Effects of Surface Defects on Photocatalytic H2O2 Production by Mesoporous Graphitic Carbon Nitride under Visible Light Irradiation

Article abstract of DOI:10.1021/acscatal.5b00408

Photocatalytic production of hydrogen peroxide (H₂O₂) from ethanol (EtOH) and molecular oxygen (O₂) was carried out by visible light irradiation (λ > 420 nm) of mesoporous graphitic carbon nitride (GCN) catalysts with different surface areas prepared by silica-templated thermal polymerization of cyanamide. On these catalysts, the photoformed positive hole oxidize EtOH and the conduction band electrons localized at the 1,4-positions of the melem unit promote two-electron reduction of O₂ (H₂O₂ formation). The GCN catalysts with 56 and 160 m² g^-1 surface areas exhibit higher activity for H₂O₂ production than the catalyst prepared without silica template (surface area: 10 m² g^-1), but a further increase in the surface area (228 m² g^-1) decreases the activity. In addition, the selectivity for H₂O₂ formation significantly decreases with an increase in the surface area. The mesoporous GCN with larger surface areas inherently contain a larger number of primary amine moieties at the surface of mesopores. These defects behave as the active sites for four-electron reduction of O₂, thus decreasing the H₂O₂ selectivity. Furthermore, these defects also behave as the active sites for photocatalytic decomposition of the formed H₂O₂. Consequently, the GCN catalysts with relatively large surface area but with a small number of surface defects promote relatively efficient H₂O₂ formation. (Chemical Equation Presented).

Full text of DOI:10.1021/acscatal.5b00408

Subscriber access provided by University of Virginia Libraries & VIVA (Virtual Library of Virginia)
Article
Effects of Surface Defects on Photocatalytic H2O2 Production by
Mesoporous Graphitic Carbon Nitride under Visible Light Irradiation
Yasuhiro Shiraishi, Yusuke Kofuji, Hirokatsu Sakamoto,
Shunsuke Tanaka, Satoshi Ichikawa, and Takayuki Hirai
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b00408 • Publication Date (Web): 09 Apr 2015
Downloaded from http://pubs.acs.org on April 12, 2015
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 9
ACS Catalysis
1
2
3
4
5
6
7
8
9
Effects of Surface Defects on Photocatalytic H O Production
2
2
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
by Mesoporous Graphitic Carbon Nitride under Visible Light
Irradiation
,†,‡
†
†
ǁ
§
Yasuhiro Shiraishi,* Yusuke Kofuji, Hirokatsu Sakamoto, Shunsuke Tanaka, Satoshi Ichikawa,
†
and Takayuki Hirai
†
Research Center for Solar Energy Chemistry, and Division of Chemical Engineering, Graduate School of Engineering
Science, Osaka University, Toyonaka 560ꢀ8531, Japan
‡
PRESTO, JST, Saitama 332ꢀ0012, Japan
ǁ Department of Chemical, Energy and Environmental Engineering, Kansai University, Suita 564ꢀ8680, Japan
§
Institute for NanoScience Design, Osaka University, Toyonaka 560ꢀ8531, Japan
Supporting Information Placeholder
ABSTRACT: Photocatalytic production of hydrogen peroxide (H O ) from ethanol (EtOH) and molecular oxygen (O ) was carried
2
2
2
out by visible light irradiation (λ >420 nm) of mesoporous graphitic carbon nitride (GCN) catalysts with different surface areas,
prepared by silicaꢀtemplated thermal polymerization of cyanamide. On these catalysts, the photoformed positive hole oxidize EtOH,
and the conduction band electrons localized at the 1,4ꢀpositions of the melem unit promote twoꢀelectron reduction of O (H O
2
2
2
2
–1
formation). The GCN catalysts with 56 and 160 m g surface areas exhibit higher activity for H O production than the catalyst
2 2
2
–1
2 –1
prepared without silica template (surface area: 10 m g ), but further increase in the surface area (228 m g ) decreases the activity.
In addition, the selectivity for H formation significantly decreases with an increase in the surface area. The mesoporous GCN
with larger surface areas inherently contain a larger number of primary amine moieties at the surface of mesopores. These defects
behave as the active sites for fourꢀelectron reduction of O , thus decreasing the H selectivity. Furthermore, these defects also
behave as the active sites for photocatalytic decomposition of the formed H O . Consequently, the GCN catalysts with relatively
2 2
O
2
2 2
O
2
2
large surface area but with small number of surface defects promote relatively efficient H O formation.
2
2
KEYWORDS: Photocatalysis; Hydrogen peroxide; Graphitic carbon nitride, Visible light; Oxygen reduction
+
+
INTRODUCTION
The h oxidize alcohol and produce aldehyde and H (eq. 1),
–
while the e promote twoꢀelectron reduction of O and produce
Hydrogen peroxide (H
emits only water as a byproduct and is widely used for pulp
bleaching, disinfection, and organic synthesis. H
attracted much attention as a new energy carrier for fuel cells,
2
O
2
) is a versatile clean oxidant that
2
H
2
O
2
(eq. 2).
R−CH
O + 2H + 2e → H O
2
1
2
O
2
has also
+
+
2
OH + 2h → R−CHO + 2H
(1)
(2)
+
−
alternative to H , because it is waterꢀsoluble and can be used
2
2
2
2
in an oneꢀcompartment cell for electricity generation. H O is
currently manufactured in industry by the anthraquinone
method that needs highꢀenergyꢀconsuming twoꢀstep oxidation
2
2
The reaction proceeds at ambient temperature without H
is potentially a safe and sustainable H synthesis. Its
efficiency is, however, very low since the selectivity for the
and
2
O
2
2
3
and hydrogenation reactions on Pdꢀbased catalysts. Recently,
H
2
O
2
synthesis with H
2
and O
2
has been studied extensively
amount of H
consumed is ~6%.
of O
2
O
2
formed relative to the amount of alcohol
This is because oneꢀelectron reduction
[superoxide ( OOH) radical formation, eq. 3] and
4
5
9−11
with Pd or Au–Pd bimetallic catalysts. This direct synthesis
quantitatively produces H O , but requires extreme care due to
●
2
2
2
the potentially explosive nature of the H /O mixed gases. A
new catalytic process capable of producing H O without H is
therefore desired.
fourꢀelectron reduction of O
predominantly on the catalysts. These suppress twoꢀelectron
reduction of O
2
(water formation, eq. 4) occur
2
2
2
2
2
(eq. 2), resulting in very low H
O
2
selectivity.
