Subscriber access provided by OCCIDENTAL COLL
Interfaces: Adsorption, Reactions, Films, Forces, Measurement Techniques, Charge Transfer, Electrochemistry, Electrocatalysis, Energy Production and Storage
Doping of Nb5+ Species at the Au-TiO2 Interface for Plasmonic Photocatalysis Enhancement Yasuhiro Shiraishi, Jun Imai, Naoki Yasumoto, Hirokatsu Sakamoto, Shunsuke Tanaka, Satoshi Ichikawa, and Takayuki Hirai Langmuir, Just Accepted Manuscript • Publication Date (Web): 27 Mar 2019 Downloaded from http://pubs.acs.org on March 27, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Doping of Nb5+ Species at the Au–TiO2 Interface for Plasmonic Photocatalysis Enhancement Yasuhiro Shiraishi,*,† Jun Imai,† Naoki Yasumoto,† Hirokatsu Sakamoto,† Shunsuke Tanaka,‡ Satoshi Ichikawa,∥ and Takayuki Hirai† †
Research Center for Solar Energy Chemistry, and Division of Chemical Engineering, Graduate School of Engineering Science, Osaka University, Toyonaka 560-8531, Japan. ‡ Department of Chemical, Energy and Environmental Engineering, Kansai University, Suita 564-8680, Japan. ∥
Institute for NanoScience Design, Osaka University, Toyonaka 560-8531, Japan.
ABSTRACT: Au nanoparticles loaded on semiconductor TiO2 absorb visible light due to their surface plasmon resonance (SPR), and inject the photogenerated hot electrons (ehot–) into the conduction band of TiO2. The separated charges promote oxidation and reduction reactions. The step that determines the rate of the plasmonic photocatalysis on the Au/TiO2 system is the ehot– injection through the Schottky barrier created at the Au–TiO2 interface. In the present work, niobium (Nb5+) oxide species were doped at the Au–TiO2 interface by loading Nb5+ onto the TiO2 surface followed by deposition of Au particles (2 wt % of TiO2). Visible light irradiation of the Au/Nb5+/TiO2 catalysts promotes aerobic oxidation of alcohols with much higher efficiency than undoped Au/TiO2. Lewis acidity of the Nb5+ species located at the interface cancels the negative charges of Au and creates the barrier with narrower depletion layer, promoting tunneling ehot– injection. Efficiency of the ehot– injection depends on the amount of Nb5+ doped. Loading small amounts of Nb5+ (~0.1 wt % of TiO2) creates mononuclear NbO4 species and shows large activity enhancement. In contrast, larger amount of Nb5+ creates aggregated polynuclear Nb2O5 species. They decrease the electron density of Au particles and weaken their SPR absorption. This suppresses the ehot– generation on the Au particles and decreases the activity of plasmonic photocatalysis. KEYWORDS: Photocatalysis · Gold nanoparticle · Surface plasmon resonance · Hot electron · Visible light
INTRODUCTION Surface plasmon resonance (SPR), a resonant oscillation of metal surface electrons by incident photons, is one attractive feature of Au nanoparticles (AuNPs).1 AuNPs strongly absorb visible region light ( >400 nm). Application of this feature to photocatalysis has therefore attracted attention for solar-driven chemical conversions.2,3 For this purpose, semiconductor TiO2 loaded with AuNPs (Au/TiO2) have extensively been studied.4 Generally, the plasmonic photocatalysis is considered to be initiated by the hot electron (ehot–) injection.5,6 Oscillation of sp band electrons of AuNPs by visible light (Scheme 1) produces ehot–. Some of the ehot– in the conduction band (CB) are injected into the TiO2 CB, while hot holes (+) remain on the AuNPs.7 These eCB– and + promote reduction and oxidation reactions, respectively. The step that determines the rate of plasmonic photocatalysis is the ehot– injection from the photoactivated AuNPs to the TiO2 CB.3 The Schottky barrier is created at the Au–TiO2 interface due to the difference of their Fermi levels. 