Exceptional Visible-Light Activities of TiO2-Coupled N-Doped Porous

Subscriber access provided by University of Idaho Library

Article

Exceptional visible-light activities of TiO2-coupled N-doped porous perovskite LaFeO3 for 2,4-dichlorophenol decomposition and CO2 conversion Muhammad Humayun, Yang Qu, Fazal Raziq, Rui Yan, Zhijun Li, Xuliang Zhang, and Liqiang Jing Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b04958 • Publication Date (Web): 21 Nov 2016 Downloaded from http://pubs.acs.org on November 22, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology 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 36

Environmental Science & Technology

1

Exceptional visible-light activities of TiO2-coupled N-doped porous perovskite

2

LaFeO3 for 2,4-dichlorophenol decomposition and CO2 conversion

3

Muhammad Humayun, Yang Qu, Fazal Raziq, Rui Yan, Zhijun Li, Xuliang Zhang and Liqiang Jing*

4

Key Laboratory of Functional Inorganic Materials Chemistry (Heilongjiang University), Ministry of Education,

5

School of Chemistry and Materials Science, International Joint Research Center for Catalytic Technology,

6

Harbin 150080, P. R. China.

7

Keywords: Porous LaFeO3, Improved visible-light photoactivity, 2,4-dichlorophenol degradation, CO2

8

conversion, Mechanism

9 10

Abstract: In this work, TiO2-coupled N-doped porous perovskite-type LaFeO3 nanocomposites as highly

11

efficient, cheap, stable and visible-light photocatalysts have successfully been prepared via wet chemical

12

processes. It is shown that the amount-optimized nanocomposite exhibits exceptional visible-light

13

photocatalytic activities for 2,4-dichlorophenol (2,4-DCP) degradation by ~ 3-time enhancement and for CO2

14

conversion to fuels by ~ 4-time enhancement, compared to the resulting porous LaFeO3 with rather high

15

photoactivity due to its large surface area. It is clearly demonstrated, by means of various experimental data,

16

especially for the .OH amount evaluation, that the obviously-enhanced photoactivities are attributed to the

17

increased specific surface area by introducing pores, to the extended visible-light absorption by doping N to

18

create surface states, and to the promoted charge transfer and separation by coupling TiO2. Moreover, it is

19

confirmed from radical trapping experiments that the photogenerated holes are the predominant oxidants in the

20

photocatalytic degradation of 2,4-DCP. Furthermore, a possible photocatalytic degradation mechanism for 2,4-

21

DCP is proposed mainly based on the resultant crucial intermediate, 2-chlorosuccinic acid with m/z=153, that

22

readily transform into CO2 and H2O. This work opens up a new feasible route to synthesize visible-light-

23

responsive high-activity perovskite-type nanophotocatalysts for efficient environmental remediation and

24

energy production.

ACS Paragon Plus Environment


25

1. INTRODUCTION

26

The efficient degradation of environmental pollutants and production of chemical fuels from

27

sustainable energy sources are the critical issues of scientific research nowadays. 1-3 The

28

chlorophenols comprise a group of organic pollutants that are highly resistant to degradation.4 In

29

particular, 2,4-DCP as the most common environmental hazardous pollutant, mainly arising from

30

the extensive use of pesticides, bactericides, insecticides, fungicides and herbicides, is potentially

31

toxic and carcinogenic, which has been listed by the US Environment Protection Agency as a

32

priority control pollutant.5,6 Many conventional techniques, such as air stripping, incineration,

33

biological methods and adsorption over activated carbon have extensively been used for the

34

degradation of 2,4-DCP.7 However, these techniques easily give rise to secondary pollutants,

35

which are even much hazardous to the environment.8,9 In addition, the increased CO2

36

concentration in atmosphere from the combustion of fossil fuels, vehicle exhaust and industrial

37

activities is the most serious environmental issue associated with global warming and climate

38

change.10 Thus, the conversion of CO2 into chemical fuels for sustainable energy production is

39

highly desirable. To achieve this goal, traditionally biological and electrocatalytic techniques

40

have widely been explored.11 However, these techniques are always involved with low

41

mechanical strength, less stability, catalyst poisoning, and electrode corrosion.12 Therefore, it is

42

much meaningful to develop alternative techniques to overcome the above mentioned shortfalls.

