Catalytic performance of carbon nanotubes supported palladium catalyst for hydrogen production from hydrogen iodide decomposition in thermochemical sulfur iodine cycle

Accepted Manuscript Catalytic performance of carbon nanotubes supported palladium catalyst for hydrogen production from hydrogen iodide decomposition in thermochemical sulfur iodine cycle

Amit Singhania, Ashok N. Bhaskarwar PII:

S0960-1481(18)30536-6

DOI:

10.1016/j.renene.2018.05.017

Reference:

RENE 10069

To appear in:

Renewable Energy

Received Date:

29 November 2017

Revised Date:

04 April 2018

Accepted Date:

04 May 2018

Please cite this article as: Amit Singhania, Ashok N. Bhaskarwar, Catalytic performance of carbon nanotubes supported palladium catalyst for hydrogen production from hydrogen iodide decomposition in thermochemical sulfur iodine cycle, Renewable Energy (2018), doi: 10.1016/j. renene.2018.05.017

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Catalytic performance of carbon nanotubes supported palladium catalyst for hydrogen

2

production from hydrogen iodide decomposition in thermochemical sulfur iodine cycle

3

Amit Singhaniaa*, Ashok N. Bhaskarwarb

4

a,bDepartment

5

New Delhi 110016, India

6

*Corresponding author email id: [email protected]

7

Abstract

8

The current work presents the synthesis of carbon nanotubes supported palladium catalyst for

9

hydrogen production from hydrogen-iodide decomposition in thermochemical water-splitting

10

sulfur-iodine (SI) cycle. XRD results showed that the Pd nanoparticles were highly dispersed

11

on the CNT support. Raman results showed that Pd(3%) possessed the highest quantity of

12

defects than other loaded Pd samples and CNT support. The order of catalytic activity for

13

hydrogen-iodide decomposition is: Pd(3%)/CNT > Pd(5%)/CNT > Pd(1%)/CNT > CNT.

14

This is due to high degree of defects present in Pd(3%)/CNT as compared to others.

15

Pd(3%)/CNT also showed an excellent stability of 100 h for the reaction. The post-

16

characterizations (BET, ICP-AES, XRD and TEM) of spent-Pd(3%)/CNT after 100 h were

17

carried out in order to find out the changes in the specific surface area, elemental analysis,

18

structure and particle size. No changes were observed in the specific surface area, elemental

19

analysis, particle size, and structure of the spent catalyst as compared to the fresh one. This

20

shows the Pd/CNT has a lot of potential of generating hydrogen in the thermochemical SI

21

cycle.

of Chemical Engineering, Indian Institute of Technology Delhi, Hauz Khas,

22

1

ACCEPTED MANUSCRIPT 23

Keywords:

Carbon

nanotubes;

24

decomposition; Sulfur Iodine cycle

Palladium;

Catalytic

activity;

Hydrogen

iodide

25 26

1. Introduction

27

The world is looking on hydrogen and its research is going in a speedy way as the researchers

28

have understood its potential as an energy carrier. Water is considered as an appropriate

29

precursor (cheap and present in abundance) to generate large quantities of hydrogen. Various

30

kinds of methods of hydrogen generation from water have been discussed in literature such as

31

photocatalysis, electrolysis, steam-reforming, biological, and thermochemical [1-4]. Each

32

method has its own advantages and disadvantages. Based on efficiency, low-temperature and

33

carbon-free criterion, thermochemical cycles are considered as good option. Among different

34

thermochemical cycles, sulfur-iodine (SI) cycle is one of the best processes of producing

35

hydrogen gas at low temperatures from water as direct decomposition of water requires more

36

than 3000 degree Celsius [5-7]. This cycle was started by General Atomics, USA in 1971 and

37

consists of the following reactions [8,9]:

38

Bunsen reaction:

39

SO2(g) + I2(s) + 2H2O(l) ↔ 2HI(aq) + H2SO4(aq) (T= 20-120°C)…(1)

40

Sulfuric-acid decomposition:

