Efficient Photocatalytic Hydrogen Generation from Ni Nanoparticle Decorated CdS Nanosheets

Article abstract of DOI:10.1021/acscatal.5b01812

High-quality, thickness-controlled CdS nanosheets (NSs) have been obtained through the thermal decomposition of cadmium diethyldithiocarbamate in octadecene. Ensembles with discrete thicknesses of 1.50, 1.80, and 2.16 nm have been made with corresponding

Full text of DOI:10.1021/acscatal.5b01812

Research Article
pubs.acs.org/acscatalysis
Efficient Photocatalytic Hydrogen Generation from Ni Nanoparticle
Decorated CdS Nanosheets
Maksym Zhukovskyi,* Pornthip Tongying, Halyna Yashan, Yuanxing Wang, and Masaru Kuno*
Department of Chemistry and Biochemistry, University of Notre Dame, 251 Nieuwland Science Hall, Notre Dame, Indiana 46556,
United States
*
S Supporting Information
ABSTRACT: High-quality, thickness-controlled CdS nanosheets (NSs) have been
obtained through the thermal decomposition of cadmium diethyldithiocarbamate in
octadecene. Ensembles with discrete thicknesses of 1.50, 1.80, and 2.16 nm have been
made with corresponding lateral dimensions on the order of 90 nm × 20 nm. These latter
values make the 1−3 nm NSs the largest 2D CdS specimens made to date using colloidal
chemistry. Associated Ni nanoparticle decorated counterparts have been made through the
photodeposition of Ni onto NSs with an average nanoparticle diameter of 6 nm.
Subsequent photocatalytic hydrogen generation measurements have compared the
performance of CdS NSs with that of their Ni NP decorated counterparts in water/
ethanol mixtures. Apparent quantum yields as large as 25% have been seen for Ni NP
decorated NSs with transient yields as large as 64% within the first 2 h of irradiation.
Results from ensemble femtosecond transient differential absorption spectroscopy reveal
that the origin of this high efficiency stems from efficient electron transfer from CdS to Ni. In this regard, the CdS/Ni
semiconductor/metal heterojunction acts to dissociate strongly bound excitons in CdS NSs, creating free carriers needed to carry
out relevant reduction chemistries.
KEYWORDS: CdS, nanosheet, nickel, nanoparticle, charge separation, hydrogen generation
1
. INTRODUCTION
Until now, though, the photocatalytic properties of 2D
2
1−24
materials have not been investigated in any great detail,
Low-dimensional cadmium chalcogenides such as CdS
quantum dots (QDs), nanorods (NRs), and nanowires
despite the aforementioned unique and intriguing photo-
physical properties. In what follows, we therefore investigate
the use of atomically thin CdS nanosheets for photocatalytic
hydrogen generation. We simultaneously investigate corre-
sponding enhancements that arise due to the use of Ni
nanoparticle cocatalysts. Of interest is how the large absorption
efficiencies, room-temperature excitonic resonances, long
(NWs) have been used extensively for photocatalytic hydrogen
1
−11
generation.
This is due to their tunable band gaps, efficient
absorption of visible light, large surface to volume ratios, and
their favorable and tunable conduction/valence band ener-
gies.
press charge recombination, have simultaneously been
investigated. They include two- or three-component semi-
conductor/semiconductor (e.g., CdSe/CdS core/shell
dot-in-rod systems ), semiconductor/metal (e.g., CdS/Au,
CdS/Pt, CdSe/CdS dot-in-rod/Pt and CdSe/CdS/Pt core/
shell NWs ), and semiconductor/molecular heterostructures
2
,12,13
Corresponding heterostructures, designed to sup-
25
lifetimes, and corresponding suppressed Auger responses of
CdS NSs affect their photocatalytic response.
5
,6
or
1
0
3
4
9,10
2. EXPERIMENTAL SECTION
6
2
.1. Materials. Cadmium diethyldithiocarbamate
7
(
CdDDTC, 99.5%) was purchased from BOC Science. 1-
Octadecene (ODE, 90%) was purchased from Aldrich. Stearic
acid (98%) was purchased from Alfa Aesar. Cadmium acetate
(e.g., CdSe/Ni-dihydrolipoic acid and CdS/[FeFe]-hydro-
8
genase complexes ).
While 0D and 1D cadmium chalcogenide nanostructures
1
−11
dihydrate (CdAc , 98%) was purchased from Acros Organics.
have been widely investigated for use as photocatalysts,
their 2D counterparts have only recently attracted attention in
this regard due to new synthetic advances, enabling the
production of high-quality, thickness-controlled NSs.
Subsequent investigations have revealed the unique optical
and electrical properties of 2D materials, which include their
strong room-temperature excitonic resonances, their large
absorption cross sections, high emission quantum yields, and
large charge carrier mobilities.
have quickly found use in light-emitting diodes, gas sensors,
2
Methanol, ethanol, and toluene were purchased from Fisher
Scientific and VWR. Unless otherwise noted, all chemicals were
used as received.
1
4−20
2
.2. Synthesis of 2.16 nm Thick CdS NSs. CdS NSs with
2
.16 nm thickness were synthesized by combining the following
in a three-neck flask: ODE (20 mL, 56.3 mmol), CdDDTC
(41.1 mg, 0.1 mmol), stearic acid (29.0 mg, 0.1 mmol), and
14
15
1
4,16
Consequently, 2D systems
17
18
Received: June 29, 2015
1
9
20
photodetectors, and field effect transistors.
©
XXXX American Chemical Society
6615
DOI: 10.1021/acscatal.5b01812
ACS Catal. 2015, 5, 6615−6623

ACS Catalysis
Research Article
CdAc (81.6 mg, 0.3 mmol). The three-neck flask was then
(200 ps fwhm, 1 MHz repetition rate) with the NS emission at
442 nm detected with a fast photomultiplier tube (overall
instrument time resolution of ∼1 ns). For TDA measurements,
NS samples were suspended in water/ethanol mixtures and
were placed in a 2 mm path length cuvette. Optical densities
2
connected to a Schlenk line, whereupon its contents were dried
and degassed at 65 °C for 30 min to remove any oxygen and
water. When complete, the reaction mixture was back-filled
with N and was heated to 220 °C for 3 h with vigorous stirring.
2
During this time, the initial translucent white slurry became
clear and eventually turned yellow due to the formation of CdS
NSs. Upon completion, the mixture was cooled to room
temperature, whereupon 10 mL of toluene was added to dilute
the ODE. CdS NSs were precipitated by adding 5 mL of
methanol to this mixture and centrifuging the resulting
suspension. Recovered sheets were washed with toluene and
methanol three additional times before being stored in toluene.
