Al2O3 surface complexation for photocatalytic organic transformations

Article abstract of DOI:10.1021/jacs.6b09934

The use of sunlight to drive organic reactions constitutes a green and sustainable strategy for organic synthesis. Herein, we discovered that the earth-abundant aluminum oxide (Al₂O₃) though paradigmatically known to be an insulator could induce an immense increase in the selective photo-oxidation of different benzyl alcohols in the Presence of a large variety of dyes and O₂. This unique phenomenon is based on the surface complexation of benzyl alcohol (BnOH) with the Brensted base sites on Al₂O₃, which reduces its oxidation potential and causes an upshift in its HOMO for electron abstraction by the dye. The surface complexation of O₂ with Al₂O₃ also activates the adsorbed O₂ for receiving electrons from the photoexcited dyes. This discovery brings forth a new understanding on utilizing surface complexation mechanisms between the reactants and earth abundant materials to effectively achieve a wider range of photoredox reactions.

Full text of DOI:10.1021/jacs.6b09934

Subscriber access provided by University of Otago Library
Article
Al2O3 Surface Complexation for Photocatalytic Organic Transformations
Wan Ru Leow, Wilson Kwok Hung Ng, Tai Peng, Xinfeng Liu, Bin Li,
Wenxiong Shi, Yanwei Lum, Xiaotian Wang, Xianjun Lang, Shuzhou Li, Nripan
Mathews, Joel W. Ager, Tze Chien Sum, Hajime Hirao, and Xiaodong Chen
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b09934 • Publication Date (Web): 14 Dec 2016
Downloaded from http://pubs.acs.org on December 14, 2016
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.
Journal of the American Chemical Society 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
Journal of the American Chemical Society
Al O Surface Complexation for Photocatalytic Organic Transfor-
mations
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
2
3
†
‡
†
‡
†
†
Wan Ru Leow, Wilson Kwok Hung Ng, Tai Peng, Xinfeng Liu, Bin Li, Wenxiong Shi, Yanwei
§
,
†
†
†
†
§,
∥
∥
Lum,‡ Xiaotian Wang, Xianjun Lang, Shuzhou Li, Nripan Mathews, Joel W. Ager, Tze Chien
,‡
,†
Sum, Hajime Hirao,* and Xiaodong Chen*
†
Innovative Center for Flexible Devices, School of Materials Science and Engineering, Nanyang Technological University,
50 Nanyang Avenue, Singapore 639798.
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
‡
School of Physical and Mathematical Sciences, Nanyang Technological University, 21 Nanyang Link, Singapore 637371.
§
Joint Center for Artificial Photosynthesis, Materials Sciences Division, Lawrence Berkeley National Laboratory, California
9
4720, United States.
∥Department of Materials Science and Engineering, University of California, Berkeley, California 94720, United States.
KEYWORDS: Photocatalysis; Al O ; surface complex; benzylalcohol oxidation
2
3
ABSTRACT: The use of sunlight to drive organic reactions constitutes a green and sustainable strategy for organic synthesis.
Herein, we discovered that the earthꢀabundant aluminum oxide (Al O ), though paradigmatically known to be an insulator, could
2
3
induce an immense increase in the selective photoꢀoxidation of different benzyl alcohols in the presence of a large variety of dyes
and O . This unique phenomenon is based on the surface complexation of benzyl alcohol (BnOH) with the Brønsted base sites on
2
Al O , which reduces its oxidation potential and causes an upshift in its HOMO for electron abstraction by the dye. The surface
2
3
complexation of O with Al O also activates the adsorbed O for receiving electrons from the photoexcited dyes. This discovery
2
2
3
2
brings forth a new understanding on utilizing surface complexation mechanisms between the reactants and earth abundant materials
to effectively achieve a wider range of photoredox reactions.
1
. INTRODUCTION
the latter requires unique photocatalysts to be synthesized for
specific reactions.
Photoredox catalysis has great potential in facilitating orꢀ
1
,2
In the past development of photocatalysis, aluminum oxide
ganic transformations to valuable products, as it is environꢀ
mentallyꢀfriendly, costꢀeffective, and easily realized by mere
household fixtures. Therefore much research has been conꢀ
ducted on photoꢀabsorbing molecules and complexes to conꢀ
vert solar energy to chemical energy. Out of these, Ruꢀbased
polypyridyl complexes have attracted immense popularity due
to their ease of synthesis, stability, and good photoredox capaꢀ
bilities. Recently, the merging of photocatalysis and other
types of catalysis e.g., organocatalysis and nickel or
gold
(
Al O ) has played a supporting rather than a central role,
2 3
serving as high surface area supports and photochemically
3
6,37
inert barriers.
This is because alumina is an electrical insuꢀ
38
3
lator with a wide band gap of 8.7 eV. However, Al O , in
2
3
particular the γꢀphase, possesses unique surface properties
such as bifunctional Brønsted and Lewis acid and base
3
9,40
4,5
sites,
activity that is absent in the pristine materials.
include Ag/Al O3 for enhanced NO reduction
which can endow supported materials with catalytic
41,42
6
,7
8,9
Examples
4
3,44
10,11
and
V O /Al O for the photoꢀoxidation of cyclohexane. Thus we
catalysis, could enable a large variety of difficult orꢀ
2
4
5
1
,12,13
ganic transformations
such as the coupling of aryl halides
2
5
2
3
8
14,15
16
propose that the redox potentials of organic reactants can be
modified through surface complexation with Al O , which can
with amino acids, tertiary anilines,
carboxylic acids,
and αꢀoxo acids. Meanwhile, selective
oxidation constitutes a class of difficult organic transforꢀ
9
,17,18
19
organoborates,
2
3
be coupled with photocatalysis to facilitate demanding organic
transformations under mild conditions.
2
0ꢀ23
mation, e.g. the oxidation of benzyl alcohols to aldehydes
and sulfides to sulfoxides,
2
4,25
which is essential for converting
In this article, we report that Al O surface complexation
2
3
raw materials into fine chemicals especially in the flavor and
can enable the highly selective oxidation of benzyl alcohols to
2
0,24,26
fragrance industry.
The typical synthesis route requires
benzaldehydes in the presence of dye, O and visible light,
2
toxic oxidants that are detrimental to health and the environꢀ
even though negligible oxidation occurred with the dye alone.