2
2
Photocatalytic H O production on semiconductor catalysts
+
–
●
2
2
O
2
+ H + e → OOH
(3)
(4)
6–8
such as titanium dioxide (TiO ) has also been studied. The
2
+
–
reactions are carried out by UV irradiation (λ <400 nm) of an
O + 4H + 4e → 2H O
2 2
O ꢀsaturated water with catalyst in the presence of electron
2
Earlier, we found that graphitic carbon nitride (GCN), a
metalꢀfree polymeric semiconductor with a graphitic stacking
and proton donor such as alcohols. Photoexcitation of the
+
−
catalyst produces the positive hole (h ) and electron (e ) pairs.
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 9
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2,13
structure of melem sheets,
promotes selective twoꢀelectron
literatures revealed that a variety of photoreactions such as H
2
30,31 32 33
14,15
reduction of O2 and efficiently produces H O .
light irradiation (λ >420 nm) of GCN with a small surface area
10 m g [GCN(10) catalyst], prepared by a simple thermal
polymerization of cyanamide, produces H O with ca. 90%
selectivity. Raman spectroscopy, electron spin resonance
ESR), and ab initio calculation revealed that the selective
twoꢀelectron reduction of O2 is ascribed to the efficient
formation of 1,4ꢀendoperoxide species on the melem unit. As
shown in Scheme 1, the photoformed e are localized at the
,4ꢀpositions of melem unit (a). The e reduces O and creates
a superoxo radical (b). This is rapidly reduced by another e at
the para position and produces 1,4ꢀendoperoxide species (c),
which is readily transformed to H
of 1,4ꢀendoperoxide (b→c) may suppresses oneꢀelectron
Visible
evolution,
alcohol oxidation, CO2 reduction , radical
34 22 35
2
2
polymerization, air purification, and pollutant degradation
proceed efficiently on the mesoporous GCN with increased
surface areas as compared to the nonporous GCN.
2
–1
12
2
2
In the present work, we synthesized GCN(x) catalysts with
2
−1
(
different surfaces areas [x (m g ) = 56, 160, and 228] by the
30,31
silicaꢀtemplated polymerization of cyanamide
and used
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
them for photocatalytic H O production with ethanol (EtOH)
2
2
–
as an electron and proton donor under visible light irradiation
(λ >420 nm). Similar photocatalytic system for alcohol
oxidation on mesoporous GCN has already been reported by
–
1
2
–
32
other group; however, there is no mention of H O formation.
2
2
2 2
O . The efficient formation
In the present study, photocatalytic activity for "EtOH
oxidation" increases with an increase in the surface area of the
●
3
2
reduction of O
reduction of O2 (water formation, eq. 4), thus promoting
selective twoꢀelectron reduction of O (eq. 2).
2
( OOH formation, eq. 3) and fourꢀelectron
catalysts, as reported. We found that GCN(56) and (160)
catalysts exhibit higher activity for "H O formation" than
2
2
2
nonporous GCN(10), but GCN(228) shows decreased activity.
In addition, the selectivity for H formation significantly
2 2
O
decreases on the catalysts with larger surface areas. ESR,
Xꢀray photoelectron spectroscopy (XPS), electrochemical
analysis, and ab initio calculation revealed that the catalysts
with larger surface areas contain a larger number of primary
amine moieties at the surface of mesopores. These surface
defects behave as active sites for fourꢀelectron reduction of O2
(
2 2
eq. 4), thus decreasing the H O selectivity. In addition, these
defects also behave as active sites for reductive decomposition
of the formed H O . Consequently, a GCN catalyst with a
2
2
relatively large surface area but with a small number of
surface defects promotes relatively efficient H O formation.
2
2
RESULTS AND DISCUSSION
Preparation and characterization of catalysts
Nonporous GCN(10) catalyst was obtained by calcination of
12
cyanamide at 823 K for 4 h under N as yellow powders. The
2
respective mesoporous GCN(56), (160), and (228) catalysts
Scheme 1. Proposed mechanism for selective formation of
H O on the photoactivated GCN surface.
30
were prepared according to literature procedure: calcination
2
2
of a mixture of cyanamide and different amounts of a Ludox
HS40 solution containing ca. 12 nm silica particles under N
823 K for 4 h followed by washing with a 4 M NH HF
2
2
at
The purpose of the present work is to improve the catalytic
activity of GCN while maintaining high H O selectivity. It is
well known that photocatalytic activity of semiconductor
materials increases with an increase in their surface area.
Some post synthesis methods for the surface area enlargement
of GCN have been proposed such as exfoliation of melem
4
solution at 298 K for 24 h produced yellow powders of the
catalysts.
2
2
16−18
Table 1 shows the properties of GCN catalysts. As shown in
Figure S1 (Supporting Information), N adsorption/desorption
2
analysis of mesoporous GCN shows typical typeꢀIV isotherm,
1
9
20,21
sheets by thermal oxidation etching, ultrasonication,
scission of the parts of the C–N bonds by chemical or thermal
and
although nonporous GCN(10) shows almost no hysteresis loop.
As shown in Figure 1, transmission electron microscopy
(TEM) images of GCN(228) clearly exhibit mesopores,
22,23
treatments.
More popular way is the direct synthesis of
GCN with large surface areas. Several methods have been
proposed such as creation of mesoporous GCN by thermal
polymerization of different C, Nꢀcontaining precursors such as
urea, thiourea, and guanidine derivatives
GCN nanowires, nanorods, or nanofibers by the treatment of
followed by thermal polymerization.
The most simple and widelyꢀaccepted way is the creation of
mesoporous GCN based on silicaꢀtemplating method reported
indicating that the silicaꢀtemplated polymerization indeed
13,30
creates mesosphere pore structure.
As shown in Figure S2
(Supporting Information), Xꢀray diffraction (XRD) patterns of
2
4−26
and creation of
the catalysts shows distinctive diffraction at 2θ = 27.4 (d =
12,13
0.325 nm), assigned to the (002) packing of melem sheets.
