Oxygen Vacancies Mediated Complete Visible ... - ACS Publications

Subscriber access provided by Washington University | Libraries

Remediation and Control Technologies

Oxygen Vacancies Mediated Complete Visible Light NO Oxidation via Side-On Bridging Superoxide Radicals Hao Li, Huan Shang, Xuemei Cao, Zhiping Yang, Zhihui Ai, and Lizhi Zhang Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b01849 • Publication Date (Web): 02 Jul 2018 Downloaded from http://pubs.acs.org on July 2, 2018

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 21

Environmental Science & Technology

1

Oxygen Vacancies Mediated Complete Visible Light NO

2

Oxidation via Side-On Bridging Superoxide Radicals

3

Hao Li,‡ Huan Shang,‡ Xuemei Cao, Zhiping Yang, Zhihui Ai* and Lizhi Zhang*

4

Key Laboratory of Pesticide & Chemical Biology of Ministry of Education, Institute of Applied &

5

Environmental Chemistry, College of Chemistry, Central China Normal University, Wuhan 430079,

6

P. R. China

7

* To whom correspondence should be addressed. E-mail: [email protected];

8

[email protected].

9

Phone/Fax: +86-27-6786 7535

10

11

12 13 14 15 16 17 18 19 20 21 22 23

ACS Paragon Plus Environment

Environmental Science & Technology 24

ABSTRACT It is of a great challenge to seek for semiconductor photocatalysts with prominent

25

reactivity to remove kinetically-inert dilute NO without NO2 emission. In this study, complete

26

visible light NO oxidation mediated by O2 is achieved over a defect-engineered BiOCl with

27

selectivity exceeding 99%. Well-designed oxygen vacancies on the prototypical (001) surface of

28

BiOCl favored the possible formation of geometric-favorable superoxide radicals (•O2-) in a side-on

29

bridging mode under ambient condition, which thermodynamically suppressed the terminal end-on

30

•O2- associated NO2 emission in case of higher temperatures, and thus selectively oxidized NO to

31

nitrate. These findings can help us to understand the intriguing surface chemistry of photocatalytic

32

NO oxidation and design highly efficient NOx removal systems.

33 34

Keywords: NO oxidation; Photocatalysis; Oxygen vacancy; Superoxide radical; Selectivity

35 36

Introduction

37

Anthropogenically-derived nitric oxide (NO), even in an infinitesimal concentration (sub-ppm or

38

ppb levels), is still regarded as the major contributor to acid rain, photochemical smog, ozone

39

depletion, and bears a principal responsibility for respiratory and cardiopulmonary diseases in highly

40

populated cities.1 Conventionally, dilute NO can be pre-concentrated by absorbents like active

41

carbon and zeolites for subsequent selective catalytic reduction (SCR) removal, while the heavy use

42

of noble metals (Pt, Ru, Pd) and reducing reagents under high operating temperatures (>500 K)

43

largely restrict its practical application.2 Driven by the stringent environment regulations,

44

semiconductors-based photocatalysis has received considerable attention in view of its potentiality

45

for efficient dilute NO removal.3,4 Unfortunately, photocatalytic NO removal still suffers from the

46

toxic intermediates generation and nitrification products handling, as NO2 with higher toxicity may

47

be generated. Although selective NO reduction kinetics with ammonia (2NH3 + NO + NO2 → 2N2 + ACS Paragon Plus Environment

Page 2 of 21

Page 3 of 21


48

3H2O) will be maximized to reduce the emission of NO2 in SCR system by unifying the ratio of

49

NO2/NO in a relatively closed chamber,2,5 NO2 is an extremely undesirable intermediate or product

50

in open photocatalytic system. Therefore, the complete NO oxidation to nitrate during photocatalytic

51

NO removal remains a great challenge, which will avoid risky NO2 emission and also provide

52

potential metabolic nitrogen for micro-organisms.

53

Unavoidable NO2 generation in contemporary photocatalytic NO removal systems is normally

54

attributed to the presence of one-oxygen reactive species such as holes (O-) and hydroxyl radicals

55

(•OH), which detrimentally lead to the partial NO oxidation (NO(g) + O-/•OH → NO2(g)). In the light

56

of this fact, tuning the formation of oxygen reactive species in the molecular form to dominantly

57

govern complete NO oxidation, particularly superoxide radicals (O2 + e- → •O2-; NO(g) + •O2- →

58

NO3-), has been proposed previously.3,6 Unfortunately, such a •O2--based solution still remains

59

debatable because some scientists found that •O2- could also cause the partial NO oxidation (NO(g)

60

→ NO2(g)/NO2(aq)-).7-9 The controversy over this issue intrinsically stems from the poor

61

understanding of mechanistic function schemes of •O2- for NO oxidation on the surface molecular

62

level, especially the influences of specific geometric structures of •O2- on the reaction selectivity

63

have never been taken into consideration.