This phenomenon can be extended to dyes of different moieꢀ
ties. The underlying mechanism is that the oxidation potential
of benzyl alcohol may be decreased through simple complexaꢀ
tion with Al O , thus rendering it susceptible to oxidation by
27ꢀ30
ment.
As for the photocatalytic redox route, the reaction
can proceed only if the redox potential of the photocatalyst is
3
greater than that of the reactants . Conventional strategies to
fulfill this condition include using wide band gap semiconducꢀ
2
3
3
1
20,24
32
tors such as TiO2
or Nb O , and modifying the catiꢀ
photocatalysts with smaller redox potentials. The complexaꢀ
tion of O with Al O also enables it to receive electrons from
2
5
3
,33,34
on
or ligands of transition metal complexes to tune the
2
2
3
3,35
redox potential. However, the former method can only utiꢀ
lize the short wavelength range of the solar spectrum, while
the photoexcited dye, thus facilitating the overall transfer of
protons and electrons from benzyl alcohol to O . This method
2
would bring forth the potential of utilizing surface complexaꢀ
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 9
tion mechanisms between the reactants and earth abundant Agilent Technology 19091Jꢀ413 capillary column (30 m ×
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
materials to achieve a wider range of photoredox reactions.
0.32 mm × 0.25 mm) using highꢀpurity N as the carrier gas
2
and chlorobenzene as the internal standard. The standard analꢀ
ysis conditions are: injector temperature 250 ºC, detector temꢀ
perature 300 ºC, column temperature ramped from 50 ºC to
300 ºC with a rate of 20 ºC min . GC–MS analysis was conꢀ
ducted with a Shimadzu GC 2010 gas chromatograph
equipped with a Shimadzu GCMSQP2010 Ultra mass specꢀ
trometer and a Restek (Rxi®ꢀ5Sil MS) capillary column (30 m
2. MATERIALS AND METHODS
ꢀ
1
2
.1 Preparation and Characterization of Al O
2 3
All solid catalysts were purchased from Sigma Aldrich and
Alfa Aesar and used without further modification. The syntheꢀ
sized Al O was prepared by solid state reaction, in which 7.5
2
3
×
0.25 mm × 0.25 mm), coupled with an electron ionization
g of aluminum nitrate nonahydrate (Al(NO ) ·9H O) dissolved
3
3
2
mass spectrometer with highꢀpurity He as the carrier gas.
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
in 50 ml deionized water was adjusted to pH 7.0 with
NH ·H O solution, dried, and then calcined at 600 ºC.
3
2
2
.3 Cyclic Voltammetry
Powder Xꢀray diffraction (XRD) was recorded by a Brukerꢀ
AXS Xꢀray diffractometer with Cu Kα irradiation (λ = 1.5418
Å). The morphologies of the samples were examined by scanꢀ
ning (SEM, JEOLꢀJSMꢀ7600F) and highꢀresolution transmisꢀ
sion electron microscopy (HRTEM, JEOLꢀJEMꢀ2100F). The
Brunauer–Emmett–Teller (BET) surface area was measured
with an ASAP 2020 apparatus (Micromeritics Instrument
Corp.). UVꢀvisible diffuse reflectance spectroscopy was used
to measure the optical properties of Al O with barium sulfate
The Al O modified electrode was fabricated according to
2
3
procedures present in literature. Basically, Al O paste was
2
3
deposited using the doctorꢀblade technique on fluorineꢀdoped
stannic oxide coated glass (sheet resistance of 15 ꢁ/sq., Nipꢀ
pon Sheet Glass Co., Ltd., Japan). The resultant layer was
dried in air at room temperature and heated at 450 °C for 1
hour before cooling to room temperature.
2
3
Cyclic voltammetry was carried out using a CHI 660D elecꢀ
trochemical workstation (CH Instruments, Inc., Austin, USA).
A threeꢀelectrode configuration was used for the measurement
of cyclic voltammetry; the Al O modified FTO glass funcꢀ
as the reference and transformed to the absorption spectra
according to the KubelkaꢀMunk relationship. The CO ꢀTPD
2
was performed on the Auto Chem 2920 (USA) apparatus. 0.10
g of the sample was loaded in a Uꢀshaped quartz tube, heated
at 550 ºC for 1 h in helium flow and then cooled to 40 ºC, in
order to saturate the sample with CO . The CO ꢀTPD spectrum
2
3
tions as the working electrode, a Pt wire as the counter elecꢀ
+
trode, and an Ag/Ag electrode as the reference. A 0.1 M soluꢀ
2
2
tion
of
tetraꢀnꢀbutylammonium
hexafluorophosphate
was then measured in the temperature range of 40–600 ºC at a
(TBAPF6) in CH CN was used as the supporting electrolyte.
3
ꢀ
1
constant heating rate of 10 ºC min . Xꢀray photoelectron specꢀ
troscopy (XPS) was performed using a Kratos Axis Ultra DLD
system with a monochromatized Al Kα source (hν = 1486.6
eV). A takeoff angle of 0º relative to the surface normal was
used to sample the maximum surface depth. The subsequent
XPS peak fitting and analysis were performed with the Casꢀ
aXPS software. Energy calibration was referenced to the C1s
peak at 284.8 eV. Fourier transform infrared (FTIR) spectrosꢀ
copy was performed on the Al O samples pelletized with KBr
Ferrocene was employed as internal standard. A glassy carbon
electrode was used as working electrode for the measurement
of oxidative potential of BnOH.
2.4 Theoretical Simulation of the BnO-Al O Surface
2
3
Complex and Dyes’ Energy Structure
All calculations are based on the Al O model proposed by
2
3
46
Digne et al. . A slab with the (110) facet exposed and dimenꢀ
sions of (16.826Å*16.136Å*25.000Å) is created to model the
BnOH adsorption, while the periodic boundary condition calꢀ
culations were conducted with VASP computational packagꢀ
2
3
using a PerkinElmer Frontier setup.
2
.2 Reaction and Post-Reaction Analysis
47ꢀ50
es.
The PBE functional and PAW method were used for
All reactants and catalysts were purchased from Sigma Alꢀ
calculation, and only the Γ point is chosen for Brillouin zone
51ꢀ53
drich and Alfa Aesar and used without further purification,
while all solvents were purchased from Fisher Chemical. The
solid catalysts were thermally preꢀtreated at 200 ºC under amꢀ
bient atmosphere prior to reaction to remove surface water and
impurities. In a typical reaction, 50 mg of Al O , 0.5 µmol of
sampling.