27−29
melamine with HNO
3
In that, mesoporous GCN show lower peak intensity than the
nonporous GCN(10) due to their low crystallinity by the
polymerization with a silica template, as observed in the
3
0,31
36
by Wang et al.
thermal polymerization of cyanamide with silica particles (ca.
2 nm) as a template followed by removal of the particles by
washing with base. This produces mesoporous GCN with very
They synthesized mesoporous GCN by
related system. Figure S3 (Supporting Information) shows
the diffuse reflectance UVꢀvis spectra of the catalysts.
Mesoporous GCN show spectra similar to those of nonporous
GCN at λ >420 nm, but show stronger absorption at λ <420
1
2
–1
large surface area (up to 380 m g ), which can easily be
controlled by the amount of silica particles added. Several
nm due to the multiple light scattering by the mesoporous
30,37
structure.
Tauc plot of the absorption data revealed that the
ACS Paragon Plus Environment

Page 3 of 9
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
band gap energies of mesoporous GCN are very similar to
those of the nonporous GCN (ca. 2.57–2.62 eV) regardless of
the presence of mesopores.
polymerization therefore creates primary amine moieties on
the surface. Subsequent removal of silica by the dissolution
with NH HF therefore left these moieties on the surface of
4
2
mesopores. These surface defects behave as the active sites for
fourꢀelectron reduction of O (water formation, eq. 4) and,
hence, decrease the selectivity for twoꢀelectron reduction of O
on the 1,4ꢀpositions of melem units (H formation, eq. 2).
Photocatalytic activity
2
Photocatalytic H O production was carried out with EtOH
2
2
14
2
as the electron and proton donor. An EtOH/water (9/1 v/v)
mixture (5 mL) containing each respective catalyst (20 mg)
was photoirradiated at λ >420 nm by a Xe lamp with magnetic
2 2
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
stirring under O (1 atm) at 298 K. All of the systems produce
CH CHO as the main oxidation product of EtOH with only
3
minor amounts of CH
3
COOH and CO
2
, as is the case for
nonporous GCN(10). Figure 2a shows the timeꢀdependent
change in the amount of CH CHO during photoreaction on the
1
4
3
respective catalysts. The GCN catalysts with larger surface
area exhibit higher activity for CH CHO formation, indicating
3
30−35
that, as reported by several literatures,
enlarged surface
area of GCN indeed enhances photocatalytic cycles. Figure 2b
shows the time profiles for H formation during the reaction.
GCN(56) and (160) catalysts exhibit higher activity for H O
2 2
O
2
2
formation than GCN(10), but GCN(228) is ineffective. Figure
2c shows the time profiles for the selectivity of the amount of
H O formed relative to the amounts of oxidation products of
2
2
EtOH {= [H
2 2 3 3 2
O ]/[CH CHO + CH COOH + (CO )/2] × 100}.
14
In the case of GCN(10), as we reported earlier, the H O
2
2
selectivity at the early stage of reaction (3 h) is 92% and
scarcely changes even after prolonged photoirradiation (~24 h).
In contrast, the H O selectivities on mesoporous GCN at 3 h
2 2
are much lower and decrease with an increase in the surface
area: 92% for GCN(10) > 72% for GCN(56) > 58% for
GCN(160) > 42% for GCN(228). This clearly indicates that
the selectivity for twoꢀelectron reduction of O decreases with
2
an increase in the surface area of catalysts.
Surface defects on the catalysts
2 2
The decrease in the H O selectivity associated with an
increase in the surface area of catalysts (Figure 2c) is ascribed
to the formation of a large number of primary amine moieties
on the surface of mesopores. As schematically shown in
Scheme 2a, this originates from the polymerization of
cyanamide at the silica surface. Acidic silica surface attracts
38
the amine moieties of cyanamide molecules. Thermal
Figure 1. Typical TEM images of GCN(228) catalyst.
Table 1. Properties of Respective Catalysts.
e
composition of N atoms / %
a
b
p
c
p
3
d
S
BET
D
/ nm
V
/ cm g
E
bg
/ eV
catalyst
2 –1
m g
–1
/
Npyridine
Ncenter
Ntertiary
Nsecondary
Nprimary
GCN(10)
GCN(56)
GCN(160)
GCN(228)
10
56
–
–
2.62
2.61
2.57
2.59
71.6
11.9
13.9
2.0
0.6
15.4
11.2
9.9
0.18
0.39
0.54
70.3
70.1
70.7
11.7
11.7
11.8
12.2
9.1
3.9
6.5
1.9
2.6
4.0
160
228
5.4
8.1
a
b
Brunauer-Emmett-Teller (BET) surface area. Barrett-Joyner-Halenda (BJH) adsorption pore size. BJH adsorption pore
c
d
volume. Band gap energies determined by a plot of the Kubelka–Munk function versus the energy of light absorbed (Figure S3,
e
2
Supporting Information). Determined by integration of N1s XPS charts (Figure 3) and CO -TPD profiles (Figure 4).
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 9
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
b 1
c
00
100
50
0
x = 228
x = 56
x = 160
x = 228
x = 10
200
x = 160
x = 56
x = 56
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50
x = 160
100
x = 10
x = 10
x = 228
0
0
0
6
12
18
24
0
6
12
18
24
0
6
12
18
24
t / h
t / h
t / h
Figure 2. Time-dependent change in the amounts of (a) CH
on the respective GCN(x) catalysts. Photoirradiation was carried out with a 2kW Xe lamp (light intensity at 420–500 nm was 26.9
3
CHO and (b) H
2
O
2
, and (c) H
2
O
2
selectivity during photoreaction
–2
2 2 2 2 3 3 2
W m ). H O selectivity (%) = [H O ] / ([CH CHO] + [CH COOH] + [CO ] /2) × 100.
Scheme 2. (a) Proposed mechanism for the formation of surface defects on the walls of mesopores. (b) Different N atoms on GCN.
