Catalytic hydrogenation of CO and CO2 in the Presence of Light Hydrocarbons
Vahid Shadravan
Bachelor of Chemical Engineering (Shahid Bahonar University of Kerman)
A thesis submitted in fulfilment of the requirements for the Degree of
Doctor of Philosophy in Chemical Engineering
School of Engineering Faculty of Engineering and Built Environment The University of Newcastle Callaghan, NSW 2308, Australia
March 2018
Statement of Originality
I hereby certify that the work embodied in the thesis is my own work, conducted under normal supervision.
The thesis contains published scholarly work of which I am a co-author. For each such work a written statement, endorsed by the other authors, attesting to my contribution to the joint work has been included.
The thesis contains no material which has been accepted, or is being examined, for the award of any other degree or diploma in any university or other tertiary institution and, to the best of my knowledge and belief, contains no material previously published or written by another person, except where due reference has been made in the text. I give consent to the final version of my thesis being made available worldwide when deposited in the University’s Digital Repository, subject to the provisions of the Copyright Act 1968 and any approved embargo.
Vahid Shadravan
Signature: Date:
ii Statement of Contribution of Others
I, the undersigned, attest that Research Higher Degree candidate, Vahid
Shadravan, has carried out the experiments, results analysis and writing of paper included in this thesis.
Professor Eric Kennedy
Signature: Date:
Professor Michael Stockenhuber
Signature: Date:
iii Acknowledgments
Firstly, I would like to express my sincere gratitude to my supervisors Professor
Eric Kennedy and Professor Michael Stockenhuber for their continuous support, meticulous supervision, patience, encouragement and immense knowledge. The life lessons that I have learnt from my supervisors are the most valuable achievements of my journey as a PhD candidate at the University of Newcastle.
I would also like to thank my friends and colleagues at the Priority Research
Centre for Energy, Dr Omid Mowla, Hadi Hosseini Amoli, Pedram Ghaseminejad
Sadr, Luke Harvey, Dr Glenn Bryant, Matthew Drewery, Jarrod Friggieri, Dr Jerry
Pui Li, Dr Sara Mosallanejad, Hamed Mootabadi, Dr Gizelle Sanchez, Dr Nasser
Ahmed Khan, Dr Khalil Ahmed, Guangyu Zhao, Penghui Yan and many others for their assistance and friendship.
I would like to take this opportunity to thank some influential people who helped me to find, develop and continue my interest in chemical engineering. Firstly, I thank Mr Neku Amal, my middle-school chemistry teacher; in his classes and laboratory demonstrations, I found my deep interests in chemical experiments and studies. I am also grateful for everything that I have learnt during my Bachelor studies at Shahid Bahonar University of Kerman from Dr Sattar Ghader, Dr Ali
Farsi and Dr Seyed Soheil Mansouri.
I would like to warmly appreciate all my friends (Ehsan, Ali, Farhad, Soheil and
Shahram) and my relatives (my uncles, aunts, grandparents and my brother’s wife, Melika) in Iran; and my friends in Australia (Omid(s), Leily, Armin(s), Neda,
Behzad, Esi, Mahmoud and the members of Newcastle Aikido) for their friendship, kind support and encouragements. My sincere thanks also goes to my
iv inspiring instructors at Newcastle Aikido (Saku Shin Kan), Sensei Drius and
Sensei Gabriel.
Finally, I would like to dedicate my thesis to my dearest father and mother (Hossein and Maliheh), my lovely brother (Saeed) and my beloved fiancé (Maryam). I appreciate their patience while I was living far from them to follow my dreams. They have been continuously a source of love and support through my life. I would have not been able to accomplish this journey without their patience, love, unconditional support and sacrifice.
v Abstract
Carbon oxides emission, as by-products of many industrial synthetic hydrocarbon processes, causes serious environmental issues and negatively affects commercialisation of some new processes (e.g. OCM). Thus, producing CO and
CO2 (COx) free (or with minimal amount of COx) synthetic hydrocarbon streams is necessary to facilitate commercialization of these new processes as large scale industrial plants. Moreover, due to the significant environmental effect of COx, it is critical to develop processes to convert COx and reduce their emission into the atmosphere. In this thesis, catalytic hydrogenation of COx in the presence of light hydrocarbons (methane, C2-C3 alkane and alkene) was studied.
The feasibility of converting COx (where the residual concentration of both CO and CO2 in the product gas stream are less than 1 ppm) without reducing the inlet concentration of feed hydrocarbons was initially investigated over a bench-mark hydrogenation catalyst (Ni/Al2O3). It is found that the inlet species were consumed and converted in different temperature ranges. For feed compositions containing COx and C1-C3, the consumption of carbon monoxide, carbon dioxide and C2/C3 paraffins was observed and their maximum conversion was attained over different temperature ranges, in the following order: CO (150 – 250 °C) <
CO2 (250 – 350 °C) < C2/C3 paraffins (275 – 400 °C). Moreover, Olefins were converted under all reaction conditions at lower temperatures (below 150 °C) due to the hydrogenation reaction which resulted in the formation of saturated hydrocarbons.
Furthermore, the studies on COx hydrogenation in the presence of light hydrocarbons were extended to the development of catalysts to enhance the total outlet concentration of light hydrocarbons in a COx-free product stream. The
vi effect of different transition metals (i.e. Fe, Co, Cu, Cr, Mn, Zn, Ru, Rh, Ag and
Cd) on the catalytic performance of a Ni/Al2O3 catalyst was studied. Different and distinct promoting or inhibiting influence was observed (e.g. maximum C2-C4 yield of production increased from 6% for Ni/Al2O3 to 12% for Ni-Mn/Al2O3). The characteristics of partially charged active sites of the catalysts were studeid by employing different techniques (i.e. in situ NO-FTIR, CO-/H2-TPD and chemisorption). It is found that the addition of transition metals to Ni/Al2O3 markedly changed the structure of the active sites on the primary catalyst. For example, addition of copper resulted in increasing the ratio of Carbon-accepting to Oxygen-accepting sites (i.e. NO linear/bent adsorption increased from 7.70 for
Ni/Al2O3 to 24.89 for Ni-Cu/Al2O3), which is probably increased the chance of linear CO adsorption that needs higher temperature for C – O cleavage. In contrast, by adding manganese to Ni/Al2O3 catalyst the ratio of electron accepting to donating sites balanced on the catalyst surface. Thus, most probably the number active carbon and hydrogen species increased on the surface.
Promoting effect of manganese on Ni/Al2O3 was further investigated. Catalyst activity measurements as well as various characterisation techniques (such as
XRD, CO and H2 chemisorption, in situ NO-FTIR and TPR) were performed for a series of Ni-Mn/Al2O3 catalysts with different nickel and manganese contents. It is considered that there is an optimum amount of Mn added (i.e. bi-metallic Ni-
Mn/Al2O3 catalyst with 8 wt% of nickel and 4 wt% of manganese) to the primary catalyst which enhanced the catalyst activity and selectivity. Moreover, the more hydrogen amount in the feed stream improved the catalyst activity for COx hydrogenation and selectivity toward C2-C4 production (i.e. maximum C2-C4 yield of production increased from 1.5% for ~9.5 kPa H2 to 6.5% for ~37.8 kPa H2 in
vii the feed stream over Ni/Al2O3). According to investigation of the catalysts’ electronic properties with different Ni and Mn contents, changes in catalytic activity (for COx hydrogenation) and selectivity (for light hydrocarbons formation) can be interpreted as being due to the effect of different electronic structure of the catalysts with variety of Ni/Mn ratios. The electrostatic properties of crystalline nickel and nickel-manganese particles was studied by computational methods
(i.e. KS-DFT).
Finally, this study continued on investigating the catalytic hydrogenation of COx in an industrial gas mixture containing light hydrocarbons. Complete removal of
COx present in an ethane offgas (ExxonMobil refinery, Altona, VIC) via catalytic hydrogenation (over Ni-Mn/Al2O3) was studied. The effects of adding extra hydrogen and pre-treatment of the feed stream on the process was analysed. It is found that the addition of hydrogen gas into the feed reduced the concentration of CO and CO2 to below the detection limits. By adding 25% and 40% of extra H2 to the feed stream no COx were detected in the outlet. Moreover, pre-treatment of the offgas using molecular sieves to remove water vapour from the feed gas stream did not affect COx hydrogenation at low temperatures (below 300 °C).
However, pre-treatment resulted in a significant reduction in CO and CO2 concentrations at temperatures above 300 °C. The results also confirmed the saturate gas plant ethane offgas can considerably deactivate the Ni-Mn/Al2O3 catalyst. The effect of ethane and ethylene in the feed gas stream on catalytic hydrogenation of low concentration CO and CO2 has also been investigated.
Ethane addition did not influence the hydrogenation of COx at 180 °C while it inhibited the hydrogenation reaction at 320 °C. On the other hand, ethylene addition inhibited CO and CO2 hydrogenation at both 180 °C and 320 °C.
viii List of Publications/Awards
Conferences:
. Vahid Shadravan, Eric Kennedy, Michael Stockenhuber, “The influence of the electronic structure of bi-metallic nickel-based catalysts on the catalytic
hydrogenation of CO and CO2”, International Symposium on Relations between Homogeneous and Heterogeneous Catalysis, 2018. . Vahid Shadravan, Eric Kennedy, Michael Stockenhuber, “Promoting effects of
manganese on Ni/Al2O3 for catalytic hydrogenation of COx in presence of light hydrocarbons”, Europa-cat, 2017. . Vahid Shadravan, Eric Kennedy, Michael Stockenhuber, “Hydrogenation of CO
and CO2, in presence of light hydrocarbons, over nickel catalysts", Catalysed Energy Storage in Chemicals, 2017.
. Vahid Shadravan, Eric Kennedy, Michael Stockenhuber, “CO and CO2 methanation in the presence of light alkanes and alkenes over transition metal- Ni alumina supported bi-metallic catalysts”, International Conference on Environmental Catalysis, 2016. Journals:
. Vahid Shadravan, Jason Waleson, Eric Kennedy, Michael Stockenhuber, “Effect
of manganese on selective catalytic hydrogenation of COx in the presence of light
hydrocarbons over Ni/Al2O3: An experimental and computational analysis”, Submitted to Journal of Catalysis, 2018. . Vahid Shadravan, Eric Kennedy, Michael Stockenhuber, “An experimental
investigation on the effects of adding a transition metal to Ni/Al2O3 for catalytic
hydrogenation of CO and CO2 in presence of light alkanes and alkenes”, Catalysis Today, Volume 307, 1 June 2018, Pages 277-285 (Online: 2017). https://doi.org/10.1016/j.cattod.2017.05.036 Awards:
. Winner of Faculty of Engineering and Built Environment (The University of Newcastle, Australia) Annual Poster Competition Prize (2016).
