Outstanding performance of recyclable amorphous MoS3 supported

Subscriber access provided by UNIV OF LOUISIANA

Remediation and Control Technologies

Outstanding performance of recyclable amorphous MoS3 supported on TiO2 for capturing high concentrations of gaseous elemental mercury: Mechanism, kinetics and application Jian Mei, Chang Wang, Lingnan Kong, Xiaoli Liu, Qixing Hu, Hui Zhao, and Shijian Yang Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.9b00464 • Publication Date (Web): 22 Mar 2019 Downloaded from http://pubs.acs.org on March 22, 2019

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

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

Environmental Science & Technology

1

Outstanding performance of recyclable amorphous MoS3 supported

2

on TiO2 for capturing high concentrations of gaseous elemental

3

mercury: Mechanism, kinetics and application

4

Jian Mei, Chang Wang, Lingnan Kong, Xiaoli Liu, Qixing Hu, Hui Zhao, Shijian Yang *

5

Jiangsu Key Laboratory of Anaerobic Biotechnology, School of Environment and Civil

6

Engineering, Jiangnan University, Wuxi 214122, P. R. China

7 8 9

1

ACS Paragon Plus Environment


10

Abstract:

11

Hg0 capture by sorbents was a promising technology to control Hg0 emission from coal-fired

12

power plants and smelters. However, the design of a high performance sorbent and the predicting

13

of the extent of Hg0 adsorption were both extremely limited due to the lack of adsorption kinetics

14

and structure-activity relationship. In this work, the adsorption kinetics of gaseous Hg0 onto

15

MoS3/TiO2 was investigated and kinetic parameters were obtained by fitting breakthrough curves.

16

According to the kinetic parameters, the removal efficiency, the adsorption rate and the capacity for

17

Hg0 capture were accurately predicted. Meanwhile, the structure-activity relationship of metal

18

sulfides for gaseous Hg0 adsorption was built. The chemical adsorption rate of gaseous Hg0 was

19

found to mainly depend on the amount of surface adsorption sites available for the physical

20

adsorption of Hg0, the amount of surface S22- available for Hg0 oxidation and gaseous Hg0

21

concentration. As MoS3/TiO2 showed a superior performance for capturing high concentrations of

22

Hg0 due to the large number of surface adsorption sites for the physical adsorption of gaseous Hg0,

23

it has promising applications in recovering Hg0 from smelting flue gas.

24

2


Page 2 of 30

Page 3 of 30


25

Table of Content

26 27

3



28

Page 4 of 30

1. Introduction

29

Due to the toxicity of Hg and its bioaccumulation in ecosystems, a global and legally-binding

30

treaty (the Minamata Convention) was implemented to prevent Hg emission into the enviroment.1

31

Hg-containing compounds, especially HgS, are often associated with coal and metal sulfide ores in

32

the earth; Hg is typically emitted during coal combustion and nonferrous metal smelting.2-4

33

Therefore, nonferrous metal smelters and coal-fired power plants are considered important regulated

34

sources in the Minamata Convention.3, 5-7

35

Hg species in coal-fired flue gas and smelting flue gas mainly present as Hgp (particulate-bound

36

mercury), Hg2+ (oxidized mercury) and Hg0 (elemental mercury).8 Hgp and Hg2+ can be removed by

37

air pollution control devices (APCDs) in modern smelters and coal-fired power plants.9, 10 However,

38

gaseous Hg0, i.e., the main Hg species emitted by coal-fired power plants and smelters to the

39

atmosphere, cannot be removed due to its high insolubility and volatility.2 Meanwhile, Hg0

40

concentration in smelting flue gas (~30 mg m-3) is 2-3 orders of magnitudes higher than that in coal-

41

fired flue gas (~100 μg m-3) as the Hg content in metal sulfides is much higher than that in coal 9-11.

42

Therefore, there is a great demand to control Hg0 emission to smelting flue gas.