64

In this study, we demonstrate that complete NO oxidation to nitrate with selectivity exceeding

65

99% can be achieved by the possible generation of side-on bridging •O2-, using a defect-engineered

66

BiOCl as the model photocatalyst. BiOCl, as a characteristic UV-light-responsive oxide material,

67

has received growing attention due to its inherent oxygen vacancies (OVs) associated

68

photocatalysis.10,11 Well-designed catalytic OVs on its prototypical (001) surface can extend the

69

photoresponse region of BiOCl to visible light, and also allow the selective and efficient activation

70

of O2 to •O2- in different geometric structures possible for atomistic level investigation and practical

71

application. ACS Paragon Plus Environment


Page 4 of 21

72 73

Experimental Section

74

Chemicals and Materials. Synthetic materials for BiOCl and other related chemicals were bought

75

from Sinopharm Chemical Reagent Co., Ltd.

76

Preparation of BiOCl. Hierarchical BiOCl nanospheres were prepared according to the method

77

reported by our group.12 First, we slowly added 3 mmol Bi(NO3)3·5H2O into 16 mL ethylene glycol

78

solution, which contained 3 mmol KCl. Then, the mixture was kept stirring until Bi(NO3)3·5H2O was

79

totally dissolved. Finally, we poured the above mixture into a Teflon-lined stainless autoclave (20

80

mL). After 12 h of reaction at 160 °C, the precipitates were centrifuged and the surface residual ions

81

were totally washed with deionized water and ethanol. The as-obtained powder, which was denoted

82

as BiOCl-OV, was dried at 60 °C in air for further use. In order to prepare BiOCl-OV with lower

83

concentration of OVs, we calcined BiOCl-OV in the air at 300 oC for different time, e. g. 30 min, 60

84

min, 90 min, 120 min, and 180 min to gradually reoxidize the OVs. When the calcination time was

85

lengthened to 3 h, we were able to obtain a nearly defect-free BiOCl counterpart sample. To prepare

86

freestanding BiOCl nanosheets with (001) surface exposed, we added 1 mmol of Bi(NO3)3·5H2O

87

into distilled water (16 mL), which contained 1 mmol KCl. pH value was then adjusted to 1 with

88

using 1 M NaOH under continuous stirring. We then poured the above mixture into a Teflon-lined

89

stainless autoclave (20 mL).13 After 24 h of reaction under 220 °C, resulting precipitates were

90

centrifuged and the surface residual ions were totally washed with deionized water and ethanol.

91

Final BiOCl powder was dried at 60 °C for further use. In order to introduce OVs on the

92

freestanding BiOCl nanosheets, 0.1 g of BiOCl was uniformly dispersed in a crucible and annealed

93

at 300 °C for 120 min under high vacuum with the heating rate being controlled at 5 oC/min.

94

First-principles density of functional theory (DFT) calculation. Theoretical calculations were

95

performed

via

DFT

+

U

calculations

(generalized


gradient

approximation

with

Page 5 of 21


96

Perdew-Burke-Ernzerhof (PBE) exchange-correlation function).14 Thickness of BiOCl(001) surface

97

was tested using a (1 × 1) surface, and was implemented by the CASTEP code. In this part of

98

calculation, we used the ultrasoft pseudopotentials and plane-wave pseudopotential approach for all

99

the atoms.15,16 Kinetic energy cutoff was set as 380 eV. During the optimization, the force and energy

100

respectively converged to 0.03 eV/Å and 10-5 eV/atom.17 For subsequent NO or O2 adsorption

101

calculation, a (2 × 2) supercell with a 20 Å vacuum was further used. This part of calculation was

102

enabled by a VASP code with a kinetic energy cutoff of 520 eV.18,19 Location of the transition states

103

(TS) was determined by nudged elastic band (NEB) method.20,21 We calculated the adsorption energy

104

was calculated as via the following equation: ∆E = E(BiOCl + adsorbate) - E(BiOCl) - E(adsorbate).

105

The E(adsorbate) and E(BiOCl) are respectively the energies of the two isolated (noninteracting)

106

subsystems, while E(BiOCl + adsorbate) is total energy of the interacting adsorbate-surface system.

107

We depicted charge density difference via: ∆ρ = ρ(BiOCl + adsorbate) - ρ(adsorbate) - ρ(BiOCl).

108

ρ(adsorbate) and ρ(BiOCl) are respectively the densities of the two noninteracting subsystems, while

109

ρ(BiOCl + adsorbate) is the density of the interacting adsorbate-surface system.

110

Materials characterization. Phase structure of the as-prepared BiOCl was determined by the

111

powder X-ray diffraction (XRD) measurements (Rigaku D/MAX-RB diffractometer, Cu Ka

112

radiation, Λ= 0.15418 nm). Surface morphology was recorded via a field-emission scanning electron

113

microscope (SEM, JEOL 6700-F) and transmission electron microscopy (TEM, JEOL JSM-2010).

