ethanol (100 mL, Sinopharm
Chemicals, >99.5%), and ultrasonicated 1 min for homogeneous
dispersion. The reactions were operated under (i) deaerated con-
dition with N2 protection; (ii) atmosphere with simply sealed; and
(iii) aerating condition with an air flow of 0.1 L min−1, respectively.
Particularly, the ethanol solvent was deaerated by ultrasonicating
and vacuuming for 30 min in advance in the case of deaerated
condition. The solution was then photoirradiated under magnetic
stirring. At a regular interval, a volume of 3–4 mL was sampled,
and then oxidated by air to obtain H2O2 from H2EAQ.
Quantitative analysis of toluene and its reaction products were
performed using gas chromatograph (Agilent GC 7890B) with a
capillary column (Agilent DB-17, 30 m × 0.25 mm × 0.5 ␮m) and
a FID detector. o-Dichlorobenzene was used as internal standard
substance for the quantitative analysis of benzaldehyde. Injector
temperature: 250 ◦C; Detector temperature: 300 ◦C. Oven temper-
ature was programmed as follows: the initial temperature was set
at 50 ◦C, followed by constant heating rate of 10 ◦C min−1 until the
final temperature of 280 ◦C was attained and held for 5 min. The
carrier gas was N2 with the flow of 1 mL min−1, and the splitting
ratio was set as 20:1.
2.2.2. Dual-phase photocatalytic reaction
2-Ethyl-9,10-anthraquinone (0.50 g) was dissolved in different
solvents (80 mL). Double-distilled water (20 mL) was then added
into the solution. The reactions were operated under aerating con-
dition with an air flow of 0.1 L min−1. The suspension was then
photoirradiated under magnetic stirring. In the first several hours,
1.0 mL of sample in aqueous phase was taken out for the deter-
mination of H2O2 and equal amount of H2O was added into the
reaction liquid to make sure the volume of aqueous phase was not
changed after sampling. When the concentration of H2O2 reached
ca. 500 mM, more than 1.0 mL of sample was taken out in every 0.5 h
to keep balance of the H2O2 concentration and stabilize it around
500 mM.
3. Results and discussion
3.1. Mono-phase photocatalytic reaction
A considerable amount of H2O2 was produced just with the help
of EAQ in the homogeneous system of ethanol (Fig. 1). The concen-
tration of H2O2 got the maximal value (2.6 mM) in 2.0 h, and then
kept almost unchanged, i.e., the relative content of H2EAQ to EAQ
kept a chemical equilibrium. In other word, the hydrogenation of
the EAQ here is limited a lot for the dynamic restriction. Specifically,
the extent of hydrogenation (conversion of EAQ to H2EAQ) only
reaches ca. 30% in this situation, which is far below that adopted in
industrial AO process (up to 60–70%) [1,18,19]. The hydrogenation
of EAQ in this system is achieved with a mild EAQ photocatalytic
process, in which H2 is substituted by the H-containing solvent
(CH3CH2OH), and the delusional danger from the violent reaction
between H2 and O2 is actually inexistent. Therefore, the photo-
catalytic hydrogenation of EAQ seems available in the presence of
O2, which suggests one-pot production of H2O2 can be possibly
2.3. The method for HPLC and LC–MS analysis
HPLC (LC-20A, Shimadzu) and LC–MS (LCMS-IT-TOF, Shimadzu)
were used to analyze EAQ and its oxidative intermediates. For
the LC part, Inert Sustain C18 column (4.6 × 250 mm, 5 ␮m) was
selected as the chromatographic column. Mobile phase condition:
A/B = H2O/CH3OH; 0–1 min, 60% B; 1–10 min, gradient change from

68
G. Xu et al. / Journal of Molecular Catalysis A: Chemical 420 (2016) 66–72
Table 1
The production capacity of the crude H2O2 aqueous solution with various H-donors
in aerating condition under full-spectrum irradiation.
b
d
System
Solvent
Ka (mM h−1
)
Ca (mM) Vc (mL h−1
)
Co (mM)
Mono-phase ethanol
54.0/21.0
/
/
/
ethyl acetate 22.5/8.3
40.3
478.0
500.0
574.0
1.0
3.0
4.0
5.0
10.6
32.5
54.5
64.5
Dual-phase
toluene
74.3/17.7
99.4/24.4
133.8/33.7
xylene
mesitylene
a
The reaction rate constant under full-spectrum/visible irradiation.