43

Hg0 capture by sorbents is a promising emission control technology.12, 13 Current sorbents can be

44

classified into three categories: carbon-based sorbents,14, 15 metal oxides 16 and metal sulfides.17 The

45

adsorption of Hg0 onto carbon-based sorbents and metal oxides is remarkably inhibited by SO2,18,

46

19

47

smelting flue gas can reach ~5%, which is approximately 2 orders of magnitudes larger than that in

48

coal-fired flue gas.11 Recently, a series of metal sulfides with an excellent resistance to SO2, for

49

example nanosized CuS,20 nanosized ZnS,21-23 natural derived pyrrhotite,17 H2S-pretreated Fe-Ti

50

spinel,24 and CuS/TiO2,11 have been developed as sorbents for Hg0 capture.

which is inevitably present in coal-fired flue gas and smelting flue gas. The SO2 concentration in

51

In many studies, the Hg0 removal efficiency is reported to describe the sorbent’s performance for

52

Hg0 capture. However, experiment conditions for Hg0 capture, for example gas hourly space

53

velocity (i.e., GHSV), adsorption time and gaseous Hg0 concentration in the inlet, are multifarious,

54

which are closely related to removal efficiency. Therefore, it is difficult to compare the Hg0

55

adsorption performance of sorbents reported in different studies.8,

56

adsorption kinetic parameters, which reflect the intrinsic properties of sorbents for gaseous Hg0 4


12, 13, 16, 22, 25, 26

Therefore,

Page 5 of 30


57

adsorption, are suggested. Based on these kinetic parameters, removal efficiency, adsorption rate

58

and capacity for Hg0 capture can be predicted. Furthermore, the structure-activity relationship of

59

sorbents for Hg0 capture can be built, which will be great benefit to further improve sorbent’s

60

performance for Hg0 capture. However, only a few studies have focused on the adsorption kinetics

61

of gaseous Hg0 onto sorbents,19, 20, 27 and the kinetic parameters were hardly obtained.

62

MoSx (MoS2 or MoS3), a typical layered transition metal dichalcogenides (TMDCs), have

63

received great interest as catalysts for hydrodesulfurization,28 hydrogen evolution,29,

64

reduction,31 and gas adsorption. 32 In this work, amorphous MoS3 supported on TiO2 was explored

65

as a recyclable sorbent for gaseous Hg0 capture. Based on the mechanism of Hg0 adsorption onto

66

MoS3/TiO2, an adsorption kinetic model was built, and kinetic parameters were obtained by fitting

67

breakthrough curves. The kinetic study showed that the adsorption rate of gaseous Hg0 onto metal

68

sulfides depended on the amount of surface adsorption sites for the physical adsorption of gaseous

69

Hg0, the amount of surface S22- for Hg0 oxidation and gaseous Hg0 concentration. Because

70

MoS3/TiO2 had a large amount of surface adsorption sites for the physical adsorption of gaseous

71

Hg0 due to its layered structure (S-Mo-S2), it showed excellent Hg0 capture performance. MoS3/TiO2

72

exhibited an advantage for capturing high concentration of Hg0, and therefore, showed promising

73

applications in recovering gaseous Hg0 from smelting flue gas.

74

2.

75

2.1 Sample preparation

30

CO2

Experimental Section

76

MoO3/TiO2 (5 wt.% MoO3) was prepared following the wet impregnation method using Degussa

77

P25 TiO2 as a support and ammonium molybdate as a precursor. After drying at 110 oC for 12 h,

78

the sample was calcined in air at 500 °C for 3 h. Then, MoS3/TiO2 (i.e., amorphous MoS3 supported

79

on P25 TiO2) was obtained by the sulfuration of MoO3/TiO2 with H2S/N2 (600 ppm) at 300 °C for

80

60 min with a GHSV of 1.2×105 cm3 g-1 h-1.33-35 Meanwhile, CuS supported on P25 TiO2 (i.e.,

81

CuS/TiO2), which has a superior performance for Hg0 capture,11 was prepared as a comparison.

