Effect of Precursor Selection on the Photocatalytic Performance of Indium Oxide Nanomaterials for Gas-Phase CO2 Reduction

Article abstract of DOI:10.1021/acs.chemmater.6b00301

Nonstoichiometric indium oxide nanoparticles, In₂O_3-x(OH)_y, have been shown to function as active photocatalysts for gas-phase CO₂ reduction under simulated solar irradiation. Herein we demonstrate that the choice of starting material has a strong effect on the photocatalytic activity of indium oxide nanoparticles. We examine three indium oxide materials prepared via the thermal decomposition of either indium(III) hydroxide or indium(III) nitrate and correlate their stability and photocatalytic activity to the number and type of defect present in the material. Further, we use ¹³CO₂ isotope-tracing experiments to clearly identify the origins of the observed carbon-containing products. Significantly, we find that the oxidizing nature of the precursor anion has a substantial impact on the defect formation within the sample. This study demonstrates the importance of surface defects in designing an active heterogeneous photocatalyst and provides valuable insight into key parameters for the precursor design, selection, and performance optimization of materials for gas-phase CO₂ reduction.

Full text of DOI:10.1021/acs.chemmater.6b00301

Article
pubs.acs.org/cm
Effect of Precursor Selection on the Photocatalytic Performance of
Indium Oxide Nanomaterials for Gas-Phase CO2 Reduction
Laura B. Hoch,† Le He,†,‡ Qiao Qiao,§,∥ Kristine Liao,† Laura M. Reyes,† Yimei Zhu,§
and Geoffrey A. Ozin*,†
†Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, Ontario M5S 3H6, Canada
§Condensed Matter Physics and Materials Science Department, Brookhaven National Laboratory, Upton, New York 11973, United
States
∥Department of Physics, Temple University, Philadelphia, Pennsylvania 19122, United States
S
* Supporting Information
ABSTRACT: Nonstoichiometric indium oxide nanoparticles, In2O3−x(OH)y,
have been shown to function as active photocatalysts for gas-phase CO2
reduction under simulated solar irradiation. Herein we demonstrate that the
choice of starting material has a strong effect on the photocatalytic activity of
indium oxide nanoparticles. We examine three indium oxide materials
prepared via the thermal decomposition of either indium(III) hydroxide or
indium(III) nitrate and correlate their stability and photocatalytic activity to
the number and type of defect present in the material. Further, we use 13CO2
isotope-tracing experiments to clearly identify the origins of the observed
carbon-containing products. Significantly, we find that the oxidizing nature of
the precursor anion has a substantial impact on the defect formation within
the sample. This study demonstrates the importance of surface defects in
designing an active heterogeneous photocatalyst and provides valuable insight
into key parameters for the precursor design, selection, and performance optimization of materials for gas-phase CO2 reduction.
vacancies can also act as important active sites in heterogeneous
catalysis reactions, facilitating reactant adsorption and con-
version.7,24 For example, in TiO2, it has been suggested that
surface oxygen vacancies act as charge traps, facilitating electron
transfer to adsorbed species, while bulk oxygen vacancies act as
recombination centers, lowering the excited state charge carrier
lifetimes and adversely affecting the photocatalytic activity.25
Surface hydroxide groups have also been shown to play a key
role in photocatalysis reactions.13,24,26,27 In many materials,
they have commonly been associated with facilitating the
adsorption of gaseous reactant species. Indeed, surface
hydroxides have a known affinity for CO2, typically forming
surface adsorbed carbonate or bicarbonate species.28 Several
groups have reported positive correlations with hydroxide
content, CO2 capture capacity, and enhanced photocatalytic
activity.24,26,27
INTRODUCTION
■
Harnessing abundant solar energy to drive the conversion of
greenhouse gas CO2 into carbon-based fuels and chemical
feedstocks − a process analogous to nature’s photosynthesis −
is an area of growing scientific focus, with the ultimate goal of
providing a sustainable alternative to traditional fossil-based
carbon.1−4 Enabling this technology would allow the capture
and conversion of CO2 emitted from energy production and
manufacturing exhaust streams into valuable products, thereby
creating significant economic and environmental benefits.5−7
Of particular interest is the development of gas-phase
heterogeneous photocatalysts based on semiconductor nano-
materials, which can efficiently harvest incident solar energy,
utilizing it to drive useful molecular transformations at its
surface.3,8,9 Nanostructured metal oxides have shown significant
promise as photocatalysts for CO2 reduction, due to their
stability, ease of synthesis, and scale-up.10−12
The role of bulk and surface defects in determining the
photocatalytic activity of metal oxide semiconductor photo-
catalysts has been an area of intense research.13−16 In metal
oxides, perhaps the most ubiquitous and well-studied type of
defect is the oxygen vacancy.17−19 These oxygen vacancies have
been shown to have a significant impact on the optical and
electronic properties.13,20 Typically acting as donor states, they
are generally cited as the primary reason for the n-type
conductivity associated with most metal oxides.21−23 Oxygen
In our previous work, we demonstrated that nonstoichio-
metric indium oxide nanoparticles, In2O3−x(OH)y, function as
active photocatalysts for CO2 reduction to CO via the reverse
water gas shift (RWGS) reaction.29 We correlated both the
photocatalytic activity and CO2 capture capacity to the oxygen
vacancy and surface hydroxide content of the In2O3−x(OH)y
Received: January 22, 2016
Revised: May 31, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.chemmater.6b00301
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
(III) chloride (anhydrous, Alfa Aesar, 99.99%) at 250 °C for 3 h. The
resulting powders were finely ground with a mortar and pestle.
High surface area indium oxide made from the indium(III) nitrate
precursor was prepared via a SiO2-templating method, illustrated in
Figure S3. Indium(III) nitrate hydrate (3 g, Sigma-Aldrich, 99.9%) was
dissolved in 30 g of Ludox HS-40 colloidal silica nanoparticles
(Aldrich, 40 wt % in water) and sonicated for 30 min to ensure
homogeneity. The solution was dried overnight at 80 °C, and the
resulting solid was directly calcined at 250 °C for 3 h. After calcination
the solid changed from white to pale yellow indicating the formation
of indium oxide. This In2O3/SiO2 mixture was then ground with an
agate mortar and pestle, and the SiO2 was removed by base etching.