41

H2SO4(g) ↔ H2O(g) + SO2(g) + ½ O2(g) (T= 800-900°C)…(2)

42

Hydrogen-iodide decomposition:

43

2HI(g) ↔ H2(g) + I2(g) (T= 400-550°C)…(3)

44

Net reaction: H2O(l) → H2(g) + O2(g)…(4)

45

2

ACCEPTED MANUSCRIPT 46

The reaction 1 is Bunsen in which mixture of hydrogen iodide and sulfuric acid is

47

produced by the reaction of sulfur dioxide, iodine and water. The reaction 2 and 3 involve the

48

decomposition of sulfuric acid and hydrogen iodide. The decomposed products sulfur dioxide

49

and iodine are again recycled to the Bunsen reaction. In this way, the process goes on and the

50

net sum is the generation of hydrogen and oxygen from water. Hydrogen-iodide

51

decomposition is a slow kinetic equilibrium-limited reaction. A small amount of catalyst is

52

necessary to enhance the rate of this reaction. Also, a temperature range of 400-550oC is

53

targeted in order to use cheap materials of construction for hydrogen-iodide decomposition

54

reaction.

55

In literature, both noble such as Pt, Pd, Ir, and Rh and non-noble such as Ni, Co, Mo,

56

and Ag based catalysts are reported [10-16]. Different supports such as ceria, zirconia,

57

alumina, activated carbon, carbon molecular sieves (CMS), graphite and carbon nanotubes

58

(CNTs) are also used by the researchers for hydrogen-iodide decomposition [17-20]. Among

59

these different supports, CNTs have not been explored much earlier by the researchers.

60

Zhang et al. [15] reported CNTs supported Ni catalyst, but this catalyst did not achieve

61

equilibrium conversions and no long-term stability test was reported for this catalyst.

62

Similarly, Wang et al. [21] reported CNTs supported Pt catalysts for hydrogen-iodide

63

decomposition. These supported Pt catalysts were more active than Ni catalysts, but Pt was

64

not stable under corrosive hydrogen iodide environment. The crystallite size of Pt increases

65

from 2 to 17.6 nm after hydrogen iodide decomposition at 550°C. In this case also, no long-

66

term stability test was reported for hydrogen-iodide decomposition. Noble metal Pt catalyst

67

has been reported as a highly active catalyst for this reaction, but it tends to agglomerate at

68

high temperatures [20,21]. Another PGM catalyst, Pd is another good option as a catalyst as it

69

is highly active and stable for this kind of reaction [22]. So, the combination of CNTs and Pd

70

is interesting for this particular reaction. Also, this combination has not been explored earlier

3

ACCEPTED MANUSCRIPT 71

according to the author’s knowledge. Here, long time-on-stream stability of 100 h has been

72

conducted on Pd/CNT, which was not done earlier using CNTs supported catalysts.

73

Supported catalysts can be synthesized by different routes such as impregnation, sol-gel, co-

74

precipitation, combustion, electroless deposition and hydrothermal [23-26].

75

The aim of the present work is to explore the catalytic activity of synthesized CNTs

76

supported Pd catalysts for hydrogen-iodide decomposition reaction. Also, a long-term test is

77

carried out for 100 h to check its stability. The data of XRD, BET, ICP-AES, TEM and

78

RAMAN have been carried out and reported here.

79 80

2. Experimental details

81

2.1. Synthesis of Pd/CNTs

82

The synthesis of Pd/CNTS was done by using the impregnation method. Appropriate amount

83

of palladium chloride was added to the 100 ml of deionized water containing CNTs (CNTs –

84

Sigma-Aldrich). 1 ml of 0.001 M sodium borohydride was added to the above mixture

85

followed by the addition of 2 ml of 1 M sodium hydroxide and 1 ml of hydrazine. The above

86

reaction mixture was vigorously stirred at 80oC on a magnetic stirrer for 3 hours. The product

87

was filtered and washed with deionized water (4-5 times) and finally dried at 110oC for 4 h.