114

Light absorption capacity of BiOCl was evaluated via a UV-visible spectrophotometer (UV-2550,

115

Shimadzu, Japan). To observe in situ catalytic NO oxidation, surface functional group change was

116

recorded with a diffuse reflectance FTIR spectra (Nicolet iS50FT-IR). Concentration of

117

photoinduced •O2- and OVs were quantitatively determined via electron paramagnetic resonance

118

(EPR, Billerica, MA) using Cu2+ as the calibrator. NO and O2 adsorption on BiOCl were evaluated

119

via temperature-programmed desorption experiments (TPD) in a quartz reactor using a TCD as ACS Paragon Plus Environment


detector. NO2-TPD was analyzed with a flow reactor connected to a mass spectrometer (UTI 100C).

121

Nitrate was measured by Thermo Scientific Dionex ionic chromatography (ICS-900). Valence state

122

of Bi was analyzed with X-ray photoelectron spectroscopy (XPS, Perkin-Elmer PHI 5000C).

123

Photocatalytic activity test. We prepared BiOCl or TiO2 films by carefully coating the aqueous

124

suspension of the corresponding photocatalysts onto a glass dish (diameter = 12 cm). A continuous

125

flow reactor was adopted to simulate practical NO removal at ppb levels under ambient conditions.

126

Volume of the rectangular reactor (30 cm × 15 cm × 10 cm [L × W × H]) was around 4.5 L. For

127

photocatalytic NO removal over BiOCl, a Xenon lamp (λ > 400 nm) or UV light (λ ~ 360 nm) was

128

used. A 480 nm monochromatic light (20 mW/cm2) was used as the light source for photocatalytic

129

NO removal over TiO2-P25. NO gas was provided from a compressed gas cylinder with a

130

concentration around 50 ppm balanced by Ar, which was further diluted to 500 ppb by an air stream.

131

A gas blender was used to mix the gas streams, simultaneously controlling the flow rate at 1 L/min.

132

To ensure the sufficient absorption of light by the as-prepared films, the lamp was placed vertically

133

above the films, which sat in the middle of the reactor. After adsorption-desorption equilibrium were

134

reached, the light source was turned on to trigger the photocatalytic reactions. The concentration of

135

NO and NO2 were monitored by a chemiluminescence NOx (the sum of NO and NO2) analyzer

136

(Teledyne, NOx analyzer, model T200).

137 138

Results and Discussion

139

DFT calculation revealed fully oxidized BiOCl(001) surface was quite inert towards NO or O2

140

adsorption, while the introduction of an OV with localized electrons remarkably enhanced its surface

141

reactivity. According to the adsorption energies and activated bond lengths, O2 in either terminal

142

end-on or side-on bridging mode interacted more strongly with the OV as compared with NO, which

143

is reasonable as O2 of lower 2π* orbital energy is a better electron-acceptor than NO (Figure S1). So, ACS Paragon Plus Environment

Page 6 of 21

Page 7 of 21


144

aerobic catalytic NO removal, especially in case of extremely low concentration of NO, is supposed

145

to take place via an Eley-Rideal mechanism, through which NO will attack the •O2- preferentially

146

adsorbed on OVs.22 A broader DFT study was then performed to explore the possible NO oxidation

147

pathways and intermediate species initiated by surface •O2- in different geometries. When NO

148

approached the terminal end-on •O2-, it was oxidized to peroxynitrite (OONO-), simultaneously

149

releasing 1.32 eV energy (pathway I). Peroxynitrite is known to be a powerful and toxic oxidant that

150

causes the direct biotoxicity of NO when NO acts as an intercellular messenger in vivo, and also a

151

transient precursor towards the NO2 formation on metal or metal oxide surfaces.23,24 As expected,

152

breaking of the O-O bond in peroxynitrite led to the release of gaseous NO2 with a barrier of 0.49 eV

153

(Figure 1a).

154 155

Figure 1. (a) Free energy change against the reaction coordinate for the oxidation of NO by •O2- on

156

BiOCl(001) surface in different geometries. (b) Geometric transition from peroxynitrite to nitrate.

157

TS represents transition state. Charge density difference and O2 partial DOS of the BiOCl(001) ACS Paragon Plus Environment


surface adsorbed with (c) O2 and (d) nitrate. The yellow and blue isosurfaces with an isovalue of

159

0.005 a.u. represent charge accumulation and depletion in the space. The vertical dashed line in the

160

DOS shows the VBM.

