Sulfur fate during lignite pyrolysis process in chemical looping

1 State Key Laboratory of Coal Combustion, School of Energy and Power. 4. Engineering, Huazhong University of Science and .... conversion of sour gas ...
0 downloads 11 Views 1MB Size
Subscriber access provided by UCL Library Services

Article

Sulfur fate during lignite pyrolysis process in chemical looping combustion environment Jinchen Ma, Chaoquan Wang, haibo zhao, and Xin Tian Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.7b03149 • Publication Date (Web): 28 Dec 2017 Downloaded from http://pubs.acs.org on January 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 free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels 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 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

2

Sulfur fate during lignite pyrolysis process in chemical looping combustion environment

3

Jinchen Ma1, ‡, Chaoquan Wang1, ‡, Haibo Zhao1*, and Xin Tian1

1

4

1

State Key Laboratory of Coal Combustion, School of Energy and Power

5

Engineering, Huazhong University of Science and Technology, Wuhan, 430074, P. R.

6

China.

7

* Email: [email protected] (Haibo Zhao)

8 9



These authors contributed equally

ABSTRACT

10

Chemical looping combustion (CLC) is a novel technology with the feature of

11

CO2 inherent separation, where the fuel is converted via lattice oxygen (instead of

12

gaseous oxygen) provided by the oxygen carrier that circulates between two

13

structurally interconnected but atmosphere-isolated reactors, i.e., fuel reactor and air

14

reactor. In the fuel reactor of in-situ gasification CLC (iG-CLC),

15

pyrolysis/gasification products of coal are oxidized by lattice oxygen in the O2-free

16

environment. Therefore, the characteristics of sulfur species evolution and distribution

17

in the coal pyrolysis products are significantly different from those in the

18

conventional combustion/gasification/pyrolysis processes.

the

19

In this study, a two-stage fluidized bed reactor was utilized to investigate the

20

reaction between the oxygen carrier and in-situ coal pyrolysis products, where the

21

coal and oxygen carrier particles are separately loaded in two reactors. In this way, the

22

influence of oxygen carrier on the coal pyrolysis process could be eliminated. As

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

23

obtained from the experiment, the distribution of sulfur species in coal pyrolysis

24

products changed significantly after being oxidized by the oxygen carrier. To be more

25

specific, the sulfur species were 70.1% H2S, 0.2% SO2, 0.8% COS and 13.1% CS2,

26

respectively, during the coal pyrolysis process in the blank experiment loaded with

27

silica sand. Whereas, the concentrations of the sulfur species (in the same order)

28

changed to 26.0%, 68.2%, 0% and 0%, respectively, once the pyrolysis products went

29

through the Fe2O3/Al2O3 oxygen carrier. The result indicates that most of the H2S,

30

COS and CS2 contents could be oxidized by the oxygen carrier to generate SO2 in the

31

CLC environment. The sulfurous gas conversion rate at the CLC experiment was

32

higher than that at blank experiment due to the fast evolution of sulfur in tar, which

33

was also converted by oxygen carrier to enhance the sulfur conversion at the CLC

34

experiment. Most of the H2S could be oxidized by the oxygen carrier to generate SO2

35

via the reaction H2S(g) + 9Fe2O3 = 6Fe3O4 + H2O(g) + SO2(g), and this has been

36

confirmed by both experiment and HSC simulation. Moreover, SEM+EDX results

37

indicated that no metallic sulfide was formed on the surface of reduced oxygen

38

carrier.

39

Keyword: chemical looping combustion, sulfur evolution, coal, pyrolysis, oxygen

40

carrier

41

1 Introduction

42

Chemical looping combustion (CLC) of coal is a novel way to realize low-cost

43

CO2 capture and cascade energy utilization at the same time1. For a CLC system, it

44

usually consists of two separated reactors: fuel reactor (FR) and air reactor (AR),

ACS Paragon Plus Environment

Page 2 of 40

Page 3 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

45

where the oxygen carrier (OC, mainly transition-state metal oxides) circulates

46

between them to transport the lattice oxygen needed for fuel conversion. During the

47

in-situ gasification CLC (iG-CLC) process, the coal undergoes two important steps

48

occurring in FR: the fast pyrolysis process of raw coal and the relatively slow

49

gasification process of residual char. The produced combustible gases (mainly CO, H2,

50

CH4, H2S, et al.) are finally oxidized by the OC to generate CO2, H2O, SO2 and other

51

by-products.

52

As almost all coals are sulfur-containing substance, the sulfur behavior is thus a

53

key issue within iG-CLC process: on the one hand, the OC could be sulfurized in the

54

presence of sulfur contaminants, resulting in the deactivation of OC2,3; on the other

55

hand, the sulfur-containing flue gases will reduce the purity of the captured CO2,

56

which demonstrates adverse effect to CO2 compression and air cleaning4. Several

57

researches have evaluated the effects of H2S on the performance of various OCs for

58

CLC or chemical looping with oxygen uncoupling (CLOU), usually using

59

H2S-containing synthesis gas as fuel. Francisco et al.5 firstly investigated the reaction

60

between Ni-based OC and H2S via HSC thermodynamic simulation and then a series

61

of experiments were conducted in a 500 Wth CLC unit. The results showed that Ni3S2

62

was formed under all experimental conditions due to the presence of H2S in CH4 in

63

the fuel reactor. Cristina et al.6 showed that the reactivity of Al2O3 supported NiO

64

oxygen carrier decreased when reacting with H2S-containing CH4 in the 500 Wth CLC

65

unit, because of the formation of nickel sulfides and sulfates. Diego et al.7

66

investigated the performance of Fe-based OC in the same 500 Wth CLC unit for

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

67

conversion of sour gas with high H2S content. The result showed that complete

68

conversion of H2S to SO2 was attained when the oxygen carrier-to-fuel ratio was

69

above 1.5 and, meanwhile the formation of iron sulfides could be avoided in FR.