Specifically, the In2O3/SiO2 powder sample was dispersed in 2 M
NaOH (10× excess by mass) and sonicated overnight (approximately
20 h). The suspension was then heated to 80 °C with a high stirring
rate and heated for an additional 24 h. The resulting sample was
separated via centrifugation and purified by washing five times with
deionized nanopure water, sonicating between washings to ensure
adequate removal of any trapped impurities. The final precipitate was
dried overnight at 70 °C in a vacuum oven and finely ground with a
mortar and pestle. Several control samples were also prepared. The
indium nitrate was replaced with the indium hydroxide precursor (0.6
g), combined with 10 g of Ludox HS-40, sonicated, dried, calcined,
etched, and purified in exactly the same way as described above.
Additionally, two indium oxide samples (0.5 g, prepared without SiO2
from standard indium nitrate and indium hydroxide precursors as
described in the previous paragraph) were placed in 50 mL of 2 M
NaOH, exposed to the same basic etching conditions and purified in
the same way as described above.
Sample films were prepared for photocatalytic testing by drop-
casting 20 mg of each sample powder - suspended via sonication in
deionized, nanopure water (3 mL) - onto 1″ × 1″ binder free
borosilicate glass microfiber filters (Whatman, GF/F, 0.7 μm).
Physical Characterization. Powder X-ray diffraction (PXRD) was
performed on a Bruker D2-Phaser X-ray diffractometer, using Cu Kα
radiation at 30 kV. Nitrogen adsorption isotherms were obtained at 77
K using a Quantachrome Autosorb-1-C, and the surface area was
determined using Brunauer-Emmet-Teller (BET) theory. Sample
morphology was determined using an aberration corrected JEOL
JEM-ARM200CF scanning transmission electron microscope
(STEM). The atomic resolution high angle annular dark field
(HAADF) and low magnification bright field images were taken at
200 kV. The diffuse reflectance of the samples was measured using a
PerkinElmer Lambda 1050 UV/Vis/NIR spectrometer equipped with
an integrating sphere with a diameter of 150 mm.
samples, suggesting both defects play a key role in the reaction
mechanism.30,31 These samples were produced via thermal
dehydroxylation of indium(III) hydroxide. During this reaction,
adjacent hydroxide groups condense to form a bridging oxide
and release water. We demonstrated that the calcination
temperature during the synthesis of these nanoparticles had a
strong effect on the amount of hydroxide groups retained
within the sample: lower temperatures resulted in less
condensation and formation of bridging oxide bonds, resulting
in the retention of more hydroxides and oxygen vacancy
defects. We observed a strong positive correlation between CO2
capture capacity and hydroxide content. However, an In(OH)3
control sample, which had the highest hydroxide content,
showed a significantly lower CO2 capture capacity, even when
corrected for surface area. This indicated that surface
hydroxides alone are not sufficient to facilitate CO2 capture
and subsequent photocatalytic reduction of CO2. However, due
to the temperature-dependence of the synthesis, which affects
the retention of both hydroxides and oxygen vacancies, it is
difficult to separate and evaluate the effects these two defects
have on the reactivity of the different samples.
In order to gain better insight into the reaction mechanism
and understand the role hydroxides and oxygen vacancies play
in the formation of an active indium oxide based photocatalyst,
we sought to make indium oxide from a non-hydroxide
containing precursor. Herein we examine the photocatalytic
activity of three different indium oxide samples prepared via the
thermal decomposition of either indium(III) hydroxide or
indium(III) nitrate. Our results demonstrate that the choice of
starting material has a significant effect on the number and type
of defects present in the material. In particular, we find that the
oxidizing nature of the precursor anion has a substantial impact
on the defect formation within the sample, which strongly
correlates with the stability and photocatalytic activity. This
work significantly advances our understanding of the role
precursor anions play in determining a material’s photocatalytic
activity and further illustrates the importance of surface defects
in designing an active material capable of facilitating gas-phase
CO2 reduction.
X-ray photoelectron spectroscopy (XPS) was performed using a
PerkinElmer Phi 5500 ESCA spectrometer in an ultrahigh vacuum
chamber with base pressure of 1 × 10−9 Torr. The spectrometer uses
an Al Kα X-ray source operating at 15 kV and 27 A. The samples used
in XPS analyses were prepared by drop-casting aqueous dispersions
onto p-doped Si(100) wafers. All data analyses were carried out using
the Multipak fitting program, and the binding energies were referenced
to the NIST-XPS database32 and the Handbook of X-ray Photo-
electron Spectroscopy.43
Gas-Phase Photocatalytic Measurements. Gas-phase photo-
catalytic rate measurements were conducted in a custom fabricated 1.5
mL stainless steel batch reactor with a fused silica view port sealed with
Viton O-rings. Briefly, the reactors were evacuated using an Alcatel dry
pump prior to being purged with H2 gas (99.9995%) at a flow rate of 6
mL/min for 1 h. The outlet valve was sealed, and the reactors were
injected with 1 atm of 13CO2 (99.9 atomic % Sigma-Aldrich) for a total
reactor pressure of 2 atm. The inlet valve was then sealed, and the
reactors were heated to 150 °C. The reactor temperatures were
controlled by an OMEGA CN616 6-Zone temperature controller, with
a thermocouple placed in contact with the sample. The pressure inside
the reactor was monitored during the reaction using an Omega PX309
pressure transducer. Reactors were irradiated with a 1000 W Hortilux
Blue metal halide bulb for a period of 16 h. Product gases were
analyzed by a flame ionization detector (FID) and thermal
conductivity detector (TCD) installed in a SRI-8610 Gas Chromato-
EXPERIMENTAL SECTION
■
Synthesis of Indium Oxide Nanoparticles. All chemicals were
used as received without any further purification. The indium(III)
hydroxide precursor was prepared by dissolving indium(III) chloride
(3.6 g, 16.2 mmol, Sigma-Aldrich, 98%) in a 3:1 solution (72 mL) of
anhydrous ethanol (Commercial Alcohols) and deionized, nanopure
water (resistivity 18.2 MΩ cm). In a separate beaker, a 3:1 mixture of
ethanol and ammonium hydroxide was prepared by combining
aqueous ammonium hydroxide (18 mL, Caledon, 28−30% adjusted
to 25 wt % with deionized water) with anhydrous ethanol (54 mL).