8,9 The overbarrier path (Scheme 1a) is the mainly accepted mechanism for the ehot– injection. In this case, the ehot– with an energy higher than the barrier height are injected, where high overbarrier enegy is the main limitation.10 Several methods have therefore been proposed to enhance the ehot– injection by lowering the height
of Schottky barrier.11,12 In contrast, the tunneling path has been reported in some Au/TiO2 systems.13,14 As shown in Scheme 1b, the ehot– are injected into the TiO2 CB by penetrating migration with much lower energy.7 Although its triggering mechanism has not been clarified yet, the tunneling ehot– injection is the key to creating highly active plasmonic photocatalysts. Recently, we have clarified one of the factors triggering the tunneling ehot– injection.15 We made AuNPs on the anatase particles of the Degussa P25 TiO2.16,17 Visible light irradiation of the Au/P25_anatase catalyst promotes aerobic oxidation with high quantum yield (7.7 % at 550 nm) via the tunneling ehot– injection, whereas other commercially-availabe TiO2 loaded with AuNPs show low quantum yields (450 nm
0.01
50
30
Au/Nb5+x/TiO2 x [wt %] =
0.1 0.3
20
Count
0.5
30
Au/Nb2O5
20
20
Nb5+0.1/Au/TiO2
10 10
0
10
20
40
60
Acetophenone formed / µmol 0 0 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Size / nm
Size / nm
Size / nm
Figure 2. Size distribution of AuNPs on the respective catalysts.
3
Au/TiO2
Au/Nb5+0.1/TiO2 Au/Nb5+0.5/TiO2
2
b 60 Acetophenone formed / µmol
0
50 40
λ >450 nm
30 20 10
dark
0
F(R∞)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
0
6
12
18
24
Time / h
1 Nb2O5
0 300
Au/Nb2O5 TiO2
400
Nb5+0.1/TiO2 Nb5+0.5/TiO2
500 600 λ / nm
700
800
Figure 4. (a) Amounts of acetophenone produced by the reaction on the respective catalysts (black) in the dark or (yellow) under >450 nm light irradiation (light intensity at 450–800 nm, 16.8 mW cm–2). (b) Change in the amounts of acetophenone formed on Au/Nb5+0.1/TiO2 in the dark or under photoirradiation, where 10 mg of catalyst was used. For all runs, acetophenone was produced solely with >99% selectivity
Figure 3. DR UV-vis spectra of the respective catalysts.
Photocatalysis. Figure 4a summarizes the amounts of acetophenone formed by aerobic oxidation of 1-phenylethanol at 298 K for 12 h in the dark (black) or under λ >450 nm light irradiation by a Xe lamp (yellow). Bare TiO2 and Nb5+0.1/TiO2 scarcely produce acetophenone (95% selectivity (Table 1). Furthermore, ICP analysis of the solution
3 ACS Paragon Plus Environment
Langmuir
Table 1. Aerobic photooxidation of various types of benzylic alcohols on Au/Nb5+0.1/TiO2.a substrate
conv. /%b
select. /%c
product
78
>99
0.1 Photocurrent density / μA cm-2
recovered after photoreaction did not detect leaching of Au and Nb species, indicating that the species are stably adhered on the surface even after the reaction.
Au/Nb5+0.1/TiO2 light on
Nb5+0.1/TiO2 TiO2 Au/Nb2O5
0
120
>99
79
99
240
360
600
Figure 5. Photocurrent response of the catalysts measured at 298 K in 0.5 M Na2SO4 under >450 nm light at a bias of 0.1 V.