70

Wang et al.8 found that Cu2S and FeS were the main sulfide products and H2S was

71

easier to react with CuO than Fe2O3 in the CLC process, using coal-derived synthesis

72

gas (also with H2S contained) as fuel. Tian et al.9 investigated the performance of

73

different types of OCs in a thermogravimetric analyzer (TGA), using coal-derived

74

synthesis gas mixed with 4042 ppm of H2S as fuel. The results showed that the OC

75

would first react with CO and H2 and then react with H2S to form metal sulfides. Shen

76

et al.10 concluded that Ni3S2 was the only metal sulfide when using NiO-based OC

77

and it was completely reversible in redox process. The reaction rate between NiO and

78

H2S was higher than that of NiO and CO. Gu et al.11 found that the iron ore could be

79

sulfurized in the presence of H2S, resulting in the formation of FeS, and the released

80

sulfurous gases mainly consisted of COS, SO2 and CS2. Moreover, they claimed that

81

H2S was mainly oxidized by Fe2O3 rather than by Fe3O4 or FeO. Ewelina et al.12

82

investigated the stability of Fe-Mn OC during CLC redox cycles with H2S-containing

83

coal-derived synthesis gas. It was found that the presence of H2S would significantly

84

decrease the reduction rates, but the oxidation rates of the reduced oxygen carrier

85

increased. Solunke et al.13 studied the redox kinetics of nanostructured Ni-BHA and

86

Cu-BHA with H2S-containing synthesis gas in TGA. The results showed that the

87

oxygen capacity was increased by the partial sulfurization of the support. Wang et

88

al.14 investigated the effects of oxygen excess numbers and temperature on the sulfur

ACS Paragon Plus Environment

Page 4 of 40

Page 5 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

89

evolution using HSC simulation.

90

In a typical coal-derived iG-CLC process, coal is first pyrolyzed and then

91

gasified, and the resultant combustible gases are finally reduced by the OC. As can be

92

speculated, the sulfur fate may be quite different during coal pyrolysis and

93

gasification processes in CLC environment (the absence of gaseous O2 and the

94

presence of OC) when compared with traditional combustion/gasification/pyrolysis

95

processes. With respect to the current research on the reaction between H2S and OC in

96

CLC, most of them are investigated by using H2S-containing synthesis gas as fuel9, 12,

97

13

98

rather complicated in the real CLC process. Therefore, a careful identification and full

99

understanding of the sulfur distribution and evolution characteristics during the

100

. Actually, the reaction characteristics and the evolution of sulfurous gases were

coal-derived CLC process are certainly required.

101

The objective of this work was to investigate the characteristics of sulfur species

102

(H2S, SO2, COS, CS2) evolution and distribution during the coal pyrolysis process

103

(which, as the first step of iG-CLC of coal, occurs very rapidly with a large proportion

104

of sulfur in coal released) in a two-stage fluidized bed reactor. Well-organized

105

experiments were conducted for the comparison of the CLC process and simple coal

106

pyrolysis process (using OC and silica sand as bed materials in the second reactor,

107

respectively) to identify the sulfur behaviors. By using the two-stage fluidized bed

108

reactor, the reaction between oxygen carrier and in-situ coal pyrolysis gases

109

(generated in the first pyrolysis reactor) can be directly evaluated and, meanwhile

110

minimizing the influence of oxygen carrier on the coal pyrolysis process since coal

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

111

and oxygen carrier particles are separately loaded in two reactors. The sulfur balance

112

was calculated based on the difference of sulfur in coal and in char. The interaction of

113

OC and sulfurous gases (H2S, COS, CS2 and SO2) was simulated using the

114

HSC-Chemistry software 6.0 to further understand the sulfur evolution in experiments.

115

Finally, the sulfur migration path in CLC of in-situ coal pyrolysis products was

116

deduced.

117

2 Experimental

118

2.1 Materials

119

A typically Chinese low-rank coal (Xiaolongtan, XLT in short) is used as solid

120

fuel. The XLT coal was dried at 105 oC in air for 10 h and sieved in range of 100-350

121

µm. The relevant XLT char was prepared in Ar atmosphere and heated to 950 oC at 15

122

o

123

prepared in the tube furnace at constant temperature of 950 oC and Ar atmosphere) are

124

listed in Table 1. The Fe-based OCs in the range of 200-300 µm were prepared by

125

sol-gel method15 with 54 wt.% Fe2O3 and 46 wt.% Al2O3.

126

2.2 Experimental procedure

C/min. The proximate and ultimate analyses of XLT coal and XLT char (the char was

127

The two-stage fluidized bed reactor (shown in Fig. 1) consists of a pyrolysis

128

reactor, where the coal releases the volatile contents, and a CLC fuel reactor, in which

129

the volatile-containing gases without cooling process are introduced to fluidize either

130

OC or silica sand. In this way, the mixing of OC and coal is avoided, which eliminates

131

the influence of OC on the coal pyrolysis process. The inner diameter of both reactors

ACS Paragon Plus Environment

Page 6 of 40

Page 7 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

132

is 35 mm, with the height of 150 mm for the coal pyrolysis reactor and 250 mm for

133

the CLC fuel reactor, respectively. The pore diameter of the distribution plate is 74

134

µm, located in the CLC fuel reactor, which could prevent the entrainment of coal/ash

135

particles into the CLC fuel reactor. For each experiment, 0.232g XLT coal was first

136

fed into the first-stage pyrolysis reactor, using Ar as fluidizing gas. Then the in-situ

137

pyrolysis products were rapidly carried into the CLC fuel reactor by Ar, where OC (or

138

silica sand as blank experiment) was loaded to investigate the distribution and

139

evolution characteristics of sulfurous gases during the CLC process of pyrolysis

140

products (or the O2-free coal pyrolysis process). An on-line Fourier Transform

141

Infrared Spectroscopy (FTIR, GASMET DX4000) was employed to measure the

142

contents of sulfurous gases (SO2, COS and CS2). The infrared absorption band of

143

FTIR was in the range of 600 - 4250 cm-1. The infrared peaks of COS, CS2 and SO2

144

were at 2051, 1527, and 1371 cm-1, respectively. H2S was measured by the online H2S

145

detector (MS600-H2S). The online gas analyzer (Gasboard Analyzer 3100) was used

146

to determine the concentrations of CO2, CO, CH4, O2 and H2, within the range of 0 -

147

10 vol.% for CO, CH4, O2 and H2 and 0-100 vol.% for CO2. The measurement

148

accuracy was 1% of full scale. To guarantee the reliability of the experiment, the

149

FTIR and the H2S detector were calibrated. The range of H2S, SO2, COS and CS2 was

150

500 ppmv with the accuracy lower than 1% of full scale, respectively. Moreover, to

151

avoid the interference of the H2O infrared peak, the outlet gas was dehydrated by the

152

electric condenser and the tar was removed by the tar removing unit before flowing

153

into the FTIR.