Note that using CdDDTC alone in the reaction does not
yield CdS NSs, given their 1:4 Cd:S stoichiometry. Additional
(OD) were ∼0.3 at the excitation wavelength (λ = 387 nm).
exc
Ultrafast TDA experiments were performed with a Clark
MXR CPA 2010 laser system employing a fiber-based
spectrometer. Experiments were conducted by exciting CdS
and CdS/Ni NSs at 387 nm (3.20 eV) with ∼150 fs pulses
2
(pump fluence 10 μJ/cm ). Transient absorption spectra were
subsequently acquired with a delayed, low-intensity white light
continuum with wavelengths between 400 and 750 nm (3.10−
1.65 eV). White light was generated by passing a fraction of the
fundamental through a CaF crystal.
2
CdAc was therefore needed to supplement Cd in CdDDTC as
2.5. Concerted Ni NP Photodeposition and Photo-
catalytic Hydrogen Generation Experiments. For photo-
catalytic H2 generation measurements, CdS NSs were
precipitated from 2 mL of a toluene stock suspension
(approximate NS concentration 1.3 × 10− M) via
centrifugation and were redispersed in 5 mL of ethanol. NSs
were then precipitated a second time and again redispersed in 5
mL of ethanol. This precipitation/ethanol resuspension step
was repeated three additional times to remove any residual
toluene. Finally, NSs were resuspended in 5 mL of a 10%/90%
water/ethanol mixture and purged with Ar to remove any
oxygen.
2
well as to make the reaction Cd rich. CdDDTC does, however,
play an important role in the NS synthesis in that on
replacement by an equivalent amount of sodium diethyldithio-
carbamate only spherical CdS nanoparticles were produced.
7
2
.3. Synthesis of 1.50 and 1.80 nm Thick CdS NSs. To
synthesize 1.50 nm (1.80 nm) NSs, 41.1 mg of CdDDTC (0.1
mmol), 72.5 mg of stearic acid (0.25 mmol), and 40.8 mg of
CdAc (0.15 mmol) (29.0 mg of stearic acid (0.1 mmol) and
2
5
4.4 mg of CdAc (0.2 mmol)) were used. All other synthetic
2
parameters/conditions were kept constant.
.4. Sample Characterization. TEM samples were
2
prepared by drop-casting dilute CdS NS toluene suspensions
onto carbon-coated copper TEM grids (Ted Pella). NSs were
analyzed with a JEOL JEM2011 TEM operating at 200 keV.
XRD samples were prepared by depositing concentrated NS
suspensions onto (0001) cut quartz substrates and allowing the
suspension to dry under ambient conditions. Powder X-ray
diffraction data (Cu Kα radiation, λ = 1.5418 Å, D8 Advance
Davinci, Bruker) were collected in 0.02° increments with a 10
s/step integration time.
In the case where Ni NPs were photodeposited onto NSs
prior to photocatalytic H generation measurements (discussed
2
more fully in the Results and Discussion), CdS sheets were
resuspended in 5 mL of a 10%/90% water/ethanol mixture
−
5
containing NiSO with concentrations ranging from 1.3 × 10
4
to 1.0 × 10− M. Solutions were then placed in quartz cells,
3
purged with Ar to remove any oxygen, and irradiated with a 405
2
nm laser diode (Coherent Obis) for 30 min (Iexc = 3.5 W/cm )
to induce Ni NP photodeposition. Briefly, under laser
excitation of CdS NSs, photogenerated excitons are created.
The existence of defects means that some fraction of excitons
will dissociate into component electrons and holes, whereupon
subsequent surface trapping of electrons becomes possible.
To estimate NS thickness-dependent conduction band
positions, voltammetry measurements were conducted. Electro-
chemical measurements were performed under dark conditions
using a three-electrode cell with AgCl as a reference electrode,
Pt as the counter electrode, and CdS NSs on FTO as the
working electrode. Measurements were conducted in 0.1 M
solutions of Na SO under nitrogen. Potentials ranged from 0.2
2
+
Such localized electrons can then reduce surface-adsorbed Ni
ions, leading to Ni nanoparticle formation.
26,27
A correspond-
ing yellow to greenish color change is observed, indicating Ni
NP deposition.
For hydrogen generation measurements, all samples were
excited with 405 nm light from the same laser used to
photodeposit Ni. At given intervals, evolved gases were
analyzed with a gas chromatograph (Thermo, Trace GC
Ultra). For excitation intensity dependent measurements, Ni
2
4
to −2.0 V with a scan rate of 20 mV/s (Princeton Applied
Research 2273). The NS conduction band potential was taken
2
+
to be the peak potential for Cd reduction in electrochemical
2
scans.
CdS NS concentrations were estimated using ICP-AES
(
PerkinElmer Optima 8000). Samples were prepared by
−7
precipitating 1 mL aliquots from a ∼1.3 × 10 M NS stock
suspension via centrifugation at 5000 rpm for 5 min.
Supernatants were discarded and recovered NSs were dried
under ambient conditions. They were then digested using 2 mL
NPs were first deposited onto CdS NSs with a NiSO
4
concentration of 2.0 × 10− M (average resulting NP diameter
4
6 nm, coverage of one NP per NS). Following this, the 405 nm
laser intensity was set to an I value between 5 and 3.5 W/
exc
2
of aqua regia, whereupon 48 mL of 5% HNO was added to
cm . Hydrogen was then detected using gas chromatography at
3
produce samples suitable for ICP analysis. ICP elemental
calibration curves were created using commercial standards
with known concentrations.
UV−visible absorption spectra of CdS NSs suspended in
toluene were recorded with a Jasco V-670 UV/visible/near-
infrared spectrophotometer coupled to a 60 mm integrating
sphere. A 1 cm cuvette was used for all studies. Time correlated
single photon counting (TCSPC) kinetics were acquired with a
Jobin Yvon Fluorocube using 371 nm pulsed LED illumination
given intervals.
2.6. Polyethylenimine (PEI) Ligand Exchange. To
conduct ligand exchange, CdS/Ni NSs were precipitated by
centrifuging their toluene suspensions. A 4 mL ethanol solution
of PEI (1.5 mM, MW 25000) was then added to the recovered
sheets. The resulting suspension was stirred for 1 h. At this
point, NSs were precipitated again via centrifugation, where-
upon they were redispersed in water/ethanol mixtures with
varying ratios of 10%/90%, 50%/50%, and 90%/10% (by
6
616
DOI: 10.1021/acscatal.5b01812
ACS Catal. 2015, 5, 6615−6623

ACS Catalysis
Research Article
Figure 1. Low- and high-magnification TEM images of (a, c) CdS and (b, d) CdS/Ni NSs. Insets give the corresponding ensemble SAED images (a,
b) and a high-resolution TEM image showing basal plane lattice fringes (top right in (c)) as well as a NS side view (bottom left in (c)). Ensemble
EDXS spectra of 2.16 nm (e) CdS and (f) CdS/Ni NSs.
volume). All ligand exchange reactions were done under
nitrogen in a glovebox to prevent Ni oxidation.
regarding the amounts of each chemical used can be found in
the Experimental Section. Thickness control was achieved by
varying the cadmium acetate to stearic acid concentration ratio
in preparations. Following degassing, the reaction vessel was
3
. RESULTS AND DISCUSSION
.1. Synthesis of CdS NSs. CdS NSs were synthesized
back-filled with N and was heated to 220 °C. The mixture was
2
3
left at 220 °C for an additional 3 h. During this time, the
mixture turned from clear to light yellow due to CdS NS
formation. The resulting product was then isolated by
centrifuging the suspension, followed by several toluene/
methanol washing steps to remove any excess surfactant.