Note: The contribution of Vahid Shadravan, as the first author of the above publications, was in the forms of employing experimental/computational mythologies, data collection/analysis and preparing the original draft of each publication.
ix Table of Contents
Statement of Originality ...... ii
Statement of Contribution of Others ...... iii
Acknowledgments ...... iv
Abstract ...... vi
List of Publications/Awards ...... ix
Table of Contents ...... x
List of Figures ...... xvi
List of Tables ...... xxvii
List of Abbreviations ...... xxx
Nomenclature ...... xxxii
Chapter 1: Introduction ...... 1
1.1 Research background ...... 2
1.2 Objectives of current study ...... 4
1.3 Thesis structure ...... 5
1.4 References ...... 8
Chapter 2: Literature review ...... 10
2.1 Abstract ...... 11
2.2 Overview ...... 12
2.3 Carbon oxides catalytic hydrogenation ...... 16
2.3.1 Introduction to methanation and FTS ...... 16
2.3.2 Different catalysts ...... 19
2.3.2.1 Different supports ...... 19
2.3.2.2 Different metals ...... 21
2.3.2.2.1 Metals other than nickel ...... 23
x 2.3.2.2.2 Nickel ...... 29
2.3.3 Thermodynamics ...... 36
2.3.4 Mechanistic understanding ...... 43
2.3.5 Summary ...... 53
2.4 References ...... 55
Chapter 3: Materials and methods ...... 71
3.1 Catalyst preparation ...... 72
3.1.1 Preparation method ...... 72
3.1.2 Materials ...... 73
3.1.2.1 Catalyst support ...... 73
3.1.2.2 Metal precursors ...... 73
3.1.3 Preparation procedure ...... 74
3.2 Catalytic measurements ...... 74
3.2.1 Experimental apparatus ...... 74
3.2.2 Analytical instruments ...... 77
3.2.3 Variable gas feed composition ...... 79
3.2.4 Reactor experiments procedure ...... 80
3.2.5 Quantitative assessment of catalyst performance ...... 81
3.3 X-ray diffraction (XRD) ...... 82
3.4 In-situ Fourier transform infrared spectroscopy (in-situ FTIR) ...... 86
3.5 Temperature programmed desorption (TPD) ...... 89
3.6 Temperature programmed reduction (TPR) ...... 93
3.7 Chemisorption (CO and H2) ...... 95
3.8 Inductively coupled plasma optical emission spectroscopy ...... 97
3.9 Nitrogen adsorption manometry ...... 98
3.10 Solid-state electrostatic potential calculations ...... 101
xi 3.11 References ...... 104
Chapter 4: On the catalytic hydrogenation of carbon oxides in the presence of C1-C3 olefins and paraffins ...... 109
4.1 Abstract ...... 110
4.2 Introduction ...... 111
4.3 Experimental ...... 113
4.3.1 Catalyst Preparation ...... 113
4.3.2 Catalyst Evaluation ...... 114
4.3.3 Catalyst Characterization ...... 116
4.4 Thermodynamic Analysis ...... 116
4.5 Results and Discussion ...... 118
4.5.1 Thermodynamic Analysis ...... 118
4.5.2 Catalyst Characterization ...... 125
4.5.3 Catalyst activity assessment ...... 126
4.5.3.1 Gas phase reactions/reactions on internal surface of the reactor (Set 1) ...... 126
4.5.3.2 C2/C3 olefins and paraffins (Set 2) ...... 128
4.5.3.3 Carbon monoxide hydrogenation (Set 3) ...... 129
4.5.3.4 Effect of C2/C3 hydrocarbon addition on CO hydrogenation (Set 4) 130
4.5.3.5 Carbon dioxide hydrogenation (Set 5) ...... 131
4.5.3.6 Effect of C2/C3 hydrocarbons addition on CO2 hydrogenation (Set 6) ...... 132
4.5.3.7 Carbon monoxide and carbon dioxide co-hydrogenation (Set 7) .... 133
4.5.3.8 Effect of C2/C3 hydrocarbons addition on CO and CO2 hydrogenation (Set 8) ...... 134
4.5.3.9 Effect of C1-C3 hydrocarbons on CO and CO2 hydrogenation (Set 9) ...... 136
xii 4.6 Conclusions ...... 137
4.7 References ...... 139
Chapter 5: An experimental investigation on the effect of adding a transition metal to Ni/Al2O3 for the catalytic hydrogenation of CO and CO2 in presence of light alkanes and alkenes ...... 146
5.1 Abstract ...... 147
5.2 Introduction ...... 148
5.3 Experimental ...... 149
5.3.1 Catalyst testing ...... 149
5.3.2 Nitric oxide adsorption Fourier Transform Infra-Red Spectroscopy (NO- FTIR) ...... 151
5.3.3 Temperature programmed desorption of H2 and CO ...... 152
5.3.4 Hydrogen and carbon monoxide chemisorption ...... 152
5.4 Results and Discussion ...... 152
5.4.1 Catalyst activity ...... 152
5.4.2 NO-FTIR ...... 155
5.4.3 Temperature programmed desorption of H2 and CO ...... 162
5.4.4 Hydrogen and carbon monoxide chemisorption ...... 166
5.5 Conclusion ...... 168
5.6 References ...... 170
Chapter 6: Effect of manganese on the selective catalytic hydrogenation of
COx in the presence of light hydrocarbons over Ni/Al2O3: An experimental and computational study...... 175
6.1 Abstract ...... 176
6.2 Introduction ...... 177
6.3 Experimental methods ...... 179
6.3.1 Catalyst testing ...... 179
xiii 6.3.2 Inductively Coupled Plasma – Optical Emission Spectroscopy (ICP- OES) ...... 182
6.3.3 X-ray Diffraction analysis (XRD) ...... 182
6.3.4 Temperature Programmed Reduction (TPR) ...... 182
6.3.5 Nitric oxide adsorption Fourier Transform Infra-Red Spectroscopy (NO- FTIR) ...... 183
6.3.6 H2 and CO chemisorption ...... 184
6.4 Computational method: Solid-state electrostatic potential ...... 185
6.5 Results and discussion ...... 186
6.5.1 Catalyst testing ...... 186
6.5.2 Inductively Coupled Plasma – Optical Emission Spectroscopy (ICP- OES) ...... 196
6.5.3 X-ray Diffraction analysis (XRD) ...... 196
6.5.4 Temperature Programmed Reduction (TPR) ...... 198
6.5.5 Nitric oxide adsorption Fourier Transform Infra-Red Spectroscopy (NO- FTIR) ...... 200
6.5.6 H2 and CO chemisorption ...... 210
6.5.7 Solid-state electrostatic potential ...... 213
6.6 Conclusions ...... 221
6.7 References ...... 223
Chapter 7: Selective conversion of CO and CO2 from a refinery offgas by catalytic hydrogenation over nickel-based catalysts ...... 236
7.1 Abstract ...... 237
7.2 Introduction ...... 238
7.3 Experimental ...... 246
7.3.1 Addition of hydrogen ...... 246
7.3.2 Feed stream drying conditions ...... 247
xiv 7.3.3 Used catalyst’s performance ...... 248
7.3.4 Effects of hydrocarbons on low concentration COx hydrogenation .... 249
7.4 Results and discussion ...... 251
7.4.1 Addition of hydrogen ...... 251
7.4.2 Feed stream drying conditions ...... 253
7.4.3 Used catalyst’s performance ...... 257
7.4.4 Effects of hydrocarbons on low concentration COx hydrogenation .... 261
7.4.4.1 Set 1 (C2/C3-LT)...... 261
7.4.4.2 Set 2 (C2H6-LT)...... 262
7.4.4.3 Set 3 (C2H4-LT)...... 263
7.4.4.4 Set 4 (C2H6-HT) ...... 264
7.4.4.5 Set 5 (C2H4-HT) ...... 265
7.5 Conclusion ...... 268
7.6 References ...... 269
Chapter 8: Conclusions and recommendations ...... 274
8.1 Conclusions ...... 275
8.2 Recommendations ...... 278
8.3 References ...... 281
Appendix A ...... 283
Section 1. Pure UFCC, effect of H2 addition and pre-treatment ...... 283
Section 2. Used and reactivated catalyst ...... 285
Highlights ...... 287
xv List of Figures
Figure 2.1. Predictions about occurring the peak and subsequent decline in the global oil production [8]...... 13
Figure 2.2. NASA’s measure of pollution in the Troposphere for (a) carbon monoxide and (b) carbon dioxide in 2012 [22, 23]...... 15
Figure 2.3. A schematic view of surface polymerization during Fischer-Tropsch synthesis to form light hydrocarbons [3]...... 18
Figure 2.4. Anderson-Schulz-Flory model to predict the product distribution [3].
...... 18
Figure 2.5. Schematic of a typical Ni-Al2O3 interaction in Ni/Al2O3 catalysts [41].
...... 20
Figure 2.6. Excerpt from the periodic table of elements (active metals for methanation and FTS are highlighted)...... 21
Figure 2.7. Activities of different supported transition metal catalysts as a function of reaction energy for dissociative carbon monoxide chemisorption [45]...... 22
Figure 2.8. Catalytic performance of Ru (5 wt%) on different oxide supports. CO conversion: solid symbols. CO2 conversion: open symbols [29]...... 24
Figure 2.9. The catalytic activity of different sizes nanosized Co3O4 catalysts [60].
...... 25
Figure 2.10. Selectivity toward light hydrocarbons for COx hydrogenation over
Mn-oxide-supported Fe catalysts prepared with two different methods [31]. .... 27
Figure 2.11. CO conversion versus reaction temperature over Rh, Ru, Pt and Pd
(5 wt%)/Al2O3 [76]...... 28
xvi Figure 2.12. Illustrating the relation between electronegativity of the support oxides and the absorption frequency of bent-type NO adsorption (NO-) [27, 84].
...... 30
Figure 2.13. Illustrating the relation between electronegativity of the support oxides and turnover frequency (TOF), chain growth probability, and CH4 selectivity [27, 84]...... 31
Figure 2.14. Illustrating the shift in XPS 3p3/2 binding energy of Co and Ni and in
Co-Ni alloy catalysts on SiO2 supports: (a) Co; (b) 75CO25Ni; (c) 50Co50Ni; (d)
25Co75Ni; and (e) Ni. The total metal loading was always 10 wt% [27, 85]. .... 32
Figure 2.15. Illustrating the infrared absorbance of linear- and bent-type NO and its ratio on the alloy composition: (○) bent-type NO; (∆) linear-type NO; (●) ratio of the absorbance of bent- to linear-type band [27, 85]...... 33
Figure 2.16. Sketch of the catalyst structure and selective reactions occurring during the synthesis of methane [87]...... 34
Figure 2.17. The calculated equilibrium constants (K) of the eight reactions involved in methanation process [38, 97]...... 38
Figure 2.18. Product compositions for CO (a) and CO2 (b) methanation at equilibrium (0.1 MPa) [38, 97]...... 39
Figure 2.19. Effects of pressure and temperature on co-methanation of carbon oxides: (a) conversion and (b) CH4 yield [97]...... 39
Figure 2.20. Comparing the ASF distribution at α = 0.7 (○) to experimental data from FTS on Co-Re/γ-Al2O3 at 20% CO conversion (♦) and 50% CO conversion
(□). n is carbon number and Wn is the weight fraction [27]...... 40
Figure 2.21. Carbon selectivity at 20 bar and H2/CO = 2.1. Experimental at 483
K for a Co-Re/γ-Al2O3: C4+ (□), CH4 (●), C3 (▲), and C2 (◊) [27, 108]...... 42
xvii Figure 2.22. A general schematic presentation of CO hydrogenation chain growth mechanism via oxygenated surface species [109]...... 44
Figure 2.23. A general schematic presentation of CO hydrogenation chain growth mechanism via deoxygenated surface species (carbene mechanism) [109]. ... 45
Figure 2.24. Schematic presentation of activation of carbon monoxide on a catalyst surface giving surface carbide (carbide theory) [46]...... 45
Figure 2.25. Schematic view of CO insertion mechanism [111]...... 46
Figure 2.26. Schematic view of enol mechanism [111]...... 47
Figure 2.27. Proposed chain growth reaction and its schematic view involving alkyl as growing chain [46, 109]...... 48
Figure 2.28. Proposed chain growth reaction and its schematic view involving alkenyl as growing chain [46, 109]...... 48
Figure 2.29. Proposed chain growth reaction and its schematic view involving alkylidene as growing chain [46, 109]...... 49
Figure 2.30. Mechanism of CO methanation proposed by van Meerten et al.
