MoO3 against

Subscriber access provided by Kaohsiung Medical University

Remediation and Control Technologies

Self-Prevention of Well-Defined-Facet Fe2O3/MoO3 against Deposition of Ammonium Bisulfate in Low-temperature NH3-SCR Yaxin Chen, Chao Li, Junxiao Chen, and Xingfu Tang Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b04621 • Publication Date (Web): 20 Sep 2018 Downloaded from http://pubs.acs.org on September 21, 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 24

Environmental Science & Technology

Self-Prevention of Well-Defined-Facet Fe2O3/MoO3 against Deposition of Ammonium Bisulfate in Low-temperature NH3-SCR Yaxin Chen,†,§ Chao Li, †,§ Junxiao Chen,† and Xingfu Tang*,†,‡ †

Department of Environmental Science & Engineering, Fudan University, Shanghai 200433, China

‡

Shanghai Institute of Pollution Control & Ecological Security, Shanghai 200092, China

§

These authors contributed equally to this work.

1

ABSTRACT

2

Low-temperature selective catalytic reduction of NO by NH3 (NH3-SCR) is a promising

3

technology for controlling NO emission from various industrial boilers, but it remains

4

challenging because unavoidable deposition of ammonium bisulfates (ABS) in the stack gases

5

containing both SO2 and H2O inevitably results in deactivation of catalysts. Here we developed a

6

stable low-temperature NH3-SCR catalyst by supporting Fe2O3 cubes on surfaces of MoO3

7

nanobelts with NH4+-intercalatable interlayers, which enables Fe2O3/MoO3 to spontaneously

8

prevent ABS from depositing on the surfaces. Using in situ synchrotron X-ray diffraction, 1H

9

magic angle spinning nuclear magnetic resonance, and temperature-programmed decomposition

10

procedure, the results demonstrate that NH4+ of ABS was initially intercalated in the interlayers

11

of MoO3, leading to a NH4+-HSO4- cation-anion separation by conquering their strong

12

electrostatic interactions, and subsequently the separated NH4+ was consumed by taking part in

13

low-temperature NH3-SCR. Meanwhile, the surface HSO4- separated from ABS oxidized the

14

reduced catalyst during the NH3-SCR redox cycle, concomitant with release of SO2 gas, thereby

15

resulting in decomposition of ABS. This work assists the design of stable low-temperature NH3-

16

SCR catalysts with strong resistance against deposition of ABS.

ACS Paragon Plus Environment

1


Page 2 of 24

17

INTRODUCTION

18

Nitrogen oxide (NO) is one of the crucial precursors for forming both particulate matters and

19

ozone in the atmosphere,1,2 and hence NO emission control is a key requirement for reducing

20

these atmospheric pollutants. Numerous efforts have been devoted to controlling NO emission,2,3

21

and selective catalytic reduction of NO by using NH3 (NH3-SCR) as reductant into N2 and H2O

22

over V2O5-based catalysts is one of the widely used technologies.1,4 In the SO2-containing flue

23

gases, SO3 produced from SO2 oxidation reacts with NH3 to form viscous ammonium bisulfate

24

(ABS), NH4HSO4, which will cause catalyst deactivation by blocking active sites when NH3-

25

SCR occurs below a critical temperature (often slightly higher than the dew point of ABS).5,6 To

26

eliminate such an ABS inhibition, to reheat the flue gases up to higher than the dew point of ABS

27

is extremely required at the expense of energy cost. Therefore, one of the important prerequisites

28

for low-temperature SCR is to develop a stable catalyst with strong resistance against deposition

29

of ABS.

30

On the basis of the reaction equation of SO3 + NH3 + H2O = NH4HSO4, one approach to

31

avoiding the ABS generation is to reduce the vapor pressure of SO3 via retarding oxidation

32

ability of catalysts to SO2. However, partial oxidation of NH3 is one of the important steps in

33

NH3-SCR, which demands catalysts to have a desirable oxidation ability to achieve high NH3-

34

SCR efficiency,7 i.e., the unfavorable oxidation of SO2 often occurs simultaneously with NH3-

35

SCR.8 Furthermore, a certain amount of SO3 is often present in flue gases, readily reacting with

36

NH3 to form ABS. To an extent, the formation of ABS in the low-temperature NH3-SCR process

37

seems unavoidable. An alternative is to alter the acid-basic properties of catalysts’ surfaces by

38

adding promoters according to an adsorption model of ABS. Phil et al.9 proposed a dual-site

39

adsorption model of ABS on V2O5-M/TiO2 (M presents a promoter) that NH4+ and HSO4- of


2

Page 3 of 24


40

ABS are respectively adsorbed as H3N-H···O-V and adjacent HO3S-O···M bonds on the surfaces

41

of V2O5-M/TiO2. The results demonstrated that weakening the O···M bonding strength of HO3S-

42

O···M by increasing the acidity of M made ABS easily desorbed, thereby strengthening

43

resistance to ABS inhibition, which was also corroborated by other reports.10,11 However, ABS

44

formation is nearly barrierless,12 and once the ABS was deposited on the surface of catalysts, the

45

viscosity of ABS (0.1~0.2 Pa.s)6 made it difficult to decompose the produced ABS. Thus, the

46

catalytic activity still gradually decreased, implying that the alteration of the catalysts’ surfaces

47

cannot essentially conquer the ABS inhibition.