Transmission electron microscopy (TEM) measurements of
resulting 2.16 nm CdS NSs show that they possess lateral
dimensions of 93(±12) × 15(±2) nm (Figure 1a,c). Lateral
sizing histograms as well as additional TEM micrographs can be
found in Figures S1 and S2 in the Supporting Information.
Their flexible, ultrathin nature is also evident in obtained TEM
images through apparent rolling and twisting. Corresponding,
through the thermal decomposition of cadmium diethyldithio-
carbamate (CdDDTC) in octadecene. The approach is the first
to use a single-source precursor in synthesizing cadmium
chalcogenide NSs and complements existing approaches that
involve the thermolysis of cadmium acetate (CdAc ) in the
2
28−30
presence of fatty acids and elemental sulfur.
What results
are CdS NSs with discrete thicknesses of 1.50, 1.80, and 2.16
nm and whose lateral dimensions are 90 nm × 20 nm. These
latter values make the produced CdS NSs the largest made to
date using colloidal chemistry.
Briefly, a mixture of CdDDTC, cadmium acetate, and stearic
acid in octadecene was degassed at 65 °C. Specific details
6
617
DOI: 10.1021/acscatal.5b01812
ACS Catal. 2015, 5, 6615−6623

ACS Catalysis
Research Article
Figure 2. (a) Representative XRD powder patterns of 1.50, 1.80, and 2.16 nm thick CdS NSs. The JCPDS stick pattern for bulk zincblende CdS
JCPDS# 65-2887) is included for reference purposes. (b) UV/visible absorption and emission spectra of 1.50, 1.80, and 2.16 nm NSs. Asterisks
(
denote absorption and emission contributions from thinner (minority) NS subpopulations within a given ensemble. (c) UV/visible absorption
spectra of 2.16 nm CdS and CdS/Ni NSs. The inset gives CdS NS tail states highlighted on a semilogarithmic graph.
albeit rare, edge-on TEM images confirm the 2.16 nm thickness
of the sheets. An example is shown in the inset of Figure 1c.
Tandem TEM-based ensemble SAED measurements reveal
apparent (200), (222), and (400) diffraction rings (inset,
Figure 1a) with associated d spacings of 0.294 nm (200), 0.170
nm (222), and 0.148 nm (400). High-resolution, basal plane
images (inset, Figure 1c) further show that the sheets exhibit
lattice spacings of ∼0.34 nm, consistent with those from the
Information summarizes all XRD-based CdS NS d spacings
along with known bulk values.
To determine the conduction band positions of resulting
NSs, electrochemical measurements were conducted as
described in the Experimental Section. CdS NS electrodes
were prepared by drop-casting samples onto fluorine doped tin
oxide (FTO) substrates. After drying, electrodes were tested
under dark conditions in 0.1 M Na SO solutions (correspond-
2
4
2
+
(111) planes of zincblende (ZB) CdS. All experimental lattice
ing pH of 6.4). Resulting voltammetry scans show Cd
parameters are larger than those found in bulk CdS. This
discrepancy likely originates from lattice strain due to excess
reduction peaks at −1.19 and −1.34 V vs the normal hydrogen
electrode (NHE) for the 2.16 and 1.80 nm NSs (Figure S6 in
the Supporting Information). These features are attributed to
each ensemble’s conduction band position given the lack of
other reduction features in the scan and the similarity of these
potentials to known conduction band positions for other
28,30
cadmium, terminating NS {100} basal planes.
dispersive X-ray spectroscopy (EDXS) analyses of the sheets
Figure 1e and Figure S3 in the Supporting Information)
Energy
(
corroborate this conclusion by showing excess cadmium to
sulfur ratios of 1.18(±0.04), 1.18(±0.06), and 1.16(±0.07) for
2
,31,32
cadmium chalcogenide nanostructures.
No reduction
1
.50, 1.80, and 2.16 nm CdS NSs.
Corresponding 1.80 nm NSs possess similar rectangular
features are seen for 1.50 nm NSs. This is consistent with
previous observations in the case of small <2.6 nm diameter
CdSe QDs, where the lack of a reduction feature can be
attributed to obscuration by electrode-related hydrogen
evolution. Notably, the proton reduction Nernst potential of
−0.38 eV versus NHE (dotted line, Figure S6) indicates that
the 2.16 and 1.80 nm CdS NSs are suitable systems for
photocatalytic hydrogen generation. Additional details about
these electrochemical measurements/conduction band position
estimates can be found in the Experimental Section.
2
shapes with associated lateral dimensions of 96(±9) × 21(±3)
nm. In contrast, 1.50 nm NSs exhibit irregular shapes,
complicating an analysis of their exact lateral dimensions.
TEM images of the 1.80 and 1.50 nm sheets and corresponding
edge-on TEM images as well as associated lateral sizing
histograms (where applicable) can be found in Figures S2, S4,
and S5 of the Supporting Information.
Independent powder X-ray diffraction (XRD) measurements
confirm that all CdS NSs adopt their cubic ZB lattice. Figure 2a
illustrates XRD powder patterns acquired from 2.16, 1.80, and
3.2. Optical Characterization of CdS NSs. Ensemble
absorption spectra of 2.16 nm thick CdS NSs show a
characteristic band edge excitonic resonance at 433 nm (2.86
eV) (Figure 2b). Analogous absorption spectra are seen for
1.80 and 1.50 nm NSs, where the first excitonic transition shifts
to the blue with decreasing NS thickness (Figure 2b). An
1.50 nm NS ensembles. For 2.16 nm NSs, characteristic
zincblende reflections are seen at 26.04° (111), 30.07° (200),
43.51° (220), 51.35° (311), and 63.32° (400) (JCPDS# 65-
2887). Associated d spacings are 0.342, 0.297, 0.208, 0.178, and
0.147 nm. These values agree with aforementioned estimates
33
apparent defect tail also exists, decaying exponentially into the
gap (inset, Figure 2c). The tail does not appear to be scattering
related, since an integrating sphere, which accounts for
from TEM-based SAED measurements. Similar reflections are
seen for 1.80 and 1.50 nm NSs with corresponding d spacings
that increase with decreasing NS thickness. This can again be
attributed to lattice expansion, stemming from the presence of
3
4
scattering losses to the absorption, has been used.