An isolated complex was used to describe the
BnO–Al O surface complex approximately, and it was asꢀ
2
3
2+
sumed that neighbouring BnOH, Ru(bpy)3 and solvent moleꢀ
cules do not exert a large influence on its energy levels. The
HOMO level of the free BnOH molecule is determined from
the vertical ionization energy, which is calculated by the
Gaussian09 package using the PBE functional with 6ꢀ31+G*
basis set. The vacuum level is used to locate the absolute enerꢀ
gy level.
2
3
Ru(bpy) Cl , and 0.1 mmol of BnOH were added to 5 mL of
3
2
CH CN in a Pyrex vessel, and allowed to stir for 30 min in the
3
dark. O was then purged into the Pyrex vessel to achieve an
2
O atmosphere of pressure 0.1 MPa. The reaction mixture was
2
then illuminated by the 7 W white light household LED, the
irradiating wavelength range of which was measured to be
400ꢀ700 nm, with two maxima at 450 and 550 nm (Figure
S10a), under magnetic stirring at 800 rpm.
The Gaussian09 package was used for the DFT calculation
of the dye molecules. The B3LYP functional with 6ꢀ31G*
basis set is used for the calculations. For Rose Bengal, the
iodide atoms were treated with the SDD basis set. To take the
solvation effect into account, the selfꢀconsistent reaction field
(SCRF) approach was used and the solvent involved is aceꢀ
Upon reaction completion, Al O was separated from the reꢀ
2
3
action mixture by centrifugation. The collected Al O was
2
3
51ꢀ53
washed twice with CH CN and dried overnight in a vacuum
tonitrile.
The energy level of Rhodamine B in vacuum is
3
oven at 60 ºC prior to characterization. The product identities
were confirmed by comparison of the gas chromatography
very low due to its positive charge (no counter ion is added to
the system). The HOMOꢀLUMO gaps are almost always largꢀ
(
GC) retention times with that of standard samples. The prodꢀ
ucts were then quantitatively analyzed through a GC (Agilent
890A) equipped with a flame ionization detector (FID) and
er that the S S transition energy predicted by TDDFT. This is
0 1
2+
the most significant for Ru(bpy)3 as it is cationic and its elecꢀ
7
2
ACS Paragon Plus Environment

Page 3 of 9
Journal of the American Chemical Society
tronic transition is MLCT, which tends to deviate a lot from its
HOMOꢀLUMO gap.
observed. This may be because the low acidity of the α proton
renders it less favourable towards the formation of H O . For
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
2
2
furfuryl alcohol 2f, although the conversion was 99.3 %, there
was no selectivity for furfural 2’f (Table S2). The combined
2.5 Time-Resolved Photoluminescence (TRPL)
2+
Al O ꢀRu(bpy) surface complex system also showed imꢀ
2
3
3
A pump wavelength of 400 nm was used for the TRPL
proved conversion in the hydroxylation of phenylboronic acid
3a to phenol 3’a (from 14.6 % to 96.0 %), but not the epoxidaꢀ
tion of styrene 3b (Table S3). The oxidation of benzylamine
measurements, which originates from the frequency doubled
TM
8
00 nm laser pulses generated from a Coherent Libra Reꢀ
generative Amplifier using a Beta Barium Borate (BBO) crysꢀ
tal. The luminescence signal was collected and then dispersed
3
c and thioanisole 3d to imine 3’c and sulfoxide 3’d were
2+
−1
easily able to proceed to completion with Ru(bpy)
3
only,
by a DK240 1/4ꢀm monochromator with 300 g mm grating.
The transient photoluminescence signal was resolved by an
hence the addition of Al O would not be necessary. Due to
2
3
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
TM
the limited scope of the article, only the mechanism of BnOH
oxidation would be examined in detail.
Optronis Optoscope streak camera system, which possesses
an ultimate temporal resolution of about 10 ps when operated
at the fastest scan speed.
3
. RESULTS AND DISCUSSION
First of all, a series of different benzyl alcohols 1a-k was
2
+
tested with Ru(bpy)3 as the photosensitizer and mild condiꢀ
tions such as irradiation with a 7 W household white light
LED and O as the oxidant (Figure 1a). The Al O used is the
2
2
3
commerciallyꢀavailable Aldrich 517747, unless otherwise
stated. XRD studies showed that the Al O is γꢀcrystalline
2
3
(
Figure S1a), while HRTEM and SEM revealed that the Al O3
2
comprises micronꢀsized mesoporous particles (Figures S2a
and S3a). The Al O also boasts a large BET surface area of
2
3
2
3
32 m /g, which is favorable for catalysis (Figure S4a and
Table S1). XPS suggests that the Al O is free of metal ions;
2
3
the characteristic peaks for metals commonly used in catalysis
i.e., Ag, Au, Co, Cu, Fe, Mn, Mo, Ni, Ru and Pd) were not
(
observed (Figure S5). It can be seen from Figure 1b that the
simple addition of Al O caused a large increase in converꢀ
2
3
sions for all 11 benzyl alcohols (52.5ꢀ88.6%, with most over
2
+
75.0%), while Ru(bpy)3 alone only afforded trace converꢀ
sions (<5.0 %). The addition of Al O has increased the turnoꢀ
2
3
ver number (TON) of benzyl alcohol (BnOH) with respect to
2
+
Ru(bpy)3 by over 30 times (Figure S6a). This discovery is
paradoxical to existing paradigms as Al O is known as an
2
3
3
8
insulator and tends to play a supporting rather than a central
role in photocatalysis.
Meanwhile, characterization of the postꢀreaction Al O sugꢀ
3
6,37
2
3
gests that only the surface of Al O was involved in the photoꢀ
2
3
catalytic reaction. The XRD spectra of the postꢀreaction Al O3
2
showed negligible change in its crystallinity (Figure S1b).
Based on the SEM and TEM images, there appears to be little
change in the morphology of Al O (Figures S2b and S3b).
Figure 1. Photocatalytic organic transformation catalyzed by the
Al
O
3
and dye surface complex system. a) Visible light driven
. b) Conversion of
different benzyl alcohols by Ru(bpy)3 with and without Al O .