48

A promising strategy is to decompose the produced ABS under low-temperature NH3-SCR

49

conditions, but it is a formidable task because it is energetically unfavorable to decompose

50

ABS.12,13 Johnston et al.13 carried out theoretical calculations and found that electrostatic

51

interactions among cations and anions of ABS are much stronger than that due to hydrogen

52

bonding and the free energy of ABS formation. This explicitly elucidates that one method to

53

decomposing ABS is to separate NH4+ from HSO4- by conquering the electrostatic interactions,

54

and thus the separated NH4+ can be used for reducing NO,14 while HSO4- can be reduced into

55

SO215 or reacts with surface H+ to form H2SO4,16 and ultimately at NH3-SCR reaction

56

temperatures, SO2 and H2SO4 leave from catalyst surfaces, leading to decomposition of ABS. A

57

motivation for this purpose originates from our recent results of successfully trapping K+ from

58

potassium salts such as K2SO4 or KCl,17 and the fact that NH4+ ion is equivalent to K+ ion18

59

allows one to trap NH4+ from ABS by designing a catalyst with abundant NH4+-trapping sites.

60

Note that α-MoO3 has a layered structure with a suitable interlayer distance and variable

61

oxidation states, which allows the intercalation of NH4+, concomitant with an energy saving of

62

~73 kJ mol-1, as shown in reaction equation: NH3(g) + 1/0.23 H0.31MoO3(s ) = 1/0.23


3


Page 4 of 24

63

(NH4)0.23H0.08MoO3(s).19 Moreover, α-MoO3 is also an important promoter of commercial NH3-

64

SCR catalysts owing to its strong adsorption ability to NH3 and resistance to SO2 poisoning.

65

Therefore, α-MoO3 should become a desired candidate for decomposing ABS via NH4+ trapping.

66

In this work, we used α-MoO3 nanobelts with a layered structure for trapping NH4+ from

67

ABS, on the surfaces of which α-Fe2O3 was supported to form a Fe2O3/MoO3 NH3-SCR catalyst.

68

Self-prevention function of Fe2O3/MoO3 was studied by operando and in situ low-temperature

69

NH3-SCR. In situ synchrotron X-ray diffraction (SXRD), 1H magic angle spinning nuclear

70

magnetic resonance (1H MAS NMR), and temperature-programmed decomposition (TPDC) of

71

ABS were used to investigate the process of ABS decomposition. This work provides a general

72

strategy to rationally design low-temperature NH3-SCR catalysts with strong resistance against

73

ABS inhibition for controlling NO emission.

74

EXPERIMENTAL SECTION

75

Catalyst Preparation. Fe2O3 was synthesized through a PVP solvethermal route.20 Briefly,

76

1.810 g Fe(NO3)39H2O and 3.584 g PVP (Mw = 40000) were dissolved in 40 mL of DMF. The

77

solution was turned into a 50 mL Teflon-lined stainless steel autoclave, which was then put into

78

an oven and heated at 180 oC for 30 h, followed by the autoclave being cooled to room

79

temperature naturally. The red precipitates were collected by centrifugation, washed with

80

deionized water and ethanol for several times, and finally dried in air at 60 oC overnight.

81

MoO3 was synthesized via a modified hydrothermal method.21 Typically, 2 g molybdenum

82

power (Aladdin, 99.5%) was added into 10 mL deionized water to form a uniform mixture, to

83

which 20 mL 30% (wt.%) H2O2 was slowly added until the solution became light-yellow after

84

stirring for 30 min, and then transferred to a Teflon-lined stainless steel autoclave and kept at

85

220 °C for 48 h. The precipitate was filtered and rinsed by deionized water and ethanol for


4

Page 5 of 24


86

several times, and finally dried at 80 °C and calcined at 400 °C for 4 h. For Fe2O3/MoO3

87

preparation, the as-prepared Fe2O3 (0.1 g) was dispersed in 20 mL deionized water, and added

88

into a 40 mL of an aqueous solution containing MoO3 nanobelts (1 g). The flurry was kept at 80

89

o

90

obtain Fe2O3/MoO3. After the Fe2O3 loading, the highly dispersed states of the Fe2O3 cubes were

91

observed on the surfaces of the MoO3 nanobelts, and the crystalline states and the morphology of

92

the Fe2O3 cubes remained almost unchanged (Figure S1). Component content of Fe2O3/MoO3

93

was obtained by x-ray fluorescence analysis (XRF), as listed in Table S1, and from the

94

experimental data, the Fe2O3 loading is calculated to be 9.9% with respect to MoO3, approaching

95

the value (10%) of the Fe2O3 loading during the preparation. To further load ABS, samples were

96

impregnated with a certain amount of ABS solution.

C until dry, and the solid was dried overnight at 105 oC and calcinated at 300 oC for 4 h to

97

Catalytic Evaluation. Operando and in situ low-temperature NH3-SCR tests were

98

performed in a fixed-bed quartz reactor (i.d. = 8 mm) under steady flow and one atmospheric

99

pressure. 1.000 g Fe2O3/MoO3 with 40-60 mesh were charged for each run, and 0.1 g pure Fe2O3

100

with 40-60 mesh were used in the in situ low-temperature NH3-SCR tests for comparison. The

101

feed gas contained 500 ppm NO, 500 ppm NH3, 3.0 vol% O2, 145 ppm SO2 (when used), 10

102

vol% H2O (when used) and balanced N2 with a total flowrate of 500 mL min-1. Concentration of

103

NO in the effluent was measured by using a Fourier transform infrared spectrometer (FTIR,

104

Thermo Scientific Antaris IGS analyzer). To achieve viable kinetic data, the influences of

105

internal and external diffusions have been eliminated before the kinetic measurements, as shown

106

in Figure S2.

