Selective hydrogen peroxide formation by titanium dioxide photocatalysis with benzylic alcohols and molecular oxygen in water

Article abstract of DOI:10.1021/cs400511q

Photocatalytic production of hydrogen peroxide (H₂O₂) on semiconductor catalysts with alcohol as a hydrogen source and molecular oxygen (O₂) as an oxygen source has attracted much attention as a potential method for safe H₂O₂ synthesis, because the reaction can be carried out without the use of explosive H₂/O ₂ mixed gases. Early reported photocatalytic systems with aliphatic alcohol as a hydrogen source, however, produce only a few millimolar levels of H₂O₂. We found that benzylic alcohols, when used as a hydrogen source for photoreaction in water with titanium dioxide (TiO ₂) photocatalyst, produce a very high concentration of H ₂O₂ (ca. 40 mM). Raman spectroscopy and electron spin resonance analysis revealed that the enhanced H₂O₂ formation is due to the efficient formation of side-on coordinated peroxo species on the photoactivated TiO₂ surface, via the reaction of benzylic alcohol and O₂. The peroxo species is readily transformed to H₂O₂, thus facilitating highly efficient H ₂O₂ production.

Full text of DOI:10.1021/cs400511q

Subscriber access provided by NORTH CAROLINA STATE UNIV
Article
Selective Hydrogen Peroxide Formation by Titanium Dioxide
Photocatalysis with Benzylic Alcohols and Molecular Oxygen in Water
Yasuhiro Shiraishi, Shunsuke Kanazawa, Daijiro Tsukamoto,
Akimitsu Shiro, Yoshitsune Sugano, and Takayuki Hirai
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/cs400511q • Publication Date (Web): 23 Aug 2013
Downloaded from http://pubs.acs.org on August 27, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth
Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 8
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
Selective Hydrogen Peroxide Formation by Titanium Dioxide
Photocatalysis with Benzylic Alcohols and Molecular Oxygen
in Water
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Yasuhiro Shiraishi,* Shunsuke Kanazawa, Daijiro Tsukamoto, Akimitsu Shiro, Yoshitsune Sugano, and
Takayuki Hirai
Research Center for Solar Energy Chemistry, and Division of Chemical Engineering, Graduate School of Engineering Science,
Osaka University, Toyonaka 560-8531, Japan
Supporting Information Placeholder
KEYWORDS: Photocatalysis; Hydrogen peroxide; Titanium dioxide; Peroxide; Surface chemistry
ABSTRACT: Photocatalytic production of hydrogen peroxide (H2O2) on semiconductor catalysts with alcohol as a hydrogen source
and molecular oxygen (O2) as an oxygen source has attracted much attention as a potential method for safe H2O2 synthesis, because
the reaction can be carried out without the use of explosive H2/O2 mixed gases. Early-reported photocatalytic systems with aliphatic
alcohol as a hydrogen source, however, produce only a few millimolar levels of H2O2. We found that benzylic alcohols, when used as a
hydrogen source for photoreaction in water with titanium dioxide (TiO2) photocatalyst, produces very high concentration of H2O2 (ca.
40 mM). Raman spectroscopy and electron spin resonance analysis revealed that the enhanced H2O2 formation is due to the efficient
formation of side-on coordinated peroxo species on the photoactivated TiO2 surface, via the reaction of benzylic alcohol and O2. The
peroxo species is readily transformed to H2O2, thus facilitating highly efficient H2O2 production.
INTRODUCTION
Ti3+(e–) + O2 → Ti4+−O−O●
(2)
Hydrogen peroxide (H2O2) is a clean oxidant that emits only
water as a byproduct and is widely used in industry for organic
synthesis, pulp bleaching, wastewater treatment, and disinfec-
tion.1 H2O2 is commercially produced by the anthraquinone
method, which has some non-green features such as high energy
utilization because of the multistep hydrogenation and oxida-
tion reactions. Recently, H2O2 production with H2 and O2 gases
The superoxo radical is transformed to a hydroperoxo species
via further reduction with e–. Protonation of the species pro-
duces H2O2.13,14
Ti3+(e–)−O−O● + H+ → Ti4+−O−OH
Ti4+−O−OH + H+ → Ti4+ + H2O2
(3)
(4)
−
4
has been studied extensively with Pd2 or Au−Pd bimetallic
These reactions proceed at atmospheric pressure and room
temperature without H2 gas and can be a clean and safe H2O2
synthesis. The amount of H2O2 produced, however, is very low
−
7
nanoparticle catalysts.5 This direct synthesis is potentially an
alternative process for H2O2 production from the viewpoint of
green and sustainable chemistry, but some care is necessary for
operation because of the potentially explosive nature of H2/O2
gas mixtures.