Symbol * represents type 1 (ordinary) reaction site [24, 121]...... 50
Figure 2.31. Mechanism of CO methanation proposed by Hayes et al. Symbol * represents type 1 (ordinary) reaction site [24, 124]...... 51
Figure 2.32. Mechanism of CO methanation proposed by Sehested et al.
Symbols * and # represent type 1 (ordinary) and type 2 (5-fold coordinated) reaction sites respectively [24, 127]...... 52
Figure 2.33. Simplified different reaction mechanisms for CO2 methanation [33].
...... 53
Figure 3.1. Pore filling during (a) impregnation and the effect of subsequent (b) drying [1]...... 72
xviii Figure 3.2. Schematic view of the experimental apparatus...... 75
Figure 3.3. Experimental apparatus used for catalyst performance experiments: a) overview of experimental setup, b) Shimadzu-GC, c) Alicat mass flow controllers, d) Labec furnace and e) Varian Micro-GC...... 76
Figure 3.4. Typical chromatograms (Top: mFID, Bottom FID)...... 78
Figure 3.5. Schematic stretching of Bragg’s law which can be derived from the triangle ABC [9]...... 83
Figure 3.6. An example of Miller indices for cubic structure [8]...... 84
Figure 3.7. Typical data spectrum of X-ray diffraction instrument (Sample: Al2O3).
...... 85
Figure 3.8. A schematic view FTIR setup in transmission mode (TIR) [21]...... 86
Figure 3.9. a) The reaction chamber; and b) the sample holder of in-situ FTIR apparatus...... 87
Figure 3.10. The in-situ Fourier transform infrared spectroscopy system...... 88
Figure 3.11. Typical analysed spectrum of in-situ FTIR instrument (Sample:
Ni/Al2O3, adsorbate: NO)...... 89
Figure 3.12. The temperature programmed desorption (TPD) system...... 91
Figure 3.13. A typical spectrum of TPD analysis (Sample: CO desorption from
Ni/Al2O3)...... 92
Figure 3.14. The temperature programmed reduction (TPR) system...... 93
Figure 3.15. A typical spectrum for temperature programmed reduction (Sample:
Ni/Al2O3)...... 94
Figure 3.16. The volumetric chemisorption system...... 96
Figure 3.17. Classification of adsorption isotherms (IUPAC) [43]...... 99
xix Figure 3.18. Typical nitrogen adsorption/desorption isotherm (Sample: Ni/Al2O3).
...... 100
Figure 4.1. Calculated equilibrium constants for possible reactions involved in the hydrogenation of carbon oxides in presence of light hydrocarbons...... 120
Figure 4.2. Equilibrium concentrations of CO, CO2, CH4 and C2-C3 calculated without solid carbon for different H2 content...... 122
Figure 4.3. Equilibrium concentrations of CO, CO2, CH4 and C2-C3 calculated with solid carbon for different H2 content...... 123
Figure 4.4. Produced solid carbon at chemical equilibrium state for different H2 content...... 124
Figure 4.5. Nitrogen adsorption/desorption isotherm linear plot for the a) Ni/Al2O3 and b) Al2O3...... 125
Figure 4.6. XRD patterns Ni/Al2O3 and Al2O3 (Blue arrows: Al2O3, green arrows:
NiO)...... 126
Figure 4.7. Concentration of ethane and ethylene versus catalyst bed temperature for set 1. The inlet values are showed as dashed lines with the same line colour for each component...... 127
Figure 4.8. Concentration of C2H6 and C3H8 versus catalyst bed temperature for set 2. The inlet values are showed as dashed lines with the same line colour for each component...... 128
Figure 4.9. Concentration of CH4, CO and C2-C3 versus catalyst bed temperature for set 3. The inlet values are showed as dashed lines with the same line colour for each component...... 129
xx Figure 4.10. Concentrations of CH4, CO and C2-C3 versus catalyst bed temperature for set 4. The inlet values are showed as dashed lines with the same line colour for each component...... 130
Figure 4.11. Concentrations of CH4 and CO2 versus catalyst bed temperature for set 5. The inlet values are showed as dotted lines with the same line colour for each component...... 131
Figure 4.12. Concentrations of CH4, CO2 and C in C2-C3 versus catalyst bed temperature for set 6. The inlet values are showed as dotted lines with the same line colour for each component...... 132
Figure 4.13. Concentrations of CH4, CO, CO2 and C2-C3 versus catalyst bed temperature for set 7. The inlet values are showed as dotted lines with the same line colour for each component...... 134
Figure 4.14. Concentrations of CH4, CO, CO2 and C2-C3 versus catalyst bed temperature for set 8. The inlet values are showed as dotted lines with the same line colour for each component...... 135
Figure 4.15. Concentrations of CH4, CO, CO2 and C2-C3 versus catalyst bed temperature for set 9. The inlet values are showed as dotted lines with the same line colour for each component...... 136
Figure 5.1. Carbon monoxide conversion versus temperature over different catalysts...... 153
Figure 5.2. Carbon dioxide conversion versus temperature over different catalysts...... 153
Figure 5.3. Maximum yield of C2-C4 hydrocarbon produced over different catalysts...... 154
xxi Figure 5.4. Concentration changes of C2 – C4 hydrocarbons over temperature by hydrogenation of CO and CO2 in presence of light hydrocarbons...... 155
Figure 5.5. IR spectra for NO adsorption (left) over Ni/Al2O3 and temprature program desorption (right)...... 157
Figure 5.6. IR spectra for NO adsorption (left) over Ni-Cu/Al2O3 and temprature program desorption (right)...... 157
Figure 5.7. IR spectra for NO adsorption (left) over Ni-Mn/Al2O3 and temprature program desorption (right)...... 158
-1 Figure 5.8. Identified IR bands around 1850 cm for Ni/Al2O3...... 160
-1 Figure 5.9. Identified IR bands around 1850 cm for Ni-Cu/Al2O3...... 160
-1 Figure 5.10. Identified IR bands around 1850 cm for Ni-Mn/Al2O3...... 161
Figure 5.11. Deconvolution of temperature programmed desorption profiles for
Ni/Al2O3 – H2 (a), CO (b) and CO2 (c)...... 164
Figure 5.12. Diagramatic representation of the interaction of CO and H2 with neighboring sites that have different electronic properties...... 167
Figure 5.13. Diagramathic representation of the effects of adding Cu and Mn on the site structure of Ni/Al2O3...... 167
Figure 6.1. CO conversion for different feed compositions over Ni/Al2O3 (Ni:Mn
12:0) catalyst...... 187
Figure 6.2. CO2 conversion for different feed compositions over Ni/Al2O3 (Ni:Mn
12:0) catalyst...... 188
Figure 6.3. Yield of production of C2-C4 hydrocarbons for different feed compositions over Ni/Al2O3 (Ni:Mn 12:0) catalyst...... 189
Figure 6.4. CO conversion for catalysts with varying Ni/Mn ratio in the feed stream with H2/Other=4 feed composition...... 190
xxii Figure 6.5. CO2 conversion for catalysts with varying Ni/Mn ratios in the feed stream with H2/Other=4 feed composition...... 191
Figure 6.6. Yield of production of C2-C4 hydrocarbons for catalysts with varying
Ni/Mn ratios in the feed stream with H2/Other=4 feed composition...... 192
Figure 6.7. Arrhenius plots for CO hydrogenation over Ni/Al2O3 and Ni-Mn/Al2O3
(Ni/Mn = 2 and 0.5) catalysts...... 193
Figure 6.8. Arrhenius plots for CO2 hydrogenation over Ni/Al2O3 and Ni-Mn/Al2O3
(Ni/Mn = 2 and 0.5) catalysts...... 194
Figure 6.9. Arrhenius plots for CO2 hydrogenation over Ni/Al2O3 and Ni-Mn/Al2O3
(Ni/Mn = 2 and 0.5) catalysts...... 195
Figure 6.10. XRD patterns for the alumina support and all catalysts. The symbols
◆, ●, ▲, ■ and ★ were used to mark the identified reflections for Al2O3, NiO,
NiMnO3, Mn2O3 and MnO2 phases respectively...... 197
Figure 6.11. Temperature programmed reduction profiles or samples with different Ni and Mn contents...... 200
Figure 6.12. In-situ FTIR spectra for a. NO adsorption over single metal Ni/Al2O3
(Ni:Mn 12:0), followed by b. temperature programmed desorption...... 203
Figure 6.13. In-situ FTIR spectra for a. NO adsorption over bi-metallic Ni-
Mn/Al2O3 (Ni:Mn 8:4), followed by b. temperature programmed desorption. ... 204
Figure 6.14. In-situ FTIR spectra for a. NO adsorption over bi-metallic Ni-
Mn/Al2O3 (Ni:Mn 6:6), followed by b. temperature programmed desorption. ... 205
Figure 6.15. In-situ FTIR spectra for a. NO adsorption over bi-metallic Ni-
Mn/Al2O3 (Ni:Mn 4:8), followed by b. temperature programmed desorption. ... 206
Figure 6.16. In-situ FTIR spectra for a. NO adsorption over bi-metallic Ni-
Mn/Al2O3 (Ni:Mn 0:12), followed by b. temperature programmed desorption. . 207
xxiii Figure 6.17. Peak deconvolution for metal – mononitrosyl species formed on nickel containing samples during in-situ NO-FTIR analysis...... 209
Figure 6.18. The calculated value of Linear/Bent ratio of metal – mononitrosyl compounds for catalysts with different Ni/Mn ratio...... 210
Figure 6.19. Chemisorbed carbon monoxide, hydrogen and CO/H ratio on each catalyst...... 213
Figure 6.20. Surface electrostatic potential Vs(r) on 0.001 au isodensity for nickel
(left) and nickel-manganese (right) particles with the size of one crystallographic unit cell...... 215
Figure 6.21. Surface electrostatic potential Vs(r) on 0.001 au isodensity for nickel
(left) and nickel-manganese (right) particles with the size of 64 crystallographic unit cell...... 217
Figure 6.22. Diagrammatic representation of the effects of adding manganese to
Ni/Al2O3 on the hydrogenation of CO and CO2 in the presence of light hydrocarbons...... 221
Figure 7.1. The process flow diagram of a typical crude oil refinery...... 241
Figure 7.2. CO concentration for feed streams containing SGPE and different additional H2 contents over Ni-Mn/Al2O3 catalyst: (a) Full scale, (b) Zoomed-in.