2
oxidation of benzyl alcohol to aldehyde with O
2
3
2
2+
Surface adsorption studies confirmed that Al O retained its
2
3
2
3
c) Conversion of BnOH with different metal oxides and
mesoporous structure, but the BET surface area was somewhat
2+
2+
Ru(bpy)3 . d) Extensiveness of the Al O and Ru(bpy) dual
2
3
3
reduced, which may be due to the centrifugation of the Al O3
2
surface complex system towards different commercial Al O
2
3
during its separation from the reaction mixture (Figure S4b
and Table S1). No significant differences were observed in the
products.
XPS spectra of Al O after reaction (Figure S7). This implies
2
3
It is interesting that the phenomenon appears to be exclusive
to Al O ; with other metal oxides such as SiO and MgO, negꢀ
that Al O was not consumed during the reaction to generate
2
3
2
3
2
active catalytic species.
ligible BnOH conversions were observed (Figure 1c). The
TON achieved with Al was >50 times higher than that with
SiO and MgO (Figure S6a). The initial rate of conversion and
turnover frequency (TOF) were relatively high with Al , but
2
+
The addition of Al O to Ru(bpy) could also significantly
2
O
3
2
3
3
improve the selective oxidation of phenol 2a to cyclohexaꢀ2 5ꢀ
dienꢀ1ꢀone 2’a (from 0 % to 14.5 %, Table S2). Although
some extent of hydroquinone 2b oxidation to quinone 2’b
2
O
3
2
decreased due to the decreasing concentration of BnOH availꢀ
able for reaction and the formation of hydroxyl groups on the
2
+
could be achieved with Ru(bpy)3 alone (21.0 %), the incluꢀ
sion of Al O could lead to a dramatic increase in yield (61.3
4
0
55ꢀ57
Al O surface when the byproduct H O
H
2
decomposes to
O (Figure S6bꢀc, Figure S8). Control studies revealed that
negligible BnOH conversion occurred in the dark and under
N (Figure S9a), which indicated that (1) the reaction proceeds
2
2
3
2
2
2
3
5
4
58
%
). However, for aliphatic alcohols such as butanol 2c, penꢀ
tanol 2d and cyclohexanol 2e (from 0.130ꢀ0.243 % to 4.88ꢀ
.42 %, Table S2), only a small improvement in yield was
8
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 9
via a photocatalytic pathway, and (2) O is essential for the
version under the 490ꢀ750 nm wavelength range (Figure
S11b).
2
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
transformation of BnOH, suggesting that O serves as a sacriꢀ
2
ficial electron accepter that receives 2 electrons and 2 protons
from BnOH to form benzaldehyde. The BnOH conversion
generally increased with the amount of Al O used, but the
3.1 Decrease of BnOH Oxidation Potential due to Benzylic
O-H dissociation
2
3
extent of increment becomes less significant for larger
amounts of Al O (Figure S9b). Figure 1d showed that the
2+
The success of the Al O ꢀRu(bpy) photocatalytic system
2
3
3
2
3
2+
and the negligible activity of the individual Ru(bpy)3 can be
explained by the decrease of the BnOH oxidation potential
through surface complexation with Al O . In Figure 1b, 2ꢀ
phenomenon also applies to other commercial Al O3 and
2
Al O synthesized via solid state reaction, but not for the nonꢀ
2
3
2
3
mesoporous commercial Al O [Aldrich 718475]. It can also
2
3
methoxybenzyl alcohol 1c (71.6%) showed a lower conversion
relative to 4ꢀmethoxybenzyl alcohol 1d (88.6%), as the methꢀ
oxy group in the ortho position provides greater steric hinꢀ
drance around the OꢀH group. The decrease in conversion with
increasing amounts of acetic acid added (Figure S12a) sugꢀ
gests that the Brønsted base sites on Al O (Figures S13aꢀb)
be observed that the basic Al O [Alfa Aesar 11503] showed
2
3
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
higher conversion compared to the neutral [Alfa Aesar 11502]
and acidic Al O [Alfa Aesar 11501], which suggests the imꢀ
2
3
portance of Brønsted basicity in enabling the reaction. The
characterizations of the aforementioned Al O samples can be
2
3
2
3
found in Figures S1ꢀ4 and Table S1.
are responsible for the surface complexation, as acetic acid
would compete with BnOH for chemisorption on Brønsted
base sites. Cyclic voltammetry conducted on BnOH in aceꢀ
tonitrile with a bare glassy carbon electrode showed an irreꢀ
35
versible BnOH oxidation peak at around +2.0 V vs Ag/AgCl .
2+
3+
As the Ru(bpy)3 /Ru(bpy)3 potential in acetonitrile is +1.29
3,13
3+
V vs SCE, BnOH cannot be oxidized by Ru(bpy)3 . Howꢀ
ever, the use of an Al O ꢀmodified electrode resulted in an
2
3
additional oxidation feature at around +0.63 V vs Ag/AgCl
(Figure 3a), which is within the oxidative potential of
3+
Ru(bpy)3 . This oxidation feature is suggestive of a chemiꢀ
sorbed BnOꢀAl O surface complex, which is oxidized at the
2
3
Al O ꢀbare electrode interface. The BnOꢀAl O surface comꢀ
2
3
2
3
plex is characterized by a lower oxidation potential and can be
3
+
easily oxidized by Ru(bpy)3 (Figure 4a, step II). No addiꢀ
Figure 2. The extensiveness of the Al O and dye surface comꢀ
2
3
tional oxidation feature was detected for the SiO ꢀ and MgOꢀ
2
plex system towards dyes of different moieties.
modified electrodes (Figure S15a), while similar oxidation
peaks are detected for other BnOH derivatives (Figure S15bꢀ
d). The addition of acetic acid caused a quenching of the oxiꢀ
dation feature at +0.63 V vs Ag/AgCl (Figure 3a), indicating
This phenomenon can also be extended to a large variety of
dyes with different moieties, i.e. naphthalene, anthracene, porꢀ
phyrin, and perylene. While none of the dyes were able to
oxidize BnOH in the absence of Al O , most were able to
that the BnOꢀAl O surface complexation occurs through OꢀH
2
3
2
3
deꢀprotonation by the Brønsted base surface sites. As little
BnOH conversion occurred when Al O was replaced by inorꢀ
achieve some BnOH conversion when Al O was added (Figꢀ
2
3
ure 2). A positive relationship can be observed between the
dye excitation wavelengths (Figure S10d) and extent of BnOH
2
3
ganic bases (Figure S12b), this suggests that surface complexꢀ
ation on Al O has a stabilizing effect on the deprotonated
2+
conversion. Ru(bpy)3 and peryleneꢀ3,4,9,10ꢀtetracarboxylic
2
3
ꢀ
BnO .