−
11
(<0.2 mM).8 Several kinds of TiO2 catalysts modified with
fluoride,15 or loaded with Au16 or Au–Ag alloy nanoparticles17
have been proposed so far; however, all of these systems still
produce only a few millimolar levels of H2O2. The low efficien-
cy for H2O2 production on the photoactivated TiO2 surface is
because the hydroperoxo species is very unstable and is easily
decomposed by further reduction with e− to inactive hydroxide
anion.12
Photocatalytic production of H2O2 with semiconductor tita-
−
11
nium dioxide (TiO2) has also attracted much attention.8 The
reaction is usually carried out by UV irradiation of TiO2 sus-
pended in an O2-saturated water with alcohol as electron and
proton donor. Photoexcitation of TiO2 produces electron (e−)
and positive hole (h+) pairs. The h+ oxidizes alcohol and pro-
duces aldehyde and protons.
Ti3+(e–)−O−OH + H+ → Ti4+−●OH + OH−
Ti3+(e–)−●OH → Ti4+ + OH−
(5)
(6)
R−CH2OH + 2h+ → R−CHO + 2H+
(1)
The e− on the TiO2 surface (surface Ti3+ site) reduces O2 and
produces a superoxo radical.12
A new approach is therefore necessary for efficient production
of H2O2 by photocatalysis.
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 8
1
2
3
4
5
6
7
All of the early-reported systems for photocatalytic H2O2 pro-
duction have employed aliphatic alcohols (e.g. ethanol) as the
highest value among the photocatalytic systems reported earli-
− −
11,15 17
er.8
−
−
11,15 17
electron and proton donor.8
Herein, we report that
Figure 1 shows the time-dependent change in the amounts of
substrate and products during photoirradiation of TiO2 with
alcohol and O2 in water. In the case with benzyl alcohol (Figure
1a), the selectivity for H2O2 formation is almost unchanged
(>30%) even after prolonged photoirradiation (~30 h), although
the rate of H2O2 formation decreases with photoirradiation
time due to the subsequent decomposition of H2O2 by UV
irradiation.18 In contrast, as shown in Figure 1b, photoreaction
with ethanol always shows much lower H2O2 selectivity (≤0.7%).
As shown by entries 6 and 7 (Table 1), substituted benzylic al-
cohols are also effective for H2O2 production. In addition, the
apparent quantum yield for H2O2 formation during photoreac-
tion with benzyl alcohol (ΦAQY (%) = [H2O2 formed × 2] / [pho-
ton number entering into the reaction vessel] × 100), deter-
mined with a 334 nm monochromatic light as a light source, is
29.1%, whereas the ΦAQY obtained during photoreaction with
ethanol is only 0.5%. These findings clearly suggest that
benzylic alcohols are indeed effective for efficient and selective
H2O2 production.
As reported,19 benzylic alcohols are strongly adsorbed onto
the TiO2 surface, and visible light excitation of the formed
charge-transferred complex promotes oxidation of alcohols.
However, as shown in Table 1 (entry 8), reaction with benzyl
alcohol and O2 on TiO2 under visible light irradiation (>420
nm) produces very small amount of H2O2, and the selectivity
for H2O2 formation (25%) is lower than that obtained under
UV irradiation (33%, entry 5). This suggests that band gap
photoexcitation of TiO2 promotes efficient and selective for-
mation of H2O2.
8
9
benzylic alcohols, when used for photoreaction in water with
TiO2 and O2, efficiently promote H2O2 formation. This
photocatalytic system produces very high concentration of H2O2
(ca. 40 mM), which is the highest concentration among the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
−
11,15
−
early reported systems.8
17 Raman spectroscopy and electron
spin resonance (ESR) analysis revealed that the enhanced H2O2
formation is because photoexcitation of TiO2 in water with
benzylic alcohols and O2 selectively produces side-on coordinated
peroxo species on the TiO2 surface. This peroxo species is readily
transformed to H2O2, resulting in efficient H2O2 production.