161

Differently, side-on bridging •O2- could directly oxidize NO into monodentate nitrate without

162

any barrier (pathway II, ∆E = -3.10 eV), which was far more thermodynamically accessible than

163

pathway I. Abstracting a proton from a neighboring hydroxyl or adsorbed water by nitrate towards

164

the formation of HNO3* was also favorable with an energy release around 1.64 eV. Moreover, energy

165

expenditure towards the diffusion of nitrate or HNO3 on the surface was only 0.20 eV, much smaller

166

than that (0.49 eV) of NO2 desorption (Figure 1a). Since geometric transformation from

167

peroxynitrite to its structural isomer of nitrate was up against a high barrier of 1.03 eV, NO oxidation

168

via pathway I or II was rather independent (Figure 1b). Such a theoretical scenario can explain the

169

controversy over the mechanistic roles of •O2- on the NO oxidation. Obviously, the side-on bridging

170

•O2- (•O2-B) mediated complete NO oxidation (pathway II: •O2-B + NO(g) → NO3-) would be prior to

171

the terminal end-on one (•O2-T) mediated partial NO oxidation (pathway I: •O2-T + NO(g) → OONO-

172

→ NO2) according to the reaction thermodynamics.

173

To uncover the remarkably promoted thermodynamics of NO oxidation by side-on bridging •O2-,

174

charge density difference calculation was adsopted to trace the interfacial charge transfer. After O2

175

was adsorbed on the OV of BiOCl(001) surface, instant charge back donation from the OV to O2

176

occurred, which was depicted by the localized electrons depletion on two Bi atoms around the OV

177

and electrons accumulation on the coordinating O2 (Figure 1c). Partial density of states (DOS) of O2

178

contained two occupied spin-up majority states and one occupied spin-down minority state below

179

the Fermi energy, as well as an empty spin-down minority states near the conduction band, revealing

180

the filling of one O2 2π* orbital toward the •O2- formation (Figure 1c).25 Along with the nitrate

181

formation, •O2- was then found to interact rigorously with NO according to a significant charge ACS Paragon Plus Environment

Page 8 of 21

Page 9 of 21


182

depletion of N atom and an extraordinary electron gaining by three O atoms, manifesting an

183

outstanding oxidative ability (Figure 1d). Meanwhile, 2π orbitals of •O2- underwent a considerable

184

broadening and were shifted below the valence band maximum (VBM) of BiOCl, indicating the

185

complete NO oxidation by side-on bridging •O2- (Figure 1d).22,26,27

186 187

Figure 2. (a) Backscattered scanning electron microscopy image of BiOCl nanospheres and (b) their

188

Barrett-Joyner-Halenda pore-size distribution plot. (c) NO-TPD and (d) O2-TPD profiles of the

189

BiOCl photocatalysts. (e) EPR spectra of BiOCl-OV under different conditions.

190 191

Motivated by the theoretical calculation results, we then synthesized specific hierarchical BiOCl

192

nanospheres assembled by (001) surface exposed nanosheets containing OVs (BiOCl-OV) (Figure

193

2a and S2). The presence and the relative concentration of OVs were estimated by XPS (Figure S3).

194

The hierarchical nanostructured morphology with small mesopores (2~10 nm), which are beneficial ACS Paragon Plus Environment


for adsorbates/products diffusion and transportation in many energy and environmental applications,

196

offers the possibility of nitrate diffusion and storage for long term NO removal (Figure 2b and

197

S4).28,29 To experimentally verify whether BiOCl-OV could achieve complete NO oxidation

198

mediated by side-on bridging •O2- on (001) surface, a series of techniques were performed to in situ

199

characterize the surface chemistry under reaction conditions. Temperature-programmed desorption

200

(TPD) was first used to detect NO and O2 adsorption on BiOCl-OV. NO-TPD spectra of BiOCl-OV

201

was dominated by two desorption at 455 K and 502 K (Figure 2c). Since hydroxyl groups and lattice

202

O atoms on clean BiOCl(001) surface were inert for direct oxidation of NO to nitrite or nitrate, these

203

two peaks were therefore ascribed to the chemical adsorption of NO on OV with *O or *N-end. As

204

for O2-TPD, we observed a wide molecular O2 desorption peak from 450 K to 700 K over

205

BiOCl-OV that surface side-on bridging •O2- were supposed to fall into this region (Figure 2d).

206

Judging from the temperature and area of the desorption peaks, O2 was supposed to be adsorbed

207

preferentially on the OVs as compared with NO, consistent with our DFT calculations. Spin-reactive

208

•O2- were then monitored by room-temperature (298 K) electron paramagnetic resonance (EPR). As

209

compared with BiOCl, BiOCl-OV possessed a typical signal at g = 2.001, corresponding to the

210

inherent OVs (Figure 2e). Irradiation of BiOCl-OV with visible light in O2 atmosphere led to the

211

formation of another anisotropic EPR peak with a g-tensor of 2.005, being ascribed to the

212

photoinduced •O2- (Figure 2e).7 Such an •O2- EPR peak was not observed for BiOCl (Figure 2e).

213

Interestingly, on exposure to NO gas, intensity of the EPR signal for •O2- completely disappeared,

214

suggesting the intimate interaction between •O2- and NO (Figure 2e).