dianhydride (PTCDA), with excitation wavelengths between
4
7
50 and 550 nm, showed extremely high conversions of
9.4(±4.4)% and 82.2(±11.1)%, respectively. Commercial
Furthermore, the decrease of BnOH oxidation through surꢀ
face complexation is validated through firstꢀprinciples calculaꢀ
tions on free BnOH and BnOH adsorbed on the most widely
exposed plane of the γꢀphase Al O surface, the (110) facet,
dyes such as Rose Bengal and Rhodamine B were also able to
effect significant conversions of 64.5(±4.6)% and
36.2(±2.6)%, respectively, but their lack of stability underꢀ
mined their photocatalytic activity. 9ꢀanthracenecarboxylic
acid (ACA) and 1,4,5,8ꢀnaphthalenetetracarboxylic dianhyꢀ
dride (NTCDA), with absorbance peaks in the range of
2
3
46
based on the Digne model (Figure S16a) . The HOMO of the
free BnOH molecule was calculated as ꢀ8.46 eV (Figure 3b).
The DOS diagrams for the adsorption of BnOH onto the AlIII
site (Figure 3c and d) show that OꢀH dissociation results in an
3
00−400 nm that is far from the LED wavelengths, yielded a
upward shift in the PDOS of BnOH and the appearance of two
additional occupied peaks (the leftꢀhand side of the black
dashed line with energy > 0 eV). The two peaks correspond to
occupied localized states arising from the molecule (Figure
S16d), and their energy levels are colored in red in Figure 3b.
These occupied states arise as BnOH draws electron density
from the surface after OꢀH dissociation to stabilize its new
alkoxide state, which causes it to become electron rich while
creating an electron deficiency in a localized area of the surꢀ
face. The electron gain by BnOH increases the electron repulꢀ
sion within the molecule and causes an upward shift in energy
level. As the CꢀH σꢀbonds are located at the valence edge, a
low conversion of 23.0(±1.7)% and 13.1(±5.3)%, respectively.
However, when the light source was switched to one with a
wavelength range that is suitable for absorption by the ACA
(
350ꢀ750 nm), the combined Al O ꢀACA surface complex
2 3
system was be able to effect a significant degree of converꢀ
sion, i.e. 78.0% in 2 h (Figure S11a). It must be noted that
ACA by itself was only able to achieve a low conversion of
2
7.3%. This suggests that different wavelengths of light can be
accessed by the Al O ꢀdye surface complex system by merely
2
3
selecting a dye with an appropriate excitation wavelength. For
2
+
instance, the use of Ru(bpy)3 with Al O showed higher conꢀ
2
3
3+
version than Degussa P25 under the 450ꢀ750 nm wavelength
range, and the use of PTCDA with Al O showed higher conꢀ
hole generated on the surface by Ru(bpy)3 oxidation and
transferred to BnOH would result in the weakening and subseꢀ
2
3
4
ACS Paragon Plus Environment

Page 5 of 9
Journal of the American Chemical Society
quent cleavage of the CꢀH bond to form benzaldehyde (Figure
4a, steps IIIꢀIV). In this way, Al O enables the oxidation of
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
2
3
BnOH through the formation of a chemisorbed surface comꢀ
plex that has a lower oxidation potential than that of
2
+
3+
Ru(bpy)3 /Ru(bpy)3
.
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
Figure 4. Proposed mechanism of visibleꢀlightꢀdriven BnOH
2
+
oxidation by the Al O and Ru(bpy) surface complex system. a)
2
3
3
2+
Initiation process involving the Ru(bpy)3 photosensitizer. b)
Subsequent chain propagation process involving two surface
complex intermediates.
In the previous section, it has been demonstrated that the reꢀ
action proceeds via a photocatalytic pathway. However, the
st
high apparent quantum yield of the reaction in the 1 hour
(15.4 %, Figure S17) appears to suggest the possibility of a
secondary chain propagation pathway (Figure 4b), in which
the carbocationic intermediate IV donates an electron to a
neighbouring chemisorbed BnOH (intermediate II). From Figꢀ
ure S17c, it can be seen that the initial quantum yield of the
reaction increases almost proportionally with the amount of
BnOH, such that a high value of 69.6 % was achieved for 1.0
mmol of BnOH. The combination of two pathways may have
enabled the reaction to proceed with high efficacy with a given
amount of light energy.
2+
Figure 3. Chemisorption on the surface of Al O decreases the
3.2 Electron Transfer from Ru(bpy)
O2
3
due to Activation of
2
3
oxidation potential of BnOH. a) Cyclic voltammogram of BnOH
with Al O ꢀmodified (red) and bare electrodes (blue), and 40 ꢂl
2
3
As a control experiment to confirm the interaction between
acetic acid (black). b) Energy levels of Al O , free BnOH and the
2
3
2+
Al O and the photosensitizer, we grafted Ru(bpy) to the
2
3
3
chemisorbed surface complex. c) DOS of BnOH physically adꢀ
sorbed to the AlIII site, and d) chemisorbed via OH dissociation.
The dashed line indicates the formation of localized occupied
states just above the VBM of Al O .
surface of Al O through –COOH groups, in order to reduce
2
3
the proximity between the two (Table S4, Column [b]). As
BnOH conversion is generally greater for the grafted
2
3
2+
2+
Ru(bpy)3 relative to the ungrafted Ru(bpy)3 , the interaction
of the dye with Al O appears to be significant for the reaction.