88

The loading amounts of Pd on CNTs were 1, 3 and 5 wt.%.

89

2.2. Characterization of Pd/CNTs catalysts

90

The BET surface area was calculated using BET machine (Micrometrics, ASAP 2010). The

91

samples were degassed at 300oC for 5 h. The crystallite size and phases were determined by

92

powder XRD machine (Rigaku X-ray diffractometer, DMAX IIIVC). The structural

93

morphology and particle size were collected on a TEM instrument (Tecnai G2-20 Twin, FEI).

94

For preparation of TEM samples, the powder catalysts were dispersed in ethanol using ultra 4

ACCEPTED MANUSCRIPT 95

sonication for 30 minutes and then 2 drops of dispersed solution were put on carbon coated

96

Cu grids. The Raman spectrometer (Horiba JY Lab RAM HR 800) was used to find out the

97

phases and defects present in the catalysts. The exact loading of Pd metal was confirmed by

98

Simultaneous ICP-AES spectrometer.

99

2.3. Activity

100

Figure 1 shows the experimental set-up for testing of catalytic activity of Pd/CNTs. The

101

catalytic activity of Pd/CNTs was performed in a fixed-bed quartz reactor of 16 mm i.d. 1

102

gram of Pd/CNTs catalyst was used. A small amount of quartz wool was put both at lower

103

and upper part of catalyst. The catalyst was sandwiched between the quartz wool. The WHSV

104

of (HI-55 wt.%) 12.9 h-1 was used along with nitrogen as carrier gas (40 ml/min.). The

105

temperature of 400-550°C was used for hydrogen-iodide decomposition. The scrubbers and

106

condenser were used to block undecomposed hydrogen iodide, iodine and water. The mixture

107

of hydrogen and nitrogen was passed through a molecular sieve 5A column and analysed by a

108

gas chromatograph.

109 110

Figure 1. Schematic diagram of testing of catalytic activity of Pd/CNTs. 5

ACCEPTED MANUSCRIPT 111 112

3. Results and discussion

113

3.1. Characterization

114

Table 1 shows the specific surface area, and actual wt.% of Pd in the Pd/CNTs. The support

115

CNTs showed a specific surface area of 221.2 m2 g-1. The Pd loading decreased the specific

116

surface area of CNTs.

117

Table 1. Specific surface area and actual wt.% of Pd in the Pd/CNTs. Theoretical Pd

Actual Pda

Specific

(wt.%)

(wt.%)

area (m2 g-1)

CNT

-

-

221.2

Pd(1%)/CNT

1.0

0.92

205.3

Pd(3%)/CNT

3.0

2.94

198.5

Pd(5%)/CNT

5.0

4.88

164.1

Catalyst

118

adetermined

surface

by ICP-AES.

119 120

The phases were determined by the powder-XRD technique. Figure 2 shows the XRD

121

patterns of synthesized Pd/CNT catalysts. CNT showed peaks at 26 and 43°, which

122

corresponded to the (002) and (101) planes of carbon. This means that it has a quasi-graphitic

123

structure. In the Pd-loaded sample, no peaks of Pd or PdO were found in the diffraction

124

patterns, which showed that the Pd particles were highly dispersed on the CNT support. The

125

high dispersion of Pd on the CNT is might be due to the special structure of the support [27]

126

because even after high Pd loading of 3 and 5 wt.% no peaks are observed.

6

ACCEPTED MANUSCRIPT

127 128

Figure 2. XRD of (a) CNT, (b) Pd(1%)/CNT, (c) Pd(3%)/CNT, and (d) Pd(5%)/CNT.

129 130

The morphological analysis was done by TEM. Figure 3 shows the TEM micrographs

131

of the CNT supported Pd catalysts. The Pd particles are spherical in shape. The particle sizes

132

of Pd(1%)/CNT and Pd(3%)/CNT were 2-6 nm and 4-8 nm respectively. On increasing the

133

Pd% further resulted into a large increase in particle size. Pd(5%) showed a particle size of

134

10-20 nm.