RESULTS AND DISCUSSION
Photocatalytic Activity for H2O2 Production. Photocatalytic
reactions were carried out as follows: water (5 mL) containing
TiO2 (50 mg) and each respective alcohol (350 mM) was
photoirradiated with magnetic stirring by a high-pressure Hg
lamp (λ >280 nm) with O2 (1 atm) at 298 K. Anatase TiO2
–
1
(JRC-TIO-1; BET surface area, 73 m2 g ; average particle size,
21 nm), kindly supplied from Catalyst Society of Japan, was
used as a catalyst. Table 1 summarizes the results obtained by
12 h photoreaction. With ethanol (entry 1), the H2O2 concen-
tration is 0.3 mM and the selectivity for H2O2 formation rela-
tive to the amount of alcohol consumed (= [H2O2 formed] /
[alcohol consumed] × 100) is only 0.4%. Other aliphatic alco-
hols are also ineffective for H2O2 formation (entries 2–4). In
contrast, as shown by entry 5, photoreaction with benzyl alco-
hol exhibits very high H2O2 selectivity (33%) and produces sig-
nificantly high concentration of H2O2 (40 mM), which is the
Table 1. Results for Photocatalytic H2O2 Production on TiO2 under Various Conditionsa
alcohol consumed
/ ꢀmol b
358.0
aldehyde or
H2O2
/ mM c
0.3
H2O2 selectivity
entry
alcohol
solvent
ketone / ꢀmol b
154.6
/ % d
0.4
1
2
ethanol
1-propanol
water
water
340.4
204.9
0.4
0.7
3
1-hexanol
water
453.4
123.5
0.3
0.3
4
5
6
7
8 e
9 f
10
11
12
13
14
a
2-phenylethanol
benzyl alcohol
p-nitrobenzyl alcohol
1-phenylethanol
benzyl alcohol
benzaldehyde
benzyl alcohol
benzyl alcohol
benzyl alcohol
benzyl alcohol
benzyl alcohol
water
water
water
water
water
771.4
257.2
8.4
5.4
604.1
587.9
628.8
85.5
1311.2 g
483.6
562.1
1284.8
420.1
198.1
518.5
397.4
53.0
39.6
33.1
40.2
4.3
32.8
28.2
32.0
25.1
water
8.6
CH3CN/water (3/7 v/v)
CH3CN/water (7/3 v/v)
CH3CN
278.3
305.6
673.6
123.3
554.9
26.1
26.7
21.1
1.6
27.0
23.8
8.2
1.9
0.8
DMF
benzotrifluoride (BTF)
1163.5
1.8
Reaction conditions: solvent (5 mL), alcohol (350 mM), catalyst (50 mg), O2 (1 atm), λ >280 nm (light intensity at 280–400 nm, 13.8
−
mW cm ), photoirradiation time (12 h). b Determined by GC. c Determined by redox titration with KMnO4 (detection limit: 0.05 mM). d
=
2
[H2O2 formed (ꢀmol)] / [alcohol consumed (ꢀmol)] × 100. Photoreaction was carried out under irradiation of visible light (λ >420 nm). f
e
Benzaldehyde (350 mM) was used as a starting material in place of alcohol. g The amount of benzaldehyde consumed.
ACS Paragon Plus Environment

Page 3 of 8
ACS Catalysis
1
2
3
4
5
6
7
8
a
40
30
20
10
0
100
50
0
b
40
30
20
10
0
100
50
0
9
10
11
12
13
14
15
16
17
18
19
20
21
22
0
6
12
18
24
30
0
6
12
18
24
30
benzyl alcohol
ethanol
1750
1500
1250
1000
750
400
300
200
100
0
1750
1500
1250
1000
750
400
300
200
100
0
CO2
CO2
benzaldehyde
acetaldehyde
H2O2
acetic acid
12 18
benzoic acid
12
H2O2
24
0
6
18
24
30
0
6
30
t / h
t / h
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Time-dependent change in (bottom) the amounts of substrates and products and (top) H2O2 selectivity and carbon balance, during
photoirradiation of TiO2 in water with O2 and (a) benzyl alcohol or (b) ethanol. Reaction conditions are identical to those in Table 1. The
carbon balance for (a) was calculated using the equation, C balance (%) = [benzaldehyde + benzoic acid + (CO2)/7] / [benzyl alcohol con-
sumed] × 100. The carbon balance for (b) was calculated using the equation, C balance (%) = [acetaldehyde + acetic acid + (CO2)/2] / [etha-
nol consumed] × 100.
Scheme 1. Proposed Mechanism for Photocatalytic Oxidation of Alcohols with O2 on the TiO2 surface.
It is also noted that anatase TiO2 is effective for this reaction,
as is observed for common photocatalytic reactions.20 As shown
in Table S1 (Supporting Information), rutile or Degussa P25
TiO2 [a mixture of anatase/rutile particles (ca. 80/20
wt/wt)],21,22 produces lower amount of H2O2 than anatase, alt-
hough the H2O2 selectivities are similar.
As shown in Table 1 (entry 9), benzaldehyde, when used as a
starting material for photoreaction with TiO2 and O2 for 12 h,
produces H2O2 (8.6 mM), which is much lower than that ob-
tained with benzyl alcohol (40 mM; entry 5). This clearly sug-
gests that photocatalytic oxidation of benzaldehyde or its pho-
toproducts is ineffective for H2O2 formation; H2O2 is efficiently
produced during photocatalytic oxidation of benzyl alcohol to
benzaldehyde.