2
3
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 9
2
+*
To understand the nature of such an interaction, timeꢀresolved
photoluminescence (TRPL) spectroscopy was used to study
the chargeꢀtransfer dynamics between the ungrafted
is almost half that of Ru(bpy)3 alone. This suggests that in
addition to the aforementioned recombination process, there is
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
2
+*
also a dominant electron transfer process from Ru(bpy)
form Ru(bpy)3 . The idea that an additive or coꢀcatalyst can
render a reaction thermodynamically feasible by accepting and
transferring electrons has been frequently proposed in literaꢀ
However, DFT calculations reveal that the CBM of
the Al O surface is too high at ꢀ3.21 eV to receive an electron
to
3
2
+
2+
3+
Ru(bpy)3 and Al O . Firstly, Ru(bpy) was photoexcited
2
3
3
2
*
using 400 nm laser pulses. The excited species Ru(bpy)3 are
generated from the metalꢀtoꢀligand charge transfer (MLCT)
process and decay exponentially with time, yielding the reꢀ
sultant TRPL signal (Figure 5a). This PL decay arises from a
combination of radiative and nonꢀradiative processes: (1) the
59
60ꢀ62
ture.
2
3
2+*
from Ru(bpy)3 . Hence, it is hypothesized that electron transꢀ
fer may not have occurred directly from Ru(bpy)3 to Al O ,
2
+
2+*
recombination of charge carriers to regenerate Ru(bpy)3 ; and
2
3
(
2) the loss of an electron to O through electron transfer.
but to a O molecule strongly complexed to the surface of
2
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
ꢀ
ꢁ
Al
O , which may not have been released in the nitrogen atꢀ
2 3
2
+
2+*
2+
Ru(bpy)3
ꢂꢃ Ru(bpy)3
→ Ru(bpy)3
2+*
(1)
mosphere (Figure 6a). This hypothesis is consistent with a
ꢀ
ꢁ
previous report that O may play a crucial role in the electron
2+
3+
ꢀ
2
Ru(bpy)3 + O ꢂꢃ Ru(bpy)3 + O ꢄ→ Ru(bpy)3 + O2
(2)
63
2
2
transfer processes of organic adsorbates at the Al O surface.
2
3
DFT calculations show that in the gas phase of O , the π and
2
x
As a similar TRPL lifetime can be observed in the absence
π
y
are half occupied and have the same energy level. However,
when O strongly complexed with the AlIII site of Al , the π
and π become energetically inequivalent and lead to a singly
of O (Figure 5b and Table 1), the recombination pathway (1)
appears to be the dominant process. Although the oxidation
potential of O is suitable for receiving electrons from
Ru(bpy)3 , there appears to be little electron transfer between
Ru(bpy)3 and O within the 300 ns timeframe in the case of
Ru(bpy)3 alone. The lack of BnOH conversion with MgO
and SiO may also be due to negligible electron transfer from
2
2
O
3
x
2
y
occupied π orbital on the surface. This results in the appearꢀ
2
2
+*
ance of a localized vacant band at ꢀ5.07 eV (Figure 6b and c),
2
+*
2+*
which is suitable for accepting electrons from Ru(bpy)
to
2
3
2
+
ꢀ
yield superoxide O
III).
(Equations 3ꢀ4 and Figure 4a, steps IIꢀ
2
2
2
+*
2+*
Ru(bpy)3
to MgO or SiO , as the lifetimes of Ru(bpy)
2
3
on
ꢀꢁ
2+
2+*
2
+*
Ru(bpy)3 + O ꢀAl ꢂꢃ Ru(bpy)3 + O ꢀAl ꢄ
(3)
(4)
2
III
2
III
MgO and SiO are largely similar to that of Ru(bpy)3 alone
(
transfer between Ru(bpy)
2
Figure 5a and Table 1). There also appears to be no electron
2+*
3+
ꢀ
Ru(bpy)3 + O ꢀAl ꢄ→ Ru(bpy) + O ꢀAl
III
2
+*
2
III
2
3
and BnOH, as no change in the
3
2+*
2
+*
The loss of electrons from Ru(bpy)3
would then yield the
lifetime of Ru(bpy)3 was observed when BnOH was added
Figure S18).
3+
oxidant species Ru(bpy)3 , which would drive the reaction
forward by removing an electron from the adsorbed BnOH to
form an αꢀcarbon radical (Figure 4a, steps IIIꢀIV). The overall
process is such that, with Al O and the photoexcited
(
2
3
2+*
Ru(bpy)3 acting in tandem as a shuttle, an electron transfer
is enabled from BnOH to O . It should be noted that the localꢀ
2
ized vacant band of the O ꢀAl complex at ꢀ5.07 eV is also
2
III
suitable for accepting electrons from the excited states of all
dyes used in Figure 2 (Figure S19a), which could account for
the efficacy of the Al O ꢀdye surface complex system.
2
3
2
+
Figure 5. Decay kinetics of TRPL signals of Ru(bpy) and
Ru(bpy)3 adsorbed on Al O , SiO , and MgO under a) ambiꢀ
ent and b) N atmosphere.
3
2
+
2
3
2
2
Table 1. Decay lifetimes of the samples measured under amꢀ
bient and N atmospheres.
2
Atmos-
Entry
Sample
Lifetime (ns)
phere
Ambient
Ambient
Ambient
Ambient
N2
2
+
1
2
3
4
5
6
Ru(bpy)3
125 ± 2
63 ± 1
2+
Ru(bpy)3 + Al O
2
3
2
+
Ru(bpy)3 + SiO2
105 ± 2
122 ± 2
115 ± 2
55 ± 1
2+
Ru(bpy)3 + MgO
2
+
Ru(bpy)3
2+
Ru(bpy)3 + Al O
N2
2
3
2
+
Figure 6. a) Scheme of proposed electron transfer between
Finally, we examine the case where Ru(bpy)3 is adsorbed
on Al O . It can be seen that under both ambient and nitrogen
conditions, the lifetime of Ru(bpy)3 in the presence of Al O3
dye and O . b) Occupied and vacant states of the singly occuꢀ
2
2
3
2
+*
2
6
ACS Paragon Plus Environment

Page 7 of 9
Journal of the American Chemical Society
pied π orbital which arose due to O and Al O surface comꢀ
(2)
(3)
Yoon, T. P.; Ischay, M. A.; Du, J. Nat. Chem. 2010, 2, 527.
Narayanam, J. M. R.; Stephenson, C. R. J. Chem. Soc. Rev.
2
2
3
2
+
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
plexation. c) Energy levels of Ru(bpy)3 , Al O , and O .
2
3
2
2
011, 40, 102.
(
4)
Maeda, K.; Sahara, G.; Eguchi, M.; Ishitani, O. ACS Catal.
2
015, 5, 1700.