107

The TPDC procedure was conducted in the reaction systems. Briefly, an ABS-deposited

108

sample of 0.6 g (40–60 mesh) was pre-heated to 100 oC in N2 for 30 min in order to remove the


5


Page 6 of 24

109

physically adsorbed water and other impurities. The sample was then heated to 600 oC at a ramp

110

of 5 oC min-1 in N2. The outlet SO2, NH3, NO, NO2, and N2O was monitored using the FTIR flue

111

gas spectrometer. The ABS reactivity behavior on MoO3 was measured via temperature

112

programmed surface reaction (TPSR) with NO in the presence of O2. First, 0.6 g ABS-deposited

113

catalyst (40–60 mesh) was exposed to a stream containing 500 ppm NO, 3% O2 and balance N2

114

at a flow rate of 500 mL min-1, and then heated from room temperature to 600 oC at a ramp of 5

115

o

116

gas spectrometer.

C min-1. The inlet and outlet concentrations of NO and SO2 were monitored using the FTIR flue

117

Catalyst Characterization. Both the room-temperature SXRD patterns and the in situ

118

SXRD patterns were recorded at BL14B of the Shanghai Synchrotron Radiation Facility (SSRF)

119

at a wavelength of 0.6883 Å. To record the in situ SXRD patterns, the sample (∼1.5 mg) was

120

loaded in a flow cell (a quartz capillary tube with a diameter of ∼1 mm) sandwiched between

121

glass wool, and then heated by following a temperature-programmed procedure at a ramp of 2 oC

122

min-1. The in situ SXRD data were collected at 2 min intervals and analyzed by using CMPR

123

software. Transmission electron microscope (TEM) and high-resolution TEM (HRTEM) were

124

conducted on a JEM 2100F transmission electron microscope. Field emission scanning electron

125

microscope (SEM) images were obtained by a JEOL JSM-6700F instrument operated at a beam

126

energy of 3 kV. 1H MAS NMR experiments were performed on a Bruker AVANCE III 400 WB

127

spectrometer at a spinning rate of 20 kHz. The chemical shifts (δ) of 1H were referenced to TMS

128

at 0 ppm. 1H MAS spectra were recorded in a spin echo pulse sequence (π/2–τ–π–τ-acquire),

129

where τ is equal to one rotor period (rotor synchronized). The excitation pulse length was 2.3 µs

130

(π/2), and typically ∼40 scans were accumulated with a 5 s delay. The specific surface areas

131

(SBET) were determined by using linear portion of Brunauer–Emmett–Teller (BET) model. The


6

Page 7 of 24


132

BET surface areas were measured by N2 adsorption at a liquid nitrogen temperature using a

133

NOVA4000e (USA, Quantachrome) automated gas sorption system.

134

RESULTS AND DISCUSSION

135

Well-Defined Facets of Fe2O3/MoO3. The morphologies and structures of Fe2O3 and MoO3

136

were studied by using TEM and XRD. Regular cube-shaped morphology of Fe2O3 particles with

137

an average size of ∼35 nm is clearly observed from the TEM image in Figure 1a. Two kinds of

138

fringes with an equal distance of ∼3.6 Å and an intersection lattice angle of 94 o can be assigned

139

to (012)Fe2O3 and (102) Fe2O3 planes of the Fe2O3 cubes, as shown in the HRTEM image viewed

140

from the [2-21]Fe2O3 direction of Figure 1b (subscripts ‘Fe2O3’ or upcoming ‘MoO3’ are used for

141

difference). By combining the above TEM observations with an upcoming XRD pattern of Fe2O3

142

(Figure 2), it is readily deduced that the Fe2O3 cube is enclosed by {012}Fe2O3, {102}Fe2O3 and

143

{112}Fe2O3 planes.22 The particular morphology of MoO3 was observed by SEM and HRTEM

144

with the typical images shown in Figure S1. MoO3 has a nanobelt-shaped morphology with a

145

width of ~200 nm and a length at the micrometer level (Figure S3a). The closest neighbor fringes

146

with distances of ∼4.0 and ∼3.7 Å with an intersection angle of ∼90o are ascribed to the

147

(100)MoO3 and (001)MoO3 planes, respectively, confirming that the electron beam is along a

148

[010]MoO3 axis (Figure S3b). Consequently, the MoO3 nanobelts are constructed by MoO3

149

monolayer (010)MoO3 sheets along the [010] direction, consistent with the previous work.23 After

150

loading Fe2O3 on the MoO3 surfaces, the morphologies of Fe2O3 and MoO3 are preserved, as

151

shown in the TEM image of Fe2O3/MoO3 (Figure 1c), the interface of which is determined by

152

using HRTEM with the electron beam perpendicular to the (010)MoO3 plane. In Figure 1d, the

153

(012)Fe2O3 and (-102)Fe2O3 planes with an intersection angle of 94o are observed, indicating that

154

the electron beam is along a [2-21]Fe2O3 axis, which is also parallel to a [010]MoO3 axis judging by


7


Page 8 of 24

155

an intersection angle of 90o between the (100)MoO3 and (001)MoO3 planes. This confirms that the

156

Fe2O3 cubes are deposited on the surfaces of MoO3, and the interface comprises the (1-12)Fe2O3

157

and (010)MoO3 planes.

Figure 1. TEM (a,c) and HRTEM (b,d) images of Fe2O3 (a,b) and Fe2O3/MoO3 (c,d).