RCT
Au/Nb5+0.5/TiO2
48
480
t/s
>99
85
Au/TiO2
0.05
0
76
Au/Nb5+0.5/TiO2
off
Rs
3000
97
CDL Au/TiO2
2000
68
97
84
99
Zʹʹ / Ω
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 9
1000 Au/Nb5+0.1/TiO2
0
0
1000 2000 3000 4000 5000 6000 Zʹ / Ω
a
81
>99
54
95 Au/Nb5+0.1/TiO2
Conditions: toluene (5 mL), (20 mg), substrate (5 mM), O2 (1 atm), time (24 h), λ >450 nm, temperature (298 K). b = [substrate converted] / [initial amount of substrate] × 100. c = [product formed] / [substrate converted] × 100.
ehot– injection. Properties of the ehot– injection from the photoactivated AuNPs to TiO2 were studied by photocurrent response analysis of the catalysts supported on a fluorine tin oxide (FTO) electrode, where the ehot– photogenerated on the AuNPs are injected into the TiO2 CB and transferred to the electrode.20 As shown in Figure 5, the photocurrent density on Au/Nb5+0.1/TiO2 is much larger than that on Au/TiO2 and Au/Nb5+0.5/TiO2, where the order is consistent with the photocatalytic activity (Figure 4a). It is noted that bare TiO2, Nb5+0.1/TiO2, and Au/Nb2O5 exhibit very low current density. The results also agree with the activity data (Figure 4a). These data indicate that the doping of relatively low amounts of Nb 5+ (~0.1 wt %) at the Au–TiO2 interface enhances the injection of ehot– from the photoactivated AuNPs to TiO2 CB and, hence, enhances plasmonic photocatalysis.
Figure 6. Nyquist plots of the catalysts measured at 298 K in 0.5 M Na2SO4 at a bias of 2 V, (closed keys) in the dark and (open keys) under >450 nm irradiation.
The properties of ehot– injection were further studied using electrochemical impedance spectroscopy (EIS), where the eCB– injected to TiO2 behave as the charge carriers.21,22 Figure 6 shows the Nyquist plots of the catalysts in the dark or under photoirradiation. The inset of this figure shows the proposed equivalent circuit model,23 where RS is the ohmic resistance, RCT is the electron transfer resistance, and CDL is the double layer capacitance, respectively. The EIS fitting parameters are shown in Table 2. In the dark (closed keys), Au/Nb 5+0.1/TiO2 shows lower RCT than Au/TiO2 and Au/Nb5+0.5/TiO2. Visible light irradiation of the catalysts (open keys) decrease R CT, indicating that the ehot– photogenerated on AuNPs are indeed injected into TiO2 CB. Au/Nb5+0.1/TiO2 shows much lower RCT than Au/TiO2 and Au/Nb5+0.5/TiO2. The results agree well with the photocatalytic activity (Figure 4a) and the photoresponse data (Figure 5). These findings again suggest that the doping small amounts of Nb5+ at the Au–TiO2 interface enhances the ehot– injection from photoactivated AuNPs to TiO2 CB. This thus promotes efficient charge separation and exhibits activity enhancement of plasmonic photocatalysis. It is well known that, in the heterojunction system, the width of depletion layer is inversely proportional to the double layer capasitance (CDL).23 As shown in Table 2, Au/Nb5+0.1/TiO2 shows much
4 ACS Paragon Plus Environment
Page 5 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
larger CDL than Au/TiO2 and Au/Nb5+0.5/TiO2, indicating that the depletion layer of Au/Nb5+0.1/TiO2 is much narrower.
a Au/TiO2
Table 2. EIS fitting parameters from the equivalent circuit of the respective catalysts.a substrate
light
RS / Ω
RCT / kΩ
CDL / μF
Au/TiO2
+
35.1
6.01
0.210
–
32.9
6.05
0.209
Au/Nb5+0.1/TiO2
Au/Nb5+0.5/TiO2
a The
+
23.9
3.04
0.331
–
23.5
5.16
0.309
+
55.2
5.96
0.134
–
53.4
6.47
0.123
Au 4f7/2
Au 4f5/2 86.6
82.9
86.8
83.1
Au/Nb5+0.1/TiO2 Au/Nb5+0.5/TiO2
83.4
87.1 Au/Nb2O5
85
90
80
Ti 2p3/2
b
458.3
Ti 2p1/2 464.1
TiO2
data are from Figure 6.