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

154

There were three steps in each experiment. The first step is used Ar as the

155

fluidizing and carrier gas during the coal pyrolysis process, with the flow rate of 2000

156

mL/min, corresponding to 8 Umf of OC (Umf, the minimum fluidized gas velocity is

157

calculated as 0.018 m/s at 950 oC and Ar atmosphere). The valve 1 was switched on

158

and the valve 2 was closed. Then the in-situ volatile is carried to the second CLC fuel

159

reactor loaded 40 g oxygen carrier. The outlet gas is measured at the out of the CLC

160

fuel reactor. After the pyrolysis process, the second step is to introduce 1600 mL/min

161

Ar and 400 mL/min O2 into the first reactor to burn out the residue char produced

162

during step 1. During the second step, the gas was cut off into the CLC fuel reactor by

163

switching on the valve 2 while closing the valve 1. The char was totally combusted in

164

the pyrolysis reactor and the outlet gas was exhausted through the outlet of the

165

pyrolysis reactor until the CO2 gas concentration was zero. The third step was

166

switched the mixture gas of 1600 mL/min Ar and 400 mL/min O2 to the CLC fuel

167

reactor to oxidize the oxygen carrier by switching on the valve 1 while closing the

168

valve 2.

169

The blank experiment was also conducted to evaluate the sulfur distribution

170

during the O2-free coal pyrolysis process. By comparing the results of the blank

171

experiment and the CLC experiment, it is possible to explore the effect of the

172

presence of Fe-based oxygen carrier on the sulfur distribution of coal pyrolysis

173

products. All the experiments have been repeated three times to ensure data reliability.

174

2.3 Data evaluation

175

The total volume of outlet gas from the CLC fuel reactor ( Fout ) was calculated

ACS Paragon Plus Environment

Page 8 of 40

Page 9 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

176

177

by Ar balance. Fout =

FAr 1 − ∑ yi

(1)

178

where FAr was the flow rate of Ar introducing into the pyrolysis reactor, yi was the

179

volume fraction of gas i (except for Ar) in the outlet of the CLC fuel reactor.

180

The volumes of the generated sulfurous gases (SO2, CS2, COS and H2S) were

181

calculated as:

182

Yi = ∫

t total

yi × Fout dt

t0

183

(2)

The mass of sulfur content in the volatile ( mS-V ) was determined by:

184

mS-V =mcoal ⋅ wS-C − mcoal ⋅ ychar ⋅ wS-Char

185

here mcoal was the mass of coal fed into the pyrolysis reactor, ychar was the char yield

186

( ychar = 0.43 through weighting the raw coal and the residue char), wS-C and wS-Char

187

were the mass fraction of sulfur in coal and char (shown in Table 1), respectively.

(3)

188

The sulfurous gases conversion ( X S ) was defined as the mole of sulfurous gases

189

(SO2, H2S, COS and CS2) released from the beginning to time t to the sulfur content

190

in XLT coal volatile, as seen in eq. 4, and the sulfurous gases conversion rate ( xS ) was

191

the differential of sulfurous gases conversion (as seen in eq. 5).

∫ =

192

XS

193

xS =

194 195

t

t0

Fout ( yH 2S + ySO2 + yCOS + 2 × yCS2 )dt 22.4 mS-V 32

dX S dt

(4) (5)

The mass balance of sulfur species during the pyrolysis process was calculated as,

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Yi ⋅ M i ⋅ 100% 22.4 ⋅ mS-V

Page 10 of 40

196

wi =

197

where wi and M i were the mass fraction and molecular weight of specie i,

198

respectively, and i represented SO2, H2S, COS and CS2.

199 200

The carbon conversion, X C , the carbon conversion rate, χ C and the CO2 yield, γ CO , were calculated in the same way as in reference16, as shown below: 2

∫ =

201

XC

202

χC =

203

(6)

ttotal

t0

mC 12

dX C dt

γ CO = 2

Fout ( yCO2 + yCH 4 + yCO )dt 22.4

(8)



ttotal

t0



ttotal

t0

(7)

Fout ( yCO2 )dt

Fout ( yCO2 + yCH4 + yCO )dt

204

where mC was defined as the mass of carbon content in the coal.

205

3 Results and Discussions

206

3.1 Thermodynamic simulation

(9)

207

H2S is the main sulfur-containing substance released during the coal pyrolysis

208

process (as indicated in Fig. 6 of Section 3.3). The active component, Fe2O3, in the

209

oxygen carrier could provide lattice oxygen to convert H2S into SO2, as R1. Moreover,

210

ferric sulfides may be formed by the reaction between H2S and Fe2O3 in CLC of coal

211

pyrolysis products, as R4, resulting in the degradation of the OC reactivity. The iron

212

oxide with different chemical valence could react with H2S in the fuel reactor as

213

follows11: H2S(g) + 9Fe2O3 = 6Fe3O4 + SO2(g) + H2O(g)

ACS Paragon Plus Environment

(R1)

Page 11 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

H2S(g) + 3Fe3O4 = 9FeO + SO2(g) + H2O(g)

(R2)

H2S(g) + 3FeO = 3Fe + SO2(g) + H2O(g)

(R3)

H2S(g) + 1/3Fe2O3 = 1/3Fe2S3 + H2O(g)

(R4)

H2S(g) + FeO = FeS + H2O(g)

(R5)

H2S(g) + Fe = FeS + H2(g)

(R6)

214

A thermodynamic simulation was conducted by using the HSC Chemistry 6.0,

215

basing on the minimization of Gibbs free energy, to understand the sulfur migration

216

behavior during the CLC of in-situ coal pyrolysis process. The thermodynamic

217

equilibrium constants of the reaction between the iron/iron oxides (Fe2O3, Fe3O4, FeO

218

and Fe) and H2S to form sulfurous gases (mainly SO2) and ferric sulfides (FeS and

219

Fe2S3) were first calculated within the temperature range of 700 - 1100 oC, as shown

220

in Fig. 2. It was noted that R1 was most thermodynamically favorable, which showed

221

the highest reactivity of Fe2O3 with H2S. The elevated temperature also posed a

222

positive effect on R1. It was speculated from these equilibrium constants K that the

223

possibility of the formation of iron sulfide (R4 - R6) was lower than that of formation

224

of SO2 (R1), and these equilibrium constants showed a decreased trend with the

225

increased temperature. Thus, the formation of iron sulfide will decrease at higher

226

temperature theoretically. A value of log K < 0 for R2 and R3 indicated that these

227

reactions were difficult to proceed. The results showed that the reaction between

228

Fe3O4/FeO and H2S impossibly occurred11.