Consequently, this represents the first observation of tail states
in ultrathin CdS NSs. Plots illustrating analogous tails for 1.80
28,30
terminal basal plane Cd.
Table S1 of the Supporting
6
618
DOI: 10.1021/acscatal.5b01812
ACS Catal. 2015, 5, 6615−6623

ACS Catalysis
Research Article
and 1.50 nm NSs can be found in the Figure S7 in the
Supporting Information.
and 1.50 nm NSs. Hence, they represent the most suitable
system for creating a photocatalyst. Photodeposition was also
chosen over a complementary thermal deposition approach,
given the better adhesion known to result between semi-
conductors and surface-deposited metals.
In brief, the photodeposition procedure works by photo-
The peak absorption cross section of 2.16 nm NSs has been
estimated using correlated TEM, UV/visible absorption
spectroscopy, and inductively coupled plasma atomic emission
spectroscopy (ICP-AES). The resulting cross section is σ
26,38
4
33 nm
=
2.9 × 10− cm with an associated molar extinction
13
2
exciting CdS NS with visible light (λ = 405 nm, Iexc = 3.5 W/
exc
7
−1
−1
2
coefficient of 7.5 × 10 M cm . Similar cross sections of
cm , 30 min exposure). This leads to exciton formation,
σ410 nm = 3.0 × 10− cm and σ
13
2
= 3.2 × 10 cm have
−13
2
whereupon the presence of surface traps results in exciton
dissociation and subsequent electron trapping. These trapped
electrons then reduce surface-adsorbed metal ions and lead to
378 nm
been found for the 1.80 and 1.50 nm NSs. A summary of these
σ values as well as details of their estimates can be found in the
Supporting Information. Additional information about these
ICP-AES measurements can also be found in the Experimental
Section.
26,27
metal NP growth on NSs.
Ni NP deposition onto CdS is
favored by the NS conduction band position (determined by
2
+
our prior electrochemical measurements) relative to the Ni /
0
39
Band edge emission is seen from 2.16 nm NSs centered at
Ni reduction potential of −0.23 V vs NHE. A yellow to
greenish color change is observed during photodeposition,
indicating Ni NP nucleation and growth. Specific details of the
photodeposition procedure can be found in the Experimental
Section.
442 nm (2.81 eV) (Figure 2b). With decreasing NS thickness,
this peak shifts to the blue and tracks the absorption. Emission
maxima are therefore seen at 419 nm (2.96 eV) and 393 nm
(3.16 eV) for 1.80 and 1.50 nm NSs, respectively. In all cases,
dominant defect-related emission, centered between 530 and
Figure 1b,d shows TEM images of resulting Ni NP decorated
CdS NSs. An analysis of the images shows that 6(±1) nm
diameter Ni NPs have been deposited onto the NSs (specific
5
70 nm (2.34−2.18 eV), corroborates the existence of tail
states first seen in the NS linear absorption.
−4
Complementary TCSPC measurements (Figure S8 in the
Supporting Information) show that the 2.16 nm CdS NSs
exhibit band edge emission decays that can be fit to
conditions: [NiSO ] = 2.0 × 10 M, λexc = 405 nm, Iexc = 3.5
4
2
W/cm ). Ensemble EDXS measurements corroborate the
deposition of Ni NPs, indicating the presence of Ni in resulting
CdS/Ni NS ensembles (Figure 1f).
triexponential decays. Corresponding time constants are τ ≈
1
0
.7 ns (53.69% area based weight), τ = 3.7 ns (32.54% area
In all cases, ensemble SAED measurements reveal that the
deposited Ni NPs adopt their cubic form. This stems from the
appearance of a diffraction ring that can be indexed to Ni’s
(200) reflection (inset, Figure 1b). Additional TEM images of
2.16 nm CdS NSs decorated with Ni NPs as well as associated
Ni NP sizing histograms can be found in the Figures S9 and
S10 in the Supporting Information.
2
based weight), and τ = 26 ns (13.77% area based weight). The
3
fastest (τ ) component is subsequently attributed to hole
1
trapping, given similar time scales reported for exciton
35
quenching due to hole trapping in CdSe nanoplatelets.
Next, since the TCSPC signal is sensitive to both electron and
hole concentrations, the latter two ns decays suggest the
existence of long-lived excitons within CdS NSs.
To alter both the deposited Ni NP size as well as surface
3
.3. CdS/Ni Photocatalyst Preparation and Character-
coverage, NiSO concentrations were varied in succession from
4
−5
−3
ization. Many low-dimensional photocatalysts employ noble
metals such as Pt to enhance their catalytic activity. This is
because noble metals possess favorable Fermi levels, which
enable them to act as effective electron acceptors while
simultaneously enhancing charge separation efficiencies. Within
the context of hydrogen generation, apparent quantum yields as
large as ∼10% have been seen for Pt NP decorated CdS NRs.
Similar values on the order of 20% have been seen for Pt NP
decorated CdSe/CdS dot-in-rod systems.
Despite these large yields, the use of noble metals makes the
large-scale use of these low-dimensional photocatalysts cost
prohibitive. Consequently, Ni has been investigated as an
alternative due to its low cost, earth abundance, and high
catalytic activity. What results are high-efficiency photocatalysts,
which include Ni nanoparticle decorated CdS NRs (53%
apparent quantum yield for hydrogen generation in water/
ethanol mixtures ), Ni NP decorated Cd Zn1−xS nanostruc-
tures (25% apparent quantum yield for hydrogen generation in
sodium sulfite ), Ni decorated CdS QDs (12.2% internal
1.3 × 10 to 1.0 × 10 M. Decreasing [NiSO ] from 2.0 ×
4
4
−
−5
10 M to 1.3 × 10 M results in experimentally identical Ni
NPs (average diameter 6(±1) nm) being deposited onto the
NSs. At low NiSO concentrations, though, not all CdS NSs
4
have Ni NPs deposited onto them (Figure S11 in the
Supporting Information). Increasing the Ni salt concentration
10
−3
to 1.0 × 10 M, in contrast, results in larger Ni NPs (average
diameter 10 (±3) nm) being deposited onto the sheets. Figure
S12 in the Supporting Information shows representative TEM
images of these particles.
9
Following Ni NP deposition, slight changes to the absorption
spectra of 2.16 nm CdS NSs are seen. Specifically, an apparent
suppression of the excitonic resonance is seen as illustrated in
Figure 2c. This observation is similar to that previously
reported for CdSe NRs and tetrapods decorated with Au
27
nanoparticles. In those cases, suppression was attributed to
the electronic mixing of states, affecting the strength of
excitonic resonances.