X-ray photoelectron spectroscopy (XPS) was used to clarifty the electronic properties of AuNPs. Figure 7a shows the charts of the catalysts at the Au 4f level. Au/TiO2 shows Au peaks at lower binding energies than Au/Nb2O5. Fermi levels of metal oxide semiconductors lie at more positive position than those of Au.24,25 This leads to an electron donation from metal oxide semiconductors to AuNPs for the Fermi level balancing and creates the Schottky barrier (Scheme 1). The Fermi level of Nb2O5 lies at more negative position than that of TiO2.26 This results in the donation of smaller amounts of electrons to AuNPs and creates AuNPs with lower electron density. As a result of this, Au/Nb2O5 shows very weak SPR absorption (Figure 3) and scarcely produce ehot–, resulting in almost no photocatalytic activity (Figure 4a). As shown in Figure 7a, Au/Nb5+0.1/TiO2 shows Au peaks at the binding energy similar to that of Au/TiO2, indicating that the doped Nb5+ scarcely affect the electron density of AuNPs. In contrast, Au/Nb5+0.5/TiO2 shows Au peaks at higher binding energies. As shown in Figure 3, Au/Nb 5+0.5/TiO2 shows SPR absorption much weaker than Au/TiO2 and Au/Nb5+0.1/TiO2. These data suggest that doping excess amounts of Nb 5+ leads to binding between Au and Nb species and decreases the electron density of AuNPs. As a result of this, Au/Nb 5+0.5/TiO2 shows weaker SPR absorption. This therefore suppresses the ehot– photogeneration and, hence, shows decreased activity of plasmonic photocatalysis (Figure 4a). As shown in Figure 7b, XPS charts of the catalysts at the Ti 2p level revealed that the positions of Ti peaks are scarcely affected by the amouts of Nb5+ doped, probably due to much smaller amounts of Nb 5+ than that of surface Ti species. These data indicate that small amounts of Nb5+ (~0.1 wt %) scarcely affect the electron density of AuNPs and maintains strong SRP absorption. In contrast, doping excess amount of Nb5+ decreases the electron density of AuNPs, resulting in weaker SPR absorption and decreased photocatalytic activity.
Au/TiO2
Au/Nb5+0.1/TiO2 Au/Nb5+0.5/TiO2 460 460
465 465
c
Nb 3d3/2 209.4
455 455
Nb 3d5/2 206.8
Au/Nb5+0.1/TiO2
209.6
206.9
Au/Nb5+0.5/TiO2 209.7
207.0
Au/Nb2O5 210 210 Binding energy / eV
205 205
Figure 7. XPS charts of the respective catalysts at (a) Au 4f, (b) Ti 2p, and (c) Nb 3d levels, respectively.
Structure of doped Nb5+ species. To clarify the structure of the doped Nb5+ species, diffuse-reflectance infrared Fourier transform spectroscopy (DRIFTS) was used at 100 K with carbon monooxide (CO) as a probe molecule.18,27 As shown in Figure 8a, the DRIFTS chart of bare TiO2 shows a strong band at 2177 cm–1 assigned to CO adsorbed on the Ti5c species and a weak band at 2156 cm–1 assigned to CO adsorbed on the – OH groups of the metal oxide surface.28 In this case, a component at 2183 cm–1 assigned to CO adsorbed on the Ti4c species does not appear. This indicates that the Lewis acidic Ti4c species15 do not exist on the surface of the present ST-21 TiO2 powders.