229

The equilibrium compositions were simulated on the basis of 1000 mol syngas

230

gas with 371.35 mol CO, 491.85mol CO2, 113.99 mol CH4, 22.75 mol H2 and 5.06

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 40

231

mol H2S, which was determined according to the coal pyrolysis products at 950 oC (as

232

shown in Fig. 4(a) and Fig. 6(a)). A value of 1.0 of the fuel to oxygen carrier ratio is

233

defined as stoichiometric mole of synthesis gas needed to reduce Fe2O3 to Fe3O4. All

234

the chemical species considered in this calculation system are listed in Table 2.

235

Fig. 3 shows the evolution of the sulfurous species, iron species and conventional

236

gas at different fuel to oxygen carrier ratios at 950 oC. CO and H2 were almost

237

completely oxidized by the oxygen carrier when the fuel to oxygen carrier ratio was

238

less than 1.0 (as shown in Fig. 3(a)), due to the relatively excessive of lattice oxygen

239

provided by Fe2O3. The concentration of CH4 was always approximate to zero due to

240

the fully decomposition of CH4 under any conditions of the fuel to oxygen carrier

241

ratio (as shown in Fig. 3(a)). With respect to the sulfurous gases (as shown in Fig.

242

3(b)), there was only SO2 generated when the fuel to oxygen carrier ratio was less

243

than 1.0, which indicated that H2S could be completely oxidized by Fe2O3 to generate

244

SO2 via R1 at 950 oC in theory. When the fuel to oxygen carrier ratio was higher than

245

1.2, S2 was generated from the decomposition of H2S (R7) and COS was generated by

246

the reaction of S2 and CO as R8. However, S2 was decreased to zero when the H2

247

concentration increased to 0.05 vol.% at the fuel to oxygen carrier ratio of 1.8. Thus,

248

the H2 was able to inhibit the decomposition of H2S in real CLC processes5,6,17. The

249

concentration of CS2 was always close to zero at any fuel to oxygen carrier ratio. H2S(g) = H2(g) + 1/2S2(g)

(R7)

CO(g) + 1/2S2(g) = COS(g)

(R8)

ACS Paragon Plus Environment

Page 13 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

250

As shown in Fig. 3(c), Fe3O4 was the main reduction form of Fe2O3 when the

251

fuel to oxygen carrier ratio was lower than 1.0. Moreover, solid sulfur compound of

252

FeS and Fe0.877S were generated when the fuel to oxygen carrier ratio was higher than

253

1.3, which was the product of reaction between FeO and H2S (R5) (shown in Fig. 2).

254

There was no Fe2S3 generated (R4) (shown in Fig. 3(c)), which could also be

255

explained by the lower thermodynamic equilibrium constants of R4 than that of R5.

256

During the CLC process, there was usually no reduction form of Fe due to the

257

thermodynamic limitation18. Thus, there was no possibility of R6 taking place in CLC

258

even though the equilibrium constant of R6 was high.

259

3.2 Conventional gases distribution

260

The carbonaceous gases and H2 distribution during the blank experiment (loaded

261

with silica sand in the CLC reactor) are shown in Fig. 4(a), and the comparative

262

experiment was conducted to investigate the conversion of the in-situ XLT coal

263

volatile products (including the carbonaceous gases, H2 and sulfurous gases, which

264

was are generated in the pyrolysis reactor and carried into the second CLC fuel

265

reactor without cooling down process) by oxygen carrier, as shown in Fig. 4(b). In

266

each experiment, either 40 g silica sand or 40 g oxygen carrier was loaded in the CLC

267

fuel reactor and 0.232 g XLT coal was fed to the pyrolysis reactor, corresponding to

268

the fuel to OC ratio of 0.5 (i.e., the OC to fuel ratio of 2.0). All the experiments were

269

repeated three times at a typical CLC operational temperature, 950 oC.

270

As shown in Fig. 4, the conventional gases during the pyrolysis process were H2,

271

CO CH4 and CO2, and the CO2 yield was 0.501 in the blank experiment. While for the

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

272

CLC experiment of in-situ pyrolysis gases, the detected carbonaceous species were

273

CO2 and a little unconverted CO, which indicated that most of combustible gases (CO,

274

CH4 and H2) was converted by oxygen carrier. A very high CO2 yield (0.963) in the

275

CLC experiment can be explained by the well-mixing of pyrolysis gas and OC at the

276

second CLC reactor, as well as high-pressure resistance of the little porous distributor,

277

resulting in the even distribution of pyrolysis gases (small or even no bubbles thereby

278

avoiding by-passing of combustible gases as possible) through the bed inventory in

279

the CLC fuel reactor. The carbon conversion reached 0.514 during the CLC

280

experiment, which was slightly higher than that in the blank experiment. This is

281

because some light hydrocarbons in tar were converted by the OC to detectable

282

carbonaceous gases. The similar carbon conversion attained in the two experiments

283

indicated that the pyrolysis process of coal was not affected by the oxygen carrier

284

loaded at the second stage CLC reactor, which demonstrated the comparability of

285

blank experiment and CLC experiment

286

3.3 Sulfurous gases distribution

287

Fig. 5 shows the evolution of sulfurous gases from the CLC fuel reactor. In the

288

blank experiment of silica sand, the H2S was rapidly released and the peak of 280

289

ppmv was reached at ca. 30 s. The CS2 was subsequently increased to 34 ppmv,

290

which can be ascribed to the decomposition of the pyrite (FeS2) or the reaction of

291

pyrite and hydro carbons, as R9. A small amount of COS was detected with a peak of

292

4 ppmv in the CO atmosphere, which was generated by the combination of S2 and

293

pyrite with CO19. There was very low concentration of SO2, in range of 0-0.6 ppmv,

ACS Paragon Plus Environment

Page 14 of 40

Page 15 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

294

which was far below the detection accuracy of FTIR (5 ppmv). Generally, the

295

released SO2 was from the decomposition of sulfates in coal19. FeS2 + C = Fe + CS2(g)

(R9)

FeS2 + CH4(g) = Fe + 2H2(g) + CS2(g)

(R10)

296

In the iG-CLC of in-situ coal volatile, the concentration of sulfurous gases from

297

the fuel reactor was significantly changed. The peak of H2S was decreased to 93 ppmv.