Page 10 of 21

Page 11 of 21


215 216

Figure 3. (a) Schematic illustration of the reaction cell for the in situ FTIR study. (b) FTIR spectra

217

of BiOCl-OV during photocatalytic NO oxidation. (c) Dynamic change of the •O2-/NO3- absorbance

218

increase along with NO absorbance peak decrease. (d) NO2-TPD profiles of BiOCl-OV under

219

different reaction conditions

220 221

Diffuse reflectance FTIR spectroscopy was further employed to check the possible reaction

222

intermediates and/or products on the BiOCl-OV in a special reaction cell (Figure 3a and Figure S5).

223

In the absence of O2, NO at trace concentration in the vacuum cell gave two distinct bands at 1091

224

cm-1 and 1630 cm-1 on the BiOCl-OV at 298 K in the dark (Figure 3b). The former band was

225

possibly arisen from the negatively-charged OV-NO species, and the latter broad band was due to

226

surface undissociated water, as specifically demonstrated by time-dependent in situ NO adsorption

227

on BiOCl in the dark (Figure S6). After sufficient O2 was pumped in, visible light illumination was ACS Paragon Plus Environment


then introduced to trigger photocatalytic reactions. Significant changes of FTIR spectra were vividly

229

observed. First, the band at 1091 cm-1 gradually decreased and became negative, accompanied by the

230

progressive increase of the band at 1003 cm-1 (Figure 3b). This band at 1003 cm-1 could be attributed

231

to the O-O stretching mode of side-on bridging O2- species, suggesting the rapid replacement of NO

232

by O2 on the OVs. Second, replaced NO was immediately converted to other nitrogen species

233

(Figure 3b). Sharp band at 1274 cm-1 was assigned to the ν3´´(split) mode of nitrate in a particular

234

monodentate state and the broad band at 1477 cm-1 was ascribed to the almost symmetrical surface

235

nitrate.30 The wide band from 1600 to 1800 cm-1 was typical of combination band associated with

236

NO and free nitrate.31 Noticeably, IR absorbance intensity of •O2- and NO3- species linearly

237

increased along with that of NO decrease, indicating that the oxidation of NO to nitrate was directly

238

mediated with photoinduced •O2- (Figure 3c). TPD was then employed to more precisely

239

differentiate the surface nitrogen species. 18O-labelled O2 was used to avoid the possible interference

240

of lattice O or pre-adsorbed O2 on NO oxidation. For photocatalytic NO oxidation under 298 K,

241

BiOCl-OV surface exhibited a recognizable NO2 desorption peak at 548 K with m/z of 48,

242

suggesting one the O atoms in NO2 was 18O-labelled (Figure 3d). Since the chemical adsorption of

243

NO2 on BiOCl-OV corresponded to a desorption peak at remarkably lower temperature of 438 K, the

244

NO2 desorption peak at 545 K was therefore arisen from 18O-labelled nitrate thermal decomposition

245

on BiOCl-OV surface (16O=N18O18O- → N16O18O) (Figure 3d).32 This observation was in consistent

246

with the FTIR result that the selective product of photocatalytic NO oxidation under ambient

247

condition was nitrate, not gaseous NO2 or surface NO2-, directly evidencing that complete NO

248

oxidation toward the formation of nitrate was the major pathway on BiOCl-OV via

249

geometric-favorable side-on bridging •O2-. It was interesting to note that NO2 desorption peak

250

around 438 K appeared during the photocatalytic NO oxidation at 335 K and 355 K, suggesting the

251

formation of chemically-adsorbed NO2 at higher reaction temperatures (Figure 3d). Apparently, the ACS Paragon Plus Environment

Page 12 of 21

Page 13 of 21


252

appearance of chemically-adsorbed NO2 was associated with a coexisted NO oxidation pathway I,

253

governed by terminal end-on •O2- according to the DFT calculations. This is reasonable because the

254

barrier of the O-O bond breaking in peroxynitrite for gaseous NO2 desorption via pathway I is only

255

0.49 eV, which could be overcame via thermal heating (Figure 1a). However, this terminal end-on

256

•O2- associated partial NO oxidation could be largely inhibited by the side-on bridging •O2- mediated

257

complete NO oxidation by performing the reaction at room temperature. We believe that the

258

proposed NO removal pathway, mainly supported by theoretical calculation and in situ TPD

259

measurements, will still require further experimental validation with sophisticated in situ

260

characterization techniques, including XPS and scanning tunneling microscope.

261 262

Figure 4. (a) Photocatalytic NO removal over the as-prepared BiOCl under visible light. (b)

263

Influence of generated •O2- on the NO oxidation kinetic constants and the NO2 concentration. (c)

264

Schematic illustration of photocatalytic NO removal on BiOCl-OV. (d) Transient photocatalytic NO ACS Paragon Plus Environment


removal on BiOCl-OV.