1
1
x
3
6
NS emission intensities are likewise affected. Specifically,
adding Ni NPs to the sheets leads to complete quenching of the
emission, whether band edge or trap related (Figure S13 in the
Supporting Information). This suggests good electronic contact
between CdS and Ni due to either efficient charge transfer or
energy transfer following photoexcitation. Subsequent hydro-
gen generation and corresponding transient differential
absorption measurements, discussed below, suggest the former.
3.4. Photocatalytic Hydrogen Generation. At this point,
hydrogen generation measurements were conducted on 2.16
37
quantum yield for hydrogen generation in glycerol ), and
dihydrolipoic acid capped CdSe QDs coupled to Ni2+ (35%
internal quantum yield for hydrogen generation in aqueous
7
solutions containing ascorbic acid ).
Consequently, to produce Ni NP decorated CdS NSs in the
current study, Ni was photodeposited onto NSs using NiSO4.
In these reactions and in all subsequent photocatalytic
measurements, the thickest 2.16 nm NSs have been used,
since they absorb further to the red than corresponding 1.80
6
619
DOI: 10.1021/acscatal.5b01812
ACS Catal. 2015, 5, 6615−6623

ACS Catalysis
Research Article
Figure 3. Apparent hydrogen generation quantum yields from 2.16 nm CdS/Ni NSs at different (a) Ni2+ concentrations (Iexc = 3.5 W/cm ), (b)
2
laser intensities, and (d) water to ethanol ratios, following ligand exchange. (c) Time-dependent hydrogen generation trajectory of CdS/Ni NSs
2
under an incident excitation intensity of Iexc = 13 mW/cm .
nm CdS and CdS/Ni NSs to investigate their photocatalytic
properties. NS suspensions were purged with argon and were
illuminated with the same 405 nm laser used to generate Ni
NPs. Illumination times were set to 18 h. Effects of the incident
Next, excitation intensities in these H generation measure-
ments were tested to see their effect on apparent quantum
2
yields. Excitation intensities were therefore varied in decreasing
2
succession from Iexc = 3500 to Iexc = 5 mW/cm using CdS/Ni
light intensity were also tested by varying I between 0.005
and 3.5 W/cm as described below. A scheme illustrating the
samples with empirically determined optimal Ni NP sizes and
surface coverages. The lower intensity limit of 5 mW/cm was
exc
2
2
H generation chemistry can be found in Figure S14 in the
2
dictated by the detection limit of the gas chromatograph. In
Supporting Information, wherein protons are obtained from
each case, corresponding H generation yields were measured
2
both H O and ethanol with the letter also acting as a hole
after 18 h of irradiation.
Resulting apparent quantum yields were seen to increase
from 4.14% (Iexc = 3.5 W/cm ) to 24.5% (I = 5 mW/cm )
with an apparent maximum of 25.3% at Iexc = 13 mW/cm . This
2
4
0,41
scavenger.
In all cases, to quantify resulting H generation
2
2
2
efficiencies, gas chromatographic measurements were con-
ducted on samples at given intervals during irradiation. From
exc
2
measured H yields as well as the number of incident photons
observed quantum yield trend is summarized in Figure 3b and
is consistent with prior studies on CdS/Ni NR photocatalysts
where decreasing excitation intensities yielded larger apparent
2
at a given excitation intensity, corresponding H generation
2
apparent quantum yields were calculated. Details of these
quantum yield estimates can be found in the Supporting
Information.
1
1
hydrogen generation quantum yields. The behavior is
attributed to differences in the kinetic order of native charge
11
To investigate any Ni NP size influences on observed
recombination versus chemical reduction processes.
hydrogen generation efficiencies, NiSO concentrations used to
deposit Ni NPs were varied between 1.3 × 10 and 1.0 × 10
M. The results of these measurements are summarized in
Finally, the performance of CdS/Ni NSs was compared to
that of their bare CdS NS counterparts (Iexc = 13 mW/cm ). Ni
4
−5
−3
2
NP decorated NSs (∼6 nm NP diameter) were seen to perform
significantly better. In the former case, optimal apparent
quantum yields of 25.3% were seen, whereas values of 0.95%
were seen for CdS NSs. The observed enhancement has several
origins and includes the presence of a metal/semiconductor
Figure 3a. They show that hydrogen generation quantum yields
2
(
λexc = 405 nm, Iexc = 3.5 W/cm , 18 h irradiation) increase
from 0.34% to 4.14% with increasing [NiSO ] between 1.3 ×
4
−
5
−4
1
0
and 2.0 × 10 M (corresponding NP diameter ∼6 nm in
4
,6,10
this range with surface coverages proportional to concentration
and reaching a limit of one NP per NS). Beyond 2.0 × 10− M,
however, quantum yields decrease, reaching a value of 0.69%
heterojunction that aids charge separation,
electron
4
43,44
accumulation on the metal,
and favorable proton affinities
45
to Ni. Together, a ∼25-fold enhancement of hydrogen
generation quantum yields results over that of bare CdS NSs.
As a further point of contrast, despite the favorable Fermi level
of Au, poor proton binding affinities, and unfavorable Au
electron discharge kinetics lead to no apparent improvements
−
3
for a corresponding NiSO concentration of 1.0 × 10
M
4
(
corresponding Ni NP diameter ∼10 nm). Although specific
45
surface coverages in this regime are difficult to determine, the
results empirically suggest that an optimal Ni NP size as well as
surface coverage exists for effective hydrogen generation (i.e., 6
43
in H2 generation efficiencies for Au NP decorated CdSe
42
5
nm NP diameter, one NP per NS).
nanowires. These results therefore suggest that CdS/Ni NS
6
620
DOI: 10.1021/acscatal.5b01812
ACS Catal. 2015, 5, 6615−6623

ACS Catalysis
Research Article
heterostructures represent a potentially useful high-efficiency
hydrogen generation photocatalyst.
At this point, Figure 3c plots the measured hydrogen
generation trajectory of an optimal CdS/Ni NS sample
−
4
(
[NiSO ] = 2.0 × 10 M resulting in 6 nm Ni NPs, a
4
2
coverage of ∼1 NP per NS, and I = 13 mW/cm ). Of note is
that the H generation rate changes during the reaction with
exc
2
larger rates seen at early times. In particular, the apparent
quantum yield, measured after 2 h of irradiation, is ∼64%,
whereas in contrast after 18 h of irradiation the apparent
quantum yield drops to 25.3%. Such temporal variations in
hydrogen generation efficiencies have been seen before in Ni
NP decorated CdS NRs, wherein initial hydrogen generation
quantum yields of 53% were observed which later decreased to
11
2
0% after 80 h of irradiation. We attribute these long time
decreases in hydrogen generation yields to partial photo-
46
47
corrosion of both CdS and Ni.