5 ACS Paragon Plus Environment
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
As shown in Figure 8b, the DRIFTS chart of Nb5+0.1/TiO2 shows an additional component at 2174 cm–1 assigned to CO adsorbed on the 4-fold coordinated Nb5+ species (Nb4c).18 This suggests that, as shown in Scheme 2a, highly dispersed mononuclear NbO4 species exist on the TiO2 surface.29,30 As shown in Figure 8c, Au/Nb5+0.1/TiO2 shows a new component at 2128 cm–1 assigned to CO adsorbed on the AuNPs surface,31 but shows almost no Nb4c component (2174 cm–1). This suggests that almost all of the surface NbO4 species are covered by the loaded AuNPs. These findings indicate that, on the Au/Nb5+0.1/TiO2 catalyst, the mononuclear NbO4 species exist at the Au–TiO2 interface (Scheme 2a).
Ti5c 2177 cm–1
a TiO2
surface –OH 2156 cm–1
2100
2150
2200
b Nb5+0.1/TiO2
Nb4c 2174 cm-1
Ti5c
Page 6 of 9
Scheme 2. Proposed Au–TiO2 interface Au/Nb5+0.1/TiO2, and (b) Au/Nb5+0.5/TiO2. a Au/Nb5+0.1/TiO2
b Au/Nb5+0.5/TiO2
AuNPs
AuNPs
mononuclear NbO4 species
aggregated Nb2O5 species
surface –OH
(a)
2100
2150
2200
for
Nb4c
Nb5c
c Au/Nb5+0.1/TiO2 Ti5c sp conduction band
AuNPs 2128 cm–1 tunneling
Nb4c 2100
2150
2200
Nb5c 2181 cm–1
ECB EF
d Nb5+0.5/TiO2
eCB –
ehot
ehot – ECB
Ti5c
Ti5c Nb4c
EF
Ti5c Nb4c
TiO2
Nb4c
Nb5c
Ti5c surface –OH
2200
Au
EVB
Nb5c TiO2
sp band
Au
wide depletion layer
2100
e Au/Nb5+0.5/TiO2
Nb Nb5c 4c
δ+
Ti5c
narrow depletion layer
2150
2200
suppressed ehot – generation
Ti5c Ti5c
δ+ sp band
EVB
surface –OH
Nb5c
Nb5c Ti5c Ti5c
Ti5c
sp conduction band
–
AuNPs 2150
2100
Wavenumber / cm–1
Figure 8. DRIFTS charts of CO adsorbed on (a) TiO2, (b) Nb5+0.1/TiO2, (c) Au/Nb5+0.1/TiO2, (d) Nb5+0.5/TiO2, or (e) Au/Nb5+0.1/TiO2 at 100 K. The black line is the obtained chart, and the gray line is sum of the respective components.
As shown in Figure 8d, the DRIFTS chart of Nb5+0.5/ TiO2 shows the components at 2177 cm–1 and 2174 cm–1 assigned to CO adsorbed on the Ti5c and Nb4c species, repectively, as is the case for Nb5+0.1/TiO2 (Figure 8b). In this case, a new strong component is observed at 2181 cm–1, which is assigned to CO adsorbed on the 5-fold coordinated Nb5+ species (Nb5c).18 This indicates that, as shown in Scheme 2b, loading larger amounts of Nb5+ leads to aggregation of the NbO4 species and creates aggregated Nb2O5 species. As shown in Figure 8e, Au/Nb5+0.5/TiO2 shows almost no component at 2181 cm–1, suggesting that the aggregated Nb2O5 are covered by the loaded AuNPs. This indicates that, as shown in Scheme 2b, the aggregated Nb2O5 exist at the Au–TiO2 interface. The formation of mononuclear NbO4 and aggregated Nb2O5 species on the respective Au/Nb5+0.1/TiO2 and Au/Nb5+0.5/TiO2
6 ACS Paragon Plus Environment
Page 7 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
catalysts is confirmed by the XPS charts at the Nb 3d level (Figure 7c). The Nb peaks appear at higher binding energy with an increase in the amount of Nb5+ doped (Au/Nb5+0.1/TiO2 < Au/Nb5+0.5/TiO2 < Au/Nb2O5) owing to the formation of aggregated Nb2O5 species.32,33 The data strongly support the formation of mononuclear NbO4 and aggregated Nb2O5 at the Au–TiO2 interface on the respective Au/Nb5+0.1/TiO2 and Au/Nb5+0.5/TiO2 catalysts. An interesting phenomenone observed in the DRIFTS charts is the almost complete coverage of the NbO4 or Nb2O5 species by the loaded AuNPs. The selective formation of AuNPs on the Nb species may originate from the deposition-precipitation method for the preparation of AuNPs,19 as depicted in Scheme 3. At a neutral pH (~7), the Au precursor (HAuCl4) exists mainly as Au(OH)3 in an equilibrium with Au(OH)4–, whereas neutral Ti-OH species form mainly on the TiO2 surface.34,35 In contrast, the Nb species contain polarized Nb=O groups. 