As shown in Figure 1a, photocatalytic reaction with benzyl
alcohol produces benzaldehyde. This is decomposed by subse-
quent photocatalytic reaction and is finally converted to CO2
(mineralization).23 GC analysis of the solution detected only a
small amount of benzoic acid. The carbon balance for the iden-
tified products (benzaldehyde, benzoic acid, and CO2) is less
than 50%, indicating that nonvolatile or thermally-degradable
ring-opening products are involved as unidentified byproducts.
Mechanism for Enhanced H2O2 Production. Photocatalytic
oxidation of alcohols with O2 on TiO2 occurs as Scheme 1.
Photoexcited TiO2 (a) transfers e− of lattice oxygen to adjacent
Ti4+, creating a charge-separated state (b). The e−–h+ pairs re-
move α-hydrogen of alcohol24 and produce an alcoholate spe-
cies (c). Subsequent photoexcitation removes β-hydrogen of the
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 8
1
2
3
4
5
6
7
8
9
species24 (d), producing a carbon radical (e). The e− on the Ti3+
on the TiO2 surface,25 produced by heterolytic cleavage of O2-
site reduces O2 and produces a superoxo radical (f).12 In the case
with benzylic alcohol, the superoxo radical combines with adja-
cent carbon radical and produces O2-bridged complex (g).
Heterolytic cleavage of the complex25 produces the aldehyde
with exchanged O atom and side-on coordinated peroxo species
(h).26 Protonation of the side-on peroxo species produces H2O2
and completes the photocatalytic cycle. In contrast, with ali-
phatic alcohol, combination of carbon radical and superoxo
radical does not occur efficiently (f), and the aldehyde is re-
leased without exchange of O atom (i).13 The superoxo radical
left on the TiO2 surface (i) is transformed to a hydroperoxo
species (j) by further reduction with e− (eq. 3). This is decom-
posed to OH− by further reduction with e− (eqs. 5 and 6). The
above mechanisms indicate that photoexcitation of TiO2 with
benzylic alcohol and O2 specifically produces side-on peroxo
species (h), via the heterolytic cleavage of O2-bridged complex
(g). This peroxo species facilitates efficient and selective H2O2
production.
bridged complex (Scheme 1g → h). The heterolytic cleavage is
confirmed by the spectrum of the TiO2 recovered after photo-
reaction with benzyl alcohol and labeled molecular oxygen
(18O2) in water. As shown in Figure 2b, new band is observed at
−
1
870 cm , which is assigned to the 16O−18O stretching vibration.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
−
1
−
1
The isotopic shift (ꢁ = 24 cm ) is close to the shift (25 cm )
calculated based on Hooke′s law for a diatomic O−O stretch-
ing.28 These data clearly suggest that, as shown in Scheme 1,
photoreaction with benzylic alcohol produces O2-bridged com-
plex (g) and its heterolytic cleavage creates side-on peroxo spe-
cies (h).25 It is noted that, as shown in Figure 2b, photoreaction
of benzyl alcohol with 18O2 shows a weak signal at 894 cm−
1
assigned to the 16O−16O side-on peroxide species. This is proba-
bly because the 16O−18O side-on peroxide species formed on the
photoactivated TiO2 surface undergo O-exchange with the
H216O molecules, as observed for related water systems.29
In contrast, as shown in Figure 2c, the TiO2 particles recov-
ered after photoreaction with ethanol and O2 in water scarcely
exhibits 16O−16O band. This suggests that photoreaction with
ethanol scarcely produces side-on peroxo species, and the result
is consistent with the very low efficiency for H2O2 production
with ethanol (Table 1, entry 1). The above Raman spectroscopy
data suggest that the formation of side-on peroxo species
(Scheme 1h) is crucial for efficient H2O2 formation. The for-
mation of a large number of side-on peroxo species with
benzylic alcohol is explained by the stabilization of carbon radi-
cal (Scheme 1f) due to the electron delocalization on the adja-
cent aromatic ring.30 This facilitates efficient coupling between
the carbon radical and superoxo species (Scheme 1g), producing
large number of side-on peroxo species.
a
894
benzyl alcohol/16O2
797
800
830
950
900
850
750
b
870
894
benzyl alcohol/18O2
830
830
797
The stabilization of the carbon radical for benzylic alcohol
depends strongly on the solvent polarity; highly polarized water
is necessary for efficient formation of O2-bridged complex. Ta-
ble 1 (entries 10−14) summarizes the results for photoreaction
with benzyl alcohol in different organic solvents. All of these
systems show H2O2 selectivity lower than that obtained in water
950
900
850
850
800
797
750
750
c
ethanol/16O2
894
900
(entry 5). Figure 3 (●) shows the relationship between the em-
950
800
N 31,32
pirical solvent polarity parameter, ET ,
of the solvents and
Wavenumber / cm −1
the H2O2 selectivity obtained by photoreaction in the respective
solvents (Table 1, entries 5, 10−14). A linear relationship indi-
cates that H2O2 is selectively produced in more polar solvents.