The final concentration of H2O2 in aqueous phase.
The output of crude H2O2 aqueous solution every one hour at equilibrium stage.
The final concentration of H2O2 in organic phase.
b
c
d
unavoidable (Fig. 1). Similar problem has also emerged in the direct
synthesis of H2O2 with H2 and O2, in which the sequential hydro-
genation of H2O2 to H2O possibly occurs after the generation of
H2O2 [21,22].
As for the aerating case (blowing air with a gas flow of
0.05 L min−1), the production of H2O2 was well maintained with
a steady formation rate of ca. 53 mM h−1 and gave a final concen-
tration of 317.5 mM in 6.0 h. When increasing the air flows to 0.1
and 0.2 L min−1, the concentration of H2O2 increased some to 324.0
and 366.5 mM (i.e., 12.5 g L−1). Surely, the supply of O2 indeed had
a great impact on the H2O2 production. In the presence of O2, the
hydrogen-transfer channal from the hydrogen-trapped EAQ under-
goes smoothly. The produced concentration of H2O2 in 6.0 h (with a
gas flow of 0.2 L min−1) is calculated to be 43.3-fold as the concen-
tration of EAQ used here. Besides, the over-hydrogenation of H2O2
is not observed with air blowing. The process to produce H2O2 in
the presence of O2 is a little different from that in deaerated condi-
tion. When in deaerated condition, excited EAQ captures H-atoms
from the H-donor (here refers to ethanol) to form H2EAQ under
irradiation. Then the H2EAQ is oxidized with O2 (when air blow-
ing at the second step) to produce H2O2 and original EAQ, which
is similar to the industrial AO process. However, in the present of
O2, when the excited EAQ molecule abstracts one H-atom from the
H-donor to form a protonated semiquinone radical [23], a rapid
reaction between the semiquinone radical and O2 would occur,
in which the semiquinone radical transports H to O2 to form the
hydroperoxyl radical (HO2•) [24–29], which subsequently dismu-
tates to form H2O2 as shown in Eqs. (1)–(3). Therefore, the repeated
use of EAQ can be achieved for the H2O2 production, which greatly
breaks through the concentration limit of EAQ.
Fig. 2. The photocatalytic H2O2 production under visible irradiation (␭ > 420 nm) in
ethanol with (a) deaerated condition, (b) simply sealed condition, and (c) aerating
with an air flow of 0.1 L min−1. Inset: zoomed curve of (a). The concentration of EAQ:
2.0 g L−1
.
Fig. 3. Dual-phase photocatalytic production of H2O2 in aerating condition
(0.1 L min−1) with (a) ethyl acetate, (b) toluene, (c) xylene, and (d) mesitylene as
H-donors. Organic phase: 80 mL; H2O: 20 mL. EAQ: 0.5 g.
hꢀ
EAQ→EAQ ∗ H-donor
→
HEAQ• + D•
(1)
(2)
(3)
achieved. Besides, the produced concentration of H2O2 in a single
cycle is severely limited by the EAQ concentration used in the tra-
ditional AO route. However, as EAQ can be circularly used in the
homogeneous photocatalytic process here, the concentration lim-
itation for producing H2O2 can possibly be solved. Therefore, the
photocatalytic reactions were also carried out in a simply sealed
flask (un-deaerated & un-aerating) as well as under an aerating
condition.
It should be noted that as the concentration of EAQ in ethanol
is limited to 8.46 mM according to its solubility, the maximal con-
centration of H2O2 could achieve 8.46 mM at most if the generation
of H2EAQ was one-off. However, as shown in Fig. 1, the concentra-
tion of H2O2 reached the maximal value of 19.0 mM at 1.0 h for
the simply sealed case, i.e., the produced H2O2 concentration lim-
itation from the EAQ solubility is broken. It is suggested that the
EAQ molecules here repeatedly reacted to trap the H-atom from
the H-donor and then liberate the H-atom to the oxygen molecules.
However, when the dissolved oxygen molecules were exhausted,
the as-formed H2O2 would be further hydrogenated to H2O by
interacting with the hydrogen-trapped EAQ. Therefore, although
the direct hydrogenation of H2O2 to H2O has a relatively high acti-
vation barrier [20], the concentration decrease of H2O2 becomes
HEAQ• + O2→EAQ + HO2
2HO2•→H2O2 + O2
•
According to the significant visible light absorbance of EAQ (Fig.