...... 252
Figure 7.3. CO2 concentration for feed compositions containing SGPE and different additional H2 contents over Ni-Mn/Al2O3 catalyst...... 253
Figure 7.4. CO concentration for feed streams containing just SGPE with different drying conditions over Ni-Mn/Al2O3: (a) Full scale, (b) Zoomed-in...... 254
Figure 7.5. CO2 concentration for feed streams containing just SGPE with different drying conditions over Ni-Mn/Al2O3 catalyst...... 255
xxiv Figure 7.6. Schematic illustration of CO and CO2 formation via catalytic steam reforming of hydrocarbons...... 257
Figure 7.7. CO concentration for simulated feed stream containing H2, COx and hydrocarbons over fresh and used Ni-Mn/Al2O3 catalyst...... 258
Figure 7.8. CO2 concentration for simulated feed stream containing H2, COx and hydrocarbons over fresh and used Ni-Mn/Al2O3 catalyst...... 259
Figure 7.9. CO and CO2 concentration for the switch experiment (Table 7.6) with simulated feed, SGPE and H2 streams (dashed lines: inlet concentrations). .. 260
Figure 7.10. CO concentration for the C2/C3-LT experiment (Table 7.8) with feed streams A and B (Table 7.7). Dashed line shows inlet concentration...... 262
Figure 7.11. CO concentration for the C2H6-LT experiment (Table 7.8) with feed streams A and C (Table 7.7). Dashed line shows inlet concentration...... 263
Figure 7.12. CO concentration for the C2H4-LT experiment (Table 7.8) with feed streams A and D (Table 7.7). Dashed line shows inlet concentration...... 264
Figure 7.13. CO and CO2 concentration for the C2H6-HT experiment (Table 7.8) with feed streams A and C (Table 7.7). Dashed line shows inlet concentration.
...... 265
Figure 7.14. CO and CO2 concentration for the C2H4-HT experiment (Table 7.8) with feed streams A and D (Table 7.7). Dashed line shows inlet concentration.
...... 266
Figure Appa. 1. CO concentration for feed streams containing UFCC and different additional H2 contents over Ni-Mn/Al2O3 catalyst (dashed lines show inlet concentrations)...... 284
xxv Figure Appa. 2. CO2 concentration for feed streams containing UFCC and different additional H2 contents over Ni-Mn/Al2O3 catalyst (dashed lines show inlet concentrations)...... 284
Figure Appa. 3. CO concentration for hydrogen-rich feed stream over used and reactivated Ni-Mn/Al2O3 catalyst (dashed lines show inlet concentrations). .... 286
Figure Appa. 4. CO2 concentration for hydrogen-rich feed stream over used and reactivated Ni-Mn/Al2O3 catalyst (dashed lines show inlet concentrations). .... 286
xxvi List of Tables
Table 2.1. Main possible reactions involved in methanation of carbon oxides [38].
...... 37
Table 3.1. Standard specifications of Sasol alumina spheres, used in this study.
...... 73
Table 3.2. Different transition metal catalyst precursors...... 73
Table 3.3. Method parameters of data acquisition by Micro-GC...... 77
Table 3.4. Method parameters of data acquisition by Shimadzu GC...... 79
Table 3.5. Estimated detection limits for each components based on S/N method.
...... 79
Table 3.6. Gas specifications...... 80
Table 4.1. Feed compositions for each set of experiments...... 115
Table 4.2. Standard properties of possible components in the system. The bracket is the reference used to source the thermodynamic for each species.
...... 118
Table 4.3. Possible chemical reactions and their standard properties...... 119
Table 4.4. Feed compositions used in thermodynamic analysis...... 121
Table 4.5. N2 adsorption/desorption results for the catalyst and its support. .. 125
Table 5.1. Feed stream composition...... 150
Table 5.2. caclulated linear-/bent-type metal – mononitrosyl for Ni/Al2O3, Ni-
Cu/Al2O3 and Ni-Mn/Al2O3 catalysts...... 161
Table 5.3. Calculated percentage of the area for each fitted peak on H2-TPD profiles of Ni/Al2O3, Ni-Cu/Al2O3 and Ni-Mn/Al2O3 catalysts...... 163
Table 5.4. Calculated percentage of the area for each fitted peak on CO-TPD profiles of Ni/Al2O3, Ni-Cu/Al2O3 and Ni-Mn/Al2O3 catalysts...... 165
xxvii Table 5.5. Calculated percentage of the area for each fitted peak on CO2-TPD profiles of Ni/Al2O3, Ni-Cu/Al2O3 and Ni-Mn/Al2O3 catalysts...... 165
Table 5.6. Chemisorbed CO and H (μmol gas/g catalyst) over Ni/Al2O3, Ni-
Cu/Al2O3 and Ni-Mn/Al2O3...... 166
Table 6.1. The labels and calculated metal contents of catalysts...... 179
Table 6.2. Composition of each feed stream ...... 181
Table 6.3. Calculated activation energies for the reaction of CO, CO2 and C2H6 with H2 over Ni:Mn 12:0, Ni:Mn 8:4 and Ni:Mn 4:8 catalysts...... 193
Table 6.4. Metal contents of each catalyst based on ICP-OES analysis...... 196
Table 6.5. Dispersion, specific surface area and particle size calculated for metal particles based on H2 chemisorption...... 211
Table 6.6. Dispersion, specific surface area and particle size calculated for nickel particles based on CO chemisorption...... 212
Table 7.1. The composition of Saturate Gas Plant Ethane Offgas mixture supplied by Mobil Refining Australia...... 246
Table 7.2. Feed stream composition with different amounts of hydrogen added to the SGPE...... 247
Table 7.3. The conditions and materials used for each pre-treatment methods.
...... 247
Table 7.4. The hydrogen-rich feed stream composition for analysing the performance of used Ni-Mn/Al2O3 catalysts...... 248
Table 7.5. The experiment details and pre-run conditions for fresh and used Ni-
Mn/Al2O3 catalysts...... 248
Table 7.6. Details of each step for the switch experiment with the simulated feed stream (Table 7.4) and SGPE...... 249
xxviii Table 7.7. Feed compositions for each feed stream in this section...... 250
Table 7.8. Details of each step for each switch experiment with the low concentration COx feed compositions (Table 7.7)...... 250
Table Appa. 1. The composition of UFCC Offgas supplied by Mobil Refining
Australia...... 283
Table Appa. 2. Feed stream composition with different amounts of hydrogen added to the UFCC...... 283
Table Appa. 3. The hydrogen-rich feed stream composition for analysing the performance of used Ni-Mn/Al2O3 catalysts...... 285
Table Appa. 4. The experiment details and pre-run conditions for used and reactivated Ni-Mn/Al2O3 catalysts...... 285
xxix List of Abbreviations
ASF Anderson-Schulz-Flory
BET Brunauer-Emmett-Teller
BJH Barrett-Joyner-Halenda
COD Crystallography Open Database
DFT Density Functional Theory
DR Dry Reforming
DRM Dry Reforming of Methane
FCC Fluid Catalytic Cracking
FID Flame Ionization Detector
FTIR Fourier-Transform Infrared
FTS Fischer-Tropsch Synthesis
GC Gas Chromatography
GGA Generalized Gradient Approximation
HCO Heavy Cycle Oil
HPS High-Pressure Separator
HS-D HaySep-D
ICP-OES Inductively Coupled Plasma Optical Emission Spectrometry
ICSD Inorganic Crystal Structure Database
IUPAC International Union of Pure and Applied Chemistry
KS-DFT Kohn-Sham Density Functional Theory
LCO Light Cycle Oil
LN2-MCT Liquid Nitrogen cooled Mercury Cadmium Telluride
LOD Limit of Detection
LOQ Limit of Quantification
xxx LPG Liquefied Petroleum Gas mFID methanizer Flame Ionization Detector
MS5A Molecular-Sieve 5 Angstrom
MTO Methanol to Olefins
OCM Oxidative Coupling of Methane
OD Outside Diameter
POX Partial Oxidation
RWGS Reverse Water Gas Shift
SC Steam Cracking
SCF Self-Consistent Field
SGPE Saturate Gas Plant Ethane
SMSI Strong Metal Support Interaction
SNG Synthetic Natural Gas
SR Steam Reforming
TC Thermal Cracking
TCD Thermal Conductivity Detector
TIR Transmission Infrared
TOF Turnover Frequency
TPD Temperature Programmed Desorption
TPR Temperature Programmed Reduction
UHV Ultra-High Vacuum
WGC Wet Gas Compressor
WGS Water Gas Shift
XPS X-ray Photoelectron Spectroscopy
XRD X-ray Diffraction
xxxi Nomenclature
Ci Concentration of i
Ai Chromatography peak area of i
RFi Chromatography response factor of i
Xr Conversion of r
Sp Selectivity of p
Yp Yield of p
TC Total mole of carbon d’ Spacing between atomic layers
θ Incident angle
λ Wavelength
φa, φb, φc Incident waves angle
φa0, φb0, φc0 Propagating waves angle h, k, l Miller indices
Ψ Wave function of the quantum system
Ĥ Electronic Hamiltonian operator
E Energy of the state
ϕi One-electron wave function (orbital)
N Number of electrons r one electron’s space coordinate
σ one electron’s spin coordinate ci Corresponding coefficient
χ Atomic orbital
μ, ν Local orbitals
Ρμν Density matrix
xxxii Ωμν Mono-electronic properties matrix
ρ̂(R) Electron density at point R
V̂(R) Electrostatic potential at point R
T Temperature
K Equilibrium constant
∆H° Standard enthalpy of reaction
∆G° Standard Gibbs free energy of reaction
∆S° Standard entropy of reaction
° ∆Hf Standard enthalpy of formation
S° Standard entropy
Gt Total Gibbs free energy of the system
푦푖 Molar fraction of i ngas Number of moles of gas adsorbed on the metal particles
Fw Formula weight of the particle
Sf Stoichiometric factor wt% Weight percent of the metal particle in the sample
Wsample Weight of sample
σm Atomic cross sectional area of metal
NA Avogadro number
ρm Metal particle density
D Metal dispersion d Particle size
S Specific surface area
ZA Charge of the nucleus of atom A
Vs(r) Electrostatic potential on surface
xxxiii
Chapter 1
Introduction
Chapter 1
1.1 Research background
Light hydrocarbons (C2-C4 alkane and alkenes) are used as feedstocks in a wide variety of chemical and petrochemical processes to synthesize products such as liquid fuels, polymers and solvents [1, 2]. Light hydrocarbons are primarily produced in petrochemical plants from carbonaceous feedstocks (mainly from fossil sources) [3, 4]. Rapid depletion of fossil hydrocarbon resources, as result of industrial and population growth and increasing energy demand, is one of the major challenges of the 21st century. Coincidently, and due to economic and environmental issues, many countries are focusing on designing processes to decrease their dependency on fossil fuel resources (oil, gas and coal) [5]. Carbon oxides emission, as industrial by-products of most processes involving hydrocarbon synthesis, invokes serious environmental issues and negatively affects commercialisation of many new processes [6, 7]. Thus, producing a CO and CO2 (COx) free (or minimal COx) synthetic hydrocarbon stream is necessary to facilitate commercialization of these new processes as large scale industrial plants [8]. Moreover, due to the significant environmental effect of COx, it is critical to develop processes to convert COx and reduce their emission into the atmosphere [9]. Designing processes to enhance efficiency and in turn reduce
COx emission has been extensively studied in recent decades. The raising interests of different industrial enterprises, applying beneficial procedures to reduce COx, will continue to drive research and development of catalysts and processes for the production of hydrocarbons from carbon oxides through a direct route [10, 11].