Figure 3 (○) shows the relationship between the ETN of solvents
and the intensity of 16O−16O band for side-on peroxo species
(IO–O) obtained by Raman analysis of TiO2 after photoreaction
with benzyl alcohol in the respective solvents (Figure S1, Sup-
porting Information). The IO–O increases with an increase in
solvent polarity, and this tendency is consistent with the H2O2
selectivity (●). These data suggest that polar solvents indeed
enhance the formation of side-on peroxo species and produce
H2O2 more efficiently. The enhanced formation of side-on
peroxo species is probably due to the stabilization of carbon
radical in polar solvents30 (Scheme 1f). This enhances the for-
mation of O2-bridged complex (Scheme 1g) and creates larger
number of side-on peroxo species (Scheme 1h), resulting in
enhanced H2O2 production in water.
Figure 2. Raman spectra of the TiO2 recovered after photoreaction
in water with benzyl alcohol and (a) 16O2 or (b) 18O2, and the TiO2
(c) recovered after photoreaction in water with ethanol and 16O2.
The reaction conditions are identical to those in Table 1.
The formation of side-on peroxo species (h) during reaction
with benzylic alcohol is confirmed by Raman spectroscopy. The
TiO2 particles recovered after photoreaction (12 h) with alcohol
and O2 were used for analysis. Figure 2a shows the Raman spec-
trum of the TiO2 particles recovered after photoreaction with
benzyl alcohol and O2 in water. Deconvolution of the spectrum
−
−
1
1
exhibits three bands at 797, 830, and 894 cm . The 797 cm
−
1
band is assigned to a first overtone of B1g (397 cm ) mode of
anatase TiO2.27 The 830 cm− band is assigned to the C–O
1
stretching vibration of the alcoholate species adsorbed on the
TiO2 surface (Scheme 1c).25 The 894 cm− band is assigned to
1
the 16O−16O stretching vibration of the side-on peroxo species
ACS Paragon Plus Environment

Page 5 of 8
ACS Catalysis
1
2
3
4
water
1
5
6
7
8
the solution recovered after photoirradiation of TiO2 with ben-
zyl alcohol and O2 in the presence of 5,5-dimethyl-1-pyrroline
N-oxide (DMPO), a spin trapping reagent. With less polar
CH3CN as a solvent (Figure 4a), distinctive signals assigned to
CH3CN/water
= 7/3 (v/v)
30
20
10
0
β
the DMPO–O2●– spin adduct33 were observed (αN = 12.7 G; αH
9
10
11
12
13
14
15
16
17
18
19
= 8.5 G; g = 2.0065). This suggests that, as shown in Scheme 1i,
the superoxo radical is produced during photoreaction with
benzylic alcohol in less polar solvents and is trapped by the
reaction with DMPO.
CH3CN
DMF
0.5
CH3CN/water
= 3/7 (v/v)
●–
Ti4+−O−O● + DMPO → Ti4+ + DMPO−O2
(7)
BTF
The intensity of the spin adduct signal decreases with an in-
crease in the solvent polarity, and almost no signal is observed
in water (Figure 4c). In contrast, as shown in Figure 3 (○), the
formation of side-on peroxo species is enhanced with an in-
crease in the solvent polarity. These data suggest that, during
photoreaction with benzyl alcohol in less polar solvents
(Scheme 1), the carbon radical (f) is rapidly removed and leaves
superoxo radical (i), whereas the photoreaction in polar water
produces O2-bridged complex (g) and results in enhanced for-
mation of side-on peroxo species (h).
0
0.5
1
N
ET
N
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. (●) Relationship between ET of solvents and the H2O2
selectivity obtained by photoreaction with benzyl alcohol and O2
(12 h) in the respective solvents (Table 1). (○) Relationship between
N
ET of solvents and the intensity (IO–O) of side-on coordinated
–
1
peroxo species (894 cm ) on the TiO2 recovered after photoreac-
tion in the respective solvents (Figure S1, Supporting Information).
IO–O = [{intensity of 894 cm– band in respective solvents} ×
1
Figure 4d shows the ESR spectrum of water obtained after
photoreaction with ethanol and O2. The signals for DMPO–
{amount of benzyl alcohol consumed in water}] / [{intensity of 894
cm–1 band in water} × {amount of benzyl alcohol consumed in re-
spective solvents}].