135 136

Figure 3. TEM micrograph of (a) Pd(1%)/CNT, (b) Pd(3%)/CNT, and (c) Pd(5%)/CNT.

7

ACCEPTED MANUSCRIPT 137 138

Raman spectroscopy was used to determine the different phases and defects present in

139

the Pd/CNT catalysts. Figure 4 shows the Raman spectra of Pd/CNT catalysts. All the

140

samples showed peaks at 1345 and 1589 cm-1 which corresponds to the defect, D and

141

graphite, G bands of carbon [28]. After loading of Pd metal, the intensity of defect peak

142

increased as compared to the CNT support, which meant the quantity of defects got

143

increased. The ratio, ID/IG was used to measure the quantity of defects present in the catalyst

144

sample. The order of ID/IG was as follows: Pd(3%)/CNT > Pd(5%)/CNT > Pd(1%)/CNT >

145

CNT. These results show that the Pd(3%)/CNT possesses the highest quantity of defects as

146

compared to CNT support.

147 148

Figure 4. Raman spectra of (a) CNT, (b) Pd(1%)/CNT, (c) Pd(3%)/CNT, and (d)

149

Pd(5%)/CNT.

150 151 152 8

ACCEPTED MANUSCRIPT 153

3.2. Catalytic activity

154

The activity of CNT supported Pd catalysts was measured for hydrogen-iodide

155

decomposition at 400-550°C. Figure 5 shows the hydrogen iodide conversion of Pd supported

156

catalysts at different decomposition temperatures. CNT showed a small conversion during the

157

tested temperatures. It showed a maximum conversion of around 10.3% at 550°C. The

158

deposition of Pd metal on the surface of CNT increased the hydrogen iodide conversion to a

159

much larger extent. Pd(1%) showed a minimum conversion of 13.5% at 400°C and maximum

160

of 20.1% at 550°C. Among different Pd catalysts, Pd(3%)/CNT possessed the highest activity

161

of 22.3 and 23.7% at 500 and 550°C respectively. The addition of further wt% of Pd in the

162

Pd/CNT catalyst i.e. with 5 wt.%, resulted in the decrease in hydrogen iodide conversions.

163

These results can be explained on the basis of the quantity of defects (ID/IG ratio). The

164

Pd(3%)/CNT possessed the highest quantity of defects than CNT support and others.

165 166

Figure 5. Hydrogen iodide conversion of CNT supported Pd catalysts.

167

9

ACCEPTED MANUSCRIPT 168

Figure 6 shows the effect of Pd wt.% in Pd/CNT on hydrogen iodide conversion at

169

500°C. As the Pd wt.% increased from 0 to 3, the conversion value also increased from 6.6%

170

to 22.3%. But, on further increase in Pd wt.% from 3 to 10, the hydrogen iodide conversions

171

decreased from 23.7% to 18.3%.

172 173

Figure 6. Effect of Pd wt. % in Pd/CNT on hydrogen iodide conversion at 500°C.

174 175

A long-term test was also studied to test the stability of Pd(3%)/CNT catalyst for 100

176

h. Figure 7 shows the stability test of Pd(3%)/CNT for 100 h. This test was performed with 1

177

g of catalyst at 500°C. The stability result showed that there was no change in hydrogen

178

iodide conversion during 100 h testing. The conversion was constant throughout the testing.

179

This shows that the Pd(3%)/CNT is highly active and stable catalyst for corrosive hydrogen-

180

iodide decomposition.

10

ACCEPTED MANUSCRIPT

181 182

Figure 7. Stability test of Pd(3%)/CNT at 500°C.

183 184

The BET, ICP-AES and XRD were carried out for the spent-Pd(3%)/CNT catalyst

185

after 100 h testing. The specific surface area of the spent catalyst was found to be 196.7 m2 g-

186

1.