Kuriki, R.; Matsunaga, H.; Nakashima, T.; Wada, K.;
4. CONCLUSION
(
5)
Yamakata, A.; Ishitani, O.; Maeda, K. J. Am. Chem. Soc. 2016, 138,
5159.
In conclusion, we employed the unconventional strategy of
utilizing Al O surface complexation to modify the oxidation
2
3
(
7
6)
4.
Cuthbertson, J. D.; MacMillan, D. W. C. Nature 2015, 519,
potential of organic reactants, which would enable the oxidaꢀ
tion of organic reactants with high oxidation potentials inacꢀ
cessible to the unassisted photocatalyst. This has been successꢀ
fully demonstrated through the photocatalytic oxidation of
benzyl alcohols to benzaldehydes. High conversion and selecꢀ
tivity can be achieved with the surface complex systems comꢀ
prising Al O and a variety of dyes, but not with these dyes
(7)
(8)
Jin, J.; MacMillan, D. W. C. Nature 2015, 525, 87.
Zuo, Z.; Ahneman, D. T.; Chu, L.; Terrett, J. A.; Doyle, A.
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
G.; MacMillan, D. W. C. Science 2014, 345, 437.
(9)
45, 433.
10)
Tellis, J. C.; Primer, D. N.; Molander, G. A. Science 2014,
3
(
Shu, X.ꢀz.; Zhang, M.; He, Y.; Frei, H.; Toste, F. D. J. Am.
2
3
Chem. Soc. 2014, 136, 5844.
11) Sahoo, B.; Hopkinson, M. N.; Glorius, F. J. Am. Chem.
Soc. 2013, 135, 5505.
alone. The formation of the chemisorbed BnOꢀAl O surface
2
3
(
complex is attributed to the strong Brønsted base sites on
Al O , which accepts protons from benzylic OꢀH. This causes
2
3
(12)
Hopkinson, M. N.; Sahoo, B.; Li, J.ꢀL.; Glorius, F. Chem.
a shift in the oxidation potential of BnOH, which enhances the
ease of subsequent electron and CꢀH proton abstraction to
Eur. J. 2014, 20, 3874.
(13)
Rev. 2013, 113, 5322.
Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Chem.
form benzaldehyde. The surface complexation of O with
2
(
(
14)
15)
Vila, C. ChemCatChem 2015, 7, 1790.
Xie, J.; Shi, S.; Zhang, T.; Mehrkens, N.; Rudolph, M.;
Al O also activates the adsorbed O for receiving electrons
2
3
2
from the photoexcited dyes. This discovery may subvert our
understanding of the role of Al O in photocatalytic reactions.
Hashmi, A. S. Angew. Chem. Int. Ed. Engl. 2015, 54, 6046.
(16) Noble, A.; McCarver, S. J.; MacMillan, D. W. J. Am.
Chem. Soc. 2015, 137, 624.
17) Primer, D. N.; Karakaya, I.; Tellis, J. C.; Molander, G. A.
J. Am. Chem. Soc. 2015, 137, 2195.
(18) Wang, F.; Li, C.; Chen, H.; Jiang, R.; Sun, L. D.; Li, Q.;
Wang, J.; Yu, J. C.; Yan, C. H. J. Am. Chem. Soc. 2013, 135, 5588.
(19) Chu, L.; Lipshultz, J. M.; MacMillan, D. W. Angew. Chem.
Int. Ed. Engl. 2015, 54, 7929.
20) Higashimoto, S.; Kitao, N.; Yoshida, N.; Sakura, T.;
Azuma, M.; Ohue, H.; Sakata, Y. J. Catal. 2009, 266, 279.
21) Ohkubo, K.; Suga, K.; Fukuzumi, S. Chem. Commun.
2006, 2018.
2
3
For instance, it may be possible for Al O to play a role beꢀ
2
3
yond that of a mere support or scaffold in the performance
(
64
enhancement in solar hydrogen production
sensitized solar cells.
and dyeꢀ
6
5,66
This discovery brings forth a new
methodology of utilizing surface complexation mechanisms
between the reactants and earthꢀabundant materials to effecꢀ
tively achieve a wider range of photoredox reactions.
(
ASSOCIATED CONTENT
Supporting Information
(
(22)
Tanaka, A.; Hashimoto, K.; Kominami, H. J. Am. Chem.
Experimental procedures, control experiments, DFT calculations,
material surface characterizations and the reaction mechanism are
included in the supporting information.
Soc. 2012, 134, 14526.
(
23)
Commun. 2014, 50, 8651.
24) Lang, X.; Leow, W. R.; Zhao, J.; Chen, X. Chem. Sci.
2015, 6, 1075.
Chen, L.; Peng, Y.; Wang, H.; Gu, Z.; Duan, C. Chem.
(
AUTHOR INFORMATION
Corresponding Authors
(25)
55, 4697.
Lang, X.; Zhao, J.; Chen, X. Angew. Chem. Int. Ed. 2016,
(
26)
Ed. Engl. 2008, 47, 9730.
27) Sun, M.; Zhang, J.; Putaj, P.; Caps, V.; Lefebvre, F.;
Pelletier, J.; Basset, J.ꢀM. Chem. Rev. 2014, 114, 981.
Zhang, M.; Chen, C.; Ma, W.; Zhao, J. Angew. Chem. Int.
*
*
chenxd@ntu.edu.sg (X.C.)
hirao@ntu.edu.sg (H.H.)
(
Author Contributions
(28)
29)
(30)
Beller, M. Adv. Synth. Catal. 2004, 346, 107.
Largeron, M.; Fleury, M. B. Science 2013, 339, 43.
Largeron, M.; Fleury, M. B. Angew. Chem. Int. Ed. Engl.
The manuscript was written through contributions of all authors.
All authors have given approval to the final version of the manuꢀ
script.
(
2012, 51, 5409.
(
31)
73.
(32)
Lang, X.; Chen, X.; Zhao, J. Chem. Soc. Rev. 2014, 43,
Notes
4
The authors declare no competing financial interests.
Shishido, T.; Miyatake, T.; Teramura, K.; Hitomi, Y.;
Yamashita, H.; Tanaka, T. J. Phys. Chem. C 2009, 113, 18713.
(33) Nagib, D. A.; Scott, M. E.; MacMillan, D. W. C. J. Am.
Chem. Soc. 2009, 131, 10875.
34) Stevenson, S. M.; Shores, M. P.; Ferreira, E. M. Angew.