43

In recent several decades, heterogeneous semiconductor photocatalysis has been regarded as

44

an environmental friendly and promising alternative technique.13 Extensive efforts have been

45

made to develop low cost, highly efficient, and stable semiconductor photocatalysts for

46

degrading 2,4-DCP into inorganic minerals and for CO2 conversion into chemical fuels.5,6,14 In

47

this regard, numerous semiconductor oxides especially TiO2, have been widely investigated.15-20


Page 2 of 36

Page 3 of 36


48

However, the ineffective utilization of solar light (only ca. 4% UV-light) and low quantum

49

efficiency still greatly impair their applications.21 Based on the large proportion of visible-light

50

(ca. 46%) in the solar spectrum, it is highly desired to develop efficient narrow-bandgap

51

photocatalysts to utilize visible light.22

52

In recent years, perovskite-type ABO3 oxides with narrow bandgap have attracted tremendous

53

attention owing to their potential applications in the field of photocatalysis, superconductors,

54

electrocatalysis, chemical sensors, CO2 conversion, etc.23 In ABO3 perovskite-type oxides, the B-

55

site metal cation is six-fold coordinated, while the A-site rare-earth element is twelve-fold

56

coordinated by oxygen anions. The BO6 octahedral units share their vertexes with each other to

57

form crystal structure back bone, while the A-site cations occupy interstitial spaces between the

58

octahedral units.24 The B-site cations and oxygen vacancies are of great importance, since

59

catalytic reactions mainly depend on the redox properties of B-site metal cations, and oxygen

60

vacancies provide adsorption and activation sites for the used substrates.25 Among the well-

61

known ABO3 perovskites, LaFeO3 (LFO) has potential applications in redox reactions, solid

62

oxide fuel cells, electronic-magnetic materials, gas sensors, electrocatalysis and photocatalysis,

63

owing to its exceeding properties, such as cheapness, high stability and nontoxicity.26 LFO has

64

been regarded as a visible-light-driven photocatalyst owing to its narrow bandgap (~ 2.0 eV), but

65

unfortunately, it usually exhibits low photocatalytic activity. This is generally attributed to the

66

low surface area, limited visible-light absorption, and weak charge carrier separation.27-29

67

To increase the surface area of LFO, several attempts have been made with certain successes

68

by decreasing its particle size and introducing porous structure via template-assisted methods.30

69

The created pores are much meaningful to improve the photocatalytic activity by providing

70

active sites and facilitating matter transport.31 It has been demonstrated in some works, including



71

ours, that the porous LFO with large surface area could be synthesized via PS-, PMMA- and

72

SBA-template assisted methods, leading to the improved photocatalytic activities.32-35 However,

73

the obtained surface area is still not satisfactory for efficient photocatalysis. Hence, for practical

74

applications, it is highly desirable to develop new templates to synthesize porous LFO with

75

rather large surface area.

76

The valence band (VB) and conduction band (CB) of LFO respectively stands at 2.2 eV and

77

0.2 eV vs SHE.34 Hence, it is expected that the photogenerated holes should possess enough

78

energy to induce oxidation reactions. Thermodynamically, it is possible to decrease the energy

79

bandgap by shifting the VB top upward. Unfortunately, related works to perovskite oxides have

80

not been reported up to date. Generally, the energy bandgaps of oxides are much large compared

81

to the corresponding nitrides, because of the relatively high VB top levels for nitrides.36,37

82

Moreover, it has been frequently employed to expand the visible-light response of wide-bandgap

83

oxides like TiO2 by doping nonmetal elements such as N, P, and S, to partially substitute crystal

84

O, in which the VB top level could be shifted upward to a certain degree.38-40 Therefore, it is

85

feasible to further expand the visible-light absorption of LFO by doping nonmetal elements, for

86

which it seems to be neglected.

87

Based on the low CB bottom level of LFO, the photogenerated electrons possess low energy to

88

thermodynamically induce reduction reactions. However, it could produce high-level-energy

89

electrons (HLEEs) under visible-light irradiation, for which it is possible to utilize for efficient

90

photocatalysis. Interestingly, it has been clearly demonstrated in our previous works that the

91

visible-light excited HLEEs of narrow-bandgap semiconductors, such as Fe2O3, BiVO4, and

92

BiFeO3, could transfer to the CB of wide-bandgap oxides, like TiO2, by which a proper-level

93

energy/platform is provided to prolong the lifetime of photogenerated electrons with enough


Page 4 of 36

Page 5 of 36


94

energy, leading to the enhanced visible-light photocatalytic activities.22,34,41 Hence, it is possible

95

to further improve the photocatalytic activities of N-doped porous LFO by coupling

96

nanocrystalline TiO2.