187

surface area. The ICP-AES analysis was done to check the change in wt.% of Pd metal in the

188

spent-Pd(3%)/CNT. The wt.% of Pd was found to be 2.93 (same as fresh one). Figure 8

189

shows the XRD of fresh-Pd(3%)/CNT and spent- Pd(3%)/CNT. The diffraction result showed

190

no change in the patterns before and after the reaction. Figure 9 shows the TEM micrograph

191

of spent-Pd(3%)/CNT. No change in particle size was observed in the spent catalyst when

192

compared to the fresh one. The particle size of spent-Pd(3%)/CNT was 4-9 nm. All these

193

results (BET, ICP-AES, XRD and TEM) confirmed the high stability of Pd-CNT catalyst for

194

hydrogen-iodide decomposition.

It is similar to the fresh-Pd(3%)/CNT. No significant change was occurred in the specific

11

ACCEPTED MANUSCRIPT

195 196

Figure 8. XRD of (a) fresh-Pd(3%)/CNT and (b) spent-Pd(3%)/CNT.

197

198 199

Figure 9. TEM micrograph of spent-Pd(3%)/CNT.

200 201

Figure 10 shows the Arrhenius plot of Pd(3%)/CNT and CNT for hydrogen-iodide

202

decomposition. The apparent activation energy was calculated by assuming a differential

203

mode of reactor of lower conversions. The apparent activation energy of CNT and

204

Pd(3%)/CNT was found to be 58.7 and 51.1 kJ mol-1. These values are matched with the

12

ACCEPTED MANUSCRIPT 205

reported literature (Pt/AC – 53.2 kJ mol-1, AC – 55.5 kJ mol-1, Pd/γ-Al2O3 – 54.3 kJ mol-1,

206

and Pt/γ-Al2O3 – 54.3 kJ mol-1) [8,19,21].

207 208

Figure 10. Arrhenius plot of Pd(3%)/CNT and CNT.

209 210

4. Conclusion

211

The CNT supported Pd catalysts were successfully prepared by the impregnation method.

212

TEM analysis gave 2-20 nm size of particles of Pd/CNT catalysts. Raman showed

213

Pd(3%)/CNT possessed highest quantity of defects as compared to other loaded-Pd samples

214

and CNT support. The hydrogen-iodide decomposition results showed that Pd(3%)/CNT

215

possessed highest catalytic activity. It gave an equilibrium conversion of 23.7% at 550°C. It

216

also showed excellent stability of 100 h for the reaction. The post-characterization results

217

further showed the high stability of Pd(3%)/CNT catalyst. No change in BET surface area,

218

wt. % of Pd, diffraction patterns and particle size were found in the spent- Pd(3%)/CNT

219

catalyst. This shows that it is highly active and stable for the decomposition reaction.

220 13

ACCEPTED MANUSCRIPT 221

Conflict of Interest

222

The author declares there is no conflict of interest.

223 224

Acknowledgements

225

The authors want to thank central facility of IIT Delhi, New Delhi, India, for granting access

226

to characterization instruments. Also, the help from ONGC Energy Centre, New Delhi, India

227

has been acknowledged here.

228 229

References

230

[1]

C. Du, H. Huang, J. Jian, Y. Wu, M. Shang, W. Song, Enhanced electrocatalytic

231

hydrogen evolution performance of MoS2 ultrathin nanosheets vis Sn doping, Appl.

232

Catal. 538 (2017) 1-8.

233

[2]

A. Singhania, S.M. Gupta, Nickel nanocatalyst ex-solution from ceria-nickel oxide

234

solid solution for low temperature CO Oxidation, J. Nanosci. Nanotechnol. 18 (2018)

235

4614-4620.

236

[3]

(2016) 48-59.

237 238

G. Colon, Towards the hydrogen production by photocatalysis, Appl. Catal. 518

[4]

A. Singhania, A.N. Bhaskarwar, D.M. Kale, N.J. Thomas, B.N. Prabhu, A. Bhardwaj,

239

B. Bhargava, D. Parvatalu, S. Banerjee, Highly active supported bimetallic (Ni-Pt)

240

catalyst for hydrogen iodide (HI) decomposition and synthesis procedure thereof. IN

241

Pat. 2259/DEL/2014 2014.