S1 and Table S2), the visible light activity of EAQ was also tested
by cutting off the UV band with an optical filter (␭ > 420 nm). The
similar tendencies of H2O2 production were observed as shown
in Fig. 2. In the case of pre-deaerated reaction, the concentration of
H2O2 increased slowly and reached the maximal value of 1.5 mM in
6.0 h. In the condition of limited O2-supply (simply sealed), the pro-
duced maximal concentration of H2O2 was enhanced to 19.9 mM in
2.0 h with a much faster rate. However, after the exhaustion of O2,
the production of H2O2 was blocked and the H2O2 concentration
turned to decrease with a slower rate compared to that with the
full-spectrum irradiation. Air-blowing gave a stable O2-supply and
enhanced the total amount of H2O2 production. The concentration
of H2O2 was stepwise accumulated to 126.2 mM after irradiation
for 6.0 h, which is about 39% of the produced H2O2 concentra-
tion obtained with full-spectrum irradiation in the same condition.
In brief, the photocatalytic production of H2O2 with EAQ exhibits

G. Xu et al. / Journal of Molecular Catalysis A: Chemical 420 (2016) 66–72
69
Fig. 4. Production of H2O2 (A) and consumption of EAQ (B) vs irradiation time in aerating condition (0.1 L min−1). The samples were taken from the reactions operated in
ethanol under (a) visible irradiation & (b) full-spectrum irradiation, and in toluene under (c) visible irradiation & (d) full-spectrum irradiation.
Scheme 1. The evolution of EAQ and impurities in the photocatalytic production of H2O2.
extremely high efficiencies under both UV and visible irradiation,
although mere visible irradiation is relatively weaker in exciting
EAQ (with the maximum absorption peak in 256 nm, see Fig. S1)
for H2O2 production.
added back in for every 0.5 h. In this way, a crude H2O2 aqueous
solution of ca. 500 mM was obtained as shown in Fig. 3.
Ethyl acetate is a poor H-donor. The H2O2 production reached
balance in 4 h and then the concentration of H2O2 kept at a low con-
centration (ca. 40 mM) with no increase. On the contrary, methyl
benzenes exhibited excellent performances. The initial formation
rates of H2O2 were estimated as 22.5, 74.3, 99.4 and 133.8 mM h−1
for ethyl acetate, toluene, xylene, and mesitylene, respectively. The
output of the crude H2O2 aqueous solution with these H-donors
(at the balance stage) were summarized and listed in Table 1 with
other concerned parameters. The concentrations of H2O2 in organic
phases were also determined and listed in Table 1.
3.2. Dual-phase photocatalytic reaction
In order to directly obtain a crude H2O2 aqueous solution of high
concentration to simulate the practical H2O2 production, the dual-
phase photocatalytic reaction with organic phase to dissolve EAQ
and aqueous phase to extract the generated H2O2 was operated
(Fig. 3). Different methyl benzenes and ethyl acetate were used
as organic solvents to dissolve EAQ as well as supply hydrogen
atoms. The produced concentration of H2O2 in aqueous phase was
monitored every 1.0 h until the concentration of H2O2 reached ca.
500 mM. Then a certain amount of H2O2 aqueous solution (more
than 1.0 mL) was taken out and an equal amount of water was
3.3. The oxidative pathway of toluene
Toluene was selected as the model H-donor here to analyze the
possible oxidative pathway of methyl benzenes by determining its
products after reaction for 8.0 h. Benzaldehyde was determined to

70
G. Xu et al. / Journal of Molecular Catalysis A: Chemical 420 (2016) 66–72
Table 2
EAQ and its oxidative intermediates identified with LC–MS.