Catalytic hydrogenation of COx (e.g. catalytic methanation and Fischer-Tropsch synthesis) to light hydrocarbons is one of the main direct routes that offer flexibility
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 2 Chapter 1
in feed composition and associated low cost process conditions [12, 13].
However, designing a catalyst with high activity and selectivity for long-term application remains elusive. Amongst the many novel catalysts studied, transition metals catalysts have been proposed for catalytic hydrogenation of COx. In this case, nickel is one of the most widely studied, as it is considered to be a relatively low cost and highly active metal for Fischer-Tropsch synthesis and COx methanation [14, 15]. Nickel, when compared to other active transition metals such as Fe and Co, is more selective for the synthesis of light hydrocarbons
(especially methane) from COx reaction with H2. One important obstacle for further development of nickel-based catalysts is its relatively rapid rate of deactivation due to formation of volatile carbonyl species and carbon deposition.
Studies on the development and modification of nickel catalysts for production of long chain hydrocarbons have continued since the pioneering work published in
1933 by Sabatier and Senderens [16, 17]. Since then, research has continued, especially focussed on the modification of nickel based catalysts to enhance its selectivity toward production of long chain hydrocarbons. On the other hand, methanation of COx over nickel catalysts has been widely studied, especially for producing synthetic natural gas from coal and biomass [18]. Different nickel- based catalysts for COx hydrogenation has been prepared and there are numerous publications available dealing with different aspects of the catalyst and the process. However, the research strategies adopted to modify the structure and electronic properties of nickel catalysts by different methods (such as utilization of metal support interaction, optimization of particle size and using different promoters) to enhance higher C2+ hydrocarbon yields has only been moderately investigated. It has been suggested that catalyst modification by
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 3 Chapter 1
preparing bi-metallic catalysts could be a fruitful pathway to improve catalytic performance of nickel catalysts (e.g. enhancing selectivity toward higher hydrocarbons production) [17]. This study focused on catalytic hydrogenation of
COx in the presence of light hydrocarbons over nickel based catalysts. The main objectives of the study are provided in the following section.
1.2 Objectives of current study
The main objective of this project is develop and understanding of the processes involved in catalytic hydrogenation of COx. The focus is in particular on processes that can reduce the residual concentration of both CO and CO2 in the product gas stream to less than 1 ppm in the presence of light hydrocarbons without reducing the inlet concentration of feed hydrocarbons. In addition, the development of catalysts which result in the enhancement of the total outlet concentration of light hydrocarbons in a COx-free product stream was studied. The catalyst development has proceeded based on experimental results and computational analysis techniques. The key steps of this research procedure are briefly explained below.
Studying the influence of light hydrocarbons on catalytic hydrogenation of
CO and CO2.
Undertaking thermodynamic simulation of a reacting system containing
H2, CO, CO2, CH4 and C2-C3 light olefins/paraffins to determine the effects
of different parameters (e.g. feed composition, temperature and etc.) on
the system’s equilibrium state.
Investigating the influence of different reaction conditions, such as feed
composition and reaction temperature on catalyst performance.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 4 Chapter 1
Analysing the performance of different modified bi-metallic nickel-based
catalysts for “complete” removal of COx and enhancing the yield of
production for light hydrocarbons in a feed stream containing hydrogen,
carbon oxides and light hydrocarbons.
Determining the properties and characteristics of nickel based bi-metallic
catalysts (e.g. oxidation state, elemental composition, crystallographic
properties, surface area, pore structure, metal dispersion, particle size
etc.) for further catalyst and process development.
Achieving insight into the nature of catalyst modification effects on the
catalyst performance by employing experimental and computational
methods such as in-situ spectroscopy analysis of catalysts surface,
density functional theory calculations, monitoring adsorption and
desorption behaviour of probed catalysts.
Studying complete removal of carbon oxides from an industrial feed
stream containing light hydrocarbons and hydrogen.
1.3 Thesis structure
The structure of this research thesis is explained in this section.
Chapter 1 provides a brief introduction on the research background of this project and outline and explain the main objectives and motivation of current study.
Previous investigations on the catalytic hydrogenation of carbon oxides are reviewed and highlighted in chapter 2. In this case, the industrial importance of synthetic hydrocarbons and the necessity of reducing COx emission is disclosed.
A general overview of carbon oxides hydrogenation processes is provided focusing on catalytic methanation and Fischer-Tropsch synthesis. The chapter
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 5 Chapter 1
covers different aspects of these processes such as catalyst development, thermodynamic analyses and mechanistic understanding.
Chapter 3 contains detailed information about different experimental and analytical methods employed in this project. In this case, description of different materials (e.g. chemicals, gases and etc.), catalyst preparation techniques and catalytic reaction analysis methods are included. Moreover, catalyst characterization methods (e.g. BET, TPR, TPD, in-situ FTIR, XRD, ICP-OES and chemisorption) are explained.
Investigation of the catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons over a benchmark nickel-based hydrogenation catalyst,
Ni/Al2O3, is presented in chapter 4. The influence of adding methane and C2-C3 alkanes and alkenes were studied as differing feed compositions. Catalyst characterization using techniques such as nitrogen adsorption manometry and
X-ray diffraction analysis are also presented. In addition, the chapter contains a thermodynamic analysis of the system’s equilibrium state at different conditions
(e.g. different reaction temperature, H2 content and etc.).
In chapter 5, the effect of adding a secondary transition metal to the alumina supported nickel catalyst for COx hydrogenation in the presence of light hydrocarbons is studied. In this case, a series of bimetallic Ni-M/Al2O3 (M: Fe,
Co, Cu, Cr, Mn, Zn, Ru, Rh, Ag, Cd) were prepared and analysed. The promoted bi-metallic catalysts were characterised by different experimental techniques.
Selected bi-metallic catalysts (Mn- and Cu-promoted) which either enhanced or inhibited the activity and selectivity of the primary Ni/Al2O3 catalyst were further studied by different techniques such as in-situ Fourier transform spectroscopy technique using nitric oxide as probe molecule.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 6 Chapter 1
Chapter 6 presents a detailed investigation on the promoting effects of manganese on Ni/Al2O3 for catalytic hydrogenation of CO and CO2 in feed stream containing light hydrocarbons. In this case, Ni-Mn/Al2O3 catalysts with different
Ni/Mn content ratio were prepared and analysed. The chapter contains catalyst activity measurements for all catalysts. Formation of different Ni-Mn species were determined by experimental techniques such as X-ray diffraction analysis. All prepared catalysts were also characterised by other methods such as chemisorption (using H2 and CO). In addition, selected catalysts were also examined by temperature programmed reduction and in-situ Fourier transform spectroscopy (using NO as probe molecule). The nature of active species containing nickel and manganese atoms were also investigated by employing density functional theory (DFT) computational method.
The possibility of applying this process (i.e. carbon oxides removal from a feed stream containing light hydrocarbons without reducing the inlet concentration of hydrocarbons), using an actual industrial gas mixture as feed stream, is presented in chapter 7. In this case, the catalytic hydrogenation process was studied using saturate gas plant ethane offgas (ExxonMobil refinery, Altona, VIC) over Ni-Mn/Al2O3 catalyst. Effects of different process conditions (such as adding extra H2) and feed stream pre-treatment were studied. In addition, the effects of ethane and ethylene on the catalytic hydrogenation of low concentration CO and
CO2 has been studied.
Chapter 8 presents the main conclusions of this research project. Some recommendations for researchers interested in continuing this field of study are also provided in the final chapter.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 7 Chapter 1
1.4 References
[1] H.M. Torres Galvis, K.P. De Jong, Catalysts for production of lower olefins from synthesis gas: A review, ACS Catalysis, 3 (2013) 2130-2149.
[2] E. Saldivar-Guerra, E. Vivaldo-Lima, Handbook of Polymer Synthesis,
Characterization, and Processing, Wiley2013.
[3] R.A. Meyers, Handbook of Petroleum Refining Processes, McGraw-Hill
Education2003.
[4] H. Jahangiri, J. Bennett, P. Mahjoubi, K. Wilson, S. Gu, A review of advanced catalyst development for Fischer-Tropsch synthesis of hydrocarbons from biomass derived syn-gas, Catalysis Science & Technology, 4 (2014) 2210-2229.
[5] P.B. Weisz, Basic Choices and Constraints on Long-Term Energy Supplies,
Physics Today, 57 (2004) 47-52.
[6] J.H. Lunsford, Catalytic conversion of methane to more useful chemicals and fuels: A challenge for the 21st century, Catal. Today, 63 (2000) 165-174.
[7] R. Razzaq, C. Li, S. Zhang, Coke oven gas: Availability, properties, purification, and utilization in China, Fuel, 113 (2013) 287-299.
[8] V. Shadravan, E. Kennedy, M. Stockenhuber, An experimental investigation on the effects of adding a transition metal to Ni/Al2O3 for catalytic hydrogenation of CO and CO2 in presence of light alkanes and alkenes, Catalysis Today,
(2017).
[9] C. Song, Global challenges and strategies for control, conversion and utilization of CO2 for sustainable development involving energy, catalysis, adsorption and chemical processing, Catalysis Today, 115 (2006) 2-32.
[10] M.A. Vannice, The Catalytic Synthesis of Hydrocarbons from Carbon
Monoxide and Hydrogen, Catalysis Reviews, 14 (1976) 153-191.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 8 Chapter 1
[11] A.Y. Khodakov, W. Chu, P. Fongarland, Advances in the development of novel cobalt Fischer-Tropsch catalysts for synthesis of long-chain hydrocarbons and clean fuels, Chemical Reviews, 107 (2007) 1692-1744.
[12] B. Jager, R. Espinoza, Advances in low temperature Fischer-Tropsch synthesis, Catalysis Today, 23 (1995) 17-28.
[13] S. Rönsch, J. Schneider, S. Matthischke, M. Schlüter, M. Götz, J. Lefebvre,
P. Prabhakaran, S. Bajohr, Review on methanation – From fundamentals to current projects, Fuel, 166 (2016) 276-296.
[14] G.H. Watson, I.C. Research, Methanation Catalysts, IEA Coal
Research1980.