82

2.2 Activity assessment

83

Hg0 adsorption was conducted on a fixed-bed microreactor with a reaction temperature of 40-100

84

°C. The GHSV for Hg0 capture was approximately 7.2×105 cm3 g-1 h-1 with 300 mL min-1 of N2 and

85

25 mg of sorbent (40-60 mesh). Gaseous Hg0 concentration was determined online by a cold vapor 5



Page 6 of 30

86

atomic absorption spectrophotometer (CVAAS, Lumex R-915+). To investigate the effect of

87

gaseous Hg0 concentration in the inlet on Hg0 adsorption, the concentration varied from

88

approximately 280 to 4400 μg m-3. A stable concentration of Hg0 was provided by an Hg permeation

89

tube (provided by Greencalm Instruments of Suzhou, China) and the Hg0 concentration was adjusted

90

by changing the temperature of water bath. To investigate the influences of chemical components

91

in the flue gas on Hg0 adsorption, 1% SO2, 8% H2O and/or 7% O2 were simulated in the flue gas.

92

Furthermore, the recyclability of MoS3/TiO2 for Hg0 capture was investigated. Spent MoS3/TiO2

93

was first thermally treated under air atmosphere (GHSV=1.5×104 h-1) at 400 oC for 40 min to desorb

94

adsorbed Hg0 and to transform MoS3/TiO2 back to MoO3/TiO2. Then, the obtained MoO3/TiO2 was

95

re-sulfurated (the same as the sulfuration processing) for another cycle of Hg0 adsorption.

96

2.3 Characterization

97

XPS (X-ray photoelectron spectra), XRD (X-ray diffraction) patterns, and BET (Brunauer-

98

Emmett-Teller) surface areas were measured using an X-ray photoelectron spectroscope

99

(ESCALAB 250Xi, Thermo), an X-ray diffractometer (AXS D8 Advance, Bruker), and a nitrogen

100

adsorption

101

programmed desorption of Hg) profiles in N2 and in air of MoS3/TiO2 after Hg0 capture were

102

performed on the microreactor at a heating rate of 10 oC min-1 with a GHSV of 2.0×106 cm3 g-1 h-1.

103

TPO (temperature programmed oxidation) profiles of MoS3/TiO2 and CuS/TiO2 were also

104

performed on the microreactor in air at a heating rate of 10 oC min-1 with a GHSV of 1.2×105 cm3

105

g-1 h-1. Gaseous SO2 concentration in the outlet was measured online by an infrared spectrometer

106

(ANTARIS, IGS Analyzer, Thermo SCIENTIFIC).

107

3. Results and discussion

108

3.1 Performance of Hg0 adsorption onto MoS3/TiO2

apparatus

(Autosorb-1,

Quantachrome),

respectively.

Hg-TPD

(temperature

109

The breakthrough curves of Hg0 adsorption onto MoS3/TiO2 under N2 atmosphere with gaseous

110

Hg0 concentration in the range of 280 to 4400 μg m-3 are shown in Figure 1.The removal efficiency

111

(i.e., η) of Hg0 adsorption onto MoS3/TiO2 gradually decreased as gaseous Hg0 concentration in the

112

inlet increased from 280 to 4400 μg m-3. Meanwhile, Figure 1 also indicates that the rate of Hg0

113

adsorption onto MoS3/TiO2 in 3 h generally increased with the increase of gaseous Hg0

114

concentration from 280 to 4400 μg m-3. Figure S1 in the Supporting Information compares the 6


Page 7 of 30


115

breakthrough curves of Hg0 adsorption onto MoS3/TiO2 and CuS/TiO2. The adsorption rate of

116

gaseous Hg0 onto MoS3/TiO2 decreased remarkably in the first 50 min (Stage I) and then

117

approximately reached a steady-state (Stage II). These phenomena were not consistent with the

118

breakthrough curves of Hg0 adsorption onto CuS/TiO2 (Figure S1b in the Supporting Information).11

119

The adsorption rate of Hg0 onto CuS/TiO2 decreased remarkably during the adsorption and the

120

breakthrough curve did not contain Stage II (i.e., the steady-state). Although the adsorption rate of

121

Hg0 onto MoS3/TiO2 in Stage I was less than that onto CuS/TiO2, the average adsorption rate of Hg0

122

onto MoS3/TiO2 in 360 min was much higher than that onto CuS/TiO2 (Figure S1 in the Supporting

123

Information).