97

Generally, the photocatalytic reactions for pollutant degradation and CO2 conversion over

98

semiconductor photocatalysts are mainly involved with a series of photophysical and

99

photochemical processes. The photophysical process is mainly concerned with the

100

photogenerated charge separation, while the photochemical process is related to the produced

101

radicals like •OH.42 Naturally, they are much crucial for efficient photocatalysis. Hence, it is of

102

great significance to deeply explore them for the detailed photocatalytic mechanisms. In addition,

103

some studies have been reported on the degradation of 2,4-DCP over narrow-bandgap oxides,

104

such as PbO2, WO3, and Fe3O4 etc.4,5,43 However, the degradation activity is desired to further

105

improve, along with a clear degradation mechanism. To the best of our knowledge, there has

106

been no previous report on the fabrication of TiO2-coupled N-doped porous perovskite-type LFO

107

photocatalyst with exceptional photoactivities for 2,4-DCP decomposition and CO2 conversion,

108

along with the detailed photocatalytic mechanism until now.

109

Herein, we report the exceptional visible-light photocatalytic activities of TiO2-coupled N-

110

doped porous LFO photocatalysts for 2,4-DCP degradation into inorganic minerals and CO2

111

conversion to chemical fuels. It is clearly demonstrated that the enhanced photoactivities are

112

attributed to the increased specific surface area, to the extended visible-light absorption, and to

113

the promoted charge transfer and separation. Interestingly, we have investigated the degradation

114

mechanism of 2,4-DCP over the newly designed photocatalyst with the help of liquid

115

chromatography tandem mass spectrometry and ion chromatography analysis techniques. It is

116

confirmed that the holes directly attack the aromatic ring of 2,4-DCP and produce 2-



117

chlorosuccinic acid (m/z=153) as a crucial intermediate, which readily transforms into CO2 and

118

H2O. This work will provide feasible routes to improve the visible-light photocatalytic activities

119

of narrow-bandgap perovskite-based nanophotocatalyts for efficient solar energy utilization to

120

purify environment and produce energy. 

121

EXPERIMENTAL SECTION

122

Materials synthesis. All the reagents were of analytical grade and used as received without

123

further purification. Deionized water was used throughout the experiments. Carbon nanospheres

124

were synthesized according to the report.44 In a typical synthesis, 7.92 g of D-glucose was

125

dissolved into 80 mL of deionized water under vigorous stirring. The solution was transferred

126

into a 100 mL Teflon-lined stainless steel autoclave and hydrothermally treated at 160 oC for 10

127

h. After cooling to room temperature naturally, the black product was centrifuged and washed

128

several times with deionized water followed by absolute ethanol and finally dried in oven at 60

129

o

130

nanospheres, with the aid of ammonia solution (25%), added dropwise to the above solution until

131

pH reached 11, as depicted in Scheme S1. LaFeO3 (LFO) nanoparticles were prepared by sol-gel

132

method.32 In a typical experiment, stoichiometric amounts (0.04 mol) of La(NO3).6H2O and

133

Fe(NO3)3.9H2O were dissolved into a mixed solvent containing 25 ml ethylene glycol (EG), 25

134

ml ethanol and 50 ml deionized water at room temperature. After continuous stirring for 4 h, the

135

mixture was dried in convection oven at 80 oC until the whole solvent was evaporated. The dry

136

powder was then calcined in air at 600 oC (5 oC min-1) for 2 h to obtain LFO nanoparticles.