[15] M.E. Dry, J.C. Hoogendoorn, Technology of the Fischer-Tropsch Process,
Catalysis Reviews, 23 (1981) 265-278.
[16] P. Sabatier, J.B. Senderens, New methane synthesis, C. R. Acad. Sci. Paris,
134 (1902) 514-516.
[17] B.C. Enger, A. Holmen, Nickel and Fischer-Tropsch Synthesis, Catal. Rev.-
Sci. Eng., 54 (2012) 437-488.
[18] J. Kopyscinski, T.J. Schildhauer, S.M.A. Biollaz, Production of synthetic natural gas (SNG) from coal and dry biomass - A technology review from 1950 to
2009, Fuel, 89 (2010) 1763-1783.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 9
Chapter 2
Literature review
Chapter 2
2.1 Abstract
In this chapter, a summary of previous research studies on catalytic hydrogenation of CO and CO2 is presented. The chapter contains a general overview on the industrial importance of light hydrocarbons and the necessity of controlling carbon oxides emission from economic and environmental perspectives. The catalytic hydrogenation of carbon oxides processes developed, Fischer-Tropsch synthesis and methanation, are highlighted.
Different scientific investigations and research publications were reviewed and focussed on three important fundamental aspects: catalysts, thermodynamics and reaction mechanism.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 11 Chapter 2
2.2 Overview
Low molecular weight saturated and olefinic hydrocarbons (especially C1-C3) are among the most widely used of all organic chemicals, and are being used for application as energy sources and industrial feed stocks. Various solvents, liquefied petroleum gas (LPG), refrigeration gases (e.g. R-290a and R-600a), synthetic rubbers, different types of plastics, polyesters and polystyrene are examples of the myriad of materials that are synthesised from light hydrocarbons or their derivatives [1-3]. Light hydrocarbons are currently produced as primary or by-products around the world at different petrochemical plants and refineries and from various feedstocks. Fluid catalytic cracking (FCC), steam cracking (SC), thermal cracking (TC), Fischer-Tropsch synthesis (FTS) and methanol to olefins
(MTO) are among the most important processes to produce light hydrocarbons.
Fossil fuel sources such as coal, oil and natural gas are primary feedstocks for direct and indirect production of light hydrocarbons [4-6]. It has been predicted that natural gas and oil resources (formed over millions of years) will be completely consumed in the next few decades (probably less than a human lifespan). Figure 2.1 shows the predicted period of time in which the peak of oil supply is probably occur for demand whose rate of growth may be up to 2 % [7-
9].
Carbon oxides are an almost inevitable unwanted by-product of hydrocarbon synthesis processes. Moreover, the presence of carbon oxides in hydrocarbon feedstocks can have deleterious effects on some processes and catalysts. In general, an undesirable concentration of carbon oxides can act as the limiting factor for commercializing a process [7, 10]. Therefore, removing or separating carbon oxides from hydrocarbon product streams is critical to enhance the
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 12 Chapter 2
economic viability of the process. Under these circumstances, it is necessary that removing carbon oxide does not reduce the concentration of value-added hydrocarbons in the stream [11].
Figure 2.1. Predictions about occurring the peak and subsequent decline in the global oil production [8].
For example, one of the processes to synthesis light hydrocarbons is direct conversion of methane using oxidative coupling (OCM). Although there are environmental and financial motivations to convert methane directly to value- added products, the process has not been commercialised yet [12]. In order to commercialise the OCM process, single-pass C2+ yield and selectivity need to be greater than 30 and 90% respectively [13]. Developing efficient catalysts for the oxidative coupling of methane that approach the techno-economic targets has been challenging [12]. One reason for the low reported C2+ yields on almost all
OCM catalysts is the formation of undesired by-products (mainly CO, CO2, and
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 13 Chapter 2
solid graphitic carbon) under the process conditions (high reaction temperature).
Calculated change in the Gibbs free energy, for several reactions that can occur in the process, shows that although the formation of C2 hydrocarbons is thermodynamically favourable (∆G < 0) the formation of CO and CO2 is even more energetically downhill [12].
In addition to the deleterious influence of CO and CO2 on the commercial development of many hydrocarbon synthesis processes, carbon oxides have also significant environmental effects. Carbon dioxide is one of the major anthropogenic greenhouse gases. The increase in the concentration of atmospheric carbon dioxide has resulted in a change in the climate around the world due to global warming [14]. Moreover, the presence of carbon monoxide in the atmosphere influences concentrations of other greenhouse gases such as methane, ozone and carbon dioxide. Figure 2.2 illustrates the distribution of carbon monoxide and carbon dioxide in the Troposphere in early 2012 [15].
Extensive research activities have been carried out, mainly focussing on the design of processes to increase efficiency and in turn reduce of carbon oxide emissions. Carbon monoxide and carbon dioxide have been used as reactants in different processes to produce various hydrocarbons such as alcohols, light olefins, synthetic natural gas and liquid fuel [16-21].
The primary focus of the current study is aimed at the development of catalysts which quantitatively remove carbon monoxide and carbon dioxide from feed streams containing light hydrocarbons, and where the process does not reduce the inlet concentration of C2+ hydrocarbons. The carbon oxides removal process is conversion of CO and CO2 by catalytic hydrogenation. Therefore, published
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 14 Chapter 2
studies on various aspects of catalytic hydrogenation of COx to light hydrocarbons is reviewed and analysed in the following sections of this chapter.
Figure 2.2. NASA’s measure of pollution in the Troposphere for (a) carbon monoxide and (b) carbon dioxide in 2012 [22, 23].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 15 Chapter 2
2.3 Carbon oxides catalytic hydrogenation
As it was pointed out in the previous section, this study has two main objectives as below:
- “Complete” removal of CO and CO2 from a product stream in the presence of light hydrocarbon product gases.
- Enhancing (or certainly not reducing) the inlet concentration of light hydrocarbons (C2+).
As two of the most efficient processes in catalytic conversion of COx to light hydrocarbons, COx methanation and Fischer-Tropsch synthesis (FTS) to light hydrocarbons have attracted researchers’ interests over last century. These processes have been extensively studied for different applications [3, 24-27].
Based on the published investigations by other research groups, the following sections provide information about methanation and FTS to light hydrocarbons1.
2.3.1 Introduction to methanation and FTS
The COx methanation is referred to the simultaneous hydrogenation of CO and
CO2 (Eq. 2.1-2). Both CO and CO2 hydrogenation reactions are exothermic and produce methane and water from COx and hydrogen [24, 28].
° CO + 3H2 → CH4 + H2O; ∆H298 K = −206.2 kJ⁄mol Eq. 2.1
° CO2 + 4H2 → CH4 + 2H2O; ∆H298 K = −165 kJ⁄mol Eq. 2.2
1 For more convenience, the abbreviation FTS refers to Fischer-Tropsch synthesis to light hydrocarbons in this thesis.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 16 Chapter 2
Even though the rate of CO2 hydrogenation reaction is generally higher than CO, preferential hydrogenation of CO prevails at lower temperatures. This has been explained as being the result of a weaker adsorption energy of CO2 compared with CO [28]. As the reaction temperature increases, CO2 adsorption become stronger and its competitive adsorption over CO increases, and thus the CO2 methanation becomes more pronounced [28]. Moreover, under specific conditions the reverse water gas shift reaction (Eq. 2.3) becomes more evident.
Although the reverse water gas shift is an endothermic process which usually occurs at higher temperature than that of CO hydrogenation, it can limit the net conversion of CO [29].
° CO2 + H2 → CO + H2O; ∆H298 K = 41.1 kJ⁄mol Eq. 2.3
In addition to methanation process, the catalytic reaction of carbon monoxide and carbon dioxide with hydrogen to produce hydrocarbons and alcohols is called
Fischer-Tropsch synthesis [30-32]. The nature of FT reaction may be considered as a catalytic surface polymerization. The product stream contains a chain of products instead of a single species. Even though the FT mechanism has been widely studied for several years, many aspects of the process remain unexplained. However, it is widely accepted that the FT reaction occurs via a surface carbide chain mechanism as shown schematically in Figure 2.3 [3].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 17 Chapter 2
Figure 2.3. A schematic view of surface polymerization during Fischer-Tropsch synthesis to form light hydrocarbons [3].
Figure 2.4. Anderson-Schulz-Flory model to predict the product distribution [3].
Anderson-Schulz-Flury (ASF) model can be applied to predict the product distribution based on a chain growth probability (α) mechanism. Chain growth
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 18 Chapter 2
probability factor can be influenced by different parameters, such as process conditions, catalyst type and the presence of chemical promoters. Figure 2.4 illustrates the ASF model product distribution as a function of chain growth probability (α) [3].
Based on the ASF model, the optimum value for chain growth probability factor to achieve maximum selectivity for light olefins is between 0.4 and 0.5. One of the most effective parameters for influencing the chain probability factor is the process temperature. Under these conditions, the process temperature should be decreased for having a product stream with low alpha values [3].
2.3.2 Different catalysts
The COx methanation and Fischer-Tropsch synthesis to light hydrocarbon processes are deemed to be relatively efficient processes for conversion of carbon oxides to valuable hydrocarbons. Both of these catalytic heterogeneous processes have been widely studied in recent years. Not surprisingly, there is a wide variety of catalysts which have been developed for these processes.
Supported catalysts, which display high levels resistance to deactivation (e.g. sintering), high surface area (i.e. higher dispersion of nano-scale active components), and reduction of costs and mechanical/morphological properties of different supports are some of the more important advantages of supported catalysts [3, 4, 33-35]. In this section, the most common and efficient supported- catalysts for both COx methanation and Fischer-Tropsch synthesis (FTS) are presented and discussed.
2.3.2.1 Different supports
The physiochemical properties of a catalyst’s support, such as chemical composition, structure and thermal stability and the presence of defect groups
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 19 Chapter 2
are important aspects to consider. The support materials and their interaction with the active metal (known as strong metal-support interaction or SMSI) can affect the catalyst’s activity, selectivity and durability. Thus, many studies have been focussed on analysing the performance of different supports in various catalysts for hydrogenation of carbon monoxide and carbon dioxide. Materials with different specific surface area, chemical composition, structure and acidity/basicity have been investigated [3, 33, 36-38]. Common supports for methanation and FTS catalysts are high surface area crystalline oxides such as alumina (Al2O3), silica (SiO2) and titania (TiO2). Amongst these, Al2O3 is most frequently used [24].
Because of different crystallographic phases (e.g. α, θ, δ, κ and γ), aluminium oxide is considered to be a complicated support compared to other oxide supports (i.e. SiO2 and TiO2) [39]. Among different Al2O3 modifications, γ- Al2O3 has higher surface area, well determined surface acid – base properties and pore structure [40].
Figure 2.5. Schematic of a typical Ni-Al2O3 interaction in Ni/Al2O3 catalysts [41].
A study on nickel catalysts supported on different commercial γ- Al2O3 reported that the properties of gamma alumina strongly affects catalytic performance
(reactants conversion and selectivity toward desired products) for CO2
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 20 Chapter 2
methanation [42]. Effects of alumina pre-treatment (e.g. calcination) on catalytic performance of nickel catalysts has been also investigated (Figure 2.5) [41]. One of the main challenges associated with the use of alumina supports for COx hydrogenation is sintering in the presence of water vapour [38].