To improve the solubility of CdS/Ni NSs in aqueous
solutions, we have additionally conducted ligand exchange to
replace native stearate ligands with polyethylenimine (PEI), a
long-chain polymer previously shown to aid the water
3
solubilization of colloidal nanostructures. Details of these
ligand exchange reactions have been provided in the
Experimental Section. In this case, measured H quantum
2
yields under the same optimized reaction conditions used
−
4
earlier ([NiSO ] = 2.0 × 10 M resulting in 6 nm Ni NPs, a
4
2
coverage of ∼1 NP per NS, and Iexc = 13 mW/cm ) are seen to
be lower by a factor of ∼3. Observed apparent quantum yields
are ∼8% in comparison to ∼25.3% before ligand exchange
(
Figure 3b vs Figure 3d). Near-identical apparent quantum
yields are also seen for different water/ethanol mixtures, as
illustrated in Figure 3d. We therefore speculate that this pre/
post ligand exchange difference arises due to unwanted charge
recombination pathways which appear due to the exchange and
which effectively compete with desired reduction chemistries.
The hypothesis is corroborated by photoluminescence
measurements on PEI passivated CdS NSs, where no emission
is observed, whether band edge or trap related (Figure S13 in
the Supporting Information).
Figure 4. TDA spectra of 2.16 nm (a) CdS and (b) CdS/Ni NSs taken
at different delays following excitation. Insets give comparisons of early
time bleach kinetics at 442 and 457 nm for CdS and CdS/Ni NSs,
respectively. (c) Band edge bleach kinetics for CdS and CdS/Ni NSs.
3
.5. Transient Differential Absorption Measurements.
Finally, to better understand relevant charge recombination and
relaxation processes in 2.16 nm CdS and CdS/Ni NSs during
photocatalytic hydrogen generation, we have conducted
transient differential absorption (TDA) spectroscopy under
actual hydrogen generation conditions. TDA is an important
spectroscopic technique in this regard, since it enables one to
unravel the fate of photogenerated carriers in nanostructures.
For low-dimensional cadmium chalcogenides, the TDA bleach
signal predominantly reflects the response of electrons rather
In all cases, measurements involve exciting samples at λ = 387 nm
exc
2
under equivalent pump fluences of 10 μJ/cm .
Interestingly, a bleach feature to the red side of the band edge
also arises at 457 nm (2.71 eV) and is attributed to the
existence of tail states. This is consistent with both the linear
absorption and the accompanying trap related emission of CdS
NSs between 530 and 570 nm (Figure 2b,c).
48−50
than holes due to their lighter effective mass.
Con-
sequently, by monitoring of the bleach dynamics of TDA
spectra, electron localization and recombination kinetics can be
studied. Specific details of these TDA measurements can be
found in the Experimental Section. In all cases, experiments
have been conducted with pump fluences in the linear regime
to exclude kinetic contributions from multiexciton dynamics
An analysis of the band edge bleach recovery kinetics in
Figure 4c shows that decays can be fit to biexponential
functions with time constants of τ = 1.7 ps (0.75% area based
1
weight) and τ = 232.6 ps (99.25% area based weight). The fast
2
decay component is attributed to ultrafast electron trapping
into defects and is corroborated by the early time kinetics of the
corresponding 457 nm bleach feature shown in the inset of
Figure 4a. Namely, the 457 nm trap-related feature exhibits a
short induction period before growth at the expense of the
(Figure S15 in the Supporting Information).
Figure 4a shows that upon excitation of bare CdS NSs at 387
2
nm (corresponding pump fluence, 10 μJ/cm ) a band edge
bleach appears within ∼0.5 ps. The bleach maximum occurs at
band edge bleach. Next, the slower τ component is attributed
2
442 nm and effectively corresponds to the lowest energy
to band edge exciton recombination, given experimentally
observed exciton recombination time scales in CdS NRs less
exciton transition in the NS absorption spectrum (Figure 2b).
An associated bleach maximum at 442 nm is |ΔA|/A433 = 0.12.
51
than 1 ns.
6
621
DOI: 10.1021/acscatal.5b01812
ACS Catal. 2015, 5, 6615−6623

ACS Catalysis
Research Article
Of note is that, at 1.5 ns, 29% of the initial bleach magnitude
remains. This shows that long-lived electrons exist in CdS NSs.
Analogous behavior has been seen in CdSe NSs with an
from efficient electron transfer from CdS to Ni. The metal/
semiconductor heterojunction is thus critical in dissociating
strongly bound excitons in 2D CdS NSs and introduces an
effective pathway for creating free carriers needed to carry out
relevant reduction chemistries.
21
associated decay constant of 10 ns or longer. The long-lived
TDA bleach in CdS NSs therefore suggests that a sizable
fraction of excitons do not trap or dissociate to undergo rapid
charge recombination, despite the presence of apparent defect-
related tail states. At the same time, accompanying hydrogen
generation experiments conducted on CdS NSs show low
quantum yields of ∼0.95%. This in turn suggests that, despite
their long lifetimes, the strong binding energy of excitons in
CdS NSs makes subsequent reduction chemistries prohibitive.
Such large binding energies stem from dielectric contrast effects
ASSOCIATED CONTENT
Supporting Information
■
*
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.5b01812.
Low- and high-magnification TEM images of 1.50, 1.80,
and 2.16 nm thick CdS NSs, lateral sizing histograms for
1.80 and 2.16 nm thick CdS NSs, ensemble EDXS
spectra of 1.50 and 1.80 nm thick CdS NSs. Comparison
of CdS NS d spacings obtained from ensemble XRD
measurements along with known bulk values, electro-
chemical scans for 1.80 and 2.16 nm CdS NSs deposited
on FTO, CdS NS tail states highlighted on a semi-
logarithmic plot, NS concentration and absorption cross
section estimates, time-correlated single photon counting
kinetics for 2.16 nm CdS NSs, TEM images and NP
sizing histograms for Ni NPs deposited onto 2.16 nm
52
which arise in low-dimensional materials. In the case of CdSe
NSs, this leads to exciton binding energies on the order of 300
5
3
meV.
Following Ni NP deposition, significant changes to the TDA
response are seen. This is apparent in Figure 4b,c, where the
band edge |ΔA|/A maximum decreases significantly to a value
of |ΔA|/A
= 0.078. Corresponding bleach kinetics also
433
become faster. In this regard, traces can be fit to triexponential
decays with a significant shortening of all time constants.
Namely, values of τ = 0.39 ps (0.7% area based weight), τ =
1
2
CdS NSs using different NiSO concentrations, emission
4
.11 ps (3.6% area based weight), and τ = 166 ps (95.7% area
4
3
spectra of 2.16 nm CdS, PEI passivated CdS, and CdS/
Ni NSs, schematic illustrating hydrogen generation from
CdS/Ni composites in water/ethanol mixtures, apparent
hydrogen generation quantum yield estimates, TDA
spectra and integrated bleach feature areas for 2.16 nm
CdS NSs obtained at different pump fluences (PDF)
based weight) are extracted, leading to only ∼8% of the initial
bleach magnitude remaining at 1.5 ns (Figure 4c).