158

Figure 2 shows the SXRD patterns of the samples. The diffractions of MoO3 and Fe2O3 can

159

be readily indexed to orthorhombic and rhombohedral structures, respectively. For the XRD

160

patterns of MoO3, three strong diffraction peaks at 2θ of 5.6, 11.3, and 17.2 o are indexed to the

161

(020)MoO3, (040)MoO3, and (060)MoO3 planes, respectively, confirming that MoO3 has a layered

162

crystal structure, i.e., α-MoO3,24 consistent with the results of TEM imaging. Owing to the

163

layered structure, MoO3 has an intercalation property, as shown a structural model in the inset of

164

Figure 2. Corner-sharing chains of MoO6 octahedra shared edges with two similar chains to form

165

layers of MoO3. The layers are stacked in a staggered arrangement along the b-axis and are held

166

together by van der Waals forces, which allow intercalation of a wide range of species without

167

drastic alteration of the MoO3 host, such as H+, NH4+, Li+, and Na+.24-27 After loading Fe2O3 on


8

Page 9 of 24


168

the MoO3 surfaces, the weak reflections assigned to Fe2O3 crystals can be discernible and no new

169

phases appear. Owing to the Fe2O3 cubes being only supported on the surfaces of MoO3

170

nanobelts, the layered MoO3 of Fe2O3/MoO3 still has the intercalation property.

Figure 2. SXRD patterns of the samples. Inset: structural model of layered MoO3.

171

Operando and In Situ ABS Deposition in Low-Temperature SCR. To study the self-

172

prevention function of Fe2O3/MoO3 when the surfaces are deposited with ABS in NH3-SCR, we

173

investigated operando and in situ deposition of ABS during the NH3-SCR processes. Figure 3a

174

shows operando temperature-programmed NH3-SCR performance of Fe2O3/MoO3 after

175

impregnating Fe2O3/MoO3 with 5 wt.% ABS with respect to the catalyst. According to the

176

structure of ABS28 and the surface area (~9 m2/g) of Fe2O3/MoO3,21,22 5 wt.% ABS is enough to

177

cover all the surfaces of Fe2O3/MoO3. After the ABS impregnation, the sample was only dried at

178

80 oC for 4 h without further annealing. We conducted two consecutive temperature-programmed

179

NH3-SCR reactions and plotted the NO conversions (XNO) as a function of reaction temperature

180

(T) in Figure 3a. In the first run, XNO increases as T increases from 100 to 280 °C at a ramp of

181

5 °C min−1, and arrives at a steady state with ∼42% of XNO during the isothermal process at


9


Page 10 of 24

182

280 °C, which is equal to XNO of Fe2O3/MoO3 without ABS pre-deposition under the same

183

conditions (Figure S4). Taking T = 280 °C lower than the dew point of ABS into account,29 the

184

deposited ABS has no influence on the activity of Fe2O3/MoO3. When the catalyst was cooled

185

down to 100 °C, the second run was subsequently conducted with the same procedure as the first

186

run. Note that the catalytic performances in both runs are almost the same as each other. Hence,

187

Fe2O3/MoO3 has the self-prevention function against the ABS deposition in low NH3-SCR.

188

As aforementioned, when SCR operates at T lower than the dew point of ABS (often T < 320

189

o

C),29 catalysts are often deactivated due to deposition of ABS produced in the NH3-SCR

190

process.6,29 We investigated in situ deposition of ABS in the low-temperature NH3-SCR process

191

in the co-presence of SO2 and H2O (145 ppm SO2, 10 vol% H2O), and T was set at 280 oC to

192

facilitate the deposition of ABS.5 XNO are controlled to be lower than 15% in order to make sure

193

that NH3-SCR occurs in the reaction kinetics regime, and as shown in Figure 3b, after more than

194

24 h for in situ deposition of ABS, the catalytic activity of Fe2O3/MoO3 is very stable and XNO

195

(∼12.5%) remains unchanged with time on stream. This demonstrates that the ABS deposition

196

has no influence on the catalytic performance of Fe2O3/MoO3, in agreement with the results

197

obtained from the operando deposition of ABS in low-temperature NH3-SCR. Furthermore, to

198

shed light on the self-prevention function of Fe2O3/MoO3, under the same conditions, in situ

199

deposition of ABS was also conducted in the low-temperature NH3-SCR process over the pure

200

Fe2O3 cubes, and the results are also shown in Figure 3b. As expected, the pure Fe2O3 cubes

201

underwent continuous deactivation with time on stream and XNO decreases from 12.8 % down to

202

10.5% within 24 h, explicitly elucidating that the self-prevention function of Fe2O3/MoO3

203

originates from MoO3 rather than Fe2O3, i.e., MoO3 possesses a function of the decomposition of


10

Page 11 of 24


204

ABS, which should be intimately associated with the layered structure that allows intercalation

205

of various ions including NH4+.27 50 300

Fe2O3/MoO3

30

250 o

XNO (%)

40

XNO

20

200

1st run 2nd run

T

10

150

1st run 2nd run

0 20

40

60

T ( C)

(a)

80

100 100

Time (min)

(b)

XNO (%)

14

12

10

Fe2O3 Fe2O3/MoO3

8 0

5

10

15

20

25

Time (h)

Figure 3. (a) Operando ABS deposition on Fe2O3/MoO3 with the 5 wt.% ABS loading in lowtemperature NH3-SCR with two consecutive temperature-programmed procedures. Reaction conditions: 500 ppm of NO, 500 ppm of NH3, 3.0 vol % O2, and balanced N2, GHSV = 33,000 h−1. (b) In situ ABS deposition on Fe2O3/MoO3 and Fe2O3 in the NH3-SCR process in the copresence of SO2 and H2O at T = 280 oC. Reaction conditions: 500 ppm of NO, 500 ppm of NH3, 3.0 vol % O2, and balanced N2, 145 ppm SO2, 10 vol% H2O, GHSV = 33,000 h−1 for Fe2O3/MoO3.