The solutions were rapidly combined, resulting in the immediate
formation of a white precipitate. To control the particle size, the
resulting suspension was immediately immersed in a preheated oil bath
at 80 °C and stirred for 10 min. The suspension was then removed
from the oil bath and allowed to cool to room temperature. The
precipitate was separated via centrifugation and washed 3 times with
deionized water. The precipitate was sonicated between washings to
ensure adequate removal of any trapped impurities and then dried
overnight at 80 °C in a vacuum oven. The dried precursor powder was
finely ground with a mortar and pestle.
Each of the indium oxide samples were made by calcining 0.8 g of
the respective precursor − either indium(III) hydroxide (prepared as
described above), indium(III) nitrate hydrate (Sigma-Aldrich, 99.9%),
indium(III) acetate (anhydrous, Acros Organics, 99.99%), or indium-
B
DOI: 10.1021/acs.chemmater.6b00301
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 1. (a) Powder X-ray diffraction patterns for samples made from different indium(III) precursors. (b-d) Scanning transmission electron
microscope (STEM) images of (b) In2O3 made from indium(III) hydroxide, (c) In2O3 made from indium(III) nitrate, and (d) high surface area
(HSA) In2O3 sample prepared from indium(III) nitrate via the SiO2 templating method.
graph (GC) with a 3′ Mole Sieve 13a and 6′ Haysep D column.
Isotope product gases were measured using an Agilent 7890A gas
chromatographic mass spectrometer (GC-MS) with a 60 m GS-
Carbonplot column fed to the mass spectrometer.
1c) exhibits larger, more faceted polydisperse nanoparticles in
the range of 20−50 nm. From these images, it is evident that
the nitrate-based In2O3 sample has a much lower surface area
than the hydroxide-based sample. To quantify this, we
measured their BET surface areas and found that there was a
stark difference: 3 m2/g for the nitrate-based sample and 125
m2/g for the hydroxide-based sample. Because the photo-
reduction of CO2 on indium oxide is a surface-mediated
reaction, the substantially lower surface area of the nitrate-based
sample could result in significantly reduced apparent photo-
catalytic activity.
In order to be able to more directly compare the nitrate- and
hydroxide-based samples, we sought to synthesize a higher
surface area nitrate-based In2O3 sample. As was demonstrated
in previous work, the calcination temperature has a significant
impact on the catalytic activity; therefore, maintaining the
calcination time and temperature is important to ensure
consistency between the samples. Because In2O3 is generally
considered to be very stable under basic conditions,33 we chose
to use SiO2 spheres as an inert template, which could then be
etched away using a strong base. In this process, illustrated
schematically in Figure S3, the indium(III) nitrate precursor is
dissolved in an aqueous suspension of SiO2 nanoparticles and
allowed to dry slowly overnight, encapsulating the SiO2
nanoparticles within the indium nitrate salt. Once dry, the
sample is then calcined for 3 h at 250 °C to convert the indium
nitrate into In2O3. To remove the SiO2, the sample is
mechanically ground into a fine powder and then dispersed
RESULTS AND DISCUSSION
■
Preparation of Indium Oxide Nanoparticles from
Different Precursors. In order to better understand the role
hydroxides play in the formation of an active In2O3-based
photocatalyst, we sought to make indium oxide from a non-
hydroxide containing precursor. To this end, we calcined four
indium(III) precursor salts with different anions (nitrate,
acetate, chloride, and hydroxide) at 250 °C for 3 h, conditions
which our previous studies have demonstrated are optimal for
producing the most active indium oxide photocatalysts.29 Of
the four indium(III) precursor salts, only indium(III) hydroxide
and indium(III) nitrate produced pure crystalline indium oxide
with no observable amorphous phases, as shown in the powder
X-ray diffraction (PXRD) patterns in Figure 1a. The indium-
(III) chloride and indium(III) acetate precursors, on the other
hand, did not produce pure indium oxide as shown in Figures
S1 and S2 in the Supporting Information.
Figures 1b and 1c show the STEM images of the indium
oxide samples produced from indium(III) hydroxide and
indium(III) nitrate, respectively. From the images it is clear
that the morphology differs significantly between the samples.
The hydroxide-based In2O3 sample (Figure 1b) is comprised of
small agglomerated nanoparticles with diameters less than 10
nm. On the other hand, the nitrate-based In2O3 sample (Figure
C
DOI: 10.1021/acs.chemmater.6b00301
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
in a strongly basic solution. In order to reduce particle
aggregation and ensure as much of the SiO2 is exposed to the
basic solution as possible, this suspension was sonicated
overnight prior to heating at 80 °C for 24 h to complete the
etching procedure. The resulting fine powder was washed
extensively with deionized water to remove impurities.
Figure 1d shows an STEM image of the high surface area
(HSA)-nitrate-based sample prepared using this method. The
highly porous structure is evident from this image, and gas
adsorption measurements indicate that the BET surface area of
the nitrate sample was successfully increased from 3 m2/g to
181 m2/g. The PXRD patterns in Figure 1a show the
comparison of the HSA-nitrate-based sample with the original
nitrate-based sample and the hydroxide-based sample. As the
PXRD pattern indicates, the HSA-nitrate sample is comprised
of pure crystalline In2O3. No evidence of any crystalline side
phases or residual amorphous SiO2 is observed. The peaks of
the HSA-nitrate sample show significant broadening relative to
the nitrate- and hydroxide-based samples, indicating a decrease
in particle size, which is consistent with the TEM data.
Conversely, both nitrate-based In2O3 samples did not
produce 13CO above the baseline level, nor were any other
C-containing products observed. Because both the original
nitrate-based sample and the HSA-nitrate-based In2O3 sample
show no evidence of photocatalytic activity, this demonstrates
that surface area alone is not responsible for the observed
absence of activity for the nitrate-based samples.