298

Meanwhile, a large amount of SO2 was quickly produced, which was mainly due to

299

the oxidation of H2S in the volatile with oxygen carrier. As concluded from the HSC

300

simulation, H2S could be totally oxidized by Fe2O3 to generate SO2 as shown in Fig.

301

3(b), with the highest equilibrium constant K. In addition, there was no COS and CS2

302

detected, which were also oxidized by the oxygen carrier, as R11 and R12. In the HSC

303

simulation, there was no H2S in the outlet gas and the SO2 was close to 100 vol.%,

304

which was calculated based on the ideal thermodynamic equilibrium condition. The

305

presence of a small amount of H2S could be explained as followed: (1) In an actual

306

reaction process, the release of volatiles in coal was rather fast at high temperatures.

307

Thus, the limited residence time of volatile matters may lead to incomplete reaction

308

between H2S and oxygen carrier. (2) The reaction between H2S and oxygen carrier

309

was after the reactions between CO, CH4 and H2 with oxygen carrier, as revealed in

310

previous work11. As indicated by HSC results shown in Fig. 2, the reaction between

311

the partially reduced oxygen carrier (mainly consisting of Fe3O4) and H2S (R2) could

312

probably not take place thermodynamically.

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 40

CS2(g) + 18Fe2O3 = 12Fe3O4 + CO2(g) + 2SO2(g)

(R11)

COS(g) + 9Fe2O3 = CO2(g) + 6Fe3O4 + SO2(g)

(R12)

313

Fig. 6 shows the total volume of sulfurous gases (H2S, SO2, CS2 and COS) from

314

the fuel reactor during the coal pyrolysis process at blank experiment and CLC

315

experiment. The results showed that SO2 in the blank experiment was 0.93±0.43 µL,

316

which was below the lower limit detection of FTIR as mentioned above. Thus, the

317

generation of SO2 could be neglected. The amounts of H2S and CS2 were 350.2±7.8

318

µL and 32.6±2.3 µL at the average value of three repeated experiments, respectively.

319

The amount of COS was 3.8±0.8 µL. The amounts of the sulfurous gases released

320

were related to the type of coal used. When the coal volatile was constantly carried

321

into the fuel reactor to react with oxygen carrier, CS2 and COS were decreased to zero,

322

indicating that CS2 and COS was totally converted by oxygen carrier as R10 and R11.

323

To be more specific, 62.8 wt.% of H2S was converted to SO2 based on the increased

324

amount of SO2 and no sulfides (Fe2S3, Fe0.877S or FeS) formation was observed

325

(proved by SEM+EDX analysis in Section 3.6). This result was consistent with the

326

thermodynamic simulation of HSC chemistry: R1 was more thermodynamically

327

favorable than R8 and R12, especially at high operational temperature. Similar results

328

have also been found by Gu et al.11 and Wang et al.8. No sulfide was generated due to

329

the high oxygen carrier to fuel ratio7. Actually, during the re-oxidization process of

330

the OC (step 3 of the CLC experiment), no SO2 was detected.

ACS Paragon Plus Environment

Page 17 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

331

3.4 Sulfurous gases conversion

332

Fig. 7 shows the carbon conversation rate (dXc/dt) as function of carbon

333

conversion for the two comparative experiments. Based on the definition of carbon

334

conversion, eq. 7, it was related to the carbonaceous gases (CO2, CO and CH4) to the

335

coal carbon ratio. Thus, the yield and amount of pyrolysis products should be the

336

same under the blank experiment and CLC experiment (where the mixing of coal and

337

OC was avoided) with the same mass of coal fed. The carbon conversion should also

338

be the same. As shown in Fig. 7, the carbon conversions of blank experiment and the

339

CLC experiment were 0.507 and 0.514, respectively, when the coal pyrolysis process

340

was completely finished. The slight difference could be attributed to the different fate

341

of tar at the CLC experiment and the blank experiment. At the blank experiment with

342

silica sand loaded, the tar was still in the gaseous phase without conversion.

343

Subsequently, the gaseous tar was cooled down at the down tail tube and then

344

removed by the tar removing unit. That is, the component in tar is not detectable by

345

the on-line gas analyser. However, in the CLC experiment, the gaseous tar could be

346

converted by OC to carbonaceous gases, which could be detected by the on-line gas

347

analyser. Thus, the carbon conversion of CLC experiment was slightly higher than

348

that of the blank experiment. Additionally, when the carbon conversion was lower

349

than 0.24, the maximum carbon conversion rate of CLC experiment was higher than

350

that of the blank experiment. The peak carbon conversion rates were also different,

351

which were 0.017 s-1 at the carbon conversion of 0.16 (CLC experiment) and the

352

0.016 s-1 at that of 0.23 (blank experiment), respectively, due to the same reason as

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

353

mentioned above. The result also indicated that the tar could be easily converted into

354

gaseous phases, resulting in the maximum peak of carbon conversion rate at earlier

355

stage under the CLC experiment. When the carbon conversion exceeded 0.24, the

356

carbon conversion rates were almost equal because that the effect of carbon in tar was

357

reduced.

358

During the coal pyrolysis process (the first reactor), the sulfur compounds

359

decomposed gradually with the increased temperature and then the sulfur immigrated

360

to the pyrolysis products, char and tar. The sulfur releasing rate was quite fast at 950

361

o

362

blank experiment, the sulfur conversion rate was shown a similar trend but lower than

363

the carbon conversion rate. However, for the CLC experiment, the sulfur conversion

364

rate was higher than the carbon conversion rate when the carbon conversion was

365

lower than 0.08. It was because that the sulfur in tar was converted by OC to

366

detectable sulfurous gases rather than being removed by the tar removing unit.