Entry
Time
Formula
m/z
Substance
Structure
1
7.793
C15H8O3
236.0860
anthraquinone-2-carbaldehyde
2
3
4
10.455
10.925
12.738
C15H8O4
C16H10O3
C16H10O3
252.0779
250.0635
250.0636
anthraquinone-2-carboxylic acid
anthraquinone-2-acetaldehyde
2-acetyl-anthraquinone
5
13.706
C16H8O5
280.1115
anthraquinone-1-oxoacetic acid
6
7
14.856
17.464
C16H8O5
280.1099
222.0703
anthraquinone-2-oxoacetic acid
2-methyl-anthraquinone
C15H10O2
8
17.729
C16H6O5
278.0940
anthraquinone-1,2-furandione
9
20.419
22.275
C16H12O2
236.0811
236.0824
EAQ
10
C16H12O2
1-ethyl-anthraquinone
be 7.37 mmol (i.e., 0.94% in mass fraction) in organic phase with
GC. However, no benzyl alcohol was detected. Further, benzoic acid
was measured to be 1.54 mmol in organic phase and 0.22 mmol in
aqueous phase by HPLC, with its distribution ratio in toluene and
water to be 6.9. These results suggested that the excited EAQ seized
H from the toluene molecule to form a benzyl radical as seen in
Eq. (4), and the benzyl radical further tansformed to benzyl alco-
hol in the presence of H2O. As a better H-donor, benzyl alcohol
was more preferential to lose its hydroxylic H by the excited EAQ,
and transformed itself into benzaldehyde. Similar results (i.e., ˛-

G. Xu et al. / Journal of Molecular Catalysis A: Chemical 420 (2016) 66–72
71
carbonylation without significant ˛-hydroxylated intermediates)
have been previously reported in the alkylbenzene oxidation in
other photocatalytic reactions [30–32]. The further oxidation of
benzaldehyde with the excited EAQ then led to the produce of
benzoic acid.
or aldehyde (substance 3, minor). Further oxidation then led to
the formation of anthraquinone-2-oxoacetic acid (substance 6).
Anthraquinone-2-oxoacetic acid was not quite stable and decom-
posed to form anthraquinone-2-carbaldehyde (substance 1, minor)
and anthraquinone-2-carboxylic acid (substance 2, major), or fur-
ther oxidized to form ˛-hydroxyl compound, which was then
esterified intramolecularly to form cyclic lactone (substance 8).
On the other hand, the oxidation of 1-ethyl-anthraquinone (sub-
stance 10) gave anthraquinone-1-oxoacetic acid (substance 5),
while the oxidation of 2-methyl-anthraquinone (substance 7) gave
anthraquinone-2-carbaldehyde (substance 1) and anthraquinone-
2-carboxylic acid (substance 2) through similar processes. All these
substances got changed only in their side chains and the structure
of 9,10-anthraquinone remained unchanged; therefore, little influ-
ence can be observed on the efficiency of H2O2 production with the
oxidation of EAQ, which conformed to results shown in Figs. 1–3.
(4)
(5)
(6)
4. Conclusions
3.4. The turnover number (TON) of EAQ and its oxidative
intermediates
In summary, we provided a photocatalytic process for H2O2 syn-
thesis with EAQ as photocatalyst and various solvents as H-donors.
Compared to industrial AO process, the pre-production of power-
intensive H2 is circumvented by using light irradiation to in-situ
produce H2EAQ (or HEAQ•) in this work. Therefore, one pot reaction
for the production of H2O2 is available by combining the indus-
trial hydrogenation step of EAQ and the oxidation step of H2EAQ
(HEAQ•) with O2 into one reactor. Further, as EAQ molecule can be
repeatedly used with high TON for H2O2 production in this photo-
catalytic reaction, the dissolution of EAQ (and H2EAQ) with great
amount is not crucial anymore. Compared with other photocat-
alytic methods, our method significantly enhances the produced
concentrations of H2O2 under both UV and visible light irradiation.
Considering that the practical H2O2 concentration in industrial AO
process is estimated as ca. 400 mM [19] before the final concen-
tration treatment, our method with the dual-phase reaction gives
a raw H2O2 concentration beyond 470 mM (max. 574.0 mM with
mesitylene), which greatly promotes the possibility for the practi-
cal production of H2O2 with the photocatalytic method.
HPLC analysis was applied to track the concentration changes of
EAQ during the reaction (Fig. 4B). It should be noted that, the con-
centration of EAQ remained almost unchanged under visible light
irradiation although O2 was bubbled through the solution. On the
other hand, EAQ slowly decreased with time and dropped by ca.