Characterization of Defect Formation as a Function of
In2O3 Precursor. In order to better understand why the
nitrate-based samples are inactive toward CO2 reduction, X-ray
photoelectron spectroscopy (XPS) measurements were per-
formed to characterize the defects present in the In2O3 samples,
as well as confirm the sample composition to determine if there
are any impurities which might be adversely impacting the
photocatalytic activity. Figure S4 shows the survey spectra for
all three samples, which are remarkably similar. No nitrogen
signal is observed in either of the nitrate-based samples
(Figures S4b and c). The N 1s peak should appear at a binding
energy between 397 and 408 eV;32 however, there is no
evidence of any signal in this region of the spectrum. This
indicates that nitrogen is sufficiently removed upon calcination
of the indium nitrate and should not affect the photocatalytic
performance of the nitrate-based samples. The only obvious
difference in these spectra is the appearance of Si 2s and 2p
peaks at 157 ev and 105 eV, respectively, in the HSA-nitrate-
based sample (Figure S4c). This is likely due to residual SiO2
nanoparticles that were completely encapsulated by In2O3 and
therefore inaccessible to the NaOH etching solution. However,
SiO2 is a commonly used inert support material, and it is
unlikely that residual SiO2 nanoparticles had a negative impact
on the photocatalytic activity of the HSA-nitrate-based
samples.36 Based on these results, we do not expect that
atomic impurities are responsible for the difference in reactivity
between the hydroxide-based and nitrate-based In2O3 samples.
Another possible reason for the difference in activity between
the samples could be related to the type and quantity of defects
within the In2O3 itself. In our previous work, we demonstrated
the effect of calcination temperature on the photocatalytic
activity of the In2O3 samples and the strong correlation of
activity to both oxygen vacancies and surface hydroxides.29 As
the calcination temperature was increased, the quantity of both
hydroxide and oxygen vacancy defects decreased, as did the
photocatalytic activity. In this study, all samples were calcined
at the same temperature, 250 °C; however as indicated by the
O 1s core level peaks shown in Figure 3, there is a substantial
difference between the three In2O3 samples. The shoulder peak
that appears at higher binding energies indicates that there is
more than one chemical state of oxygen present in the
structure. Indeed, the O 1s peak can be deconvoluted into three
distinct peaks: the main oxide peak at 529.9 eV and two
additional peaks at 531.0 and 532.1 eV (Figure 3b-d). The peak
at 531.0 eV is commonly attributed to the presence of oxygen
vacancies in the structure.37,38 The shift to higher binding
energy for this oxygen vacancy peak relative to the main oxide
peak is a result of the change in oxygen atom interaction with
indium centers that are more reduced in character because they
are surrounded by less than 6 oxygen atoms (due to the oxygen
vacancies). The peak at 532.1 eV is generally attributed to
surface OH groups.37 From these plots, one can obtain a
qualitative understanding of the relative differences in defect
type and quanitity in the three In2O3 samples. As shown in
Figure 3a, the regular nitrate-based and the hydroxide-based
samples have shoulder peaks with similar intensities; however,
Investigating the Effects of Precursor Choice on
Photocatalytic Activity. Figure 2 shows the photocatalytic
Figure 2. Photocatalytic performance of the three indium oxide
samples made from different precursors under 800 W m−2 (0.8 suns)
illumination at 150 °C. All values for carbon monoxide were confirmed
using 13C isotope tracing.
activity of the three In2O3 samples under 800 W·m−2 (0.8 suns)
illumination at 150 °C under 2 atm of 13C-labeled 13CO2 and
H2 in a 1:1 ratio, which is stoichiometric for the reverse water
gas shift reaction. 13C-labeling is essential in order to prove that
the source of the carbon containing products originates from
the reactant CO2 and not from carbon contamination on the
catalyst’s surface or in the reactor. Of the three samples, only
the In2O3 made from the hydroxide-based precursor showed
activity for 13CO2 reduction, producing pure 13C-labeled CO at
a rate of 1.38 μmol·gcat−1·h−1 with no evidence of any other C-
containing products. This indicates that 13CO is the sole
product of CO2 reduction, originating from the reactant 13CO2
gas and not from adventitious carbon contamination. For
reference, this reaction rate is consistent with 13CO production
rates reported for other single component metal oxide materials
under similar conditions; for instance 1.61 μmol·gcat−1·h−1 for
MgO34 and 0.56 μmol·gcat−1·h−1 for ZrO2.35
D
DOI: 10.1021/acs.chemmater.6b00301
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 3. Comparison of XPS spectra for In2O3 samples prepared from either indium hydroxide or indium nitrate precursors. (a) Overlay of the O
1s peaks for all samples. Qualitative deconvolution of the O 1s peaks for In2O3 prepared from (b) indium(III) hydroxide, (c) indium(III) nitrate, and
(d) high surface area indium(III) nitrate. The main peak at 529.9 eV is attributed to indium oxide. The two additional shoulder peaks at 531.0 and
532.1 eV are attributed to oxygen vacancies in the structure and surface OH groups, respectively.
the primary difference between them is their shape. The nitrate-
based sample has a higher signal at higher binding energies,
indicating a greater contribution from hydroxides (Figure 3c),
while the hydroxide-based sample has a higher intensity at
intermediate binding energies, indicating a greater contribution
from oxygen vacancies (Figure 3b). Indeed the hydroxide-based
sample has the highest contribution from oxygen vacancies of
all three samples that were measured. The HSA-nitrate-based
sample has the highest shoulder peak overall and has a
substantial contribution in the hydroxide region (Figure 3d),
which is not surprising given that this sample was prepared by
etching in a strong base for 2 days. It should be noted that Si−
O bonds arising from the encapsulated SiO2 nanoparticles may
also contribute to the O 1s shoulder peak. However, as shown
in the XPS survey spectrum (Figure S4c), the Si signal is very
small relative to the In and O signals, so the contribution of Si−
O bonds to the O 1s peak would likely also be very small.
Further, the signal arising from Si−O bonds tends to be at
higher binding energies (∼532.5−533.0 eV) than what we
observed for the shoulder peak, which indicates that while there
may be some contribution from Si−O bonds, it is not likely the
dominant source of the O 1s shoulder peak.
These results, comparing the different defect populations in
the hydroxide-based and nitrate-based samples, indicate that
the lower oxygen vacancy content in the nitrate-based sample
relative to the hydroxide-based sample could be responsible for
the observed difference in photocatalytic activity. Indeed, many
studies have demonstrated that an increase in oxygen vacancy
content is strongly correlated to an increase in photocatalytic
activity in metal oxide materials.13,39 Even in relatively small
quantities, oxygen vacancies can play a key role in determining
the activity of heterogeneous catalysts by not only acting as
active sites for the reaction to take place but also altering the
electronic properties of the material.21,40,41 Therefore, although
the difference in oxygen vacancies population is small, it could
explain why the hydroxide-based sample has a much higher
photocatalytic activity.