The formation of surface defects on the mesoporous GCN is
confirmed by XPS analysis. Figure 3 show the XPS charts for
N1s level of respective GCN catalysts. All of the charts can be
deconvoluted into the three components. As reported and as
shown in Scheme 2b, these components are assigned to sp
hybridized N atoms of melem units (Npyridine) at 398.7 eV
(pink), trigonal N atoms of melem center (Ncenter) and tertiary
amine N atoms (Ntertiary) of melem terminal at 400.2 eV (blue),
and primary (Nprimary) and secondary amine (Nsecondary) N atoms
of melem terminal at 401.3 eV (green). The respective charts
clearly show that contribution of the green component (Nprimary
from the Nprimary sites increases with an increase in the surface
area of catalysts. This indicates that GCN with larger surface
areas indeed possess larger number of Nprimary sites. The
composition of respective N atoms on the GCN catalysts
(Scheme 2b) can be determined by the integration of the
deconvoluted components on the XPS (Figure 3) and
39
2
2
CO ꢀTPD data (Figure 4) and are summarized in Table 1. The
composition of primary amine moieties (Nprimary) on the
nonporous GCN(10) is only 0.6%, but the mesoporous GCN
with larger surface area have larger number of Nprimary [up to
4% for GCN(228)]. The results indicate that, as shown in
Scheme 2a, primary amine moieties are indeed produced on
the surface of mesoporous GCN catalysts.
+
Nsecondary) increases with an increase in the surface area of
catalysts. This suggests that, as shown in Scheme 2a, thermal
polymerization of cyanamide in the presence of silica particles
indeed creates a large number of primary and secondary amine
moieties.
Properties of one-electron reductions of O2
The lower H O selectivity on the mesoporous GCN is
because they promote fourꢀelectron reduction of O
formation, as eq. 4). Oneꢀelectron reduction of O ( OOH
radical formation, eq. 3) is not involved in the selectivity
decrease. This is confirmed by ESR analysis with 5,5ꢀdimethyl
2
2
2
(water
●
The formation of a large number of primary amine moieties
on the GCN catalysts with larger surface areas is confirmed by
2
temperatureꢀprogrammed desorption (TPD) analysis with CO2.
Figure 4 shows the CO
All of the profiles can be deconvoluted into three desorption
peaks, assigned to the CO molecules desorbed from Npyridine
174 °C, pink), Nsecondary (224 °C, cyan), and N sites
2
ꢀTPD profiles for respective catalysts.
1
ꢀpyrroline Nꢀoxide (DMPO) as a spin trapping reagent. An
EtOH/water (v/v 9/1) mixture (5 mL) was photoirradiated with
respective GCN catalysts (20 mg) and DMPO (0.1 mmol).
Figure 5 shows the ESR spectra of the solutions recovered
2
(
(
primary
4
0
301 °C, purple), respectively. The amount of CO desorbed
2
ACS Paragon Plus Environment

Page 5 of 9
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
after photoreaction for 3 min. All of the solutions exhibit
g = 2.0067
●
distinctive signals assigned to the DMPO– OOH spin adduct
αN = 13.5
αHβ = 9.8
β
H
14,41
(αN = 13.5 G; α = 9.8 G, g = 2.0067).
The signal
intensities on the respective catalysts are very similar,
●
indicating that similar numbers of OOH radicals are produced
x = 228
x = 160
x = 56
on these catalysts. The results suggest that oneꢀelectron
reduction of O (eq. 3) is not involved in the decreased H O
2 2 2
selectivity on the mesoporous GCN catalysts.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Npyridine
398.7
x = 10
x = 56
x = 10
Ncenter+ Ntertiary
400.2
Nprimary+ Nsecondary
3
400
3420
3440
3460
401.3
Magnetic field / G
396
398
400
402
396
398
400
402
Figure 5. ESR spectra measured at 298 K for the solutions
recovered after photoirradiation of respective GCN(x)
catalysts in an EtOH/water/O system. Photoirradiation was
2
performed for 3 min under the conditions identical to those
in Figure 2 with DMPO (0.1 mmol).
Binding energy / eV
Binding energy / eV
x = 160
x = 228
Properties of four-electron reduction of O2
The lower H O selectivity on mesoporous GCN is ascribed
to the enhanced fourꢀelectron reduction of O . Electrochemical
2
analysis with a rotating disk electrode confirms this. Figure 6a
shows the linear sweep voltammograms of GCN(10) and (228)
catalysts measured on a rotating disk electrode in a buffered
2
2
396
398
400
402
396
398
400
402
Binding energy / eV
Binding energy / eV
aqueous solution (pH 7) under O
2
atmosphere at different
Figure 3. XPS charts for N1s levels of GCN(x) catalysts. Black
line is the obtained chart, and gray line is sum of the
deconvoluted components. The respective components are
represented in Scheme 2b.
42,43
rotating speed.
The diagnostic Koutecky–Levich plots of
the data obtained at the constant electrode potential (–0.4 V)
are illustrated in Figure 6b, and the slope of the plots obtained
by linear regression was used to estimate the average number
4
4,45
Npyridine
174
of electrons (n) involved in the overall reduction of O
The plots were interpreted on the following equations, where j
is the measured current density, j is the kinetic current density,
and ω is the electrode rotating speed (rpm), respectively:
2
.
x = 10
x = 56
Nsecondary
k
2
24
Nprimary
301
–
1
= jk–1 + B ω
–1 –1/2
j
(5)
(6)
–1/6
2/3
100
200
300
400
100
200
300
400
B = 0.2nFν CD
Temperature / °C
Temperature / °C
–
1
F is the Faraday constant (96485 C mol ), ν is the kinetic
2
–1
viscosity of water (0.01 cm s ), C is the bulk concentration of
–
6
–3
x = 160
x = 228
O in solution (1.3 × 10 mol cm ), and D is the diffusion
2
–5 2 –1 46
2 k
coefficient of O (2.7 × 10 cm s ), respectively. The j
values obtained on the GCN(10) and GCN(228) catalysts
determined by the intercept of the Koutecky–Levich plots are
.79 and 1.19 mA cm , respectively. The larger j
GCN(228) indicates that O is reduced more efficiently due to
the larger surface area. The n value for GCN(10), determined
by the slope, is 2.07, indicating that the GCN catalyst with a
small surface area indeed selectively promotes twoꢀelectron
–2
0
k
value of
2
4
7
100
200
300
400
100
200
300
400
Temperature / °C
Temperature / °C
48
reduction of O (n = 2). This is consistent with high H O
2
2
2
Figure 4. CO -TPD profiles for respective GCN(x) catalysts.