137

Carbon-templated porous LaFeO3 (LFO-C) and amino-functionalized carbon-templated porous

138

LaFeO3 (LFO-AC) nanoparticles were also prepared by the same method, but to the precursor

139

solution 0.2 g of un-functionalized (for LFO-C) and amino-functionalized carbon nanospheres

C for 12 h. Same procedure was followed for the synthesis of amino-functionalized carbon


Page 6 of 36

Page 7 of 36


140

(for LFO-AC) were added under vigorous stirring. To remove carbon template, the dried powder

141

was first calcined in air at 400 oC (temp ramp 1oC min-1) for 2 h and then annealed at 600 oC (5

142

o

143

6 and 8h, nitrogen-doped LFO-AC samples (XN-LFO-AC), in a typical experiment, 1g freshly

144

prepared LFO-AC powder was taken and placed in a quartz horizontal reactor. The reactor was

145

sealed so as to prevent the gas flow to outside. Then, nitrogen gas was passed through a saturator

146

system filled with aqueous ammonia solution. The ammonia vapor was transported into the

147

reactor using N2 as carrier gas and temperature of the system was fixed at 500 oC. To fabricate

148

TiO2 coupled nitrogen-doped LFO-AC nanocomposites (YT/6N-LFO-AC), for each sample, 1 g

149

of the amount optimized 6N-LFO-AC powder was dispersed into a mix solvent (50 % vol/vol) of

150

ethanol and deionized water, and to that 2, 4, 6 and 8% by mass of TiO2 was added under

151

vigorous stirring for 4 h. Subsequently, the mixtures were dried in oven at 80 °C and finally

152

annealed in N2 at 450 °C for 2 h.

153

The detail experimental analysis is provided in the supporting information.

C min-1) for 2 h to obtain the product. To prepare different ammonolysis reaction time i.e. 2, 4,

154 155 156

Results and discussion 

Effects of porous structure

157

The XRD patterns of carbon nanospheres (Figure S1A) exhibit a weak and broad diffraction

158

peak at 2θ of 25o, which is ascribed to (002) plane of graphiticcarbon.45 From SEM micrograph

159

(Figure S1B), it is obvious that the as-prepared carbon nanospheres have ordered morphology

160

with average particle size of ~ 200 nm. The resultant non-porous LaFeO3 (LFO), carbon-

161

templated LaFeO3 (LFO-C), and amino-functionalized carbon-templated LaFeO3 (LFO-AC)



162

exhibit high crystallinity based on the XRD patterns (FigureS2A), and the diffraction peaks are

163

well-indexed to the orthorhombic phase of perovskite-type LaFeO3 (JCPDS card no. 37-1493).46

164

According to the literature,47 the orthorhombic structure is much favorable for photogenerated

165

charge transfer and separation. The UV-vis diffuse reflectance spectra of LFO, LFO-C and LFO-

166

AC samples (Figure S2B) show absorption edges at about 620 nm, which is attributed to its

167

electronic transition from the valance band to conduction band (O2p→Fe3d), and the optical

168

absorption across the energy band gap is about 2.0 eV, based on the widely accepted energy

169

bandgap equation: Eg = 1240/λ.48 The TEM micrograph of LFO (Figure S3A) shows an average

170

particle size of 100 nm, and the HRTEM image (Figure S3B) reveals that the lattice fringes at

171

(121) plane with d-spacing 0.28 nm correspond to the orthorhombic phase of LFO. The TEM

172

image of LFO-AC (Figure S3C) illustrates that the sample exhibits porous structure with average

173

particle size of 100 nm, and the HRTEM micrograph (Figure S3D) reveals that the pores exhibit

174

average diameter of ~ 15 nm.

175

From N2 adsorption-desorption isotherm curves (Figure 1A), it is clear that LFO exhibits non-

176

porous structure, while LFO-C and LFO-AC samples exhibit hysteresis loops, which are the

177

characteristics of porous structure.49 It is prominent that the resulting LFO exhibits small BET

178

surface area (4.5 m2 g-1), while LFO-C (17.1 m2.g-1) and LFO-AC (30.4 m2.g-1) show large

179

surface areas. The pore size distribution curves of LFO-C and LFO-AC samples (Figure 1B)

180

reveal that the pores exhibit average diameter of ~ 15nm. This is in good agreement with the

181

TEM results.

182 183


Page 8 of 36

Page 9 of 36


184

185 186

Figure 1 N2 adsorption-desorption isotherm curves (A), pore diameter dispersion (B),

187

photocatalytic degradation rates for 2,4-DCP (C), and fluorescence spectra related to •OH

188

amounts (D) of non-porous LaFeO3 (LFO), carbon-templated LaFeO3(LFO-C), and amino-

189

functionalized carbon-templated LaFeO3 (LFO-AC) samples.