367

Therefore, the sulfur conversion rate in CLC experiment was higher than that in blank

368

experiment. When the carbon conversion exceeded to 0.37, the sulfur conversion rates

369

were almost the same in two comparison experiments. It was explained by that the

370

release of tar was nearly finished and the CS2 was not formed at the end of coal

371

pyrolysis process as shown in Fig. 5. In addition, the conversion of H2S to SO2 had no

372

effect on the sulfur conversion rate based on the definition of sulfur conversion rate

373

(eq. 4). The sulfur conversion reached 0.46 after the pyrolysis gas went through the

374

oxygen carrier, compared to the value of 0.32 attained in the blank experiment, as

C. Fig. 8 shows the sulfur conversion rate as a function of sulfur conversion. In the

ACS Paragon Plus Environment

Page 18 of 40

Page 19 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

375

shown in Fig. 8. Once more, the higher sulfurous gases conversion rate of CLC

376

experiment was explained by that the sulfur-containing species in tar was converted

377

and oxidized into gaseous sulfur contents by the oxygen carrier.

378

3.5 Sulfur balance

379

A “rough” sulfur balance was calculated based on the sulfur in volatile ( mS-V ) as

380

eq. 3. The sulfur balance of blank experiment was 84.1% because of some sulfur

381

(such as the thiophene sulfide) remained in pyrolytic semi-char and tar20, and the tar

382

was removed before the gas being introduced into FTIR. Thus, the sulfur in tar was

383

not detected by FTIR, and resulted in the lower sulfur balance in the blank experiment.

384

The mass fraction of H2S, COS, CS2 and SO2 was 70.1wt.%, 0.8 wt.%, 13.1 wt.%, 0.2

385

wt.%, respectively. The sulfur balance of CLC experiment was 94.2 wt.% and the

386

mass percentage of SO2 was 68.2 wt.%. There was 26.0 wt.% of residual H2S in the

387

outlet gas because of the limited oxidization activity of Fe-based OC with H2S21 and

388

the fast releasing rate of H2S22, as shown in Fig. 9.

389

3.6 Characterization of the oxygen carrier

390

The surface morphology and element distribution of both the fresh and used

391

oxygen carrier were characterized by environmental scanning electron microscope

392

(ESEM, FEI Quanta 200) coupled with energy dispersive X-ray spectroscopy (EDX,

393

FEI Quanta 200). As shown in Fig. 10, the structure of the oxygen carrier was

394

maintained well without sintering or agglomeration. This result is consistent with that

395

obtained in our previous publications8,22. The EDX results show that S element was

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

396

not detected on the surface of both the fresh and used oxygen carrier, implying that

397

the S-containing species generated by coal pyrolysis can mostly be oxidized by the

398

oxygen carrier into sulfur oxides (SO2) as R1. The obtained results from the EDX

399

analysis agreed well with the simulation results shown in Fig. 3.

400

4 Discussion

401

Fig. 11 presents the sulfur migration paths during coal pyrolysis process in

402

chemical looping combustion environment. The sulfur contents in the coal could

403

migrate to the gaseous sulfur in volatile, tar as well as solid sulfur in char. At the

404

operation temperature of CLC (950 oC), the organic sulfur and pyrite of inorganic

405

sulfur were the main pyrolysis species during the coal pyrolysis process23, which

406

released as the gaseous products (H2S, CS2, COS). The sulfate was difficult to

407

decompose, thus it belongs to stable non-combustible sulfur. The reducible

408

sulphurous gases (H2S, CS2 and COS) could be converted into SO2 by the oxygen

409

carrier. During the coal pyrolysis process, the migration path of the sulfur in tar could

410

be in two paths: on the one side, the tar could be desulfurized under the H2/CO/CH4

411

atmosphere, which was similar to the hydro-desulfurization process of coal tar24, and

412

then the sulphurous gases were oxidized by the OC; on the other side, the tar was

413

directly reacted with OC and then the sulfur was released due to the C-S bond fission.

414

The detailed migration paths of the sulfur in tar should be further investigated. Solid

415

sulfur remained in char also needs further investigation afterwards.

ACS Paragon Plus Environment

Page 20 of 40

Page 21 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

416

5 Conclusions

417

The sulfur distribution and evolution during the CLC of in-situ coal pyrolysis

418

products were investigated in the two-stage fluidized bed reactor, which was used to

419

evaluate the direct reaction between oxygen carrier and in-situ coal pyrolysis gases

420

without cooling down process and avoiding the contact between coal and OC to

421

eliminate the effect of OC on the coal pyrolysis process. The sulfur distribution was

422

effected by the Fe-based OC when compared with the blank experiment. The main

423

conclusions were drawn as below:

424

1) The sulfurous gases can be oxidized by the OC in the CLC environment, resulting

425

in quite different distribution compared with the blank experiment. In the absence

426

of OC, H2S is the main sulfurous pyrolysis product of XLT lignite, accounting for

427

70.1 wt.% of all sulphurous gases, and the rests are 0.76 wt. % of COS, 0.2 wt.%

428

of SO2 and 13.1 wt. % of CS2. However, the fraction of these sulfurous gases in

429

the CLC environment were changed to 26.0 wt.% and 68.2 wt. % for H2S and

430

SO2, respectively.

431

2) The sulfurous gas conversion rates in the blank experiment and CLC experiment

432

were different when the carbon conversion was lower than 0.08. The difference

433

could be attributed to the conversion of sulfur in tar by the OC in the CLC

434

experiment, which accelerated the “phenomenological” sulfur conversion rate.

435

The sulfur balance was up to 94.2% at the CLC experiment because of the

436

detectable sulfurous gases converted from tar.

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

437

3) In the presence of OC, 63 % of H2S in lignite pyrolysis gases was oxidized by the

438

active phase of OC (Fe2O3) to generate SO2, via the main reaction H2S(g) +

439

9Fe2O3 = 6Fe3O4 + H2O(g) + SO2(g). COS and CS2 were nearly completely

440

oxidized by Fe2O3.

441

4) The sulfur migration paths during coal pyrolysis process in CLC environment can

442

be concluded as: gaseous sulfur species (H2S, CS2, COS) and the sulfur content in

443

tar were first released and then oxidized by the oxygen carrier to generate SO2.