2
selectivity (ca. 90%) obtained by photoreaction experiments
(Figure 2c). In contrast, the n value for GCN(228) is 2.72,
which is much larger than that of GCN(10). This indicates that
the GCN catalyst with a larger surface area indeed promotes
Black line is the obtained profile, and gray line is sum of the
deconvoluted components. The respective components are
represented in Scheme 2b.
fourꢀelectron reduction of O (n = 4). This is also consistent
2
with the decreased H
Figure 2c).
2 2
O selectivity during photoreaction
(
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 9
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
4
twoꢀelectron reduction of O2. In contrast, on the mesoporous
GCN catalysts with larger surface areas, a large number of
primary amine moieties on the melem terminal behave as the
a
0
active sites for fourꢀelectron reduction of O . This results in
decreased H O selectivity. This is confirmed by ab initio
2 2
calculations of the conjugated melem models (Figure 7) based
on the timeꢀdependent density functional theory (TDꢀDFT),
performed within the Gaussian 03 program. The electronic
transitions of all of the single, double, and triple melem
2
−
0.5
GCN(10
)
Rotating speed
・400 rpm
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
・800 rpm
models are mainly contributed by HOMO → LUMO (S →
0
・
・
1200 rpm
1600 rpm
−
1
S1) transitions. Interfacial plots for the respective orbitals are
summarized in Figure 7. The electrons on HOMO for the
double and triple melem models are located at the
2,6ꢀpositions of the melem unit. This suggests that these
nitrogen atoms on the melem units behave as the oxidation
GCN(228)
−
1.5
15
−0.5
0
sites, as is the case for single melem model.
E / V vs. Ag/AgCl
In the case of single melem model (a), as we reported
15
b
earlier, the LUMO electrons are located at the 1,4ꢀpositions
of the melem. This suggests that conduction band electrons are
indeed localized on these atoms, and they behave as the sites
1
1
1
.6
.4
.2
1
GCN(10)
n = 2.07
Jk = 0.795
for twoꢀelectron reduction of O . In this case, LUMO electrons
2
are also located at the terminal primary amine moiety. This
indicates that conduction band electrons are also localized on
these sites, and they also behave as the reduction sites. In the
case of double melem model (b), LUMO electrons are also
located at the 1,4ꢀpositions of melem and the terminal primary
amine moieties. In this case, the secondary amine moiety
situated between two melem moieties does not possess
electron distribution. This suggests that the secondary amine
moiety does not act as the reduction sites. In the case of triple
melem model (c), LUMO electrons are also not located at the
tertiary amine moiety. The above facts involving (i) large
distribution of LUMO electrons onto the primary amine
2
e– process
e– process
4
GCN(228)
n = 2.72
Jk = 1.19
0.02
0.04
0.06
ω–1/2 / rpm–1/2
2 2
moieties, and (ii) the decrease in H O selectivity with an
increase in the amount of primary amine moiety, strongly
suggest that the primary amine moieties on the surface of
mesopores behave as the active sites for fourꢀelectron
2
. As a result of this, GCN catalysts with larger
surface area exhibits decreased H O selectivity.
Figure 6. (a) Linear-sweep voltammograms of GCN(10) and
GCN (228) catalysts measured on a rotating disk electrode at
reduction of O
different rotating speed. (b) The Koutecky–Levich plots of
the data obtained at the constant electrode potential (–0.4
V).
2
2
Photocatalytic decomposition of H O
2
2
As shown in Figure 2c, during the photocatalytic H O
production, the H O selectivity obtained on the GCN(10)
2 2
2
2
4
9
Recently, Zheng et al. performed electrochemical analysis
for O reduction on the mesoporous GCN with a large surface
area (250 m g ), prepared by thermal polymerization of
cyanamide in the presence of a SBAꢀ15 mesoporous silica as a
template followed by the removal of template by washing with
catalyst scarcely changes even after prolonged irradiation (ca.
90%). In contrast, the H O selectivity on the mesoporous
2
2
2
2
−1
GCN catalysts decreases with the irradiation time, and the
decrease is more apparent for the catalysts with larger surface
areas. The results clearly indicate that the mesoporous GCN
with larger surface areas promote subsequent decomposition
of the formed H O . Stirring the H O solutions in the dark
2 2 2 2
condition with respective catalysts or under photoirradiation
without catalyst under N atmosphere scarcely changes the
H O amount. This suggests that H O is photocatalytically
decomposed on the mesoporous GCN. As reported for several
semiconductor photocatalytic systems,
NH
similar to that obtained in the present study. This supports the
enhanced fourꢀelectron reduction of O on GCN with a large
surface area. The change in multiꢀelectron reduction properties
4 2
HF . This material also exhibits a large n value (2.6),
2
2
of O is associated reasonably well with an increase in the
2
2
2
2
2
number of primary amine moieties (Nprimary) on the GCN
catalysts. These findings therefore imply that these defects
sites may contribute to the enhanced fourꢀelectron reduction of
4
4,50
2 2
H O is reductively
decomposed by the donation of conduction band electrons as
follows:
O2.
Photoreduction properties of the surface defects
+
–
H
2
O
2
+ 2H + 2e → 2H
2
O
(7)
As shown in Scheme 1, conduction band electrons formed
on the nonporous GCN catalyst with a small surface area are
localized at the 1,4ꢀpositions of the melem unit. They reduce
and rapidly produce 1,4ꢀendoperoxide species, which is
readily transformed to H O . This thus promotes selective
O
2
2
2
ACS Paragon Plus Environment

Page 7 of 9
ACS Catalysis
1
2
3
a
b
c
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
LUMO
LUMO
LUMO
ELUMO = –1.03 eV
ELUMO = –1.82 eV
ELUMO = –1.74 eV
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4.98 eV
4.25 eV
4.3 eV
HOMO
HOMO
HOMO
EHOMO = –6.01 eV
EHOMO = –6.12 eV
EHOMO = –5.99 eV
Figure 7. Interfacial plots of main orbitals for (a) single, (b) double, and (c) triple melem-conjugated models, calculated at the
DFT level (B3LYP/6-31G(d)).