190

Figure 1C shows the visible-light catalytic activity of LFO, LFO-C and LFO-AC samples for

191

2,4-DCP degradation. It is demonstrated that LFO-C and LFO-AC samples exhibit enhanced



Page 10 of 36

192

photoactivities as compared to the traditional LFO. Hence, it is suggested that the porous

193

structure serves large surface area and active sites for catalytic reactions. To verify this, the

194

coumarin fluorescent method was used to detect the amount of produced hydroxyl radicals (•OH),

195

in which the coumarin could easily react with the formed •OH and produce luminescent 7-

196

hydroxy-coumarin. As demonstrated earlier,50 the •OH amount could effectively reveal the

197

separation of photogenerated charges in the photocatalysis. As expected, it is confirmed based on

198

Figure 1D that the •OH amounts produced by LFO-C and LFO-AC are much obvious as

199

compared to the traditional LFO. Hence, it is deduced that the increase in the surface area is

200

directly related with the enhanced photoactivities.

201



Effects of doping N

202

The XRD patterns of nitrogen doped XN-LFO-AC samples are shown in Figure S4A. From

203

the enlarged views of the intense peaks (Figure 2A inset), it is clear that the diffraction peaks at

204

2θ=32.2o are slightly shifted toward higher 2θ values compared to that of LFO-AC, suggesting

205

that the nitrogen atoms are successfully incorporated into the crystal lattice and partially replaced

206

oxygen ones. This is in good agreement with the previous reports about N-doped BaTiO3 and

207

La2Ti2O7 perovskites.51,52 Moreover, the diffraction patterns of XN-LFO-AC samples reveal that

208

the substitution of oxygen by nitrogen does not change the crystal phase and crystallinity of

209

LFO-AC. According to the previous reports,53,54 the ionic radius of O2- (1.4 Ǻ) is almost the

210

same as ionic radius of N3- (1.5 Ǻ), which can effectively facilitate the anion substitution inside

211

the crystal structure of oxides. It is well known that the nitride ion (N3-) has a higher charge than

212

oxide ion (O2-). When N3- is incorporated in the crystal lattice of oxides, the expansion of

213

electronic cloud increases as a result the inter-electronic repulsion decreases. This higher


Page 11 of 36


214

electrical charge of N3- leads to a larger crystal field splitting that result in the formation of new

215

compounds with transition metals in higher oxidation states.

216

Further, it is suggested that the electronegativity of O2- (3.5) is slightly higher than that of N3-

217

(3.07),55 which leads to the difference in the covalence of cation-anion bonds. The strong

218

covalent metal-nitrogen bonding reduces the band gap of oxides and extends its visible-light

219

absorption range. In order to investigate the optical absorption behavior of the resultant LFO and

220

XN-LFO-AC samples, UV-visible diffuse reflectance spectra (UV-Vis DRS) were measured as

221

shown in Figure 2A. It is clear that LFO-AC exhibits the absorption feature of a typical

222

semiconductor originated by the electron transitions from its VB occupied by O2p orbitals to its

223

CB formed by La3d orbitals. Compared to LFO-AC sample, the reflection band edges of XN-

224

LFO-AC are slightly shifted toward longer wavelength direction with the increase in N content.

225

This is attributed to the formed surface states related to N near the VB top of LFO-AC. Since the

226

potential energy of N 2p orbital is higher than that of the O 2p, hence, the band gap of XN-LFO-

227

AC is effectively reduced and exhibit greater absorbance as compared to LFO-AC sample. From

228

UV-vis DRS spectra, the calculated band gap of 6N-LFO-AC is ~1.82 eV. This is further

229

confirmed by the valance band XPS spectra (Figure 2B) since the valence position is shifted

230

upward from 2.2 to 2.02 eV. This result is in good agreement with the N dopedTiO2.56 TEM

231

image of 6N-LFO-AC sample (Figure S4B) reveals that nitrogen doping does not affect the

232

morphology of LFO-AC sample. HRTEM image (Figure S4C) shows an average pore size of 15

233

nm. The EDX spectrum of 6N-LFO-AC (Figure S4D), along with the atomic percentage

234

composition, shows that a certain amount of N is doped into LFO.