206

Disintegration of ABS via Separating NH4+ from HSO4-. Next, we focused on the

207

disintegration of ABS over the MoO3 nanobelts. Owing to the layered structure of α-MoO3, the


11


Page 12 of 24

208

interactions between interlayers originate from van der Waals force and hydrogen bonds.30

209

Atomic vibrations in the interlayers are relative to temperatures, i.e., Debye-Waller attenuation

210

occurs as temperature increases.31 Slade et al.27 evidenced such a Debye-Waller attenuation when

211

NH4+ inserted into the MoO3 interlayers by monitoring the intensity of the neutron scattering as a

212

function of temperature. Similarly, we carried out temporal in situ SXRD measurements of

213

MoO3 after pre-depositing 5 wt.% ABS on the surfaces, followed by two consecutive

214

temperature-programmed procedures. Figure S5 depicted the in situ SXRD patterns of ABS-

215

deposited MoO3. At room temperature, the diffractions can be indexed to α-MoO3, except for

216

some diffractions due to ABS. In the first run, the diffractions due to ABS gradually disappear as

217

temperature increases and approaches the melting point (147 oC) of ABS. In the second run, no

218

diffractions due to ABS appears in the SXRD patterns even at room temperature, and all the

219

diffractions are only indexed to α-MoO3, indicating that ABS has left from the α-MoO3 surfaces

220

after the first run. Fe2O3/MoO3 behaves similar under the same conditions according to the ex

221

situ SXRD results, as shown in Figure S6.


12

Page 13 of 24


Figure 4. (a) Counter maps of the (020)MoO3 diffractions of the temporal in situ SXRD patterns as a function of T. After the first run, the sample was cooled to 30 oC, and then the second run was conducted. (b) Comparison of the alterations in intensity of the (020)MoO3 diffractions for both runs.

222

The diffraction intensity of (0l0)MoO3 is often sensitive to the temperature when the MoO3

223

interlayers were inserted by other ions,27 and thus we plotted the (020)MoO3 diffraction intensity

224

as a function of temperature in Figure 4a. For both runs, the intensity of (020)MoO3 becomes weak

225

as temperature increases. To distinguish the subtle difference in both runs, Figure 4b displays the

226

intensity of (020)MoO3 as a function of temperature. Obviously, it can be divided into three stages

227

in the first run. At the first stage, a rapid decrease in intensity takes place as temperature

228

increases from 30 to 150 oC, which is due to the vaporization of intercalated water (as evidenced

229

in upcoming Figure 5). At the second stage in the temperature of 150-280 oC, the intensity of

230

(020)MoO3 is gradually weak, and again decreases rapidly at the third stage where the temperature


13


Page 14 of 24

231

increases up to 280 oC. For comparison, to eliminate the Debye-Waller attenuation due to the

232

temperature effect, the second in situ SXRD measurement was performed with the same

233

procedure, and a gradual decrease in intensity occurs in the whole temperature window, which is

234

typically characteristic of the Debye-Waller attenuation. Owing to the melting point of ABS

235

being 147 oC, ABS melts and NH4+ diffuses into the MoO3 interlayers at T > 150 oC. By

236

comparing both in situ SXRD patterns, it is possible that at the first run NH4+ cations were

237

inserted into the MoO3 interlayers at T > 150 oC,27 which make the layered structure of α-MoO3

238

relatively regular due to strong electrostatic reactions between intercalated NH4+ and the MoO3

239

interlayers, thereby resulting in the less rapid decrease in intensity at T = 140-280 oC. Over 280

240

o

241

according to the previous work.30

C, the return rapid decrease in intensity possibly results from deammonization and ammonolysis

5 ABS/MoO3-80 4 8

Intensity (10 )

Bulk H2O

MoO3

NH4+

3

H2O

2

OH-

1

0 3

ABS/MoO3-200

8

Intensity (10 )

ABS/MoO3-450

2

1

0 15

10

5

0

-5

δ (ppm)

Figure 5. 1H MAS NMR spectra of MoO3 and ABS/MoO3-T (T stands for annealing temperature).


14

Page 15 of 24


242

We further confirmed the disintegration of ABS by using 1H MAS NMR, and the results are

243

shown in Figure 5. For MoO3 without ABS deposition, a strong symmetric peak appears at δ =

244

4.5 ppm, which is assigned to the hydrogen resonance of interlayered H2O. Two weak peaks at δ

245

= ~1.3 and ~0.8 ppm can be assigned to solitary H2O molecules and OH- groups, respectively.32

246

For ABS-deposited MoO3 after annealing at 80 oC (denoted as ABS/MoO3-80), two new resonant

247

features appear at δ = ~6.8 and ~5.0 ppm, which can be respectively assigned to the hydrogen

248

resonance of NH4+ and H2O in ABS.33 For ABS/MoO3-200, the peak due to NH4+ shifts down to

249

δ = ~6.5 ppm, indicating that the electronic density of H of NH4+ increases, and the peak due to

250

H2O of ABS disappears. This corroborates that NH4+ cations have been inserted in the MoO3

251

interlayers judging from the chemical downshift and the fact that the acidity of H2SO4 is much

252

stronger than that of α-MoO3, consistent with the in situ SXRD results. As for ABS/MoO3-450,

253

the peaks due to NH4+ totally disappears, concomitant with the appearance of the peaks due to

254

solitary H2O molecules and OH-, demonstrating the oxidation of partial NH4+ with the lattice

255

oxygen.34,35 Therefore, the results from in situ SXRD patterns and 1H MAS NMR spectra

256

elucidate that MoO3 with the layered structure can spontaneously separate NH4+ from ABS at T >

257

150 oC, thus leading to the disintegration of ABS.