124

3.2 Mechanism of Hg0 adsorption onto MoS3/TiO2

125

The XPS spectra of MoS3/TiO2 after Hg0 adsorption in the spectral region of Hg 4f mainly

126

centered at 104.2 and 100.2 eV (Figure 2a), which were not consistent with that of Hg0 at 100.0 eV

127

and those of HgO at 105.0 and 101.0 eV.11, 36, 37 Hg-TPD in N2 of MoS3/TiO2 after Hg0 capture

128

exhibited a Hg0 desorption peak at approximately 220 oC (Figure 3a). However, a new Hg0

129

desorption peak appeared at 350 oC when Hg-TPD was performed in air. The Hg-TPD profiles were

130

consistent with the characteristics of HgS decomposition.11 The Hg0 desorption peak at

131

approximately 220 oC was attributed to HgS decomposition, whereas that at approximately 350 oC

132

was assigned to HgSO4 decomposition, which resulted from HgS oxidation. These suggest that the

133

Hg species adsorbed on MoS3/TiO2 was mainly HgS. Therefore, the adsorption of Hg0 onto

134

MoS3/TiO2 involved chemical adsorption, which generally followed the Mars-Maessen mechanism

135

(i.e., gaseous Hg0 first physically adsorbed on the surface, which was then oxidized by the active

136

surface S species to HgS). 12, 17 To ascertain the active S species for Hg0 oxidation, XPS spectra of

137

MoS3/TiO2 before and after Hg0 adsorption were compared (Figure 2).

138

The Mo 3d binding energies on MoS3/TiO2 mainly centered at approximately 235.0, 232.2, 231.8

139

and 229.0 eV (Figure 2b). The binding energies at 231.8 and 235.0 eV were assigned to Mo 3d 3/2

140

and Mo 3d 5/2 of MoO2, respectively.38 The binding energies at 229.0 and 232.2eV were assigned

141

to Mo 3d 3/2 and Mo 3d5/2 of MoSx, respectively. 39 Furthermore, the binding energy at 226.1 eV

142

was assigned to S 2s of S2- or S22-.28 The S 2p binding energies on MoS3/TiO2 mainly centered at

143

approximately 169.6, 168.5, 164.4, 162.9 and 161.6 eV (Figure 2c). The binding energies at 169.6 7



Page 8 of 30

144

and 168.5 eV were assigned to SO42-,11 and that at 164.4 eV was assigned to S0.17 The binding

145

energies at 162.9 and 161.6 eV were attributed to S22- in MoS3 and S2- in MoS3 or MoS2, respectively.

146

28, 33, 39

147

2:1, suggesting that the molybdenum sulfide on MoS3/TiO2 was mainly MoS3.

The ratio of the area corresponding to S22- to that corresponding to S2- was approximately

148

After the adsorption of approximately 4400 μg m-3 of Hg0 at 60 oC for approximately 6 h, no new

149

binding energies appeared on MoS3/TiO2 in the spectral regions of Mo 3d and S 2p (Figures 2d and

150

2e). The percentages of Mo and S species on MoS3/TiO2 before and after Hg0 adsorption resulting

151

from the XPS analyses are shown in Table 1. The percentage of S22- on MoS3/TiO2 obviously

152

decreased after Hg0 adsorption, whereas, the percentage of S2- on MoS3/TiO2 obviously increased.

153

Furthermore, the percentages of other S species including S0 and SO42- did not vary remarkably.