235

To further confirm the surface functional groups and elemental states of the LFO-AC and XN-

236

LFO-AC samples, XPS measurement was carried out as shown in Figure S5. The typical survey



Page 12 of 36

237

spectra of LFO-AC and XN-LFO-AC samples (Figure S5A) reveal the presence of La3d, Fe2p,

238

O1s, N2p and C1s. The high-resolution spectrum of La3d is shown in Figure S5B. One can see

239

that two pairs of peaks located at 833.6 and 850.5 eV are observed and attributed to the La

240

3d5/2and La 3d3/2 energy levels, respectively, and each peak presents a satellite at 4.0 eV higher,

241

and the spin-orbit splitting of La 3d5/2 and La 3d3/2levels is 16.9 eV, indicating that there are La3+

242

ions in the oxide form. This is consistent with the previous report.57

243

The binding energies of Fe2p (Figure S5C), at 710 and 723.6 eV respectively, correspond to

244

the +3 oxidation state of Fe. The binding energies of O1s (Figure S5D) at 529.5 eV and 531.5 eV

245

respectively, are attributed to the lattice oxygen (OL) and hydroxyl oxygen (OH).58 The XPS

246

peaks of N1s (Figure S5E) are broad and located at binding energy of ~ 399 eV. It is obvious that,

247

as the ammonolysis reaction time prolongs, the intensity of N1s peak also increases. Our results

248

are in good agreement with the previous reports about N-doped TiO2.59,60 Therefore, it is

249

confirmed that nitrogen is successfully incorporated in the crystal lattice of LFO-AC by

250

substituting oxygen.

251

The N2 adsorption/desorption isotherm curves of XN-LFO-AC samples (Figure S6)

252

demonstrates that the BET surface of LFO-AC is slightly increased after doping nitrogen. Figure

253

2C demonstrates the visible-light photocatalytic activities of XN-LFO-AC samples for 2,4-DCP

254

degradation. It is clear that the LFO-AC exhibits weak photocatalytic activity. However, it is

255

remarkably enhanced after doping nitrogen, and the highest activity is observed for 6N-LFO-AC.

256

The stability and recyclability test of 6N-LFO-AC sample (Figure S7) for 2,4-DCP degradation

257

after five successive recycles reveal that the photcatalyst does not show any obvious loss in the

258

photoactivity. Hence, it is confirmed that the 6N-LFO-AC photocatalyst is highly stable during

259

the photocatalytic oxidation of 2,4-dichlorophenol.


Page 13 of 36


260

261 262

Figure 2 UV-vis DRS spectra with an inset of enlarged views of XRD patterns (A) of LFO-AC

263

and XN-LFO-AC. Valance band XPS spectra (B) of LFO-AC and 6N-LFO-AC. Photocatalytic

264

degradation rates for 2,4-DCP (C), and fluorescence spectra related to •OH (D) of LFO-AC and

265

XN-LFO-AC.

266 267

The photogenerated charge carrier transfer and separation in semiconducting solid materials

268

can be directly analyzed through a highly sensitive and non-destructive atmosphere-controlled



Page 14 of 36

269

surface photovoltage spectroscopy (SPV) technique.61,62 To investigate the effects of dopant-

270

introduced surface states on the photogenerated charge properties of XN-LFO-AC samples, SPV

271

measurements were performed. From Figure S8A, it is clear that LFO-AC exhibits a weak SPV

272

signal in N2 atmosphere, and it is remarkably enhanced after doping N and the highest signal is

273

observed for 6N-LFO-AC. This explains the role of N-introduced surface states for trapping

274

photogenerated holes and promoting charge separation.63 In order to clarify this, the SPV

275

responses of LFO-AC and 6N-LFO-AC samples were measured in different atmospheres. In

276

general, for LFO-AC, the presence of oxygen is essential for SPV response to occur, since the

277

photoinduced electrons will be captured by the surface adsorbed O2, and the respected holes will

278

preferentially diffuse to the testing electrode surface and give rise to SPV signal.64 As expected,

279

LFO-AC exhibits a strong SPV signal in O2 atmosphere (Figure S8B). In contrast, 6N-LFO-AC

280

displays a much stronger SPV signal in N2 atmosphere (Figure S8C). This is because that the

281

trapping effects of N-introduced surface states for photogenerated holes are much obvious than

282

those of the adsorbed O2 for the related electrons. Hence, it is believable that the 6N-LFO-

283

ACexhibits a weak SPV signal in air since the photoinduced electrons and holes would

284

respectively be trapped by the adsorbed O2 and the N-introduced surface states.63 Based on the

285

SPV results, it is concluded that the N-introduced surface states could effectively trap the

286

photogenerated holes, and much favorable for the photogenerated charge separation.