18 Electrostatic interaction between Au(OH)3 and the Nb=O groups may therefore create Au species adsorbed onto the Nb species behaving as an Au growth site. Thermal reduction of Au species followed by their migration may lead to a growth of AuNPs there, resulting in the selective formation of AuNPs around the Nb species. Although we cannot provide direct evidence for the above mechanism at the present stage, almost complete coverage of the Nb species by AuNPs (Figure 8) indicates that the interaction between the Au species and the Lewis acidic Nb species proposed in Scheme 3 may probably be involved in this phenomena. Scheme 3. Proposed mechanism for selective formation of AuNPs on the Nb oxide species.
–H2O
Lewis acidity of the species18 makes the surface charge of TiO2 more positive and cancel the negative charge of AuNPs. This creates a narrow depletion layer, as confirmed by the larger CDL in the EIS analysis (Table 2) and, hence, facilitates tunneling ehot– injection. In contrast, on the Au/Nb5+0.5/TiO2 catalyst, doping larger amount of Nb5+ creates aggregated Nb2O5 (Nb5c) species (Scheme 2b). This leads to binding between AuNPs and the Nb2O5 species and decreases the electron density of AuNPs. As a result of this, the catalyst exhibits weaker SPR absorption (Figure 3) and shows decreased activity of plasmonic photocatalysis (Figure 4a). CONCLUSION We fabricated Au/Nb5+/TiO2 plasmonic photocatalysts doped with Nb5+ species at the Au–TiO2 interface, and found that they efficiently catalyze aerobic oxidation of alcohols under visble light. The doped Lewis acidic species cancel the negative charges of AuNPs, creating a narrower depletion layer. This triggers tunneling ehot– injection and enhances the activity of plasmonic photocatalysis. The highest ehot– injection efficiency is achieved by the doping of relatively small amounts of Nb5+ (~0.1 wt%), where mononuclear NbO4 species exist at the Au– TiO2 interface. Apparent quantum yield for the photocatalysis (4.4% at 550 nm) is lower than that of our Au/P25_anatase catalyst (7.7% at 550 nm),15 which employs anatase particles of Degussa P25 TiO2 containing a large number of surface Lewis acidic 4-fold coordinated Ti4+ (Ti4c) species. This indicates that the efficiency for tunneling ehot– migration in the present system is lower than that in the Au/P25_anatase system. The results presented here, however, reveal that doping ″acidic species″ at the Au–TiO2 interface is an efficient way for the enhancement of plasmonic photocatalysis. So far, several stragegies have been proposed for plasmonic photocatalysis such as Ohmic contact between the plasmonic nanoparticles and semiconductor,38 transfer of hot holes (+) formed on the plasmonic metal to semiconductor,39 and electric field enhancement around the plasmonic metal.40 The doping acidic species at the Au–semiconductor interface may become a new powerful strategy for plasmonic photocatalysis enhancement.