154

These suggest that S22- on MoS3/TiO2 was the active S species for the oxidation of physically

155

adsorbed Hg0, which was consistent with Hg0 capture by pyrrhotite,17 CuS/TiO2

156

pretreated Fe-Ti spinel.24,

157

approximately described as (see Figure S2 in the Supporting Information):12

158

Hg 0 (g)  Hg 0 (ad)

(1)

159

S22 +Hg 0 (ad)  S2 +HgS

(2)

160

3.3 Kinetics of Hg0 adsorption

161

3.3.1 Primary kinetic model

162 163

40

11

and H2S-

Therefore, the chemical adsorption of Hg0 onto MoS3/TiO2 can be

According to Reaction 2, the kinetics of Hg0 adsorption can be described as: 27



d[Hg 0(g) ] dt



d[Hg 0(ad) ] dt



d[S22 ]  k[S22 ][Hg 0 (ad) ] dt

(3)

164

where k, [S22-] and [Hg0(ad)] are the kinetic constant of Reaction 2, and the amounts of S22- and

165

Hg0 physically adsorbed on the surface, respectively. The physical adsorption of Hg0 was related to

166

the number of adsorption sites on the surface. Therefore, [Hg0(ad)] can be approximately described

167

as the product of the amount of surface adsorption sites for the physical adsorption of Hg0 (i.e., [φ])

168

and the coverage ratio of the adsorption sites by Hg0 physically adsorbed (i.e., θ). Then, Equation 3

169

can be transformed as:

170



d[Hg 0(g) ] dt



d[Hg 0(ad) ] dt



d[S22 ]  k[S22 ] [ ] dt 8


(4)

Page 9 of 30


171

During Hg0 adsorption, the amounts of adsorption sites and S22- on the surface would both

172

gradually decrease. If the amount of S22- on the surface was much higher than that of the adsorption

173

sites, the decrease of surface S22- during Hg0 adsorption can be approximately neglected. Then, the

174

amount of the adsorption sites during Hg0 adsorption (i.e., [φ]t) can be approximately described

175

as:19, 27

176

[ ]t  [ ]0 exp( k[S22 ]0  t )

177 178 179

(5)

Hence, Equation 4 can be transformed as:



d[Hg 0(g) ] dt

 k[S22 ]0  [ ]0 exp(k[S22 ]0  t )  A exp( Bt )

(6)

where,

180

A=k[S22 ]0  [ ]0

(7)

181

B =k[S22 ]0 

(8)

182

According to Equations 7 and 8, the amount of the adsorption sites on the surface can be

183

calculated as:

184

[ ]0 =A / B

185

3.3.2 Verification of the kinetic model for CuS/TiO2

(9)

186

To obtain the kinetic parameters of gaseous Hg0 adsorption, the breakthrough curves of Hg0

187

adsorption onto MoS3/TiO2 and CuS/TiO2 were fitted according to Equation 6. Hg0 adsorption onto

188

CuS/TiO2 fitted well with Equation 6, and the kinetic parameters are listed in Table 2. SO2-TPO

189

(Figure 3b) indicates that the amount of S species on CuS/TiO2 was approximately 1.35 mmol g-1.

190

The ratio of S22- on CuS/TiO2 to the total S species was approximately 0.33.11 Therefore, the amount

191

of S22- on CuS/TiO2 was approximately 0.22 mmol g-1, which was 2.9-7.3 times the amount of

192

adsorption sites on CuS/TiO2 (i.e., A/B in Table 2). This indicates that the amount of S22- on

193

CuS/TiO2 was much higher than that of adsorption sites on CuS/TiO2 for the physical adsorption of

194

gaseous Hg0, which was consistent with the precondition of Equation 5. According to Equation 6

195

and the kinetic parameters of CuS/TiO2 in Table 2, 90% of the breakthrough point of Hg0 adsorption

196

onto CuS/TiO2 was predicted to appear at approximately 635, 524, 410 and 300 min at 40, 60, 80