287

To further reveal the enhanced charge separation and photocatalytic activities of XN-LFO-AC

288

samples, the fluorescent spectra related to the •OH amounts were measured, as shown in Figure

289

2D. It is clear that the amount of •OH produced by LFO-AC is greatly enhanced after doping

290

nitrogen. Interestingly, 6N-LFO-AC exhibits the largest amount of •OH. This is further

291

confirmed by the photoelectrochemical I-V curves, as shown in Figure S9A. It is well


Page 15 of 36


292

demonstrated that LFO-AC exhibits a weak photocurrent density response under visible-light

293

irradiation. However, after doping N, it is greatly enhanced, and the largest photocurrent

294

response is observed for 6N-LFO-AC. This is similar to the transient photocurrent response

295

(Figure S9B). Moreover, the transient photocurrent results further reveal that the resulting XN-

296

LFO-AC samples possess good photostability. The increased charge separation is also confirmed

297

by the electrochemical impedance spectra (EIS) measured under visible-light irradiation, as

298

shown in Figure S9C. According to the previous report,65 the smaller arc radii always exhibits

299

high charge separation. It is clear that the capacitive arc radii of XN-LFO-AC samples are

300

remarkably decreased, as compared to that of LFO-AC. Interestingly, the 6N-LFO-AC exhibits

301

the smallest arc radii, indicating that the charge recombination is significantly reduced.

302

Obviously, these PEC results are in good agreement with the SPV results. Based on the above

303

results, it is deduced that the N-introduced surface states play a vital role in the enhanced charge

304

separation, which is responsible for the improved visible-light activities of XN-LFO-AC samples

305

for 2,4-DCP degradation.

306



Effects of couplingTiO2

307

The XRD patterns of YT/6N-LFO-AC samples (Y represents different mass ratio %age of

308

TiO2 coupled) are shown in Figure S10A. The relative peaks at θ=25in YT/6N-LFO-AC

309

nanocomposites are attributed to the coupled TiO2.34 The peak intensities gradually enhanced

310

with the increase in mass ratio percentage of TiO2 coupled. The introduction of TiO2 does not

311

affect the crystal phase and crystallinity of 6N-LFO-AC. The UV-vis DRS spectra of YT/6N-

312

LFO-AC samples (Figure S10B) show that the introduction of TiO2 does not change the bandgap

313

of 6N-LFO-AC. The TEM image of 6T/6N-LFO-AC (Figure S10C) shows that small size (~10

314

nm) TiO2 nanoparticles are well dispersed on the surfaces of 6N-LFO-AC. The HRTEM image



Page 16 of 36

315

(Figure S10D) reveals that the heterojunction exists between TiO2 and 6N-LFO-AC, since the

316

lattice fringes at (121) plane with d-spacing 0.28 nm corresponds to LFO, while the latter at (101)

317

plane with d-spacing 0.35 are attributed to TiO2.34 Further, the nitrogen adsorption–desorption

318

isotherm curves (Figure S11) reveal that the surface area of 6T/6N-LFO-AC is enhanced (36.6

319

m2 g-1), as compared to 6N-LFO-AC. This is attributed to the small size TiO2 coupled.

320

Figure 3A displays the visible-light photocatalytic activities of YT/6N-LFO-AC samples for

321

2,4-DCP degradation. It is clear that the visible-light activities are greatly improved after

322

coupling TiO2. It is prominent that, as the amount of coupled TiO2 increases, the photocatalytic

323

activity also enhances. Interestingly, the 6T/6N-LFO-AC sample exhibits the highest

324

photoactivity. Further, if the amount of TiO2 exceeds a certain limit, the photocatalytic activity

325

begins to decrease. This is because that the coupled TiO2 covers the photocatalystsurface.66,67 The

326

stability and recyclability test of 6T/6N-LFO-AC for 2,4-DCP degradation (Figure S12) reveal

327

that the photcatalyst is highly stable.

328

The PEC I-V curves of YT/6N-LFO-AC samples (Figure 3B) demonstrate that the

329

photocurrent density response of 6N-LFO-AC sample is obviously enhanced after coupling TiO2,

330

and the largest photocurrent response is observed for 6T/6N-LFO-AC. From the fluorescence

331

spectra related to the produced •OH (Figure 3C), it is obvious that the amount of •OH produced

332

by 6N-LFO-AC sample is greatly enhanced after coupling TiO2, and the largest amount is

333

observed for 6T/6N-LFO-AC sample. This is consistent with the PEC results. To confirm this,

334

the SPV responses of YT/6N-LFO-AC samples were measured in N2 atmosphere, as shown in

335

Figure 3D. It is clear that the SPV response of 6N-LFO-AC sample is obviously improved after

336

coupling TiO2, and the significant response is observed for 6T/6N-LFO-AC nanocomposite.