1.24 mM in 6 h under full-spectrum irradiation in the presence of
O2. The decomposition of EAQ should be due to its oxidation by
singlet oxygen [33]. It has been reported that an energy transfer
between the excited EAQ molecule and oxygen molecule would
occur under irradiation [34], which leads to the generation of sin-
glet oxygen. Academically, only ca. 1.24 mM, i.e. 14.7% of EAQ,
changed with a H2O2 production of 324.0 mM, which gave a TON
of 260 in the lifetime of EAQ under full-spectrum irradiation. As for
the case of visible irradiation, no apparent loss (0.04 mM, 0.5%) of
EAQ was observed for H2O2 production of 126.2 mM with a TON
of 3023. The reason for the apparent difference of TONs between
these two cases may be due to that UV irradiation is more effective
in producing 1O2, which then reacts with EAQ, leading to the oxida-
tion of EAQ. By changing the solution from ethanol to toluene, the
TONs for the H2O2 production turn into to 509 under full-spectrum
irradiation and 568 under visible irradiation, with an EAQ loss of
0.26 mM and 0.08 mM in toluene, respectively. The enhancement
of the TON of EAQ in toluene solution under full-spectrum irradi-
ation should be due to the weaker polarity of toluene compared
to ethanol, which suppressed the formation of 1O2 with stronger
polarity. However, the low efficiency of H2O2 production led to only
a little increase of TON when under visible irradiation.
Correspondingly, the oxidative intermediates of EAQ were ana-
lyzed by LC–MS (Figs. S2–S12). Table 2 listed the main substances
in EAQ solution identified by LC–MS. Besides EAQ itself, there
are 10 substances existing. Substances 7 and 10 are identified
as 2-methyl-anthraquinone and 1-ethyl-anthraquinone (isomer of
EAQ), respectively. Both of them appeared in the raw EAQ solution
with low contents (both below 1.0%) and then disappeared after
the photocatalytic reaction. Presumably, they may come from the
industrial synthetic process of EAQ with phthalic anhydride and
ethylbenzene through Friedel-Crafts reaction [35,36].
Scheme 1 depicted the possible oxidative pathways of EAQ.
Unlike the case in industrial process, EAQ molecules here suf-
fered from oxidation to form oxidized products, although these
changes did not bring obvious harm to the practical efficiency
of H2O2 production. During photocatalytic reaction, the ethyl
in the excited EAQ was attacked by 1O2 (O2 excited by EAQ*)
[34,37] and formed corresponding ketone (substance 4, major)
Acknowledgements
This work was supported by the National Nature Science
Foundations of China (21177039) and the Innovation Program of
Shanghai Municipal Education Commission (13ZZ042).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molcata.2016.04.
009.
References
[1] J.M. Campos-Martin, G. Blanco-Brieva, J.L.G. Fierro, Angew. Chem. Int. Ed. 45
(2006) 6962–6984.
[2] Y. Nomura, T. Ishihara, Y. Hata, K. Kitawaki, K. Kaneko, H. Matsumoto,
ChemSusChem 1 (2008) 619–621.
[3] J.S. Jirkovskyı´, I. Panas, E. Ahlberg, M. Halasa, S. Romani, D.J. Schiffrin, J. Am.
Chem. Soc. 133 (2011) 19432–19441.
[4] S. Siahrostami, A. Verdaguer-Casadevall, M. Karamad, D. Deiana, P. Malacrida,
B. Wickman, M. Escudero-Escribano, E.A. Paoli, R. Frydendal, T.W. Hansen,
Nat. Mater. 12 (2013) 1137–1143.
[5] E.N. Ntainjua, M. Piccinini, S.J. Freakley, J.C. Pritchard, J.K. Edwards, A.F. Carley,
G.J. Hutchings, Green Chem. 14 (2012) 170–181.
[6] P. Biasi, F. Menegazzo, F. Pinna, K. Eränen, T.O. Salmi, P. Canu, Chem. Eng. J.
176 (2011) 172–177.
[7] M. Teranishi, S.-i. Naya, H. Tada, J. Am. Chem. Soc. 132 (2010) 7850–7851.
[8] D. Tsukamoto, A. Shiro, Y. Shiraishi, Y. Sugano, S. Ichikawa, S. Tanaka, T. Hirai,
ACS Catal. 2 (2012) 599–603.

72
G. Xu et al. / Journal of Molecular Catalysis A: Chemical 420 (2016) 66–72
[9] V.M. Daskalaki, P. Panagiotopoulou, D.I. Kondarides, Chem. Eng. J. 170 (2011)
433–439.