266 267

To confirm the above results and discussions, we thus employed the as-prepared BiOCl-OV film

268

to remove dilute NO (500 ppb) in a continuous flow reactor at 298 K (Figure S7). Defect-free BiOCl

269

could not remove NO efficiently under visible light irradiation (λ > 400 nm). As for BiOCl-OV, we

270

observed a fast drop of NO concentration in 5 min, followed by a slower decrease with up to 70%

271

NO removal efficiency being achieved within 15 min (Figure 4a). As expected, only 4 ppb NO2 was

272

on-line detected, much lower than the regulation standard (0.24 ppm, GB/T18883-2002) of indoor

273

NO2 and also suggesting most of NO was directly oxidized to nitrate with selectivity exceeding 99%.

274

This high selectivity towards the nitrate formation was further verified by calculating the nitrogen

275

balance via ion chromatography (Figure S8). Photocatalytic NO removal rates were linearly related

276

with •O2- in different concentrations without generating detectable NO2 (Figure 4b). Together with

277

reactive species trapping experiment, we further proposed that photocatalytic NO oxidation to nitrate

278

was directly related to •O2- generated via BiOCl intraband excitation, through which both holes (O-)

279

and •OH could be avoided (Figure 4c, S9 and S10). Interestingly, for commercial TiO2-P25,

280

photocatalytic NO2 emission could also be inhibited after the introduction of OVs (Figure S11 and

281

S12). We believe the dominant reactive surfaces, including the rutile(110) surface and anatase(101)

282

surface, which allow the formation of side-on bridging reactive oxygen species, are responsible for

283

promoted complete NO oxidation (Figure S13).27,28,33,34 Meanwhile, along with reaction temperature

284

increase from 298 K to 375 K, the selectivity for photocatalytic NO2 generation over BiOCl-OV was

285

gradually increased from 0.8% to 22.6%, agreeing well with the TPD results (Figure S14). In

286

comparison with the poor long-term stability of free-standing BiOCl-OV single crystalline

287

nanosheets with limited surface area and hierarchicality (Figure 4d and S15), BiOCl-OV

288

nanospheres were stable for continuous NO removal, highlighting the beneficial effects of ACS Paragon Plus Environment

Page 14 of 21

Page 15 of 21


289

mesopores for nitrate diffusion and storage (Figure S3). XPS analysis revealed Bi also maintained

290

most of its valence state after repeated use. Although the area percentage of Bi(3−x)+ XPS peaks at

291

163.8 and 158.4 eV of BiOCl-OV was slightly decreased from 23.5% to 20.3% (Figure S16a and

292

S16b), this slight decrease of OVs could be easily recovered via thermal annealing under high

293

vacuum at 250 oC for 10 min (Figure S16c).35 The stored nitrate after hours of use could be easily

294

washed down, exhibiting a nitrate storage capacity of 7.3 × 10-3 mol g-1 h-1, much larger than that of

295

active carbon (0.10 × 10-3 mol g-1 h-1) under the same condition.

296 297

Environmental Implications.

298

Complete removal of dilute NO without NO2 emission is a great challenge. Aiming to avoid partial

299

NO oxidation and gain mechanistic insight into the intrinsic catalytic roles of •O2-, we have utilized a

300

defect-engineered BiOCl to realize the complete oxidation of NO under visible light. It was

301

demonstrated that well-designed catalytic OVs on the prototypical (001) surface of BiOCl favored

302

the possible formation of geometric-favorable •O2- in a side-on bridging mode under ambient

303

conditions, which thermodynamically suppressed the terminal end-on •O2- associated NO2 emission

304

in case of higher temperatures, and thus selectively oxidized NO to nitrate. These findings will

305

provide the instructive information on exploring the intriguing surface chemistry of photocatalytic

306

NO oxidation and developing highly efficient NOx removal systems.

307 308

AUTHOR INFORMATION

309

Corresponding Author

310

*Phone/Fax: +86-27-6786 7535; [email protected]; [email protected]

311

Author Contributions

312

‡

H.L. and H. S. contributed equally to this work. ACS Paragon Plus Environment


Notes

314

The authors declare no competing financial interest

315 316

Acknowledgements: This work was supported by The National Key Research and Development

317

Program of China (2016YFA0203000), National Natural Science Funds for Distinguished Young

318

Scholars (21425728), National Science Foundation of China (21477044 and 51472100), 111 Project

319

(B17019), Self-Determined Research Funds of CCNU from the Colleges’ Basic Research and

320

Operation of MOE (CCNU16A02029). We also thank the National Supercomputer Center in Jinan

321

for providing high performance computation.

322 323

ASSOCIATED CONTENT

324 325

Supporting Information: NO and O2 adsorption model on prototypical BiOCl(001) surface;

326

reactive species trapping experiment; organic pollutants degradation profiles; photocatalytic NO

327

removal over TiO2 with OVs; surface area and charge of the catalysts; influence of temperatures on

328

photocatalytic NO removal; characterization of free-standing BiOCl single-crystalline nanosheets

329

(PDF)

330 331

References

332

(1) Taylor, K. C. Nitric Oxide Catalysis in Automotive Exhaust Systems. Catal. Rev. 1993, 35 (4),

333

457–481.