Page 17 of 36


337

It is clearly shown that a proper amount of TiO2 coupled is beneficial for the charge transfer

338

and separation. However, an excess amount of TiO2 is not useful. Based on the SPV responses of

339

6T/6N-LFO-AC measured in different environments (Figure 3D inset), it is further confirmed

340

that the N-introduced surface states could capture the photogenerated holes. Therefore, it is

341

deduced that the photogenerated charge separation of LFO-AC could be greatly enhanced after

342

doping N, and then coupling TiO2, along with its visible-light expansion. This is well responsible

343

for the obviously-improved visible-light photocatalytic activities after coupling TiO2.

344 345

Figure 3 Photocatalytic degradation rates for 2,4-DCP (A), PEC I-V curves (B) and fluorescence

346

spectra related to the •OH amount (C) of 6N-LFO-AC and YT/6N-LFO-AC samples. SPV

347

responses in N2 (D) of YT/6N-LFO-AC samples with inset SPV responses of 6T/6N-LFO-AC in



348

Page 18 of 36

different environment.

349 350



Photoactivity enhancement mechanism

351

To further investigate the enhanced photoactivities, the visible-light photocatalytic activities of

352

LFO, LFO-AC, 6N-LFO-AC and 6T/6N-LFO-AC samples were evaluated for CO2 reduction in

353

water without any co-catalyst. From Figure 4A, it is seen that LFO exhibits a weak

354

photocatalytic activity for CO2 conversion to CH4 (~ 20 μmol) and CO (~ 35 μmol), and for

355

producing O2 (~ 60 μmol) after 8 h irradiation, while LFO-AC displays enhanced activity by

356

producing CH4 (~ 40 μmol), CO (~ 65 μmol) and O2 (~ 115 μmol). As expected, the 6N-LFO-

357

AC and 6T/6N-LFO-AC samples exhibit much high photoactivities, especially the 6T/6N-LFO-

358

AC with produced CH4 (~ 110 μmol), CO (~ 150 μmol) and O2 (~ 230 μmol). Interestingly, the

359

photoactivity for CO2 reduction in our experiments is much higher than those in other works.68,69

360

The enhanced photocatalytic activity results are also supported by the RhB degradation rates

361

under visible-light irradiation for 1.5h (Figure 4B).

362

To well understand the charge transfer and separation processes in 6T/6N-LFO-AC, a

363

mechanism schematic is designed as depicted in Figure 5, along with the main photocatalytic

364

reactions. According to our previous report,34 the band gap of anatase TiO2 is 3.2 eV. It’s VB and

365

CB tops respectively, lies at 2.9 eV and -0.3 eV. While for LFO-AC (Eg = 2.0 eV), the VB and

366

CB respectively, stands at 2.2 eV and 0.2 eV, which can be excited by the photon energy

367

(λ≤620nm). The energy difference between the VB top of LFO and the CB top of TiO2 is 2.5 eV,

368

corresponding to the light energy of approximately λ≤496 nm based on the equation: λ = 1240/Eg,

369

where Eg stand for the energy bandgap. As for 6N-LFO-AC, the calculated band gap is 1.82 eV,

370

which can be excited by photon energy (λ≤680nm).


Page 19 of 36


371

Figure 4 Product amounts evolved during the photocatalytic CO2 conversion (A), and

372

photocatalytic degradation rates for RhB (B) over LFO, LFO-AC, 6N-LFO-AC and 6T/6N-LFO-

373

AC samples under visible-light irradiation.

374 375

For 6T/6N-LFO-AC nanocomposite, the energy difference between the VB of 6N-LFO-AC

376

and CB of TiO2 is 2.32 eV, related to the light energy of approximately λ≤534nm. Thus, it is

377

acceptable that, when 6T/6N-LFO-AC nanocomposite is irradiated under visible-light

378

(400nm

Exceptional Visible-Light Activities of TiO2-Coupled N-Doped Porous

Recommend Documents