242

[5]

A. Singhania, A.N. Bhaskarwar, D.M. Kale, N.J. Thomas, B.N. Prabhu, A. Bhardwaj,

243

B. Bhargava, D. Parvatalu, S. Banerjee, Vanadia supported Pt catalyst and use thereof

244

for hydrogen iodide decomposition in sulphur-iodine (SI) cycle for hydrogen

245

production, IN Pat. 2308/DEL/2014 2014.

14

ACCEPTED MANUSCRIPT 246

[6]

from HI decomposition reaction of SI cycle, Int. J. Energ. Res. 39 (2015) 484-493.

247 248

D. Tyagi, S. Varma, S.R. Bharadwaj, Pt/Zirconia catalyst for hydrogen generation

[7]

W.C. Cho, C.S. Park, K.S. Kang, C.H. Kim, K.K. Bae, Conceptual design of sulfur-

249

iodine hydrogen production cycle of Korea Institute for Energy Research. Nucl. Eng.

250

Des. 239 (2009) 501-507.

251

[8]

A. Singhania, A.N. Bhaskarwar, Development of catalysts for hydrogen production

252

from hydrogen-iodide decomposition in thermochemical water-splitting sulfur-iodine

253

cycle: A review. Catal. Rev. Sci. Eng. accepted, in press, (2017) 1-44. DOI:

254

10.1080/01614940.2017.1366189

255

[9]

A. Singhania, A.N. Bhaskarwar, Performance of activated carbon supported Ni, Co,

256

and Ni-Co catalysts for hydrogen iodide decomposition in a thermochemical water-

257

splitting sulfur-iodine cycle. Energ. Technol. 6 (2018) 1-9.

258

[10]

A. Singhania, V.V. Krishnan, A.N. Bhaskarwar, B. Bhargava, D. Parvatalu, S.

259

Banerjee, Catalytic performance of bimetallic Ni-Pt nanoparticles supported on

260

activated carbon, gamma-alumina, zirconia, and ceria for hydrogen production in

261

sulfur-iodine thermochemical cycle. Int. J. Hydrogen Energ. 41 (2016) 10538-10546.

262

[11]

A. Singhania, A.N. Bhaskarwar, P. Damaraju, S. Banerjee, B. Bhargava, Catalytic

263

decomposition of hydrogen iodide over nanocrystalline ceria promoted by transition

264

metal oxides for hydrogen production in sulfur-iodine thermochemical cycle, Catal.

265

Lett. accepted, in press, (2018). DOI 10.1007/s10562-018-2345-0.

266

[12]

L. Wang, P. Zhang, S. Chen, J. Xu, Overview of the development of catalysts for HI

267

decomposition in the iodine-sulfur thermochemical cycle at INET, Nucl. Eng. Des.

268

271 (2014) 60-63.

269

[13]

A. Singhania, V.V. Krishnan, A.N. Bhaskarwar, B. Bhargava, D. Parvatalu, S. 15

ACCEPTED MANUSCRIPT 270

Banerjee, Hydrogen production from the hydrogen iodide over nanosized nickel-oxide-

271

zirconia catalysts prepared by solution-combustion techniques, Catal. Commun. 93

272

(2017) 5-9.

273

[14]

P. Favuzza, C. Felici, L. Nardi, P. Tarquini, A. Tito, Kinetics of hydrogen iodide

274

decomposition over activated carbon catalysts in pellets, Appl. Catal. B: Environ. 105

275

(2011) 30-40.

276

[15]

Y. Zhang, G. Fu, Z. Wang, X. Lin, L. Wang, M. Kuang, R. Whiddon, K. Cen, HI

277

decomposition over carbon-based and Ni-impregnated catalysts of the sulfur-iodine

278

cycle for hydrogen production, Ind. Eng. Chem. Res. 54 (2015) 1498-1504.