258

Sustainable Decomposition of ABS. Finally, we investigated the possibility for sustainable

259

decomposition of ABS under low-temperature NH3-SCR conditions. Although MoO3 can

260

conquer the strong electrostatic interactions to disintegrate ABS into spatially separated NH4+

261

and HSO4-, this process will stop if the NH4+-trapping sites of the interlayers are fully occupied.

262

Furthermore, the left surface HSO4- anions (or H2SO4)15,16 also have a critical inhibition to NH3-

263

SCR if they cannot be moved from the catalytic surfaces after accumulation to a certain extent.

264

For this purpose, we pre-deposited 0.5 wt.% ABS on the surfaces of MoO3 and a commercial


15


Page 16 of 24

265

V2O5-WO3/TiO2 NH3-SCR catalyst for reference, and then conducted the TPDC procedures.

266

Figure 6 shows the TPDC of ABS pre-deposited on the surfaces of MoO3 and V2O5-WO3/TiO2.

267

As expected in Figure 6a, NH3 does not release from the surfaces of V2O5-WO3/TiO2 until T

268

reaches ∼320 oC, slightly higher than the dew point of ABS (~290 oC),29 indicating that V2O5-

269

WO3/TiO2 cannot prevent ABS from depositing on its surface at low temperatures. According to

270

the nitrogen balance, an amount of nitrogen calculated from released NH3 is much lower than

271

that in ABS, indicative of the occurrence of NH3 oxidation by the catalyst (Figure S7), and thus

272

the reduced catalysts were oxidized by HSO4-, simultaneously leading to a SO2 release. 60

(a)

NH3

V2O5-WO3/TiO2

SO2

40

0 60

(b)

MoO3

NH3 SO2

40 20 0

(c)

480

MoO3

20

NO

10

460 440

SO2 420

0 100

200

300

400

500

NO Concentration (ppm)

NH3 or SO2 Concentration (ppm)

20

600

o

T ( C) Figure 6. TPDC (a,b) and TPSR (c) profiles of V2O5-WO3/TiO2 (a) and MoO3 (b,c) with deposited 0.5 wt.% ABS. Conditions: heating rate: 5 oC min-1; 0.6 g sample, N2 500 mL min-1 for TPDC or 500 ppm NO + 3% O2 in N2 at 500 mL min-1 for TPSR.


16

Page 17 of 24


273

Note that for MoO3, the onset of the NH3 release starts with ~200 oC and reaches a

274

maximum at ~260 oC, as shown in Figure 6b. Meanwhile, NH3 should be oxidized by MoO3

275

according to the nitrogen balance calculation (Figure S8 and Table S1). To substantiate the

276

occurrence of the NH3 oxidation during the above process, we carried out the TPSR procedure in

277

the co-presence of NO and O2. As shown in Figure 6c, obviously, the separated NH4+ from ABS

278

do take part in low-temperature NH3-SCR to release the NH4+-trapping sites in the interlayers.

279

Similarly, the TPSR curve of Fe2O3/MoO3 is similar to that of MoO3, suggesting that the ABS

280

decomposition process does occur over MoO3 rather than Fe2O3 in low-temperature NH3-SCR

281

(Figure S9). It is commonly accepted that NH3 adsorbed on the Brönsted acid sites is partially

282

oxidized by catalyst, and then reacts with NO to finish NH3-SCR, and the reduced catalyst

283

subsequently was oxidized by O2 to complete a redox cycle.7 Owing to the oxidation ability of

284

SO42- stronger than O2, the separated HSO4- from ABS oxidizes the reduced catalyst to release

285

SO2 gas. As a consequence, ABS can be successfully decomposed by MoO3 in low-temperature

286

NH3-SCR via two consecutive steps: (i) spatial separation of NH4+ from HSO4- by conquering

287

the strong electrostatic interactions present in ABS; (ii) consumption of separated NH4+ from

288

HSO4- via a redox cycle with assist of MoO3 to make sure the sustainable decomposition of ABS

289

in low-temperature NH3-SCR. Therefore, MoO3 with the layered structure is a versatile support

290

to guarantee the self-prevention of Fe2O3/MoO3 catalysts from ABS deposition in low-

291

temperature NH3-SCR.

292

In conclusion, Fe2O3/MoO3 had the self-prevention function against the ABS deposition in

293

low-temperature NH3-SCR, which originated from MoO3 with the layered structure that trapped

294

NH4+ from ABS, leading to the decomposition of ABS at low temperature. The decomposing

295

process of ABS on the catalyst surface was proved by using in situ synchrotron X-ray diffraction,


17


Page 18 of 24

296

1

297

(i) spatial separation of NH4+ from HSO4- by conquering the strong electrostatic interactions

298

present in ABS; (ii) consumption of separated NH4+ from HSO4- via a redox cycle with assist of

299

MoO3 to make sure the sustainable decomposition of ABS in low-temperature NH3-SCR. This

300

work may provide a new strategy for the design and fabrication of stable low-temperature NH3-

301

SCR catalysts applied in various industrial boilers for NO emission control.