Because the population of defects clearly has a strong impact
on photocatalytic activity of indium oxide, in order to
determine if the population changes under reaction conditions
(e.g., if for example the CO2 were to react stoichiometrically to
fill oxygen vacancies), we performed an additional control
experiment where we measured the XPS spectra before and
after the active hydroxide-based sample was placed in the
photocatalytic reactor. As shown in Figure S5, even after 4 days
of continuous reaction, no significant change was observed in
the O 1s shoulder peak, confirming that the population of
defects does not change under reaction conditions as is
expected for a catalytic material.
Investigating the Effect of Precursor Choice on
Sample Stability under Basic Conditions. Because the
synthesis of the HSA-nitrate sample necessitated strongly basic
etching conditions to remove the SiO2 template, it is important
to establish if this treatment could have adversely affected the
sample’s reactivity. As a control, a sample of hydroxide-based
In2O3 was subjected to identical etching conditions with the
intent to test its photocatalytic activity and compare it to the
HSA-nitrate-based sample. However, as the powder X-ray
diffraction pattern in Figure 4a indicates, the In2O3 sample,
which originally was comprised of pure crystalline In2O3 prior
to etching, was converted into pure crystalline In(OH)3
following the etching procedure. This was quite a surprising
E
DOI: 10.1021/acs.chemmater.6b00301
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 4. Powder X-ray diffraction patterns before and after base treatment of calcined powder samples containing (a) In2O3 only − from indium
hydroxide, (b) In2O3 only − from indium nitrate, (c) In2O3 + SiO2 − from indium hydroxide, and (d) In2O3 + SiO2 − from indium nitrate. The y-
axis of all four graphs represents intensity in arbitrary units.
result because the HSA-nitrate-based sample withstood this
treatment very well and showed no evidence of In(OH)3
formation. Figure 4d shows the PXRD patterns of the HSA-
nitrate sample before and after base etching to remove its SiO2
template. The diffraction pattern before etching shows a broad
amorphous peak around 22 degrees 2θ, originating from the
amorphous SiO2 nanoparticles, but the rest of the pattern is
comprised of pure crystalline In2O3. The diffraction pattern
after etching also shows pure crystalline In2O3 with no evidence
of any other crystalline or amorphous phases. This raises two
questions: (1) how does the presence of SiO2 affect sample
stability under etching conditions and (2) is there something
chemically different about the nitrate-based In2O3 relative to
the hydroxide-based In2O3.
To answer these questions, we subjected two additional
control samples to the same basic etching conditions: a pure
nitrate-based In2O3 sample (made without SiO2) and a
hydroxide-based sample that was combined with SiO2, dried,
and calcined in exactly the same way as the HSA-nitrate-based
sample. The result of etching the pure nitrate-based sample is
shown in Figure 4b. While the nitrate-based In2O3 started out
as a highly crystalline In2O3 sample, after etching the diffraction
pattern shows a mixture of crystalline In2O3 and crystalline
In(OH)3. This demonstrates that the pure nitrate-based sample
is not as stable as the HSA-nitrate-based sample that was etched
in the presence of SiO2. The result of etching the second
hydroxide-based control sample is shown in Figure 4c. The
diffraction pattern before etching is completely amorphous,
despite increasing the data collection time to enhance the
signal-to-noise ratio. It is possible that the presence of the SiO2
template inhibited the formation of the In2O3, though this is
unlikely as the sample turned pale yellow after calcination,
which is indicative of In2O3 formation. It is also possible that
the In2O3 particles are so small, due to the presence of the SiO2
template, that they do not diffract well enough to be seen over
the amorphous background. Regardless of whether In2O3 was
or was not formed, it is very interesting to note that, as shown
in the diffraction pattern after etching, the sample does not
form crystalline In(OH)3 under the basic etching conditions (as
opposed to the hydroxide-based In2O3 sample made without
SiO2 templating mentioned previously). Instead, this weakly
crystalline pattern is almost identical to the starting diffraction
pattern of the original indium hydroxide precursor (Figure S6).
From these data, we can conclude that SiO2 is preferentially
etched over indium-containing species, and the presence of
SiO2 seems to inhibit the formation of crystalline In(OH)3,
likely by preferentially consuming hydroxide ions.
Additionally it is evident that nitrate-based samples are
significantly more stable than hydroxide-based samples under
these conditions. A possible explanation for this enhanced
stability of nitrate-based samples under basic conditions could
be explained by the differences in the thermal decomposition
reactions of the indium nitrate and indium hydroxide
precursors. When indium hydroxide decomposes, it releases
water via eq 1
2In(OH)3 → In2O3 + 3H2O
(1)
However, when indium nitrate decomposes, it releases a
significantly more oxidizing gas mixture, specifically NO2 and
O2, as shown in eq 2
F
DOI: 10.1021/acs.chemmater.6b00301
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 5. (a) Schematic illustration of a nitrogen dioxide molecule, evolved from the decomposition of indium(III) nitrate, oxidizing a surface
oxygen vacancy of the resulting indium oxide material. Schematic illustration of a hydroxide attack on (b) a hydroxylated indium oxide surface with
no surface vacancies, representing the nitrate-based samples, and (c) a hydroxylated surface with oxygen vacancies, representing the hydroxide-based
samples.
3
XPS, that both the nitrate-based and hydroxide-based samples
2In(NO3)3 → In2O3 + 6NO2 + /2O2
(2)
have a significant quantity of hydroxide groups. However, as
our previous study suggested, surface hydroxide groups alone
are not enough to explain the photocatalytic activity, and
instead both surface oxygen vacancies and hydroxide groups are
required in an active photocatalyst.29,30 The HSA-nitrate-based
sample has a comparable surface area and an even higher
concentration of surface hydroxide groups than the hydroxide-
based sample; however, it shows no evidence of photocatalytic
activity for CO2 reduction. Its stability under basic etching
conditions indicates that the surface of this sample is very stable
and likely lacks the coordinatively unsaturated indium sites
(oxygen vacancies) needed to facilitate CO2 reduction. The
hydroxide-based sample, on the other hand, has both hydroxide
groups as well as an increased amount of these coordinatively
unsaturated indium surface sites, making it unstable under basic
etching conditions and active as a photocatalyst for CO2
reduction. This stark difference in photocatalytic activity
between indium oxide samples prepared from different
precursors provides further evidence to support the hypothesis
that a combination of surface defects − both oxygen vacancies
and hydroxides − are crucial to preparing an active surface
capable of driving photocatalytic CO2 reduction.