444

There was also a part of unconverted H2S released in the fuel reactor.

445

Acknowledgments

446

These authors were supported by “National Key R&D Program of China

447

(2016YFB0600801) and National Natural Science Foundation of China (51606077

448

and 51522603)”. Meanwhile, the staff from the Analytical and Testing Center,

449

Huazhong University of Science and Technology, are appreciated for the related

450

sample analytical work.

451

References

452

1) Ishida, M.; Jin, H. A new advanced power-generation system using chemical-looping combustion.

453

Energy 1994, 19, 415-422.

454

2) Chuang, S.; Dennis, J. ; Hayhurst, A.; Scott, S. Development and performance of Cu-based

455

oxygen carriers for chemical-looping combustion. Combust. Flame 2008, 154, 109-121.

456

3) Siriwardane, R.; Tian, H.; Simonyi, T.; Poston, J. Synergetic effects of mixed copper-iron oxides

457

oxygen carriers in chemical looping combustion. Fuel 2013, 108, 319-333.

ACS Paragon Plus Environment

Page 22 of 40

Page 23 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

458

4) Adánez-Rubio, I.; Abad, A.; Gayán, P.; García-Labiano, F.; Diego, L.; Adánez, J. The fate of

459

sulphur in the Cu-based Chemical Looping with Oxygen Uncoupling (CLOU) Process. Appl. Energy

460

2014, 113, 1855-1862.

461

5) Garcíalabiano, F.; Diego, L.; Gayán, P.; Adánez, J.; Abad, A.; Dueso, C. Effect of fuel gas

462

composition in chemical-looping combustion with Ni-based oxygen carriers. 1. fate of sulfur. Ind.

463

Eng. Chem. Res. 2009, 48 (5), 2499-2508.

464

6) Dueso, C.; Izquierdo, M.; García-Labiano, F.; Diego, L.; Abad, A.; Gayán, P.; Adánez, J. Effect of

465

H2S on the behaviour of an impregnated NiO-based oxygen-carrier for chemical-looping combustion

466

(CLC). Appl. Catal. B Environ. 2012, 126 (38), 186-199.

467

7) Diego, L.; García-Labiano, F.; Gayán, P.; Abad, A.; Cabello, A.; Adánez, J.; Sprachmann, G.,

468

Performance of Cu- and Fe-based oxygen carries in a 500 Wth CLC unit for sour gas combustion with

469

high H2S content. Int. J. Greenhouse Gas Control 2014, 28, 168-179.

470

8) Wang, K.; Tian, X.; Zhao, H. Sulfur behavior in chemical-looping combustion using a copper ore

471

oxygen carrier. Appl. Energy 2016, 166, 84-95.

472

9) Ksepko, E.; Siriwardane, R.; Tian, H.; Simonyi, T.; Scia̧Żko, M. Comparative investigation on

473

chemical looping combustion of coal-derived synthesis gas containing H2S over supported NiO oxygen

474

carriers. Energy Fuels 2010, 24, 4206-4214.

475

10) Shen, L.; Gao, Z.; Wu, J.; Xiao, J. Sulfur behavior in chemical looping combustion with NiO/Al

476

2O3

477

11) Gu, H.; Shen, L.; Xiao, J.; Zhang, S.; Song, T.; Chen, D. Evaluation of the effect of sulfur on

478

iron-ore oxygen carrier in chemical-looping combustion. Ind. Eng. Chem. Res. 2013, 52, 1795-1805.

479

12) Tian, H.; Simonyi, T.; Poston, J.; Siriwardane, R. Effect of hydrogen sulfide on chemical looping

oxygen carrier. Combust. Flame 2010, 157, 853-863.

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

480

combustion of coal-derived synthesis gas over bentonite-supported metal-oxide oxygen carriers. Ind.

481

Eng. Chem. Res. 2009, 48, 8418-8430.

482

13) Solunke, R. D.; Veser, G. Nanocomposite oxygen carriers for chemical-looping combustion of

483

sulfur-contaminated synthesis gas. Energy Fuels 2009, 23, 4787-4796.

484

14) Wang, B.; Yan, R.; Dong, H. L.; Liang, D. T.; Zheng, Y.; Zhao, H.; Zheng, C. Thermodynamic

485

investigation of carbon deposition and sulfur evolution in chemical looping combustion with syngas.

486

Energy Fuels 2008, 22, 1012-1020.

487

15) Zhao, H.; Mei, D.; Ma, J.; Zheng, C. Comparison of preparation methods for iron–alumina

488

oxygen carrier and its reduction kinetics with hydrogen in chemical looping combustion. Asia-Pacific J.

489

Chemi Eng. 2014, 9 (4), 610-622.

490

16) Ma, J.; Tian, X.; Zhao, H.; Bhattacharya, S.; Rajendran, S.; Zheng, C. Investigation of two

491

hematites as oxygen carrier and two low-rank coals as fuel in chemical looping combustion. Energy

492

Fuels 2017, 31, 1896-1903.

493

17) Cabello, A.; Abad, A.; Gayán, P.; Diego, L.; Garcíalabiano, F.; Adánez, J. Effect of operating

494

conditions and h2s presence on the performance of CaMg0.1Mn0.9O3−δ perovskite material in chemical

495

looping combustion (CLC). Energy Fuels 2014, 28, 1262-1274.

496

18) Cabello, A.; Dueso, C.; García-Labiano, F.; Gayán, P.; Abad, A.; Diego, L.; Adánez, J.

497

Performance of a highly reactive impregnated Fe2O3/Al2O3 oxygen carrier with CH4 and H2S in a 500

498

Wth CLC unit. Fuel 2014, 121, 117-125.

499

19) Levy, J.; White, T. The reaction of pyrite with water vapour. Fuel 1988, 67, 1336-1339.

500

20) Hu, H.; Zhou, Q.; Zhu, S.; Meyer, B.; Krzack, S.; Chen, G. Product distribution and sulfur

501

behavior in coal pyrolysis. Fuel Proces. Technol. 2004, 85, 49-861.