9



Page 10 of 30

197

and 100 oC, respectively. Furthermore, the capacity of CuS/TiO2 for Hg0 capture (i.e., Q) can be

198

calculated as:

199

Q   k[S22 ]0  [ ]0 exp(k[S22 ]0  t )dt  [ ]0



(10)

0

200

Therefore, the capacities of CuS/TiO2 for Hg0 capture were predicted as 15.2, 12.3, 9.0 and 6.1

201

mg g-1 at 40, 60, 80 and 100 oC, respectively. To verify the kinetic model of Hg0 adsorption onto

202

CuS/TiO2, Hg0 adsorption onto CuS/TiO2 at 100 oC was performed over 10 h (Figure S3 in the

203

Supporting Information). The breakthrough curve indicates that Hg0 adsorption onto CuS/TiO2 at

204

100 oC had nearly reached saturated adsorption. Approximately 90% of the breakthrough point was

205

reached at 300 min, and the capacity of CuS/TiO2 for Hg0 capture was approximately 6.0 mg g-1;

206

these were both close to their predicted values from the kinetic model.

207

3.3.3 Modification of the kinetic model for MoS3/TiO2

208

However, Hg0 adsorption onto MoS3/TiO2 did not follow Equation 6. The relationship between

209

the rate of Hg0 adsorption onto MoS3/TiO2 and the adsorption time was plotted in logarithmic

210

coordinates (Figure S4 in the Supporting Information). If the kinetics of Hg0 adsorption followed

211

Equation 6, there would be an excellent linear relationship in Figure S4. However, Figure S4 seems

212

to show a superposition of two types of linear relationships. This suggests two types of adsorption

213

sites on MoS3/TiO2 for the physical adsorption of Hg0 (i.e., selective adsorption sites and non-

214

selective adsorption sites). 41-43 Therefore, Equation 6 can be transformed as:

215



216



d[Hg 0(g) ] dt

s

 k[S22 ]0 1[s ]0 exp(k1[S22 ]0 t )

(11)

n

 k[S22 ]0  2 [n ]0 exp(k 2 [S22 ]0 t )

(12)

d[Hg 0(g) ] dt

217

where [φs]0, [φn]0, θ1 and θ2 are the amounts of selective adsorption sites and non-selective

218

adsorption sites on the surface, and the coverage ratios of selective adsorption sites and non-

219

selective adsorption sites by physically adsorbed Hg0, respectively.

220

Then, the kinetic of Hg0 adsorption can be described as:

 221

d[Hg 0(g) ]



d[Hg 0(g) ]

s 

d[Hg 0(g) ] n

dt dt dt 2 2  k[S2 ]0 1[s ]0 exp(k1[S2 ]0 t )+k[S22 ]0  2 [ n ]0 exp(k 2 [S22 ]0 t ) =A1 exp( B1t )  A2 exp( B2t ) 10


(13)

Page 11 of 30


222

Where,

223

A1 =k[S22 ]0 1[s ]0

(14)

224

A2 =k[S22 ]0  2 [n ]0

(15)

225

B1 =k[S22 ]0 1

(16)

226

B2 =k[S22 ]0  2

(17)

227

According to Equations 14-17, the amounts of selective adsorption sites and non-selective

228

adsorption sites on the surface can be calculated as:

229

[s ]0 =A1 / B1

(18)

230

[n ]0 =A2 / B2

(19)

231

To obtain the kinetic parameters of Hg0 adsorption onto MoS3/TiO2, the breakthrough curves

232

were fitted. The correlation coefficients in Table 2 show that the breakthrough curves of Hg0

233

adsorption onto MoS3/TiO2 fitted well with Equation 13.

234

SO2-TPO (Figure 3b) indicates that the amount of S species on MoS3/TiO2 was approximately

235

0.55 mmol g-1. XPS analysis exhibits that the ratio of S22- on MoS3/TiO2 to the total of S species

236

was approximately 0.52 (Table 1). Therefore, the amount of S22- on MoS3/TiO2 was approximately

237

0.14 mmol g-1, which was 2 orders of magnitude larger than the amount of selective adsorption sites

238

on MoS3/TiO2 (i.e., A1/B1 in Table 2). This was consistent with the precondition of Equation 11.