The GCN(10) catalyst (20 mg) was added to an EtOH/water
mixture (9/1 v/v, mL) with 200 µmol H O and
photoirradiated (λ >420 nm) under N atmosphere. As shown
in Figure S4 (Supporting Information), photoirradiation leads
to a decrease in the H O amount. As shown in Table 1, the
GCN(10) catalyst contains very small amount of primary
amine moieties. This indicates that the 1,4ꢀpositions of the
melem unit behave as the active sites for photocatalytic
decomposition of H O (eq. 7). However, as shown in Figure
of primary amine moieties is very important for efficient
photocatalytic H O production.
5
2
2
2
2
2
EXPERIMENTAL SECTION
2
2
GCN(10). Cyanamide (9.0 g) was added to a porcelain cup
and calcined under N flow at 823 K for 4 h with the heating
2
−1
rate being 2.3 K min . Grinding of the resultant gave yellow
powders of GCN(10).
GCN(x) [x = 56, 160, and 228]. Cyanamide (3.0 g) was
mixed with different amount of a Ludox HS40 solution (40%
solution of ca. 12 nm silica particles; 3.6, 7.5, and 11.3 g,
respectively) and stirred at 333 K for 12 h. The resultant was
calcined under N flow at 823 K for 4 h with the heating rate
being 2.3 K min . The resulting powders were stirred in a
NH HF solution (4 M) at 298 K for 24 h to remove the silica
template particles. The obtained powders were recovered by
centrifugation and washed thoroughly with water until the pH
of the solution became ca. 7.0. They were then washed with
EtOH and dried at 343 K in vacuo for 12 h.
Photoreaction. Each respective catalyst (20 mg) was added
to an EtOH/water mixture (9/1 v/v, 5 mL) within a Pyrex glass
tube (φ 12 mm; capacity, 20 mL), and the tube was sealed with
a rubber septum cap. The catalyst was dispersed well by
2
2
2c, H O selectivity on the GCN(10) catalyst scarcely changes
even after prolonged photoirradiation. This means that, during
photoirradiation with O , the 1,4ꢀpositions of melem units
2
2
2
predominantly reduces O
2 2 2
and scarcely decomposes H O . The
significant decrease in the H O selectivity with time on the
2
2
2
−1
mesoporous GCN catalysts is therefore probably due to the
photocatalytic decomposition of H O on the primary amine
moieties. On these sites, basic primary amine moieties
attract the acidic H molecules. This may promote rapid
H O decomposition on the mesoporous GCN catalysts with
4
2
2
2
51,52
2
O
2
2
2
larger surface areas.
CONCLUSION
Mesoporous GCN prepared by thermal polymerization of
cyanamide with silica nanoparticles as a template followed by
removal of the particles were used for photocatalytic H
ultrasonication for 5 min, and O was bubbled through the
2
2
O
2
solution for 5 min. The respective tube was immersed in a
53
production from EtOH and O in water under irradiation of
temperatureꢀcontrolled water bath (298 ± 0.5 K) and
2
visible light. The GCN catalysts with larger surface areas
inherently contain a larger number of primary amine moieties
on the surface of mesopores. Selectivity for H
via twoꢀelectron reduction of O2 by the conduction band
electrons localized on the 1,4ꢀpositions of the melem unit
decreases with an increase in the surface area. This is because
the primary amine moieties behave as the active sites for
photoirradiated with magnetic stirring using a 2 kW Xe lamp
(USHIO Inc.). A 20 wt % NaNO solution was used as a
54
2
55
O
2 2
formation
filter to give light wavelengths at λ >420 nm. The light
−2
intensity at 420–500 nm was determined to be 26.9 W m .
After the photoreaction, the gasꢀphase products were analyzed
by GCꢀTCD (Shimadzu, GCꢀ8A). The catalyst was then
recovered by centrifugation. The H O amount in the resulting
2
2
fourꢀelectron reduction of O
also behave as the sites for reductive decomposition of the
formed H , probably due to the strong interaction with
acidic H O . The obtained results suggest that creation of GCN
2
. These basic primary amine sites
solution was determined by a redox titration with KMnO4.
Other liquidꢀphase products were quantified by GC−FID
(Shimadzu, GCꢀ2010).
ESR analysis. The spectra were recorded at the X band
using a Bruker EMXꢀ10/12 spectrometer with a 100 kHz
magnetic field modulation at a microwave power level of 10.5
2
O
2
2
2
catalysts with a large surface area but with a smaller number
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 9
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mW, where microwave power saturation of the signals does
ACKNOWLEDGMENT
5
6
not occur. The magnetic field was calibrated with a
1,1′ꢀdiphenylꢀ2ꢀpicrylhydrazyl (DPPH) as a standard. The
measurements were carried out as follows: catalyst (20 mg)
was suspended in an EtOH/water mixture (9/1 v/v, 5 mL)
containing DMPO (0.1 mmol) within a Pyrex glass tube (φ12
mm; capacity, 20 mL), and the tube was sealed with a rubber
This work was supported by the Grant-in-Aid for Scientific
Research (No. 26289296) from the Ministry of Education,
Culture, Sports, Science and Technology, Japan (MEXT), and
by the Precursory Research for Embryonic Science and
Technology (PRESTO) from Japan Science and Technology
Agency (JST).
septum cap. After ultrasonication (3 min) and O bubbling (5
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
min), the tube was photoirradiated at λ >420 nm with
magnetic stirring for 3 min. After the photoirradiation, the
catalyst was recovered by centrifugation, and the resulting
solution was subjected to ESR analysis.
REFERENCES
(1) Campos-Martin, J. M.; Blanco-Brieva, G.; Fierro, J. L. G.
Angew. Chem., Int. Ed. 2006, 45, 6962–6984.
(2) Shaegh, S. A. M.; Nugyen, N-T.; Ehteshami, S. M. M.; Chan,
S. H. Energy Environ. Sci. 2012, 5, 8225–8228.
(3) Sandelin, F.; Oinas , P.; Salmi, T.; Paloniemi, J.; Haario, H.
Ind. Eng. Chem. Res. 2006, 45, 986–992.
Electrochemical analysis. The rotating disk electrode
analysis was performed on a computerꢀcontrolled CHI600D
advanced electrochemical system with a threeꢀelectrode cell.
An Ag/AgCl electrode and a Pt wire electrode were used as
4
7,49
(
4) Liu, Q.; Bauer, J. C.; Schaak, R. E.; Lunsford. J. H. Angew.