334

(2) Kim, C. H.; Qi, G.; Dahlberg, K.; Li, W. Strontium-Doped Perovskites Rival Platinum Catalysts

335

for Treating NOx in Simulated Diesel Exhaust. Science (80-. ). 2010, 327 (5973), 1624–1627.

336

(3) Lasek, J.; Yu, Y. H.; Wu, J. C. S. Removal of NOx by photocatalytic processes. J. Photochem. ACS Paragon Plus Environment

Page 16 of 21

Page 17 of 21


337

Photobiol. C Photochem. Rev. 2013, 14 (1), 29–52.

338

(4) Cui, W.; Li, J.; Dong, F.; Sun, Y.; Jiang, G.; Cen, W.; Lee, S. C.; Wu, Z. Highly Efficient

339

Performance

340

SrO-Clusters@Amorphous Carbon Nitride. Environ. Sci. Technol. 2017, 51 (18), 10682–10690.

341

(5) Wang, W.; McCool, G.; Kapur, N.; Yuan, G.; Shan, B.; Nguyen, M.; Graham, U. M.; Davis, B.

342

H.; Jacobs, G.; Cho, K.; Hao, X. Mixed-Phase Oxide Catalyst Based on Mn-Mullite (Sm, Gd)Mn2O5

343

for NO Oxidation in Diesel Exhaust. Science 2012, 337 (6096), 832–835.

344

(6) Dong, F.; Wang, Z.; Li, Y.; Ho, W.-K.; Lee, S. C. Immobilization of Polymeric g-C3N4 on

345

Structured Ceramic Foam for Efficient Visible Light Photocatalytic Air Purification with Real Indoor

346

Illumination. Environ. Sci. Technol. 2014, 48 (17), 10345–10353.

347

(7) Hashimoto, K.; Wasada, K.; Toukai, N.; Kominami, H.; Kera, Y. Photocatalytic oxidation of

348

nitrogen monoxide over titanium (IV) oxide nanocrystals large size areas. J. Photochem. Photobiol.

349

A Chem. 2000, 136 (1), 103–109.

350

(8) Yamamoto, A.; Mizuno, Y.; Teramura, K.; Hosokawa, S.; Tanaka, T. Noble-Metal-Free NO x

351

Storage over Ba-Modified TiO2 Photocatalysts under UV-Light Irradiation at Low Temperatures.

352

ACS Catal. 2015, 5 (5), 2939–2943.

353

(9) Ding, X.; Ho, W.; Shang, J.; Zhang, L. Self doping promoted photocatalytic removal of no under

354

visible light with Bi2MoO6: Indispensable role of superoxide ions. Appl. Catal. B Environ. 2016,

355

182, 316–325.

356

(10)

357

Reactivity, Selectivity, and Perspectives. Angew. Chemie Int. Ed. 2018, 57 (1), 122–138.

358

(11)

359

Fenton Chemistry: Surface Structure Dependent Hydroxyl Radicals Generation and Substrate

360

Dependent Reactivity. Environ. Sci. Technol. 2017, 51 (10), 5685–5694.

and

Conversion

Pathway

of

Photocatalytic

NO

Oxidation

on

Li, H.; Li, J.; Ai, Z.; Jia, F.; Zhang, L. Oxygen Vacancy-Mediated Photocatalysis of BiOCl:

Li, H.; Shang, J.; Yang, Z.; Shen, W.; Ai, Z.; Zhang, L. Oxygen Vacancy Associated Surface



(12)

Zhang, X.; Ai, Z.; Jia, F.; Zhang, L. Generalized One-Pot Synthesis, Characterization, and

362

Photocatalytic Activity of Hierarchical BiOX (X = Cl, Br, I) Nanoplate Microspheres. J. Phys.

363

Chem. C 2008, 112 (3), 747–753.

364

(13)

365

BiOCl Single-Crystalline Nanosheets. J. Am. Chem. Soc. 2012, 134 (10), 4473–4476.

366

(14)

367

Phys. Rev. Lett. 1996, 77 (18), 3865–3868.

368

(15)

369

Payne, M. C. First-principles simulation: ideas, illustrations and the CASTEP code. J. Phys.

370

Condens. Matter 2002, 14 (11), 2717–2744.

371

(16)

372

Phys. Rev. B 1990, 41 (11), 7892–7895.

373

(17)

374

1976, 13 (12), 5188–5192.

375

(18)

376

using a plane-wave basis set. Phys. Rev. B 1996, 54 (16), 11169–11186.

377

(19)

378

method. Phys. Rev. B 1999, 59 (3), 1758–1775.

379

(20)

380

for finding saddle points and minimum energy paths. J. Chem. Phys. 2000, 113 (22), 9901–9904.

381

(21)

382

Finding transition states in complex energy landscapes. J. Chem. Phys. 2015, 142 (2), 24106.

383

(22)

384

mechanisms of NO oxidation catalyzed by Cu2O(111). Appl. Surf. Sci. 2014, 316 (1), 416–423.