Designing an Active Photocatalyst for CO2 Reduction.
As with the synthesis of any nanomaterial, there are many
parameters that can be varied that have a substantial effect on
the surface, morphology, and optoelectronic properties of the
resulting photocatalytic material, ultimately determining its
activity. The results of this study add to our increasing
understanding of the necessary design parameters needed to
make an active, single-component photocatalyst. Not only must
the semiconductor nanomaterial absorb solar light efficiently,
but it must also have a surface which is optimized to interact
favorably with reactant molecules. In this study and in our
previous work,29−31 we have demonstrated that both surface
oxygen vacancies and hydroxide groups are needed to capture
and convert CO2 to CO. In order to produce an active material
which contains both of these defects, carefully controlling the
calcination temperature and selecting an appropriate precursor
are critical. In general, lower calcination temperatures allow for
the retention of more defects within the material. In the case of
indium oxide, 250 °C, which is just above the hydroxide-to-
As the indium nitrate decomposes and begins forming In2O3,
the NO2 that gets released can oxidize surface vacancies that are
formed in the sample, as illustrated in Figure 5a. NO can easily
be reoxidized to NO2 under our reaction conditions,42 via eq 3
NO + 1/2O → 2NO2
(3)
2
and continue to oxidize the surface of the nitrate-based In2O3
sample. Without this coordinatively unsaturated indium site, it
is now much more difficult for the surface to react with
hydroxide ions during the etching procedure, as illustrated in
Figure 5b. XPS data for both nitrate-based samples (Figure 3c
and d) indicate a significant contribution to the O 1s core level
peak from oxygen vacancies; however, if these vacancies are
buried within the bulk of the sample, they would not be
accessible to react with the evolving gases and therefore would
be retained in the sample, even under these oxidizing
conditions. It is unlikely that all vacancies would be consumed
via this oxidation mechanism, which could explain why the
nitrate-based sample (without SiO2) was still partially
converted to In(OH)3 under the etching conditions.
The hydroxide-based sample, on the other hand, is not
subjected to such oxidizing conditions during its formation and
retains a significantly higher proportion of its oxygen vacancies.
This makes these samples much more susceptible to attack by
hydroxide ions during the etching procedure, as illustrated in
Figure 5c. This could explain why neither one of the hydroxide-
based In2O3 samples could withstand the basic etching
conditions.
Correlation of Surface Defects to Photocatalytic
Activity. The oxidizing nature of the nitrate-based In2O3
formation environment may also help to explain the significant
difference in reactivity between the nitrate-based and
hydroxide-based samples. Much in the same way that a
coordinatively unsaturated indium site appears necessary to
facilitate the conversion of In2O3 into In(OH)3 under basic
conditions, this higher energy unsaturated site could also
function as the active site for CO2 reduction. As our previous
studies have demonstrated, higher defect concentrations in
hydroxide-based In2O3 corresponded to significantly higher
photocatalytic activity.29 In this study we have shown, using
G
DOI: 10.1021/acs.chemmater.6b00301
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
oxide transition temperature, appears to produce the most
active sample, likely because it maximizes defects. In terms of
precursor selection, as demonstrated herein, metal hydroxides
appear to be the best choice for making hydroxylated, oxygen
vacancy-rich nanomaterials. In general, any nonoxidizing
precursor anion could be a good candidate; however, one
should take care when using carbon-containing anions because,
as demonstrated with indium(III) acetate, substantial carbon
contamination can result, complicating the analysis of the
products of the photocatalytic reaction. Other parameters,
which we have not yet examined but certainly would play a role
in optimizing photocatalytic activity of a metal oxide nanoma-
terial, include reaction time, atmosphere, and pressure during
the synthesis of the material.
AUTHOR INFORMATION
Corresponding Author
*E-mail: gozin@chem.utoronto.ca.
Present Address
■
‡Institute of Functional Nano & Soft Materials (FUNSOM),
Jiangsu Key Laboratory for Carbon-Based Functional Materials
& Devices, Soochow University, 199 Ren’ai Road, Suzhou,
215123 Jiangsu, PR China.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
CONCLUSIONS
ACKNOWLEDGMENTS
■
■
In this study we have demonstrated that the choice of precursor
has a strong effect on the photocatalytic activity of In2O3
nanoparticle samples. The oxidizing nature of the precursor
anion used in the nanoparticle synthesis plays a significant role
in the defect formation within the sample. Precursors, such as
indium nitrate, that release oxidizing gases will produce In2O3
nanoparticles with a reduced number of surface oxygen
vacancies, rendering this material inactive as a photocatalyst.
On the other hand, precursors such as indium hydroxide, which
produce more mild gaseous byproducts, allow the retention of
more of these surface oxygen vacancies and correlate to higher
photocatalytic activities. We have further confirmed that while
hydroxide groups may be important for the overall CO2
reduction reaction, simply increasing the concentration of
hydroxide groups alone will not make a more active catalyst.
These results demonstrate that the choice of precursor has a
substantial impact on the population of surface defect sites
generated during In2O3 synthesis, which plays an important
role in determining the final photocatalytic activity. This study
provides valuable insight into key parameters for the precursor
selection, materials design, and performance optimization of
heterogeneous gas-phase CO2 reduction photocatalysts.
G.A.O. is a Government of Canada Research Chair in Materials
Chemistry and Nanochemistry. Financial support for this work
was provided by the Ontario Ministry of Research Innovation
(MRI); Ministry of Economic Development, Employment and
Infrastructure (MEDI); Ministry of the Environment and
Climate Change (MOECC); Connaught Innovation Fund;
Connaught Global Challenge Fund; and Natural Sciences and
Engineering Research Council of Canada (NSERC). Q.Q. was
supported by CCDM, an EFRC funded by the U.S. DOE-BES
(#DE-SC0012575) for her TEM work, and Y.Z. was supported
by DOE-BES, Materials Science and Engineering Division
(#DE-SC0012704).
REFERENCES
■
(1) Olah, G. A.; Prakash, G. K. S.; Goeppert, A. Anthropogenic
Chemical Carbon Cycle for a Sustainable Future. J. Am. Chem. Soc.