302

ASSOCIATED CONTENT

303

Supporting Information. Some related tables and figures. This material is available free of

304

charge via the Internet at http://pubs.acs.org.

305

AUTHOR INFORMATION

306

Corresponding Author

307

*(X.T.) Phone: +86-21-65642997; fax: +86-21-65643597; e-mail: [email protected].

308

Notes

309

The authors declare no competing financial interest.

310

ACKNOWLEDGMENTS

311

This work was financially supported by the NSFC (21477023 and 21777030). The SXRD

312

measurements were conducted at the SSRF.

313

REFERENCES

H MAS NMR, and temperature-programmed decomposition of ABS, which included two steps:


18

Page 19 of 24


314

(1) Busca, G.; Lietti, L.; Ramis, G.; Berti, F. Chemical and mechanistic aspects of the selective

315

catalytic reduction of NOx by ammonia over oxide catalysts: A review. Appl. Catal. B 1998, 18

316

(1-2), 1-36.

317

(2) Pârvulescu, V. I.; Grange, P.; Delmon, B. Catalytic removal of NO. Catal. Today 1998, 46,

318

233-316.

319

(3) Roy, S.; Hegde, M. S.; Madras, G. Catalysis for NOx abatement. Appl. Energ. 2009, 86, 2283-

320

2297.

321

(4) Lietti, L.; Nova, I.; Ramis, G.; Dall’Acqua, L.; Busca, G.; Giamello, E.; Forzatti, P.; Bregani,

322

F. Characterization and reactivity of V2O5-MoO3/TiO2 de-NOx SCR catalysts. J. Catal. 1999, 187

323

(2), 419-435.

324

(5) Matsuda, S.; Kamo, T.; Kato, A.; Nakajima, F.; Kumura, T.; Kuroda, H. Deposition of

325

ammonium bisulfate in the selective catalytic reduction of nitrogen oxides with ammonia. Int.

326

Eng. Chem. Prod. Res. Dev. 1992, 21, 48-52.

327

(6) Zhou, C.; Zhang, L.; Deng, Y.; Ma, S. Research progress on ammonium bisulfate formation

328

and control in the process of selective catalytic reduction. Environ. Prog. Sustain. 2016, 35 (6),

329

1664-1672.

330

(7) Topsøe, N. Mechanism of the selective catalytic reduction of nitric oxide by ammonia

331

elucidated by in situ on-line fourier transform infrared spectroscopy. Science 1994, 265, 1217-

332

1219.

333

(8) Svachula, J.; Alemany, L. J.; Ferlazzo, N.; Forzatti, P.; Tronconi, E. Oxidation of SO2 to SO3

334

over honeycomb deNOxing catalysts. Ind. Eng. Chem. Res. 1993, 32, 826-834.


19


Page 20 of 24

335

(9) Phil, H. H.; Reddy, M. P.; Kumar, P. A.; Ju, L. K.; Hyo, J. S. SO2 resistant antimony

336

promoted V2O5/TiO2 catalyst for NH3-SCR of NOx at low temperatures. Appl. Catal. B. 2008, 78

337

(3-4), 301-308.

338

(10) Ye, D.; Qu, R.; Song, H.; Gao, X.; Luo, Z.; Ni, M.; Cen, K. New insights into the various

339

decomposition and reactivity behaviors of NH4HSO4 with NO on V2O5/TiO2 catalyst surfaces.

340

Chem. Eng. J. 2016, 283, 846-854.

341

(11) Ye, D.; Qu, R.; Zheng, C.; Cen, K.; Gao, X. Mechanistic investigation of enhanced

342

reactivity of NH4HSO4 and NO on Nb- and Sb-doped VW/Ti SCR catalysts. Appl. Catal. A

343

2018, 549, 310-319.

344

(12) Li, L.; Kumar, M.; Zhu, C.; Zhong, J.; Francisco, J. S.; Zeng, X. C. Near-barrierless

345

ammonium bisulfate formation via a loop-structure promoted proton-transfer mechanism on the

346

surface of water. J. Am. Chem. Soc. 2016, 138, 1816-1819.

347

(13) DePalma, J. W.; Doren, D. J.; Johnston, M. V. Formation and growth of molecular clusters

348

containing sulfuric acid, water, ammonia, and dimethylamine. J. Phys. Chem. A 2014, 118 (29),

349

5464-5473.

350

(14) Wang, X.; Du, X.; Zhang, L.; Chen, Y.; Yang, G.; Ran, J. Promotion of NH4HSO4

351

decomposition in NO/NO2 contained atmosphere at low temperature over V2O5-WO3/TiO2

352

catalyst for NO reduction. Appl. Catal. A. 2018, 559, 112-121.

353

(15) Li, P.; Liu, Q.; Liu, Z. Behaviors of NH4HSO4 in SCR of NO by NH3 over different cokes.

354

Chem. Eng. J. 2012, 181-182, 169-173.

355

(16) Baltin, G.; Köser, H.; Wendlandt, K. P. Sulfuric acid formation over ammonium sulfate

356

loaded V2O5-WO3/TiO2 catalysts by DeNOx reaction with NOx. Catal. Today 2002, 75, 339-345.