Both the suppressed |ΔA|/A value as well as the faster
4
33
decay kinetics are attributed to interactions between CdS and
Ni. Possible processes include ultrafast energy transfer, an effect
21
previously seen in Pt NP decorated CdSe NSs and ultrafast
charge transfer from CdS to Ni. The latter scenario is
supported by the above photocatalytic hydrogen generation
measurements, wherein dramatic enhancements to experimen-
tal hydrogen generation quantum yields are seen. In this regard,
efficient energy transfer should result in no enhancement, as it
represents a parasitic loss process for excitations.
AUTHOR INFORMATION
Corresponding Authors
E-mail: mzhukovs@nd.edu.
E-mail: mkuno@nd.edu.
■
*
*
Notes
Consequently, we conclude that the presence of metal/
semiconductor heterojunctions in CdS/Ni NSs leads to the
efficient dissociation of long-lived and strongly bound excitons.
This, in turn, suggests that the combination of several intrinsic
features of 2D CdS NSs, namely, their large absorption
efficiencies, their strong exciton binding energies, and, more
importantly, their long exciton lifetimesall coupled to the
presence of a metal/semiconductor heterojunctionresults in
efficient hydrogen generation.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by the Army Research Office
(W911NF-12-1-0578) and by the Center for Sustainable
Energy at Notre Dame (ND Energy). P.T. thanks the Royal
Thai Government Scholarship for financial support. We also
thank the Notre Dame Integrated Imaging Facility (NDIIF)
and the Notre Dame Radiation Laboratory for use of their
facilities. Chemical analyses were conducted at the Center for
Environmental Science and Technology (CEST) at the
University of Notre Dame.
4
. CONCLUSION
In summary, we have synthesized high-quality, thickness-
controlled, CdS NSs through the thermal decomposition of
CdDDTC, a single-source precursor. Thicknesses between 1.50
and 2.16 nm have been produced with lateral dimensions larger
than those currently made through the exclusive thermolysis of
REFERENCES
■
(1) Sakamoto, M.; Xiong, A.; Kanakubo, R.; Ikeda, T.; Yoshinaga, T.;
Maeda, K.; Domen, K.; Teranishi, T. Chem. Lett. 2012, 41, 1325−
1327.
CdAc . Obtained conduction band potentials and absorption
2
cross sections are useful not just within the context of
photocatalytic applications described herein but also within
the context of other energy-related applications. Corresponding
Ni NP decorated NSs have been made through the
photodeposition of Ni. Subsequent photocatalytic hydrogen
generation measurements of both CdS and CdS/Ni NSs reveal
long time apparent quantum yields as large as 25%. Ever larger
yields, as high as 64%, have been seen during the initial stages
(
4
(
2) Zhao, J.; Holmes, M. A.; Osterloh, F. E. ACS Nano 2013, 7,
316−4325.
3) Ben-Shahar, Y.; Scotognella, F.; Waiskopf, N.; Kriegel, I.; Dal
Conte, S.; Cerullo, G.; Banin, U. Small 2015, 11, 462−471.
4) Berr, M.; Vaneski, A.; Susha, A. S.; Rodríguez-Fernan
Doblinger, M.; Jackel, F.; Rogach, A. L.; Feldmann, J. Appl. Phys. Lett.
010, 97, 093108.
5) Tongying, P.; Plashnitsa, V. V.; Petchsang, N.; Vietmeyer, F.;
(
́
dez, J.;
̈
̈
2
(
of H generation. Ensemble TDA spectroscopy illustrates that
Ferraudi, G. J.; Krylova, G.; Kuno, M. J. Phys. Chem. Lett. 2012, 3,
2
the improved photocatalytic efficiency of CdS/Ni NSs stems
3234−3240.
6
622
DOI: 10.1021/acscatal.5b01812
ACS Catal. 2015, 5, 6615−6623

ACS Catalysis
Research Article
(
6) Tongying, P.; Vietmeyer, F.; Aleksiuk, D.; Ferraudi, G. J.;
Krylova, G.; Kuno, M. Nanoscale 2014, 6, 4117−4124.
7) Han, Z.; Qiu, F.; Eisenberg, R.; Holland, P. L.; Krauss, T. D.
Science 2012, 338, 1321−1324.
8) Wilker, M. B.; Shinopoulos, K. E.; Brown, K. A.; Mulder, D. W.;
King, P. W.; Dukovic, G. J. Am. Chem. Soc. 2014, 136, 4316−4324.
9) Amirav, L.; Alivisatos, A. P. J. Phys. Chem. Lett. 2010, 1, 1051−
054.
10) Wu, K.; Chen, Z.; Lv, H.; Zhu, H.; Hill, C. L.; Lian, T. J. Am.
Chem. Soc. 2014, 136, 7708−7716.
11) Simon, T.; Bouchonville, N.; Berr, M. J.; Vaneski, A.; Adrovic,
A.; Volbers, D.; Wyrwich, R.; Doblinger, M.; Susha, A. S.; Rogach, A.
L.; Jackel, F.; Stolarczyk, J. K.; Feldmann, J. Nat. Mater. 2014, 13,
013−1018.
12) Vietmeyer, F.; McDonald, M.; Kuno, M. J. Phys. Chem. C 2012,
(37) Wang, J. J.; Li, Z. J.; Li, X. B.; Fan, X. B.; Meng, Q. Y.; Yu, S.; Li,
C. B.; Li, J. X.; Tung, C. H.; Wu, L. Z. ChemSusChem 2014, 7, 1468−
1475.
(
(38) Bamwenda, G. R.; Tsubota, S.; Nakamura, T.; Haruta, M. J.
Photochem. Photobiol., A 1995, 89, 177−189.
(
(39) Forouzan, F.; Richards, T. C.; Bard, A. J. J. Phys. Chem. 1996,
1
(
00, 18123−18127.
(
1
(
40) Korzhak, A. V.; Ermokhina, N. I.; Stroyuk, A. L.; Bukhtiyarov, V.
K.; Raevskaya, A. E.; Litvin, V. I.; Kuchmiy, S. Y.; Ilyin, V. G.; Manorik,
P. A. J. Photochem. Photobiol., A 2008, 198, 126−134.
(41) Lu, H.; Zhao, J.; Li, L.; Gong, L.; Zheng, J.; Zhang, L.; Wang, Z.;
(
́
Zhang, J.; Zhu, Z. Energy Environ. Sci. 2011, 4, 3384−3388.
42) Nakibli, Y.; Kalisman, P.; Amirav, L. J. Phys. Chem. Lett. 2015, 6,
265−2268.