2011, 133, 12881−12898.
(2) Thomas, J. M. Heterogeneous Catalysis and the Challenges of
Powering the Planet, Securing Chemicals for Civilised Life, and Clean
Efficient Utilization of Renewable Feedstocks. ChemSusChem 2014, 7,
1801−1832.
(3) Kim, D.; Sakimoto, K. K.; Hong, D.; Yang, P. Artificial
Photosynthesis for Sustainable Fuel and Chemical Production.
Angew. Chem., Int. Ed. 2015, 54, 3259−3266.
(4) Centi, G.; Quadrelli, E. A.; Perathoner, S. Catalysis for CO2
Conversion: A Key Technology for Rapid Introduction of Renewable
Energy in the Value Chain of Chemical Industries. Energy Environ. Sci.
2013, 6, 1711.
(5) Liu, A.; Gao, J.; He, L. New and Future Developments in Catalysis:
Activation of Carbon Dioxide; Elsevier B.V.: 2013.
(6) Izumi, Y. Recent Advances in the Photocatalytic Conversion of
Carbon Dioxide to Fuels with Water and/or Hydrogen Using Solar
Energy and Beyond. Coord. Chem. Rev. 2013, 257, 171−186.
(7) Zhang, N.; Ciriminna, R.; Pagliaro, M.; Xu, Y.-J. Nanochemistry-
Derived Bi2WO6 Nanostructures: Towards Production of Sustainable
Chemicals and Fuels Induced by Visible Light. Chem. Soc. Rev. 2014,
43, 5276−5287.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.chemma-
ter.6b00301.
Description and characterization (powder X-ray diffrac-
tion, transmission electron microscopy, photocatalytic
testing, and UV−visible reflection spectroscopy) of other
precursors tested (indium(III) acetate and indium(III)
chloride). Schematic illustration of the SiO2 templated
synthesis of high surface area In2O3 from the indium(III)
nitrate precursor. X-ray photoelectron spectroscopy
(XPS) survey spectra of In2O3 made from hydroxide,
nitrate, and high surface area nitrate. Comparison of the
XPS O 1s peaks for the hydroxide-based sample before
being placed in the photocatalytic reactor and after
continuous reaction for 4 days. Comparison of powder
X-ray diffraction patterns for In2O3 + SiO2 made from
indium hydroxide after the basic etching procedure and
the amorphous indium hydroxide precursor used to
make the hydroxide-based In2O3 (PDF)
(8) Herron, J. A.; Kim, J.; Upadhye, A. A.; Huber, G. W.; Maravelias,
C. T. A General Framework for the Assessment of Solar Fuel
Technologies. Energy Environ. Sci. 2015, 8, 126−157.
(9) Sternberg, A.; Bardow, A. Power-to-What? − Environmental
Assessment of Energy Storage Systems. Energy Environ. Sci. 2015, 8,
389−400.
́
(10) Navalon, S.; Dhakshinamoorthy, A.; Alvaro, M.; Garcia, H.
Photocatalytic CO2 Reduction Using Non-Titanium Metal Oxides and
Sulfides. ChemSusChem 2013, 6, 562−577.
(11) White, J. L.; Baruch, M. F.; Pander, J. E., III; Hu, Y.; Fortmeyer,
I. C.; Park, J. E.; Zhang, T.; Liao, K.; Gu, J.; Yan, Y.; Shaw, T. W.;
Abelev, E.; Bocarsly, A. B. Light-Driven Heterogeneous Reduction of
H
DOI: 10.1021/acs.chemmater.6b00301
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Carbon Dioxide: Photocatalysts and Photoelectrodes. Chem. Rev.
2015, 115, 12888−12935.
(32) NIST-XPS database, version 3.5. http://srdata.nist.gov/xps/
(accessed June 1, 2016).
(33) Sheets, W. C.; Mugnier, E.; Barnabe, A.; Marks, T. J.;
́
Poeppelmeier, K. R. Hydrothermal Synthesis of Delafossite-Type
Oxides. Chem. Mater. 2006, 18, 7−20.
(34) Teramura, K.; Iguchi, S.; Mizuno, Y.; Shishido, T.; Tanaka, T.
Photocatalytic Conversion of CO2 in Water over Layered Double
Hydroxides. Angew. Chem., Int. Ed. 2012, 51, 8008−8011.
(35) Yoshida, S.; Kohno, Y. A New Type of Photocatalysis Initiated
by Photoexcitation of Adsorbed Carbon Dioxide on ZrO2. Catal. Surv.
Jpn. 2001, 4, 107−114.
(36) Chorkendorff, I.; Niemantsverdriet, J. W. Concepts of Modern
Catalysis and Kinetics; Wiley-VCH Verlag GmbH & Co KGaA:
Weinheim, Germany, 2003.
(37) Chang, W.-C.; Kuo, C.-H.; Juan, C.-C.; Lee, P.-J.; Chueh, Y.-L.;
Lin, S.-J. Sn-Doped In2O3 Nanowires: Enhancement of Electrical Field
Emission by a Selective Area Growth. Nanoscale Res. Lett. 2012, 7, 684.
(38) Janowitz, C.; Scherer, V.; Mohamed, M.; Krapf, A.; Dwelk, H.;
Manzke, R.; Galazka, Z.; Uecker, R.; Irmscher, K.; Fornari, R.;
Michling, M.; Schmeißer, D.; Weber, J. R.; Varley, J. B.; Van de Walle,
C. G. Experimental Electronic Structure of In2O3 and Ga2O3. New J.
Phys. 2011, 13, 085014.
(39) Habisreutinger, S. N.; Schmidt-Mende, L.; Stolarczyk, J. K.
Photocatalytic Reduction of CO2 on TiO2 and Other Semiconductors.
Angew. Chem., Int. Ed. 2013, 52, 7372−7408.
(40) Poznyak, S. K.; Golubev, A. N.; Kulak, A. I. Correlation between
Surface Properties and Photocatalytic and Photoelectrochemical
Activity of In2O3 Nanocrystalline Films and Powders. Surf. Sci. 2000,
454−456, 396−401.
(12) Sastre, F.; Puga, A. V.; Liu, L.; Corma, A.; Garcia, H. Solar Light
Complete Photocatalytic CO2 Reduction to Methane by H2 at near
Ambient Temperature. J. Am. Chem. Soc. 2014, 136, 6798−6801.