●–
O2 spin adduct were observed (αN = 15.1 G; αHβ = 9.5 G; g =
2.0061), with signals assigned to DMPO–●OH spin adduct (αN
β
= 14.2 G; αH = 14.0 G; g = 2.011),34 which is produced by oxi-
dation of water on the photoactivated TiO2 (OH– + h+
→
ꢀ
−
DMPO−O2
●–
●OH).35 The formation of the DMPO–O2 spin adduct signal
suggests that, during photoreaction with aliphatic alcohol
(Scheme 1), the carbon radical (f) is rapidly removed and leaves
superoxo radical (i), resulting in very low efficiency for H2O2
formation. These data fully support the proposed mechanism
(Scheme 1). Photoexcitation of TiO2 with alcohol and O2 pro-
duces a set of carbon radical and superoxo radical (f). In water
with benzylic alcohol, the carbon radical is stabilized and trans-
formed to O2-bridged complex. This produces a large number
of side-on peroxo species and efficiently produces H2O2. In
contrast, with aliphatic alcohol or with benzylic alcohol in less
polar solvents, the carbon radical is unstable and is removed
rapidly. This leaves superoxo radical to be decomposed by sub-
sequent reaction, resulting in very low efficiency for H2O2 for-
mation. These findings suggest that benzylic alcohol and water
are necessary for efficient and selective H2O2 formation.
g = 2.0065
α
N = 12.7
αHβ = 8.5
a
benzyl alcohol in CH3CN
b
benzyl alcohol
in CH3CN/water (7/3 v/v)
c
benzyl alcohol in water
DMPO−ꢀ OH
g = 2.011
α
N = 14.2
αHβ = 14.0
d
ethanol in water
αHβ = 9.5
N = 15.1
α
g = 2.0061
ꢀ
−
DMPO−O2
CONCLUSION
3400
3420
Magnetic field / G
3440
We found that photoexcitation of anatase TiO2 in water with
benzylic alcohol and O2 efficiently promotes H2O2 formation.
This system successfully produces very high concentration of
H2O2 (ca. 40 mM), which is much higher than those of early
reported photocatalytic systems with aliphatic alcohols. The
enhanced H2O2 formation is due to the efficient formation of
side-on peroxo species produced via the reaction of benzylic
alcohols and O2 in water. The present H2O2 selectivity is ca.
33%, and further improvement of the selectivity is necessary.
Nevertheless, the concept presented here based on the selective
creation of side-on coordinated peroxo species on the TiO2
surface may contribute to the design of more efficient and selec-
Figure 4. ESR spectra (298 K) of DMPO spin adduct in the solu-
tions obtained by photoirradiation of TiO2 with benzyl alcohol and
O2 in (a) CH3CN, (b) CH3CN/water (7/3 v/v) mixture, and (c)
water, and (d) the solution obtained by photoirradiation of TiO2
with ethanol and O2 in water. Reaction conditions: solvent (5 mL),
TiO2 (50 mg), alcohol (1.75 mmol), DMPO (0.1 mmol), O2 (1 atm),
photoirradiation time (15 min).
The photoreaction mechanism summarized in Scheme 1 is
supported by ESR analysis. Figure 4 shows the ESR spectra of
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 8
1
2
3
4
5
6
7
8
tive photocatalytic systems for H2O2 production and may open
a new strategy towards clean and safe H2O2 synthesis without
H2 gas.
ASSOCIATED CONTENT
Raman spectra of TiO2 recovered after photoreaction with benzyl
alcohol in different solvents (Figure S1), results of photocatalytic
H2O2 production on various TiO2 (Table S1). This material is
available free of charge via the Internet at http://pubs.acs.org.
9
EXPERIMENTAL SECTION
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Photoreaction. TiO2 (50 mg) was suspended in a solution (5
mL) containing each respective alcohol (1.75 mmol) within a
glass tube (φ12 mm; capacity, 20 mL), and the tube was sealed
with a rubber septum cap. The catalyst was dispersed well by
ultrasonication for 5 min, and O2 was bubbled through the
solution for 5 min. The tube was photoirradiated at λ >280 nm
with magnetic stirring using a 450 W high pressure Hg lamp
(USHIO Inc.).17 Visible light (λ >420 nm) irradiation was car-
ried out with an aqueous NaNO2 (20 wt%) solution as a filter.36
The temperature of solution was kept at 298 ± 0.5 K with a
digitally-controlled water bath.21 The gas-phase product was
analyzed by GC-TCD (Shimadzu; GC-14B). The catalyst was
recovered by centrifugation, and the liquid-phase product was
analyzed by GC-FID (Shimadzu; GC2010A). The substrate and
product concentrations were calibrated with authentic samples.