20

Page 21 of 24


357

(17) Huang, Z.; Gu, X.; Wen, W.; Hu, P.; Makkee, M.; Lin, H.; Kapteijn, F.; Tang, X. A “smart”

358

hollandite deNOx catalyst: Self-protection against alkali poisoning. Angew. Chem. Int. Ed. 2013,

359

52, 660-664.

360

(18) Chen, L.; Lam, S.; Zeng, Q.; Amal, R.; Yu, A. Effect of cation intercalation on the growth of

361

hexagonal WO3 nanorods. J. Phys. Chem. C. 2012, 116, 11722-11727.

362

(19) Dickens, P. G.; Hibble, S. J.; James, G. S. The preparation and thermochemistry of the

363

ammonium hydrogen insertion compound (NH4)0.23H0.08MoO3. Solid State Ionics 1986, 20, 213-

364

216.

365

(20) Zheng, Y.; Cheng, Y.; Wang, Y.; Bao, F.; Zhou, L.; Wei, X.; Zhang, Y.; Zheng, Q. Quasicubic

366

γ-Fe2O3 nanoparticles with excellent catalytic performance. J. Phys. Chem. B 2006, 110, 3093-

367

3097.

368

(21) Yao B.; Huang, L.; Zhang, J.; Gao, X.; Wu, J.; Cheng, Y.; Xiao, X.; Wang, B.; Li, Y.; Zhou,

369

J. Flexible transparent molybdenum trioxide nanopaper for energy storage. Adv. Mater. 2016, 28,

370

6353-6358.

371

(22) Patra, A. K.; Kundu, S. K.; Bhaumik, A.; Kim, Dukjoon. Morphology evolution of single-

372

crystalline hematite nanocrystals: Magnetically recoverable nanocatalysts for enhanced facet-

373

driven photoredox activity. Nanoscale 2016, 8 (1), 365-377.

374

(23) Zhou, L.; Yang, L.; Yuan, P.; Zou, J.; Wu, Y.; Yu, C. α-MoO3 nanobelts: A high performance

375

cathode material for lithium ion batteries. J. Phys. Chem. C 2010, 114, 21868-21872.

376

(24) Sun, Y.; Wang, J.; Zhao, B.; Cai, R.; Ran, R.; Shao, Z. Binder-free α-MoO3 nanobelt

377

electrode for lithium-ion batteries utilizing van der Waals forces for film formation and

378

connection with current collector. J. Mater. Chem. A 2013, 1, 4736-4746.


21


Page 22 of 24

379

(25) Dong, Y.; Xu, X.; Li, S.; Han, C.; Zhao, K.; Zhang, L.; Niu, C.; Huang, Z.; Mai, L.

380

Inhibiting effect of Na+ pre-intercalation in MoO3 nanobelts with enhanced electrochemical

381

performance. Nano Energy 2015, 15, 145-152.

382

(26) Zhou, C. X.; Wang, Y. X.; Yang, L. Q.; Lin, J. H. Syntheses of hydrated molybdenum

383

bronzes by reduction of MoO3 with NaBH4. Inorg. Chem. 2001, 40, 1521-1526.

384

(27) Slade, R. C. T.; Pressman, H. A. Neutron scatteringt investigation of hydrogenic species in

385

the ammonium molybdenum bronze (NH4)0.24H0.03MoO3, J. Mater. Chem. 1994, 4(4), 501-508.

386

(28) Lim, A. R.; Han, T. J.; Jung, J. K.; Park, H. M. 1H spin-lattice relaxation in a NH4HSO4

387

single crystal. J. Phys. Soc. Jpn 2002, 71, 2268-2270.

388

(29) Thøgersen, J. R.; Slabiak, T.; White, N. Ammonium bisulphate inhibition of SCR catalysts.

389

Frederikssund: Haldor Topsoe Inc, 2007.

390

(30) Wang, X.; Nesper, R.; Villevieille, C.; Novák, P. Ammonolyzed MoO3 nanobelts as novel

391

cathode material of rechargeable Li-ion batteries. Adv. Energy Mater. 2013, 3, 606-614.

392

(31) Lapujoulade, J.; Perreau, J.; Kara, A. The thermal attenuation of elastic scattering of helium

393

from copper single crystal surfaces. Surf. Sci. 1983, 129, 59-78.

394

(32) Lunk, H. J.; Hartl, H.; Hartl, M. A.; Fait, M. J. F.; Shenderovivh, I. G.; Feist, M.; Frisk, T.

395

A.; Daemen, L. L.; Mauder, D.; Eckelt, R.; Gurinov, A. A. “Hexagonal molybdenum trioxide”—

396

Known for 100 years and still a fount of new discoveries. Inorg. Chem. 2010, 49, 9400-9408.

397

(33) Ratcliffe, C. I.; Ripmeester, J. A.; Tse, J. S. NMR chemical shifts of dilute 1H in inorganic

398

solids. Chem. Phys. Lett. 1985, 120, 427-432.

399

(34) Casagrande, L.; Lietti, L.; Nova, I.; Forzatti, P.; Baiker, A. SCR of NO by NH3 over TiO2-

400

supported V2O5–MoO3 catalysts: Reactivity and redox behavior. Appl. Catal. B. 1999, 22 (1),

401

63-77.


22

Page 23 of 24


402

(35) Kosaki, Y.; Miyamoto, A.; Murakami, Y. Oxidation of ammonia with lattice oxygen of metal

403

oxides by pulse reaction technique. B. Chem. Soc. Jpn. 1979, 52, 617-618.


23


Page 24 of 24

Table of Contents


24

MoO3 against

Recommend Documents