2.3.2.2 Different metals
More than a century after the discovery of CO hydrogenation by Sabatier and
Senderens [43], many transition metals (mainly in group VIIIB) have been found to be active for methanation and FTS. These metals are highlighted in Figure 2.6.
In this case, different catalytic systems have been developed, based on metals that demonstrate COx hydrogenation activity. Among these metals Ru, Fe, Co and Ni are considered to be sufficiently active for this purpose. It is reported that the intrinsic activity of these four main metals for COx hydrogenation is: Ru > Fe
> Ni > Co [16, 26, 32].
Figure 2.6. Excerpt from the periodic table of elements (active metals for methanation and FTS are highlighted).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 21 Chapter 2
In commercial applications, Fe and Co-based catalysts are used for FTS while
Ni-based catalysts are used mostly in methanation plants. Ruthenium, because of its scarcity and high price, is rarely used as a commercial catalyst [20, 44].
It is well known that the COx hydrogenation process involves CO, CO2 and H2 adsorption and dissociation. Many reports have confirmed that the reaction rate depends on the rate of carbon monoxide dissociation [38, 45, 46]. The reaction energy for dissociative adsorption of carbon monoxide at 550 K has been calculated for some of COx hydrogenation active metals (Figure 2.7). It is found that there is a volcano-type relation between the CO dissociation energy and methanation activation energy [45]. These experimental investigations were in good agreement with the predictions by other researchers [47].
Figure 2.7. Activities of different supported transition metal catalysts as a function of reaction energy for dissociative carbon monoxide chemisorption [45].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 22 Chapter 2
2.3.2.2.1 Metals other than nickel
Ruthenium-based catalysts are well known and widely studied for application in both methanation and FT synthesis. Ru is one of the most active metals for methanation and FTS, especially for reactions at low temperatures [48].
However, it is not commonly used in industrial scale plants due to its very high price compared to other catalysts. For instance, ruthenium costs almost 120 times more than nickel on a mass basis [24].
Many researchers have investigated the influence of process conditions such as reaction temperature, gas hourly space velocity and catalyst preparation methods on Ru/Al2O3 catalysts [49-51]. The effect of catalyst particle size and the extent of metal loading have been described elsewhere by others [48, 50, 52, 53]. It is found, based on kinetic measurements, that COx methanation rates are higher with larger particle size of the active Ru phase supported on either alumina- or titania [48]. The optimum metal loading on the support could vary, due to the effects of the preparation and catalyst testing parameters [52, 53]. It has been reported that catalysts with lower metal loading are preferable as they are active over a wider range of temperatures [52]. A Ru/TiO2 nano-catalyst (2.5 nanometre) with 0.8 wt% metal loading was found to produce a 100 % yield of methane at low temperature (160 °C), and no evidence of catalyst deactivation was detected for this catalyst over 170 hours time-on-stream analysis. A direct correlation between COx hydrogenation activity and Ru particle size was concluded [54].
The effect of different supports on the selectivity and activity of Ru-based catalysts have been published (Figure 2.8) [29, 48, 55, 56]. Some results confirm the higher activity of TiO2 for methanation reaction in comparison with other oxide supports [48, 55].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 23 Chapter 2
Figure 2.8. Catalytic performance of Ru (5 wt%) on different oxide supports. CO
conversion: solid symbols. CO2 conversion: open symbols [29].
Cobalt is a widely studied and active metal for COx hydrogenation, especially in
FTS. However, due to its higher price compared to Ni, it hasn’t been widely used for industrial methanation processes [57]. The relationship between the oxidation state of cobalt and methane selectivity has been reported. It is found that incomplete reduction of cobalt catalysts or even presence of small quantities of
Co3O4 results in a higher selectivity toward CH4 production [58, 59]. Smaller particle sizes of Co metal, and higher level of dispersion of the metal particles were reported to enhance complete conversion of CO at low temperatures
(Figure 2.9) [60].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 24 Chapter 2
Figure 2.9. The catalytic activity of different sizes nanosized Co3O4 catalysts [60].
Iron-based catalysts, due to their low selectivity for methane production, were mainly used in FTS process [24]. The supported iron-based catalysts were studied extensively and developed in past century. Publications on iron catalysts which display high selectivities toward light hydrocarbons have been reported.
Iron oxide catalysts, modified by four rare metal oxides on heat treated gamma alumina as support, were analysed [61]. Although carbon deposition rates were extensive under certain catalysts testing conditions, these gamma alumina supported catalysts displayed a high selectivity to higher molecular weight olefinic products. Alumina-supported iron catalysts exhibited, with the performance of the catalysts being highly dependent on whether a metal nitrate or organometallic precursors was used for their preparation [3]. Different organometallic precursors of Fe on alumina have been studied [62] while other researchers [63] synthesized catalysts by using iron-nitrate as precursor. Furthermore, Fe/Al2O3 catalysts
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 25 Chapter 2
promoted with sodium and sulfur displayed stable activity, with limited sintering and a low coking rate [64]. K- and Mn-promoted iron catalysts were studied using silica gel as support and metal loading achieved using the impregnation method
[65, 66]. To compare the effect of the precursor, the catalysts were synthesized with metal nitrates and complex carbonyls. It was found that organometallic complexes would result in a higher level of metal iron dispersion compared to when metal nitrates were used as precursors. An investigation into the activity of ruthenium-iron bimetallic silica-supported catalysts showed that a high olefin selectivity was achieved [66]. However, thermogravimetric analysis disclosed that carbon deposition for this catalyst was more extensive than that observed for a
Fe/SiO2 catalyst. Moreover, other bimetallic silica-supported catalysts containing
Fe and Ni were prepared and studied by other groups [67]. Contrary to a bulk catalyst with the same composition, the supported catalyst containing lower levels of nickel content exhibited higher selectivities toward lower olefins. The applications of zeolites [68-72] and alumina phosphates [73] as support materials for iron-based Fischer-Tropsch synthesis have been explored by a several research groups. Some of the zeolite-supported potassium-promoted catalysts exhibited higher light olefins selectivity compared to that of catalysts with similar metal composition and prepared without zeolite supports [68, 69]. Moreover,
Kang et al. [70] designed bifunctional zeolite-supported catalysts in which syngas was transformed to a primary straight hydrocarbon chain on the FT active sites of catalyst, followed by chain cracking on the acid sites in the zeolite. They prepared their catalysts by impregnating Fe and Cu nitrates along with an aqueous solution of potassium carbonate on three different zeolites (ZSM-5, mordenite and beta-zeolite). Furthermore, the same group [71] studied the
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 26 Chapter 2
influence of Si/Al ratio of ZSM-5 on the activity and selectivity of the catalysts.
Other inorganic oxidic materials such as ZnO [74], ZrO2, TiO2, MgO [72], MnO and MnO2 (Figure 2.10) [31, 75] also have been studied for production of light hydrocarbons from syngas. A Pd-promoted iron catalyst on ZnO displayed moderate light olefin and high methane selectivities [74]. Bimetallic Fe-Mn catalysts, on oxide supports including SiO2, Al2O3, ZrO, TiO2 and MgO, have been investigated [72], where the impregnation method was adopted for catalyst preparation. The main focus of their experiments was to elucidate the influence of the catalyst support properties on catalytic performance. It is concluded that basic supports display improved performance toward light olefins selectivity compared to that of acidic supports.
Figure 2.10. Selectivity toward light hydrocarbons for COx hydrogenation over Mn- oxide-supported Fe catalysts prepared with two different methods [31].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 27 Chapter 2
Other groups have studied and compared the COx methanation activity of alumina-supported catalysts containing noble metals including ruthenium, palladium, platinum and rhodium (Figure 2.11) [76]. It is concluded that Ru and
Rh have high activity toward the COx hydrogenation, while Pd and Pt show lower level of intrinsic activity.
Figure 2.11. CO conversion versus reaction temperature over Rh, Ru, Pt and Pd (5
wt%)/Al2O3 [76].
As it is illustrated from the Figure 2.11, Ru has the highest activity during the carbon monoxide hydrogenation. However, due to the (undesirable) reverse water gas shift reaction, with all metals some carbon monoxide will always remain during the CO and CO2 methanation. Silica- and alumina-supported silver catalysts have been tested [77]. Good performances of the silver catalyst have been reported during the COx hydrogenation reaction. It is concluded that the
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 28 Chapter 2
supported silver catalysts activity is strongly dependent on the catalyst preparation method.
2.3.2.2.2 Nickel
Nickel is known as one of the most active metals for COx hydrogenation, which has been widely used for methanation and FTS processes as the primary metal or promoter [27, 33]. However, due to the formation of volatile Ni-carbonyl species and subsequent loss of metal from the catalyst, nickel-based catalysts are not as common as Co and Fe catalysts or Fischer-Tropsch synthesis [27]. On the other hand, the higher selectivity for formation of light hydrocarbons (especially methane) makes nickel a preferred choice for industrial scale methanation plant
[78]. The performance of nickel catalysts during COx catalytic hydrogenation depends on various factors such as the extent of nickel loading, dispersion, particle size, and metal-support interaction (MSI) and the properties of the support. These factors have been extensively studied [27, 79, 80]. One of the important factors which has been investigated by many groups is the metal- support interaction and its effects on the hydrogenation process [27, 81]. It has been reported that using different supports (e.g. Al2O3, SiO2 and TiO2) significantly changes the selectivity of the hydrogenation reaction and the ratio of chemisorbed CO/H. The existence of strong metal-support interactions (SMSI) to some extent was found for all of the highly dispersed Ni/Al2O3, Ni/SiO2 and
Ni/TiO2. Moreover, it is also observed that preparation method can affect the
SMSI. Thus, it was concluded that the nature of interaction is an important contributing factor [82].
The influence of changing electronic structure of a catalyst on its performance by changing metal-support interactions (by changing or modifying the support) or
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 29 Chapter 2
adding promoters to form alloys have been investigated [83-85]. Nitric oxide has been used as probe molecule to analyse the electronic structure of catalysts.
Electron back donation from the metal to the anti-bonding orbital of nitric oxide molecule results in the bent-type adsorption of nitric oxide (NO-). Electron donation from nitric oxide molecule to the metal forms a linear-type NO adsorption species (NO+) [86]. It is found that the extent of electron back-donation from metal to NO is decreased by enhancing the electronic MSI [84]. Figure 2.12 shows the relationship between the electronegativity of different supports and infrared frequency of bent-type NO absorption. Furthermore, the correlation between electronegativity and catalyst performance parameters (such as turnover frequency, methane selectivity and chain growth probability) suggests that there is an optimal value for electronic metal-support interaction (Figure 2.13) [84].
Figure 2.12. Illustrating the relation between electronegativity of the support oxides and the absorption frequency of bent-type NO adsorption (NO-) [27, 84].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 30 Chapter 2
Figure 2.13. Illustrating the relation between electronegativity of the support oxides and
turnover frequency (TOF), chain growth probability, and CH4 selectivity [27, 84].