43) Wood, A.; Giersig, M.; Mulvaney, P. J. Phys. Chem. B 2001, 105,
810−8815.
44) Subramanian, V.; Wolf, E. E.; Kamat, P. V. J. Am. Chem. Soc.
004, 126, 4943−4950.
45) Norskov, J. K.; Bligaard, T.; Logadottir, A.; Kitchin, J. R.; Chen,
̈
(
2
(
8
(
2
(
̈
1
(
1
(
(
16, 12379−12396.
13) Alivisatos, A. P. Science 1996, 271, 933−937.
14) Tessier, M. D.; Mahler, B.; Nadal, B.; Heuclin, H.; Pedetti, S.;
Dubertret, B. Nano Lett. 2013, 13, 3321−3328.
15) She, C.; Fedin, I.; Dolzhnikov, D. S.; Demortier
D.; Pelton, M.; Talapin, D. V. Nano Lett. 2014, 14, 2772−2777.
16) Ding, Y.; Singh, V.; Goodman, S. M.; Nagpal, P. J. Phys. Chem.
Lett. 2014, 5, 4291−4297.
17) Chen, Z. Y.; Nadal, B.; Mahler, B.; Aubin, H.; Dubertret, B. Adv.
Funct. Mater. 2014, 24, 295−302.
18) Lorenzon, M.; Christodoulou, S.; Vaccaro, G.; Pedrini, J.;
Meinardi, F.; Moreels, I.; Brovelli, S. Nat. Commun. 2015, 6, 6434.
19) Lhuillier, E.; Robin, A.; Ithurria, S.; Aubin, H.; Dubertret, B.
Nano Lett. 2014, 14, 2715−2719.
20) Lhuillier, E.; Pedetti, S.; Ithurria, S.; Heuclin, H.; Nadal, B.;
Robin, A.; Patriarche, G.; Lequeux, N.; Dubertret, B. ACS Nano 2014,
, 3813−3820.
21) Wu, K.; Li, Q.; Du, Y.; Chen, Z.; Lian, T. Chem. Sci. 2015, 6,
049−1054.
22) Wu, K.; Li, Q.; Jia, Y.; McBride, J. R.; Xie, Z.-X.; Lian, T. ACS
Nano 2015, 9, 961−968.
23) Sigle, D. O.; Zhang, L.; Ithurria, S.; Dubertret, B.; Baumberg, J.
J. J. Phys. Chem. Lett. 2015, 6, 1099−1103.
24) Voiry, D.; Yamaguchi, H.; Li, J.; Silva, R.; Alves, D. C. B.; Fujita,
J. G.; Pandelov, S.; Stimming, U. J. Electrochem. Soc. 2005, 152, J23−
(
̀
e, A.; Schaller, R.
J26.
(46) Matsumoto, H.; Sakata, T.; Mori, H.; Yoneyama, H. J. Phys.
(
Chem. 1996, 100, 13781−13785.
(47) Zhang, L.; Liu, Q.; Aoki, T.; Crozier, P. A. J. Phys. Chem. C
2015, 119, 7207−7214.
(
(48) Klimov, V. I. J. Phys. Chem. B 2000, 104, 6112−6123.
(49) Puthussery, J.; Lan, A.; Kosel, T. H.; Kuno, M. ACS Nano 2008,
2, 357−367.
(
(
(50) Zhukovskiy, M. A.; Stroyuk, A. L.; Shvalagin, V. V.; Smirnova,
N. P.; Lytvyn, O. S.; Eremenko, A. M. J. Photochem. Photobiol., A 2009,
203, 137−144.
(
(
(
51) Wu, K.; Zhu, H.; Lian, T. Acc. Chem. Res. 2015, 48, 851−859.
52) Muljarov, E. A.; Zhukov, E. A.; Dneprovskii, V. S.; Masumoto, Y.
8
(
1
(
Phys. Rev. B: Condens. Matter Mater. Phys. 2000, 62, 7420−7432.
(53) Benchamekh, R.; Gippius, N. A.; Even, J.; Nestoklon, M. O.;
Jancu, J.-M.; Ithurria, S.; Dubertret, B.; Efros, A. L.; Voisin, P. Phys.
Rev. B: Condens. Matter Mater. Phys. 2014, 89, 035307.
(
(
T.; Chen, M.; Asefa, T.; Shenoy, V. B.; Eda, G.; Chhowalla, M. Nat.
Mater. 2013, 12, 850−855.
(
25) Baghani, E.; O’Leary, S. K.; Fedin, I.; Talapin, D. V.; Pelton, M.
J. Phys. Chem. Lett. 2015, 6, 1032−1036.
26) Alemseghed, M. G.; Ruberu, T. P. A.; Vela, J. Chem. Mater.
011, 23, 3571−3579.
27) Mokari, T.; Rothenberg, E.; Popov, I.; Costi, R.; Banin, U.
Science 2004, 304, 1787−1790.
28) Li, Z.; Qin, H.; Guzun, D.; Benamara, M.; Salamo, G.; Peng, X.
Nano Res. 2012, 5, 337−351.
29) Bouet, C.; Mahler, B.; Nadal, B.; Abecassis, B.; Tessier, M. D.;
(
2
(
(
(
Ithurria, S.; Xu, X.; Dubertret, B. Chem. Mater. 2013, 25, 639−645.
(
(
30) Li, Z.; Peng, X. J. Am. Chem. Soc. 2011, 133, 6578−6586.
31) Amelia, M.; Impellizzeri, S.; Monaco, S.; Yildiz, I.; Silvi, S.;
Raymo, F. M.; Credi, A. ChemPhysChem 2011, 12, 2280−2288.
32) Poznyak, S. K.; Osipovich, N. P.; Shavel, A.; Talapin, D. V.; Gao,
M.; Eychmuller, A.; Gaponik, N. J. Phys. Chem. B 2005, 109, 1094−
100.
33) Pan, A.; Liu, D.; Liu, R.; Wang, F.; Zhu, X.; Zou, B. Small 2005,
, 980−983.
34) Bastin, J. A.; Mitchell, E. W. J.; Whitehouse, J. Br. J. Appl. Phys.
959, 10, 412−416.
35) Kunneman, L. T.; Schins, J. M.; Pedetti, S.; Heuclin, H.;
(
̈
1
(
1
(
1
(
Grozema, F. C.; Houtepen, A. J.; Dubertret, B.; Siebbeles, L. D. A.
Nano Lett. 2014, 14, 7039−7045.
(
36) Stroyuk, A. L.; Raevskaya, A. E.; Korzhak, A. V.; Kotenko, I. E.;
Glebov, E. M.; Plyusnin, V. F.; Kuchmii, S. Ya. Theor. Exp. Chem.
009, 45, 12−22.
2
6
623
DOI: 10.1021/acscatal.5b01812
ACS Catal. 2015, 5, 6615−6623