Jiang, J.; Zhao, K.; Xiao, X.; Zhang, L. Synthesis and Facet-Dependent Photoreactivity of

Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple.

Segall, M. D.; Lindan, P. J. D.; Probert, M. J.; Pickard, C. J.; Hasnip, P. J.; Clark, S. J.;

Vanderbilt, D. Soft self-consistent pseudopotentials in a generalized eigenvalue formalism.

Monkhorst, H. J.; Pack, J. D. Special points for Brillouin-zon integrations. Phys. Rev. B

Kresse, G.; Furthmüller, J. Efficient iterative schemes for ab initio total-energy calculations

Kresse, G.; Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave

Henkelman, G.; Uberuaga, B. P.; Jónsson, H. A climbing image nudged elastic band method

Zarkevich, N. A.; Johnson, D. D. Nudged-elastic band method with two climbing images:

Sun, B. Z.; Xu, X. L.; Chen, W. K.; Dong, L. H. Theoretical insights into the reaction


Page 18 of 21

Page 19 of 21


385

(23)

Wang, H.-F.; Guo, Y.-L.; Lu, G.; Hu, P. NO Oxidation on Platinum Group Metals Oxides:

386

First Principles Calculations Combined with Microkinetic Analysis. J. Phys. Chem. C 2009, 113

387

(43), 18746–18752.

388

(24)

389

hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide

390

and superoxide. Proc. Natl. Acad. Sci. 1990, 87 (4), 1620–1624.

391

(25)

392

Molecular Oxygen Activation of BiOCl Single-Crystalline Nanosheets. J. Am. Chem. Soc. 2013, 135

393

(42), 15750–15753.

394

(26)

395

Properties. ChemPhysChem 2005, 6 (9), 1911–1916.

396

(27)

397

Anatase TiO2 (101) to Adsorbed O2. J. Am. Chem. Soc. 2013, 135 (24), 9195–9199.

398

(28)

399

of hierarchical micro–mesoporous materials. Chem. Soc. Rev. 2016, 45 (12), 3439–3467.

400

(29)

401

with Hierarchical Bismuth Oxybromide Nanoplate Microspheres under Visible Light. Environ. Sci.

402

Technol. 2009, 43 (11), 4143–4150.

403

(30)

404

the species arising during nitrogen dioxide adsorption on titania (anatase). Langmuir 1994, 10 (2),

405

464–471.

406

(31)

407

Pt/K/γ-A12O3 NOx adsorber catalysts. Appl. Catal. B Environ. 2005, 58 (3–4), 245–254.

408

(32)

Beckman, J. S.; Beckman, T. W.; Chen, J.; Marshall, P. A.; Freeman, B. A. Apparent

Zhao, K.; Zhang, L.; Wang, J.; Li, Q.; He, W.; Yin, J. J. Surface Structure-Dependent

Tilocca, A.; Selloni, A. O2 and Vacancy Diffusion on Rutile(110): Pathways and Electronic

Li, Y.-F.; Selloni, A. Theoretical Study of Interfacial Electron Transfer from Reduced

Schneider, D.; Mehlhorn, D.; Zeigermann, P.; Kärger, J.; Valiullin, R. Transport properties

Ai, Z.; Ho, W.; Lee, S.; Zhang, L. Efficient Photocatalytic Removal of NO in Indoor Air

Hadjiivanov, K.; Bushev, V.; Kantcheva, M.; Klissurski, D. Infrared spectroscopy study of

Toops, T. J.; Smith, D. B.; Partridge, W. P. Quantification of the in situ DRIFT spectra of

Mikhaylov, R. V.; Lisachenko, A. A.; Shelimov, B. N.; Kazansky, V. B.; Martra, G.; ACS Paragon Plus Environment


Coluccia, S. FTIR and TPD Study of the Room Temperature Interaction of a NO–Oxygen Mixture

410

and of NO2 with Titanium Dioxide. J. Phys. Chem. C 2013, 117 (20), 10345–10352.

411

(33)

412

Rev. B 2004, 70 (19), 193410.

413

(34)

414

anatase TiO2 (101) with subsurface defects. Phys. Chem. Chem. Phys. 2010, 12 (40), 12956.

415

(35)

416

Photocatalysts with High Photo-Activities and Photosensitivities. Chem. Commun. 2011, 47 (17),

417

4947.

Wang, Y.; Pillay, D.; Hwang, G. S. Dynamics of oxygen species on reduced TiO2(110) Phys.

Aschauer, U.; Chen, J.; Selloni, A. Peroxide and superoxide states of adsorbed O2 on

Xing, M.; Zhang, J.; Chen, F.; Tian, B. An Economic Method to Prepare Vacuum Activated

418 419


Page 20 of 21

Page 21 of 21 420


TOC Art Figure

421


Oxygen Vacancies Mediated Complete Visible ... - ACS Publications

Recommend Documents