H2O2 concentration was determined by the redox titration with
KMnO4.23
AUTHOR INFORMATION
Corresponding Author
shiraish@cheng.es.osaka-u.ac.jp
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
This work was supported by the Grant-in-Aid for Scientific
Research (No. 23360349) from the Ministry of Education,
Culture, Sports, Science and Technology, Japan (MEXT).
REFERENCES
(1) Campos-Martin, J. M.; Blanco-Brieva, G.; Fierro, J. L. G. Angew.
Chem., Int. Ed. 2006, 45, 6962–6984.
(2) Lunsford, J. H. J. Catal. 2003, 216, 455–460.
(3) Blanco-Brieva, G.; Cano-Serrano, E.; Campos-Martin, J. M.;
Fierro, J. L. G. Chem. Commun. 2004, 1184–1185.
(4) Melada, S.; Rioda, R.; Menegazzo, F.; Pinna, F.; Strukul, G. J.
Catal. 2006, 239, 422–430.
Quantum yield determination. Photoreaction was carried
out using a water (2 mL) containing alcohol (0.7 mmol) and
TiO2 (20 mg) within a glass tube (φ 12 mm; capacity, 20 mL).
After ultrasonication and O2 bubbling, the tube was
photoirradiated with stirring using a Xe lamp (USHIO Inc.)23
for 3 h, where the incident light was monochromated by a 334
nm band-pass glass filter (Asahi Techno Glass Co.; LX334). The
full-width at half-maximum (fwhm) of the light was 9 nm. The
photon number entered into the reaction vessel was deter-
mined with a spectroradiometer USR-40 (USHIO Inc.).37
(5) Choudhary, V. R.; Gaikwad, A. G.; Sansare, S. D. Angew. Chem.,
Int. Ed. 2001, 40, 1776–1779.
(6) Landon, P.; Collier, P. J.; Papworth, A. J.; Kiely, C. J.; Hutchings,
G. J. Chem. Commun. 2002, 2058–2059.
(7) Edwards, J. K.; Solsona, B.; Ntainjua, E.; Carley, A. F.; Herzing,
A. A.; Kiely, C. J.; Hutchings, G. J. Science 2009, 323, 1037–1041.
(8) Kormann, C.; Bahnemann, D. W.; Hoffmann, M. R. Environ. Sci.
Technol. 1988, 22, 798–806.
(9) Cai, R.; Kubota, Y.; Fujishima, A. J. Catal. 2003, 219, 214–218.
(10) Goto, H.; Hanada, Y.; Ohno, T.; Matsumura, M. J. Catal. 2004,
225, 223–229.
(11) Hirakawa, T.; Nosaka, Y. J. Phys. Chem. C 2008, 112, 15818–
15823.
(12) Nakamura, R.; Imanishi, A.; Murakoshi, K.; Nakato, Y. J. Am.
Chem. Soc. 2003, 125, 7443–7450.
(13) Hirakawa, T.; Koga, C.; Negishi, N.; Takeuchi, K.; Matsuzawa, S.
Appl. Catal. B 2009, 87, 46–55.
(14) Hirakawa, T.; Daimon, T.; Kitazawa, M.; Ohguri, N.; Koga, C.;
Negishi, N.; Matsuzawa, S.; Nosaka, Y. J. Photochem. Photobiol. A 2007,
190, 58–68.
(15) Maurino, V.; Minero, C.; Mariella, G.; Pelizzetti, E. Chem.
Commun. 2005, 2627–2629.
(16) Teranishi, M.; Naya, S.; Tada, H. J. Am. Chem. Soc. 2010, 132,
7850–7851.
(17) Tsukamoto, D.; Shiro, A.; Shiraishi, Y.; Sugano, Y.; Ichikawa,
S.; Tanaka, S.; Hirai, T. ACS Catal. 2012, 2, 599–603.
(18) Li, X.; Chen, C.; Zhao, J. Langmuir 2001, 17, 4118–4122.
(19) Higashimoto, S.; Kitao, N.; Yoshida, N.; Sakura, T.; Azuma, M.;
Ohue, H.; Sakata, Y. J. Catal. 2009, 266, 279–285.
(20) Tanaka, K.; Capule, M. F. V.; Hisanaga, T. Chem. Phys. Lett.
1991, 187, 73–76.
(21) Tsukamoto, D.; Shiraishi, Y.; Sugano, Y.; Ichikawa, S.; Tanaka,
S.; Hirai, T. J. Am. Chem. Soc. 2012, 134, 6309–6315.
(22) Sugano, Y.; Shiraishi, Y.; Tsukamoto, D.; Ichikawa, S.; Tanaka,
S.; Hirai, T. Angew. Chem. Int. Ed. 2013, 52, 5295–5299.