Chem., Int. Ed. 2008, 47, 6221–6224.
5) Edwards, J. K.; Solsona, B.; Ntainjua, E. N.; Carley, A. F.;
the reference and counter electrode, respectively.
The
working electrode was prepared according to the procedure
49
(
described in literature: the respective GCN catalysts (20 mg)
were dispersed in water (5 mL) by ultrasonication. The
suspension (20 µL) was put onto a Pt disk electrode and dried
at room temperature. Linear sweep voltammograms were
Herzing, A. A.; Kiely, C. J.; Hutchings, G. J. Science 2009, 323,
1037–1041.
(6) Teranishi, M.; Naya, S.; Tada, H. J. Am. Chem. Soc. 2010,
1
32, 7850–7851.
(7) Maurino, V.; Minero, C.; Mariella, G.; and Pelizzetti, E.
obtained in an O
2
ꢀsaturated 0.1 M phosphate buffer solution
–1
(
pH 7) with a scan rate being 10 mV s at different rotating
Chem. Commun. 2005, 2627–2629.
8) Tsukamoto, D.; Shiro, A.; Shiraishi, Y.; Sugano, Y.;
Ichikawa, S.; Tanaka, S.; Hirai, T. ACS Catal. 2012, 2, 599−603.
9) Cai, R.; Kubota, Y.; Fujishima, A. J. Catal. 2003, 219,
214−218.
(10) Goto, H.; Hanada, Y.; Ohno, T.; Matsumura, M. J. Catal.
speed (400–1600 rpm). After each scan, O2 was babbled
through the electrolyte for 5 min to saturate O2.
(
Calculation details. All of the calculations were performed
with tight convergence criteria at the DFT level with the
Gaussian 03 package, using the B3LYP/6ꢀ31G(d) basis set for
all atoms. The excitation energies and the oscillator strengths
were calculated by TDꢀDFT at the same level for optimization
using the polarizable continuum model (PCM). Cartesian
coordinates for the respective single, double, and triple
melemꢀconjugated models are summarized in the end of
Supporting Information.
(
2004, 225, 223−229.
(11) Hirakawa, T.; Nosaka, Y. J. Phys. Chem. C 2008, 112,
15818−15823.
(12) Wang, X.; Maeda, K.; Thomas, A.; Takanabe, K.; Xin,
G.;Carlsson, J. M.; Domen, K.; Antonietti, M. Nat. Mater. 2009, 8,
76−80.
(13) Thomas, A.; Fischer, A.; Goettmann, F.; Antonietti, M.;
Müller, J. O.; Schlögl, R.; Carlsson, J. M. J. Mater. Chem. 2008, 18,
Other analysis. XRD measurements were carried out on a
Philips X′PertꢀMPD spectrometer. Diffuse reflectance UV−vis
spectra were measured on a Vꢀ550 UVꢀvis spectrophotometer
4
893−4908.
14) Shiraishi, Y.; Kanazawa, S.; Sugano, Y.; Tsukamoto, D.;
Sakamoto, H.; Ichikawa, S.; Hirai, T. ACS Catal. 2014, 4, 774−780.
15) Shiraishi, Y.; Kanazawa, S.; Kofuji, Y.; Sakamoto, H.;
(
(
(
JASCO Corp.) equipped with Integrated Sphere Apparatus
ISVꢀ469) using BaSO4 as a reference. XPS charts were
(
measured on a JEOL JPSꢀ9000MX spectrometer using Mg Kα
radiation as the energy source. N adsorption/desorption
Ichikawa, S.; Tanaka, S.; Hirai, T. Angew. Chem., Int. Ed. 2014, 53,
13454–13459.
2
analysis and CO ꢀTPD measurements were carried out on an
2
(
16) Wang, J.; Ma, J.; Li, X.; Li, Y.; Zhang, G.; Zhang, F.; Fan, X.
Chem. Commun. 2014, 50, 14237−14240.
17) Shiraishi, Y.; Saito, N.; Hirai, T. J. Am. Chem. Soc. 2005, 127,
2820-12822.
18) Li, L.; Methira, K.; Sebastian W. P.; Morgan S.; Ulrich, W.;
AUTOSORBꢀ1ꢀC/TCD analyzer (Yuasa Ionics Co., Ltd.).
TEM observations were performed using an FEI Tecnai G2
(
57
2
0ST analytical electron microscope operated at 200 kV.
1
(
ASSOCIATED CONTENT
Supporting Information
Ullrich, S.; Dominik, E. Chem. Commun. 2010, 46, 7620–7622.
(19) Niu, P.; Zhang, L.; Liu, G.; Cheng, H.-M. Adv. Funct. Mater.
2012, 22, 4763–4770.
(20) Zhang, X.; Xie, X.; Wang, H.; Zhang, J.; Pan, B.; Xie, Y.; J.
Am. Chem. Soc. 2013, 135, 18−21.
(21) Schwinghammer, K.; Mesch, M. B.; Duppel, V.; Ziegler, C.;
Senker, J.; Lotsch, B. V.; J. Am. Chem. Soc. 2014, 136, 1730−1733.
(22) Sano, T.; Tsutsui, S.; Koike, K.; Hirakawa, T.; Teramoto,
Y.; Negishi, N.; Takeuchi, K.; J. Mater. Chem. A 2013, 1,
6489–6496.
N2 adsorption/desorption isotherms (Figure S1); XRD patterns
Figure S2); diffuse reflectance UVꢀvis spectra (Figure S3);
results of photocatalytic decomposition of H (Figure S4);
(
2
O
2
Cartesian coordinates for single, double, and triple melem models.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTOR INFORMATION
(23) Dong, F.; Wanga, Z.; Sun, Y.; Ho W-K.; Zhang, H. J.
Colloid Interface Sci. 2013, 401, 70–79.
Corresponding Author
(
24) Martin, D. J.; Qiu, K.; Shevlin, S. A.; Handoko, A. D.; Chen,
X.; Guo, Z.; Tang, J. Angew. Chem., Int. Ed. 2014, 53, 9240–9245.
25) Zhang, G.; Zhang, J.; Zhang, M.; Wang, X. J. Mater. Chem.
012, 22, 8083–8091.
shiraish@cheng.es.osakaꢀu.ac.jp
(
Notes
2
The authors declare no competing financial interest.