279

[16]

A. Singhania, V.V. Krishnan, A.N. Bhaskarwar, B. Bhargava, D. Parvatalu, Hydrogen-

280

iodide decomposition over Pd-CeO2 nanocatalyst for hydrogen production in sulfur-

281

iodine thermochemical cycle, Int. J. Hydrogen Energ. 43 (2018) 3886-3891.

282

[17]

A. Singhania, A.N. Bhaskarwar, TiO2 as a catalyst for hydrogen productionfrom

283

hydrogen-iodide in thermo-chemical water-splitting sulfur-iodine cycle. Fuel 221

284

(2018) 393-398.

285

[18]

A. Singhania, A.N. Bhaskarwar, Effects of rare earth (RE- La, Pr, Nd) metal-doped

286

ceria nanoparticles on catalytic hydrogen iodide decomposition for hydrogen

287

production, Int. J. Hydrogen Energ. 43 (2018) 4818-4825.

288

[19]

water-splitting processes, Catal. Rev. 22 (1980) 325-369.

289 290

D.R. O’Keefe, J.H. Norman, D.G. Williamson, Catalysis research in thermochemical

[20]

L. Wang, D. Li, P. Zhang, S. Chen, J. Xu, The HI catalytic decomposition for the lab-

291

scale H2 producing apparatus of the iodine-sulfur thermochemical cycle, Int. J.

292

Hydrogen Energ. 37 (2012) 6415-6421.

293 294

[21]

L. Wang, Q. Han, D. Li, Z. Wang, J. Chen, S. Chen, P. Zhang, B. Liu, M. Wen, J. Xu, Comparisons of Pt catalysts supported on active carbon, carbon molecular sieve, 16

ACCEPTED MANUSCRIPT 295

carbon nanotubes and graphite for HI decomposition at different temperatures, Int. J.

296

Hydrogen Energ. 38 (2013) 109-116.

297

[22]

D. Li, L. Wang, P. Zhang, S. Chen, J. Xu, HI decompsoition over active carbon

298

supported binary Ni-Pd catalysts preapred by electroless plating, Catal. Commun. 37

299

(2013) 32-35.

300

[23]

A. Singhania, S.M. Gupta, Low temperature CO oxidation over Cu and Pt co-doped

301

ZrO2 nanoparticles synthesized by solution combustion techniques, Beilstein J.

302

Nanotechnol. 8 (2017) 1547-1552.

303

[24]

temperature CO oxidation, Beilstein J. Nanotechnol. 8 (2017) 264-271.

304 305

[25]

A. Singhania, S.M. Gupta, Low temperature CO oxidation: Effect of the second metal on activated carbon supported Pd catalysts, Catal. Lett. 148 (2018) 946-952.

306 307

A. Singhania, S.M. Gupta, Nanocrystalline ZrO2 and Pt-doped ZrO2 catalysts for low-

[26]

A. Singhania, High surface area M (M= La, Pr, Nd, and Pm)-doped ceria

308

nanoparticles: Synthesis, characterization and activity comparison for CO oxidation,

309

Ind. Eng. Chem. Res. 56 (2017) 13594-13601.

310

[27]

L. Li, Z.H. Zhu, Z.F. Yan, G.Q. Lu, L. Rintoul, Catalytic ammonia decomposition over

311

Ru/carbon catalysts: the importance of the structure of carbon support, Appl. Catal. A

312

320 (2007) 166-172.

313 314

[28]

Y. Wang, D.C. Alsmeyer, R.L. Mccreery, Raman spectroscopy of carbon materials: Structural basis of observed spectra, Chem. Mater. 2 (1990) 557-563.

315

17

ACCEPTED MANUSCRIPT

Highlights 1. Pd/CNT is explored first time for hydrogen production from hydrogen-iodide decomposition. 2. Pd(3%)/CNT achieves equilibrium conversions of 22.3 and 23.7% at 500 and 550oC respectively. 3. Pd(3%)/CNT shows an excellent time-on-stream stability of 100 h.

1