(13) Pan, X.; Yang, M.-Q.; Fu, X.; Zhang, N.; Xu, Y.-J. Defective
TiO2 with Oxygen Vacancies: Synthesis, Properties and Photocatalytic
Applications. Nanoscale 2013, 5, 3601−3614.
(14) Polarz, S.; Strunk, J.; Ischenko, V.; Van Den Berg, M. W. E.;
Hinrichsen, O.; Muhler, M.; Driess, M. On the Role of Oxygen
Defects in the Catalytic Performance of Zinc Oxide. Angew. Chem., Int.
Ed. 2006, 45, 2965−2969.
(15) Nakamura, I.; Negishi, N.; Kutsuna, S.; Ihara, T.; Sugihara, S.;
Takeuchi, K. Role of Oxygen Vacancy in the Plasma-Treated TiO2
Photocatalyst with Visible Light Activity for NO Removal. J. Mol.
Catal. A: Chem. 2000, 161, 205−212.
(16) Nowotny, J. Titanium Dioxide-Based Semiconductors for Solar-
Driven Environmentally Friendly Applications: Impact of Point
Defects on Performance. Energy Environ. Sci. 2008, 1, 565−572.
(17) Liu, L.; Gao, F.; Zhao, H.; Li, Y. Tailoring Cu Valence and
Oxygen Vacancy in Cu/TiO2 Catalysts for Enhanced CO2 Photo-
reduction Efficiency. Appl. Catal., B 2013, 134−135, 349−358.
(18) Walsh, A. Surface Oxygen Vacancy Origin of Electron
Accumulation in Indium Oxide. Appl. Phys. Lett. 2011, 98, 261910.
(19) Ye, J.; Liu, C.; Mei, D.; Ge, Q. Active Oxygen Vacancy Site for
Methanol Synthesis from CO2 Hydrogenation on In2O3 (110): A DFT
Study. ACS Catal. 2013, 3, 1296−1306.
(20) Liu, G.; Hoivik, N.; Wang, K.; Jakobsen, H. Engineering TiO2
Nanomaterials for CO2 Conversion/Solar Fuels. Sol. Energy Mater. Sol.
Cells 2012, 105, 53−68.
(21) Lany, S.; Zakutayev, A.; Mason, T. O.; Wager, J. F.;
Poeppelmeier, K. R.; Perkins, J. D.; Berry, J. J.; Ginley, D. S.;
Zunger, A. Surface Origin of High Conductivities in Undoped In2O3
Thin Films. Phys. Rev. Lett. 2012, 108, 016802.
(22) Zhang, K. H. L.; Egdell, R. G.; Offi, F.; Iacobucci, S.; Petaccia,
L.; Gorovikov, S.; King, P. D. C. Microscopic Origin of Electron
Accumulation in In2O3. Phys. Rev. Lett. 2013, 110, 056803.
(23) King, P. D. C.; Veal, T. D. Conductivity in Transparent Oxide
Semiconductors. J. Phys.: Condens. Matter 2011, 23, 334214.
(24) Habisreutinger, S. N.; Schmidt-Mende, L.; Stolarczyk, J. K.
Photocatalytic Reduction of CO2 on TiO2 and Other Semiconductors.
Angew. Chem., Int. Ed. 2013, 52, 7372−7408.
(41) Gan, J.; Lu, X.; Wu, J.; Xie, S.; Zhai, T.; Yu, M.; Zhang, Z.; Mao,
Y.; Wang, S. C. I.; Shen, Y.; Tong, Y. Oxygen Vacancies Promoting
Photoelectrochemical Performance of In2O3 Nanocubes. Sci. Rep.
2013, 3, 1021.
(42) Holleman, A. F.; Wiberg, E. Inorganic Chemistry; Academic
Press: San Diego, CA, USA, 2001.
(43) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. The
Handbook of X-ray Photoelectron Spectroscopy; Physical Electronics Inc.:
Eden Prairie, 1995.
(25) Wang, G.; Wang, H.; Ling, Y.; Tang, Y.; Yang, X.; Fitzmorris, R.
C.; Wang, C.; Zhang, J. Z.; Li, Y. Hydrogen-Treated TiO2 Nanowire
Arrays for Photoelectrochemical Water Splitting. Nano Lett. 2011, 11,
3026−3033.
(26) Ahmed, N.; Shibata, Y.; Taniguchi, T.; Izumi, Y. Photocatalytic
Conversion of Carbon Dioxide into Methanol Using Zinc−Copper−
M(III) (M = Aluminum, Gallium) Layered Double Hydroxides. J.
Catal. 2011, 279, 123−135.
(27) Choi, S.; Drese, J. H.; Jones, C. W. Adsorbent Materials for
Carbon Dioxide Capture from Large Anthropogenic Point Sources.
ChemSusChem 2009, 2, 796−854.
(28) Auroux, A.; Gervasini, A. Microcalorimetric Study of the Acidity
and Basicity of Metal Oxide Surfaces. J. Phys. Chem. 1990, 94, 6371−
6379.
(29) Hoch, L. B.; Wood, T. E.; O’Brien, P. G.; Liao, K.; Reyes, L. M.;
Mims, C. A.; Ozin, G. A. The Rational Design of a Single-Component
Photocatalyst for Gas-Phase CO2 Reduction Using Both UV and
Visible Light. Adv. Sci. 2014, 1, 1400013.
(30) Ghuman, K. K.; Wood, T. E.; Hoch, L. B.; Mims, C. A.; Ozin, G.
A.; Singh, C. V. Illuminating CO2 Reduction on Frustrated Lewis Pair
Surfaces: Investigating the Role of Surface Hydroxides and Oxygen
Vacancies on Nanocrystalline In2O3−x(OH)y. Phys. Chem. Chem. Phys.
2015, 17, 14623−14635.
(31) Ghuman, K. K.; Hoch, L. B.; Szymanski, P.; Loh, J. Y. Y.;
Kherani, N. P.; El-Sayed, M. A.; Ozin, G. A.; Singh, C. V. Photoexcited
Surface Frustrated Lewis Pairs for Heterogeneous Photocatalytic CO2
Reduction. J. Am. Chem. Soc. 2016, 138, 1206.
I
DOI: 10.1021/acs.chemmater.6b00301
Chem. Mater. XXXX, XXX, XXX−XXX