thermal reduction
AuNPs AuNPs growth
Au
Mechanism for ehot– injection enhancement. The high ehot– injection efficiency on the Au/Nb5+0.1/TiO2 catalyst originates from the tunneling mechanism, where the ehot– penetrate through the Schottky barrier.13–15 As shown in Scheme 1a, on undoped Au/TiO2, AuNPs are charged negatively by electron transfer from TiO2. The positively-charged wide depletion layer is created at the TiO2 surface and suppresses the tunneling ehot– injection.36,37 On the Au/Nb5+0.1/TiO2 catalyst, doping small amount of Nb5+ creates the mononuclear NbO4 (Nb4c) species at the Au–TiO2 interface (Scheme 2a). The
EXPERIMENTAL SECTION General. All reagents were purchased from Wako, Tokyo Kasei, and Sigma-Aldrich and used as received. Anatase ST21 TiO2 was kindly supplied from Ishihara Sangyo Co., Ltd. (Japan). Catalyst preparation. Nb5+x/TiO2 [x (wt %) = Nb5+/TiO2 × 100; x = 0.01, 0.1, 0.3, and 0.5] were prepared as follows. The TiO2 (0.5 g) was stirried in 1 M HCl (10 mL) at 363 K for 3 h to remove the impurities. The powders were washed with water, and dried in vacuo at room temperature for 12 h. The powders (0.2 g) and NbCl5 (1.2, 11.9, 35.6, or 59.3 mg) were suspended in water (20 mL), and their pH were adjusted to ~12 with NaOH pelets. After stirring the suspension for 1 h at 363 K, the powders were recovered by centrifugation, washed with water, and dried in vacuo at room temperature for 12 h. Inductivelycoupled plasma (ICP) analysis indicated that no Nb5+ remained in the resulting solution, suggesting that all of the Nb 5+ added were loaded on TiO2. Au/Nb5+x/TiO2 catalysts were prepared
7 ACS Paragon Plus Environment
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
as follows. The Nb5+x/TiO2 (0.15 g) were added to water (50 mL) containing HAuCl4·4H2O (6.4 mg). Its pH was adjusted to ~7 with 0.1 mM NaOH, and the solution was stirred at 353 K for 3 h. The powders were washed with water, and dried in vacuo for 12 h. Then, the powders were calcined in air with a heating rate of 2 K min−1 and a holding time of 2 h at 673 K. ICP analysis indicated that no Au and Nb species remained in the resulting solutions after the Au loading, suggesting that all of the Au species added were successfully loaded onto the solids without leaching of the Nb species. The mole fraction of the Nb and Au species on the catalysts were threfore calculated using their feed amounts. The Au: Nb: Ti mole ratios on the Au/Nb5+0.1/TiO2 and Au/Nb5+0.5/TiO2 catalysts were therefore determined to be 0.805 : 0.085 : 99.110 and 0.805 : 0.425 : 98.770, respectively. Photocatalysis. Catalyst was added to a Pyrex tube (, 10 mm; capacity, 20 mL) containing toluene with alcohol, and the tube was sealed with a rubber septum cap. The catalyst was dispersed by ultrasonication, and O2 was bubbled for 5 min. The tube was left in a water bath and photoirradiated at λ > 450 nm with magnetic stirring by a 2 kW Xe lamp with a glass filter.41,42 After the reactions, the catalyst was recovered by centrifugation, and the solution was analysed by GC-FID. Analysis. Electrochemical analysis was performed in a three-electrode cell using 0.5 M Na2SO4 as an electrolyte with a Pt wire and an Ag/AgCl electrode as the counter and reference electrode, respectively. The working electrode was prepared according to a previously described procedure. 15 EIS measurements were carried out in the frequency range of 10 mHz to 100 kHz.43 DRIFTS charts were obtained on a FT/IR610 system (JASCP Corp.).44 The sample (30 mg) was added to the cell, evacuated (