239

However, A2/B2 (i.e., the amount of non-selective adsorption sites on MoS3/TiO2 predicted from the

240

kinetic model) was approximately equal to the amount of S22- on MoS3/TiO2, which violated the

241

precondition of Equation 12.

242

If the amount of non-selective adsorption sites on MoS3/TiO2 was much higher than the amount

243

of S22- on the surface, the decrease of non-selective adsorption sites on the surface can be

244

approximately neglected during the adsorption. Therefore, Equation 12 was transformed as:

245



246 247

d[Hg 0(g) ] dt

n

 k[S22 ]0  2 [ n ]0 exp(k 2 [ n ]0 t )  A2 exp( B2t )

(20)

Where,

A2 =k[S22 ]0  2 [n ]0

(21)

11



248

B2 =k[n ]0  2

249 250

Page 12 of 30

(22)

Then,

A2 / B2 =[S22 ]0

(23)

251

Table 2 shows that A2/B2 of MoS3/TiO2 was approximately equal to the amount of S22- on

252

MoS3/TiO2 (i.e., approximately 0.14 mmol g-1), which was consistent with Equation 23. This

253

demonstrates that the amount of non-selective adsorption sites on MoS3/TiO2 was much higher than

254

that of S22- on the surface.

255

Then, the kinetic equation of Hg0 adsorption onto MoS3/TiO2 was revised as:

 256

d[Hg 0(g) ]



d[Hg 0(g) ]

s 

d[Hg 0(g) ] n

dt dt dt 2 2  k[S2 ]0 1[s ]0 exp(k1[S2 ]0 t )+k[S22 ]0  2 [ n ]0 exp(k 2 [ n ]0 t )

(24)

=A1 exp(B1t )  A 2 exp(B2t ) 257

According to B1 in Table 2, achieving 90% of the saturated adsorption onto the selective

258

adsorption sites with a Hg0 concentration of approximately 280 μg m-3 required approximately 550-

259

700 min. However, only 41-57 min were required when the gaseous Hg0 concentration increased to

260

approximately 4400 μg m-3. According to B2 in Table 2, achieving 90% of the saturated adsorption

261

onto the non-selective adsorption sites with a Hg0 concentration of approximately 280 μg m-3

262

required longer than 380 h, and required 30-45 h with a Hg0 concentration of approximately 4400

263

μg m-3. Because the chemical adsorption of gaseous Hg0 onto selective adsorption sites only

264

continued for a short time compared with the chemical adsorption of gaseous Hg0 onto non-selective

265

adsorption sites, the decreased S22- within the short time due to Hg0 adsorption onto the non-selective

266

adsorption sites on MoS3/TiO2 can be approximately neglected. Therefore, the chemical adsorption

267

of gaseous Hg0 onto non-selective adsorption sites and that onto non-selective adsorption sites were

268

approximately independent.

269

Table 2 shows that A2 of MoS3/TiO2 was 2.3-8.4 times that of A1 of MoS3/TiO2, suggesting that

270

the rate of chemical adsorption of Hg0 onto non-selective adsorption sites was first much higher

271

than that onto selective adsorption sites. Meanwhile, B1 of MoS3/TiO2 was one order of magnitudes

272

larger than B2, suggesting that the decrease of the rate of the chemical adsorption of gaseous Hg0

273

onto selective adsorption sites during the chemical adsorption was much remarkable than that onto 12


Page 13 of 30


274

selective adsorption sites. The kinetic model also indicates that the contribution of the chemical

275

adsorption of Hg0 onto selective adsorption sites to the rate of Hg0 adsorption was less than

Outstanding performance of recyclable amorphous MoS3 supported

Recommend Documents