Chem. Int. Ed. Engl. 2015, 54, 6506.
ACKNOWLEDGMENT
We gratefully acknowledge financial support from Singapore
Ministry of Education Tier 1 (RG130/14) and the Singapore NRF
through the SingaporeꢀBerkeley Research Initiative for Sustainaꢀ
ble Energy (SinBeRISE) CREATE Programme. T.C.S. acknowlꢀ
edges the financial support from Singapore Ministry of Education
Tier 2 (MOE2013ꢀT2ꢀ1ꢀ081 and MOE2014ꢀT2ꢀ1ꢀ044). H.H. is
grateful for a Nanyang Assistant Professorship and a JSTꢀ
PRESTO grant.
(
(35)
5517.
CanoꢀYelo, H.; Deronzier, A. Tetrahedron Lett. 1984, 25,
(
36)
Kim, W.; Tachikawa, T.; Majima, T.; Choi, W. J. Phys.
Chem. C 2009, 113, 10603.
37) Canlas, C. P.; Lu, J.; Ray, N. A.; GrossoꢀGiordano, N. A.;
(
Lee, S.; Elam, J. W.; Winans, R. E.; Van Duyne, R. P.; Stair, P. C.;
Notestein, J. M. Nat. Chem. 2012, 4, 1030.
(38)
REFERENCES
Ealet, B.; Elyakhloufi, M. H.; Gillet, E.; Ricci, M. Thin
Solid Films 1994, 250, 92.
(1)
Devery III, J. J.; Stephenson, C. R. J. Nature 2015, 519, 42.
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 9
(
39)
Shimizu, K.ꢀi.; Sugino, K.; Sawabe, K.; Satsuma, A. Chem.
(55)
Wang, Q.; Zhang, M.; Chen, C.; Ma, W.; Zhao, J. Angew.
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
Eur. J. 2009, 15, 2341.
Chem. Int. Ed. Engl. 2010, 49, 7976.
(
3
(
40)
393.
41)
Trueba, M.; Trasatti, S. P. Eur. J. Inorg. Chem. 2005, 2005,
(56)
Angew. Chem. Int. Ed. Engl. 2014, 53, 2951.
(57) Zhang, M.; Wang, Q.; Chen, C.; Zang, L.; Ma, W.; Zhao, J.
Angew. Chem. Int. Ed. Engl. 2009, 48, 6081.
(58) Suh, M.; Bagus, P. S.; Pak, S.; Rosynek, M. P.; Lunsford,
J. H. J. Phys. Chem. B 2000, 104, 2736.
(59) Asbury, J. B.; Ellingson, R. J.; Ghosh, H. N.; Ferrere, S.;
Nozik, A. J.; Lian, T. J. Phys. Chem. B 1999, 103, 3110.
Liang, S.; Wen, L.; Lin, S.; Bi, J.; Feng, P.; Fu, X.; Wu, L.
Torres Galvis, H. M.; Bitter, J. H.; Khare, C. B.;
Ruitenbeek, M.; Dugulan, A. I.; de Jong, K. P. Science 2012, 335,
35.
(42)
H. H.; Stair, P. C. Science 2012, 335, 1205.
43) Korhonen, S. T.; Beale, A. M.; Newton, M. A.;
Weckhuysen, B. M. J. Phys. Chem. C 2011, 115, 885.
44) ThibaultꢀStarzyk, F.; Seguin, E.; Thomas, S.; Daturi, M.;
Arnolds, H.; King, D. A. Science 2009, 324, 1048.
(45) Teramura, K.; Tanaka, T.; Yamamoto, T.; Funabiki, T. J.
Mol. Catal. A: Chem. 2001, 165, 299.
46) Digne, M.; Sauteta, P.; Raybaud, P.; Euzenc, P.; Toulhoatd,
H. J. Catal. 2004, 226, 54.
8
Lu, J.; Fu, B.; Kung, M. C.; Xiao, G.; Elam, J. W.; Kung,
(
(60)
(61)
Riener, M.; Nicewicz, D. A. Chem. Sci. 2013, 4, 2625.
Zhang, J.; Chen, J.; Zhang, X.; Lei, X. J. Org. Chem. 2014,
(
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
79, 10682.
(62)
Kranz, D. P.; Griesbeck, A. G.; Alle, R.; PerezꢀRuiz, R.;
Neudörfl, J. M.; Meerholz, K.; Schmalz, H.ꢀG. Angew. Chem. Int.
Ed. 2012, 51, 6000.
(
(63)
Flockhart, B. D.; Scott, J. A. N.; Pink, R. C. J. Chem. Soc.
(
(
47)
48)
Kresse, G.; Furthmüller, J. Phys. Rev. B 1996, 54, 11169.
Kresse, G.; Furthmüller, J. Comput. Mater. Sci. 1996, 6, 15.
Kresse, G.; Hafner, J. Phys. Rev. B 1993, 47, 558.
Kresse, G.; Hafner, J. Phys. Rev. B 1994, 49, 14251.
Blöchl, P. E. Phys. Rev. B 1994, 50, 17953.
Faraday Trans. 1966, 62, 730.
(64)
Li, F. T.; Liu, S. J.; Xue, Y. B.; Wang, X. J.; Hao, Y. J.;
(49)
Zhao, J.; Liu, R. H.; Zhao, D. Chem. Eur. J. 2015, 21, 10149.
(
(
(
(
50)
51)
52)
53)
(65)
Snaith, H. J. Science 2012, 338, 643.
(66) Carnie, M. J.; Charbonneau, C.; Davies, M. L.; Troughton,
Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.;
Kresse, G.; Joubert, D. Phys. Rev. B 1999, 59, 1758.
Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett.
J.; Watson, T. M.; Wojciechowski, K.; Snaith, H.; Worsley, D. A.
Chem. Commun. 2013, 49, 7893.
1
996, 77, 3865.
(54)
27, 714.
Roth, H. G.; Romero, N. A.; Nicewicz, D. A. Synlett 2016,
8
ACS Paragon Plus Environment

Page 9 of 9
Table of Contents
Journal of the American Chemical Society
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
We discovered that the oxidation potentials of benzyl alcohols and O can be reduced through surface complexation with
2
Al O , thereby allowing its photocatalytic transformation to benzaldehydes in the presence of different dyes.
2
3
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
9
ACS Paragon Plus Environment