Raman spectroscopy. Raman spectra were measured on a
confocal Raman microscope (LabRAM HR-800, HORIBA).
YAG laser (532 nm line) was used as the excitation source,
where the laser power was 100 mW and the total data accumu-
lation time was 30 s. The samples were prepared as follows:
After photoreaction, TiO2 particles were recovered by centrifu-
gation and dried at room temperature in vacuo. They were
mounted on a microscope slide and subjected for analysis. The
–
1
peak intensities were normalized to the peak at 797 cm for a
first overtone of B1g mode for anatase TiO2.
ESR measurement. ESR spectra were recorded at the X-band
using a Bruker EMX-10/12 spectrometer with a 100 kHz mag-
netic field modulation at a microwave power level of 10.5 mW,
where microwave power saturation of the signals does not oc-
cur.22 The magnetic field was calibrated using a 1,1′-diphenyl-2-
picrylhydrazyl (DPPH) as standard. The measurement was car-
ried out as follows: TiO2 (50 mg) was suspended in a solution (5
mL) containing each respective alcohol (1.75 mmol) and
DMPO (0.1 mmol) within a glass tube (φ12 mm; capacity, 20
mL), and the tube was sealed with a rubber septum cap. After
ultrasonication (5 min) and O2 bubbling (5 min), the solution
was photoirradiated (λ >280 nm) for 15 min with magnetic
stirring. The catalyst was recovered by centrifugation, and the
resulting solution was subjected to analysis.
ACS Paragon Plus Environment

Page 7 of 8
ACS Catalysis
1
2
3
4
5
(23) Tsukamoto, D.; Ikeda, M.; Shiraishi, Y.; Hara, T.; Ichikuni, N.;
Tanaka, S.; Hirai, T. Chem. Eur. J., 2011, 17, 9816–9824.
(24) Shishido, T.; Miyatake, T.; Teramura, K.; Hitomi, Y.; Yamashita,
H.; Tanaka, T. J. Phys. Chem. C 2009, 113, 18713–18718.
(25) Zhang, M.; Wang, Q.; Chen, C.; Zang, L.; Ma, W.; Zhao, J.
Angew. Chem. Int. Ed. 2009, 48, 6081–6084.
(26) Jeske, P.; Haselhorst, G.; Weyhermüeller, T.; Wieghardt, K.;
Nuber, B. Inorg. Chem. 1994, 33, 2462–2471.
(27) Tompsett, G. A.; Bowmaker, G. A.; Cooney, R. P.; Metson, J.
B.; Rodgers, K. A.; Seakins, J. M. J. Raman Spectrosc. 1995, 26, 57–62.
(28) Tsilomelekis, G.; Boghosian, S. J. Phy. Chem. C 2011, 115,
2146–2154.
(29) Risley, J. M.; Van Etten, R. L. J. Am. Chem. Soc. 1981, 103,
4389–4392.
(30) Ito, O.; Matsuda, M. J. Am. Chem. Soc. 1982, 104, 568–572.
(31) Reichardt, C. Chem. Rev. 1994, 94, 2319–2358.
(32) Johnson, B. P.; Khaledi, M. G.; Dorsey, J. G. Anal. Chem. 1986,
58, 2354–2365.
(33) Wang, W.; Ng, T. W.; Ho, W. K.; Huang, J.; Liang, S.; An, T.;
Li, G.; Yu, J. C.; Wong, P. K. Appl. Catal. B 2013, 129, 482–490.
(34) Lin, F.; Zhang, Y.; Wang, L.; Zhang, Y.; Wang, D.; Yang, M.;
Yang, J.; Zhang, B.; Jiang, Z.; Li, C. Appl. Catal. B 2012, 127, 363–370.
(35) Hirakawa, T.; Nosaka, Y.; Langmuir 2002, 18, 3247–3254.
(36) Shiraishi, Y.; Sugano, Y.; Ichikawa, S.; Hirai, T. Catal. Sci.
Technol. 2012, 2, 400–405.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(37) Koizumi, H.; Shiraishi, Y.; Tojo, S.; Fujitsuka, M; Majima, T.;
Hirai, T. J. Am. Chem. Soc. 2006, 128, 8751–8753.
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 8
1
2
3
4
5
6
7
8
SYNOPSIS TOC
Selective Hydrogen Peroxide Formation by Titanium Dioxide Photocatalysis with Benzylic Alcohols and Molecular Oxygen in
Water
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Yasuhiro Shiraishi,* Shunsuke Kanazawa, Daijiro Tsukamoto, Akimitsu Shiro, Yoshitsune Sugano, and Takayuki Hirai
ACS Paragon Plus Environment