[23] B. Hulme, E. Land, G. Phillips, J. Chem. Soc. Faraday Trans. 1 68 (1972)
1992–2002.
[10] Y. Shiraishi, S. Kanazawa, D. Tsukamoto, A. Shiro, Y. Sugano, T. Hirai, ACS
Catal. 3 (2013) 2222–2227.
[24] Y. Zhang, K.A. Simon, A.A. Andrew, R. Del Vecchio, N.V. Blough, Environ. Sci.
Technol. 48 (2014) 12679–12688.
[11] Y. Shiraishi, S. Kanazawa, Y. Sugano, D. Tsukamoto, H. Sakamoto, S. Ichikawa,
T. Hirai, ACS Catal. 4 (2014) 774–780.
[25] A.E. Alegrıı´a, A. Ferrer, G. Santiago, E. Sepúlveda, W. Flores, J. Photochem.
Photobiol. A 127 (1999) 57–65.
[12] Y. Isaka, S. Kato, D. Hong, T. Suenobu, Y. Yamada, S. Fukuzumi, J. Mater. Chem.
A 3 (2015) 12404–12412.
[26] S. Garg, A.L. Rose, T.D. Waite, Geochim. Cosmochim. Acta 75 (2011)
4310–4320.
[13] J. Sheng, X. Li, Y. Xu, ACS Catal. 4 (2014) 732–737.
[14] Y. Shiraishi, S. Kanazawa, Y. Kofuji, H. Sakamoto, S. Ichikawa, S. Tanaka, T.
Hirai, Angew. Chem. Int. Ed. 53 (2014) 13454–13459.
[15] S. Kato, J. Jung, T. Suenobu, S. Fukuzumi, Energy Environ. Sci. 6 (2013)
3756–3764.
[27] Y. Song, G.R. Buettner, Free Radic. Biol. Med. 49 (2010) 919–962.
[28] K. Nawara, P. Krysinski, G. Blanchard, RSC Adv. 2 (2012) 4059–4061.
[29] B. Armitage, Chem. Rev. 98 (1998) 1171–1200.
[30] K. Ohkubo, K. Suga, K. Morikawa, S. Fukuzumi, J. Am. Chem. Soc. 125 (2003)
12850–12859.
[16] T. Hirakawa, Y. Nosaka, J. Phys. Chem. C 112 (2008) 15818–15823.
[17] Y. Cheng, L. Wang, S. Lu¨ , Y. Wang, Z. Mi, Ind. Eng. Chem. Res. 47 (2008)
7414–7418.
[18] E. Santacesaria, M. Di Serio, A. Russo, U. Leone, R. Velotti, Chem. Eng. Sci. 54
(1999) 2799–2806.
[19] Q. Chen, Chem. Eng. Process. 47 (2008) 787–792.
[20] J. Li, A. Staykov, T. Ishihara, K. Yoshizawa, J. Phys. Chem. C 115 (2011)
7392–7398.
[21] J.K. Edwards, B. Solsona, E. Ntainjua, A.F. Carley, A.A. Herzing, C.J. Kiely, G.J.
Hutchings, Science 323 (2009) 1037–1041.
[31] B. Mühldorf, R. Wolf, Chem. Commun. 51 (2015) 8425–8428.
[32] I. Saito, K. Tamoto, T. Matsuura, Tetrahedron Lett. 20 (1979) 2889–2892.
[33] H.P. Gemoets, Y. Su, M. Shang, V. Hessel, R. Luque, T. Noël, Chem. Soc. Rev. 45
(2016) 83–117.
[34] K. Mothilal, J.J. Inbaraj, R. Gandhidasan, R. Murugesan, J. Photochem.
Photobiol. A 162 (2004) 9–16.
[35] R. Xu, X. Guo, G. Wang, J. Liu, Z. Zhang, H. Liu, Catal. Lett. 107 (2006) 149–154.
[36] H. Naeimi, S.S. Brojerdi, Polycycl. Aromat. Compd. 34 (2014) 504–517.
[37] J.J. Inbaraj, M.C. Krishna, R. Gandhidasan, R. Murugesan, Biochim. Biophys.
Acta Gen. Subj. 1472 (1999) 462–470.
[22] V.R. Choudhary, C. Samanta, P. Jana, Appl. Catal. A 317 (2007) 234–243.