ACS Paragon Plus Environment

Page 9 of 9
ACS Catalysis
1
2
3
(26) Shi, L.; Liang, L.; Wang, F.; Maa, J.; Sun, J. Catal. Sci.
Technol. 2014, 4, 3235–3243.
27) Tahir, M.; Cao, C.; Butt, F. K.; Butt, S.; Idrees, F.,
1397−1399.
(42) Jiang, H.; Zhu, Y.; Feng, Q.; Su, Y.; Yang, X.; Li, C. Chem.
Eur. J. 2014, 20, 3106–3112.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
Mahmood, N.; Ali, Z.; Aslam, I.; Tanveer, M.; Rizwana, M.;
Mahmooda, T. J. Mater. Chem. A 2013, 1, 13949–13955.
(43) Guo, Z.; Liu, H.; Jiang, C.; Zhu, Y.; Wan, M.; Dai, L.; Jiang,
L. Small 2014, 10, 2087–2095.
(44) Sheng, H.; Ji, H.; Ma, W.; Chen, C.; Zhao, J. Angew. Chem.,
Int. Ed. 2013, 52, 9686–9690.
(45) Lin, L.; Zhu, Q.; Xu, A. J. Am. Chem. Soc. 2014, 136,
11027–11033.
(46) Liu, Y. Liu, H.; Wang, C.; Hou, S-X.; Yang, N. Environ. Sci.
Technol. 2013, 47, 13889−13895.
(47) Li, S.; Wu, D.; Liang, H.; Wang, J.; Zhuang, X.; Mai, Y.; Su,
S.; Feng, X. ChemSusChem 2014, 7, 3002–3006.
(48) Sun, B.; Ou, Z.; Meng, D.; Fang, Y.; Song, Y.; Zhu, W.;
Solntsev, P. V.; Nemykin, V. N.; Kadish, K. M. Inorg. Chem. 2014,
53, 8600−8609.
(49) Zheng, Y.; Jiao, Y.; Chen, J.; Liu, J.; Liang, J.; Du, A.; Zhang,
W.; Zhu, Z,; Smith, S. C.; Jaroniec, M.; Lu, C. Q.; Qiao, S. Z. J. Am.
Chem. Soc. 2011, 133, 20116–20119.
(50) Sheng, J.; Li, X.; Xu, Y. ACS Catal. 2014, 4, 732−737.
(51) Di, Y.; Wang, X.; Thomas, A.; Antonietti, M.
ChemCatChem 2010, 2, 834–838.
(52) Zhao, L.; Bacsik, Z.; Hedin, N.; Wei, W.; Sun, Y.;
Antonietti, M.; Titirici, M.-M. ChemSusChem 2010, 3, 840–845.
(53) Sugano, Y.; Shiraishi, Y.; Tsukamoto, D.; Ichikawa, S.;
Tanaka, S.; Hirai, T. Angew. Chem., Int. Ed. 2013, 52, 5295–5299.
(54) Shiraishi, Y.; Sugano, Y.; Tanaka, S.; Hirai, T. Angew.
Chem., Int. Ed. 2010, 49, 1656–1660.
(
28) Gao, J.; Zhou, Y.; Li, Z.; Yan, S.; Wang, N.; Zou, Z.
Nanoscale 2012, 4, 3687–3692.
29) Tahir, M.; Cao, C.; Butt, F. K.; Butt, S.; Idrees, F.; Ali, Z.;
(
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Aslam, I.; Tanveer, M.; Mahmoodc, A.; Mahmoodc, N.
CrystEngComm 2014, 16, 1825–1830.
(
30) Wang, X.; Maeda, K.; Chen, X.; Takanabe, Kazuhiro,
Domen, K.; Hou, Y.; Fu, X.; Antonietti, M. J. Am. Chem. Soc.
009, 131, 1680–1681.
31) Chen, X.; Jun, Y-S.; Takanabe, K.; Maeda, K.; Domen, K.;
2
(
Fu, X.; Antonietti, M.; Wang, X. Chem. Mater. 2009, 21,
4093–4095.
(32) Su, F.; Mathew, S. C.; Lipner, G.; Fu, X.; Antonietti, M.;
Blechert, S.; Wang, X. J. Am. Chem. Soc. 2010, 132, 16299–16301.
(
33) Maeda, K.; Kuriki, R.; Zhang, M.; Wang, X.; Ishitani, O. J.
Mater. Chem. A 2014, 2, 15146–15151.
34) Kiskan, B.; Zhang, J.; Wang, X.; Antonietti, M.; Yagci, Y.
ACS Macro Lett. 2012, 1, 546−549.
35) Lee, S. C.; Lintang, H, O.; Yuliati, L. Chem. Asian J. 2012, 7,
(
(
2139–2144.
(36) Xu, J.; Wu, H-T.; Wang, X.; Xue, B.; Li, Y-X.; Cao, Y. Phys.
Chem. Chem. Phys. 2013, 15, 4510–4517.
(
37) Li, G.; Zhang, D.; Yu, J.-C. Chem. Mater. 2008, 20,
983–3992.
38) Guo, C.; Holland, G. P. J. Phys. Chem. C 2014, 118,
5792−25801.
(39) Ding, Z.; Chen, X.; Antonietti, M.; Wang, X.
3
2
(
(55) Shiraishi, Y.; Sugano, Y. Ichikawa, S.; Hirai, T. Catal. Sci.
Technol., 2012, 2, 400–405.
(56) Tsukamoto, D.; Shiraishi, Y.; Sugano, Y.; Ichikawa, S.;
Tanaka, S.; Hirai, T. J. Am. Chem. Soc. 2012, 134, 6309−6315.
(57) Shiraishi, Y.; Tanaka, K.; Shirakawa, E.; Sugano, Y.;
Ichikawa, S.; Tanaka, S.; Hirai, T. Angew. Chem., Int. Ed. 2013, 52,
8304–8308.
ChemSusChem 2011, 4, 274–281.
40) Zhu, J.; Xiao, P.; Li, H.; Carabineiro, S. A. C. ACS Appl.
Mater. Interfaces 2014, 6, 16449−16465.
41) Harbour, J. R.; Hair, M. L. J. Phys. Chem. 1978, 82,
(
(
ACS Paragon Plus Environment