ACS Paragon Plus Environment

Page 24 of 40

Page 25 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

502

21) Pérez-Vega, R.; Adánez-Rubio, I.; Gayán, P.; Izquierdo, M. T.; Abad, A.; García-Labiano, F.;

503

Diego, L.; Adánez, J. Sulphur, nitrogen and mercury emissions from coal combustion with CO2 capture

504

in chemical looping with oxygen uncoupling (CLOU). Int. J. Greenhouse Gas Control 2016, 46, 28-38.

505

22) Ma, J.; Zhao, H.; Tian, X.; Wei, Y.; Rajendran, S.; Zhang, Y.; Bhattacharya, S.; Zheng, C.

506

Chemical looping combustion of coal in a 5 kWth interconnected fluidized bed reactor using hematite

507

as oxygen carrier. Appl. Energy 2015, 157, 304-313.

508

23) Gu, Y.; Yperman, J.; Vandewijngaarden, J.; Reggers, G.; Carleer, R. Organic and inorganic sulphur

509

compounds releases from high-pyrite coal pyrolysis in H2 , N2 and CO2 : Test case Chinese LZ coal.

510

Fuel 2017, 202, 494-502.

511

24) Dong, L.; Wen, L.; Gao, X.; Cui, L.; Yang, X.; Zhao, P. A study on hydrodesulfurization process

512

of coal tar. J. Northwest University 2010, 40, 447-450.

513

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

514

Table and Fig. Captions

515

Table 1. Proximate and ultimate analyses of XLT coal and XLT char

516

Table 2. Chemical species considered in the HSC calculation for Fe-based oxygen

517

carrier

518

Fig. 1. A schematic view of the two-stage fluidized bed reactor

519

Fig. 2. Equilibrium constants of reactions between iron oxides and H2S

520

Fig. 3. Conventional gases (a), main sulfurous gases (b) and iron species (c) as a

521

function of fuel to oxygen carrier ratio at 950 oC

522

Fig. 4. Comparison of conventional pyrolysis gases: (a) blank experiment (b) CLC

523

experiment

524

Fig. 5. Sulfurous gas concentrations from the outlet of the CLC fuel reactor (the

525

second reactor): loaded with silica sand (a) or oxygen carrier (b)

526

Fig. 6. Distribution of sulfurous gases (a) blank experiment (b) CLC experiment

527

Fig. 7. Sulfurous gases conversion rate and carbon conversion rate vs. carbon

528

conversion

529

Fig. 8. Sulfur conversion rate vs sulfur conversion in blank experiment and CLC

530

experiment

531

Fig. 9. Sulfur balance of sulfurous gases in blank experiment and CLC experiment

532

Fig. 10. SEM+EDX image of fresh oxygen carrier (a) and used oxygen carrier (b)

533

Fig. 11. Sulfur migration paths in iG-CLC of coal

534 535

ACS Paragon Plus Environment

Page 26 of 40

Page 27 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

536

Table 1

537

Proximate and ultimate analyses of XLT coal and XLT char Proximate analysis (wt%, ad)a

Ultimate analysis (wt%, ad)

Type M

V

FC

A

C

H

N

S

Ob

XLT coal

10.59

49.56

33.40

6.45

45.52

4.20

1.28

0.60

31.36

XLT char

3.64

6.45

77.79

12.12

78.60

0.60

1.17

0.68

3.19

538

a

539

content was determined by difference

M, moisture content; V, volatile matters; A, ash content; FC, fixed carbon; ad, air-dried basis. b The O

540

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 40

541

Table 2

542

Chemical species considered in the HSC calculation for Fe-based oxygen carrier

Gas species

Solid oxide species

H2(g), H2O(g), CO(g), CO2(g),

Solid

Solid

Simple

sulfur

carbon

substance

species

species

species

FeCO3, FeO, Fe3O4, Fe0.877S,

CH4(g), C(g), C2(g), C3(g), C4(g),

Fe2O3,FeSO4, Fe2(SO4)3, FeSO4

Fe3C, FeS2(M),

C5(g), H2S(g), SO2(g), SO3(g), *H2O, FeSO4 COS(g), CS2(g), S(g), S2(g), S3(g),

C,C(A),

FeS, FeS2,

C(D), Fe, Fe3C(B)

Fe2S3,

S

*4H2O, Fe7S8

S4(g), S5(g), S6(g), S7(g), S8(g)

543 544

a.

FeSO4*7H2O

g is the gaseous state; C(A) and C(D) are amorphous and diamond carbon, respectively; Fe3C(B) is β-iron carbide and FeS2(M) is marcasite.

545 546 547 548 549

ACS Paragon Plus Environment

Page 29 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

550 551

Fig. 1. A schematic view of the two-stage fluidized bed reactor

552

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

553 554

Fig. 2. Equilibrium constants of reactions between iron oxides and H2S

555

ACS Paragon Plus Environment

Page 30 of 40

Page 31 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

556

557

558

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

559 560

Fig. 3. Conventional gases (a), main sulfurous gases (b) and iron species (c) as a

561

function of fuel to oxygen carrier ratio at 950 oC

562

ACS Paragon Plus Environment

Page 32 of 40

Page 33 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

563

564

565 566

Fig. 4. Comparison of conventional pyrolysis gases: (a) blank experiment (b) CLC

567

experiment

568

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

569 570 571

572 573

Fig. 5. Sulfurous gas concentrations from the outlet of the CLC fuel reactor (the

574

second reactor): loaded with silica sand (a) or oxygen carrier (b)

575

ACS Paragon Plus Environment

Page 34 of 40

Page 35 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

576 577

Fig. 6. Distribution of sulfurous gases: (a) blank experiment, (b) CLC experiment

578

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

579 580

Fig. 7. Sulfurous gases conversion rate and carbon conversion rate vs. carbon

581

conversion.

582

ACS Paragon Plus Environment

Page 36 of 40

Page 37 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

583 584

585 586

Fig. 8. Sulfur conversion rate vs sulfur conversion in blank experiment and CLC

587

experiment.

588

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

589 590

Fig. 9. Sulfur balance of sulfurous gases in blank experiment and CLC experiment.

591

ACS Paragon Plus Environment

Page 38 of 40

Page 39 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

592 593

Fig. 10. SEM+EDX image of fresh oxygen carrier (a) and used oxygen carrier (b)

594

ACS Paragon Plus Environment

Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

595 596

Fig. 11. Sulfur migration paths in iG-CLC of coal

597

ACS Paragon Plus Environment

Page 40 of 40