Similar to the effects of MSI on electronic structure of Ni catalysts, influences of alloy formation on the electronic structure has been studied. For instance, NO
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 31 Chapter 2
adsorption infrared spectroscopy and X-ray photoelectron spectroscopy (3p3/2) were employed (Ishihara et al.) to study the alloying effect of Ni-Co bimetallic catalysts on the catalytic performance [85]. An electronic interaction of Ni and Co in the outer shell orbital was discovered (Figure 2.14), and as a result of this alloying interaction it is suggested that new active site was formed. Moreover, the nitric oxide infrared spectroscopy results showed the ratio of linear- to bent-type absorbed NO changed as a result of alloy formation (Figure 2.15) [85].
Figure 2.14. Illustrating the shift in XPS 3p3/2 binding energy of Co and Ni and in Co-Ni
alloy catalysts on SiO2 supports: (a) Co; (b) 75CO25Ni; (c) 50Co50Ni; (d) 25Co75Ni; and (e) Ni. The total metal loading was always 10 wt% [27, 85].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 32 Chapter 2
Figure 2.15. Illustrating the infrared absorbance of linear- and bent-type NO and its ratio on the alloy composition: (○) bent-type NO; (∆) linear-type NO; (●) ratio of the absorbance of bent- to linear-type band [27, 85].
Among various supported nickel catalysts investigated for COx hydrogenation process, Ni/Al2O3 is one of the most extensively investigated compositions because of its high performance and low cost [87, 88].
The existence of three distinctive phases on Ni/Al2O3 was reported for Ni loading around 10 wt%. The origin of these phases is a result of the presence of different types of reducible nickel particles on the catalyst (Figure 2.16).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 33 Chapter 2
Figure 2.16. Sketch of the catalyst structure and selective reactions occurring during the synthesis of methane [87].
These particles are reducible at different temperatures and exhibit distinctly different performance for the hydrogenation reaction [89, 90]. It is reported that amorphous NiO particles with high dispersion have higher activity for methanation reaction due to the weak metal-support interaction (resulting in higher reducibility) [91].
In general, the COx hydrogenation reaction is highly sensitive to the catalyst structure [92]. The role of nickel atomic step sites as active sites has been suggested, where the relation between high activity and high dispersion was explained to arise as a result of the existence of additional step sites. These step sites are possessed by Ni lower coordination numbers in highly dispersed nickel catalysts [93, 94].
In addition to activity and selectivity, one of the important factors to study is the stability of nickel catalysts. Deactivation of the nickel supported catalysts by coke/carbon formation is well known [78]. Catalyst deactivation can occur by the
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 34 Chapter 2
following three common routes: blocked active metal surface of the catalyst, blocked pores or voids of the catalyst and physically disintegrated catalyst support [95]. In case of coke formation on nickel catalysts, three types of carbon have been identified: whisker, encapsulating and pyrolytic carbon [96]. Under same reaction conditions, carbon formation is thermodynamically favoured during
CO hydrogenation compared to CO2 hydrogenation [97]. Moreover, the presence of light hydrocarbons in the product stream is possibly a result of the formation of whisker carbon decomposing during COx hydrogenation [87]. The design and development of coke-resistant nickel-based catalysts has been investigated by many researchers [78]. In general, there are two common ways to prevent coke formation: optimising the process conditions (e.g. H2 ratio) and catalyst modification [98-100].
Another common reason for deactivation of nickel-based catalysts is sulphur poisoning. The gas-phase sulfur containing compounds, especially H2S, when present even in small quantities (ppm level) for a sufficient period of time cause significant losses in activity for methanation and hydrocarbon synthesis over transition metal (e.g. Ni, Co and Fe) catalysts [101, 102]. This deactivation effect has been studied comprehensively and it is reported that the poisoning effect is mainly due to impacts of sulphur coverage on reactants adsorption on the catalyst’s active sites. Adsorbed sulfur appears to poison hydrogen adsorption in supported nickel by a blocking mechanism. In this case, at complete coverage a simple blocking or geometrical effect is adequately proposed to explain the poisoning since the ensembles of Ni required for dissociation of H2 are no longer accessible at the surface. Moreover, it is also reported that adsorbed sulfur diminishes the amount of strongly adsorbed CO and the formation of nickel tetra-
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 35 Chapter 2
carbonyl is catalysed by the presence of adsorbed sulfur [103]. Although the negative impact of sulphur was detected and reported in the literature, different catalyst recovery/regeneration procedures were also studied. In this case, catalyst activity can be partially recovered by the following most notably procedures: (i) increasing reaction temperature; (ii) oxidizing and activating the deactivated sample [102].
2.3.3 Thermodynamics
Thermodynamic analysis of COx catalytic hydrogenation process has been undertaken [27, 38]. Possible reactions occurring during COx methanation process is illustrated in Table 2.1. This reaction list contains main methanation reactions (R1 and R2 for CO and CO2 respectively) and some of the well-known side reactions.
Among the noted side reactions, CO disproportionation (Boudouard) reaction
(R4) is possibly the most important reaction [104]. Another important side reaction is water-gas shift (WGS) reaction (R5), which is relevant due to the generation of water during COx hydrogenation [105]. It is reported that water vapour can cause changes on the surface of the catalyst and has a direct effect on the catalytic performance [106]. It is believed that R1, R2 and R4 can be considered as independent reactions and the other reactions can be invoked as a linear combination of R1, R2 and R4 [38].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 36 Chapter 2
Table 2.1. Main possible reactions involved in methanation of carbon oxides [38].
Figure 2.17 displays the estimated equilibrium constants for reactions of Table
2.1 [97]. Based on the equilibrium constants, because of their exothermic nature, all reactions are favoured at lower temperature ranges. It is evident that relatively mild reaction conditions (low temperature and low pressure) are thermodynamically suitable for COx hydrogenation. However, finding a catalyst that can reach equilibrium state at low temperature is a significant challenge for this reaction [38].
A comprehensive thermodynamic analysis of COx methanation process has been recently published [97]. The article includes a detailed thermodynamic investigation on the effect of different process conditions such as feed composition (H2/COx ratio), temperature, pressure and feed stream additives on the equilibrium state of COx hydrogenation. Parameters such as reactants conversion, carbon deposition, products selectivity and yield were calculated and compared [97]. Figure 2.18 shows the product composition at equilibrium state for CO and CO2 hydrogenation.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 37 Chapter 2
Figure 2.17. The calculated equilibrium constants (K) of the eight reactions involved in methanation process [38, 97].
It is found that the reaction thermodynamically favours the formation methane from COx hydrogenation. Carbon dioxide hydrogenation is more selective toward methane production compared to CO hydrogenation [107]. It is also observed that formation of solid carbon or carbon deposition is not thermodynamically favourable at low temperatures (below 450 °C) [97].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 38 Chapter 2
Figure 2.18. Product compositions for CO (a) and CO2 (b) methanation at equilibrium (0.1 MPa) [38, 97].
Figure 2.19. Effects of pressure and temperature on co-methanation of carbon oxides:
(a) conversion and (b) CH4 yield [97].
These researchers also investigated the simultaneous methanation of CO and
CO2 (Figure 2.19). It is found that CO conversion is much higher than CO2 conversion especially at temperatures below 600 °C. By increasing temperature,
CO2 conversion increased while CO conversion decreased sharply. To explain
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 39 Chapter 2
this, the dominance of water-gas shift reaction at higher temperatures is invoked
[97].
Although product distribution models for Fischer-Tropsch synthesis (e.g.
Anderson-Shulz-Flory) are usually derived from kinetic analysis, some thermodynamic-based modifications to the models have been reported [27, 108-
110]. The underlying explanation for applying these modifications arises as a result of the significant difference between experimental results and theoretical models such as Anderson-Shulz-Flory especially for C1-C3 range (Figure 2.20)
[27].
Figure 2.20. Comparing the ASF distribution at α = 0.7 (○) to experimental data from
FTS on Co-Re/γ-Al2O3 at 20% CO conversion (♦) and 50% CO conversion (□). n is
carbon number and Wn is the weight fraction [27]. One model for Fischer-Tropsch synthesis product distribution was developed by
Bell [110]. In this model, thermodynamic analysis of oxidation and reduction of the active metal present on the catalyst were accounted. In addition, the effect of hydride and carbide formation were also included through the Gibbs-free energy
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 40 Chapter 2
modifications. Based on this investigation, it was concluded that iron and cobalt catalysts have oxides with Gibbs-free energy of formation that enables equilibrium CO2/CO and H2O/H2 ratios in the range in which chain growth is thermodynamically stable. A similar conclusion was not drawn for other metals
(e.g. nickel, platinum and palladium). It was found that formation of methane as major product is thermodynamically favourable for metals such as nickel [110].
However, later re-calculations based on Bell’s model found some modifications were necessary because methane was the only product formed over both bulk nickel and cobalt [27]. As noted by Bell the CO2/CO ratio, which is thermodynamically stable, should be modified even by cobalt to form higher hydrocarbons [110]. Therefore, the ratio of CO2/CO was modified by adjusting
Co/CoO ratio (by adjusting Gibbs-free energy of formation for CoO through: Co +
CO2 → CoO + CO) in the re-calculations [27]. Figure 2.21 illustrates hydrocarbon selectivity during FTS over cobalt-based catalyst. It was found that the deviation for predicting light hydrocarbons between ASF and experimental data can be explained thermodynamically [27]. Further modifications to improve the model by adjusting hydride and carbide formation has been also formulated since the publication of the original study by Bell [110]. In conclusion, it is suggested in literature that thermodynamic-based models are accurate for predicting product distribution when compared to ASF model [27]. This means that in addition to the chain-propagation as a polymerisation reaction and the catalyst effects on kinetics, the carbon number distribution is also largely determined by the influence of local thermodynamic properties and the attainment of a system to partial equilibrium [27].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 41 Chapter 2
Figure 2.21. Carbon selectivity at 20 bar and H2/CO = 2.1. Experimental at 483 K for a
Co-Re/γ-Al2O3: C4+ (□), CH4 (●), C3 (▲), and C2 (◊) [27, 110].
In case of nickel catalysts, which is thermodynamically unfavourable to form higher hydrocarbons, it is feasible to improve selectivity toward higher hydrocarbons production by modifying CO2/CO ratio through adjustment of
Gibbs-free energy of oxidation and reduction [27]. To adjust Gibbs-free energy of oxidation and reduction, the bulk nickel catalyst can be modified by using different suitable supports or promoters [27].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 42 Chapter 2
2.3.4 Mechanistic understanding
In this section, some of the key mechanisms developed to describe CO and CO2 hydrogenation are presented. Reaction mechanism for both methanation and
Fischer-Tropsch synthesis are reviewed and discussed.
It is well known that FTS is a catalytic polymerisation reaction that occurs on the surface of the catalyst. Therefore, the reaction mechanism is defined in terms of three main steps: chain initiation, chain propagation (or chain growth) and chain termination (or desorption) [46, 111]. There are many different contributions in the literature to explain the mechanism and reaction kinetics of these three steps.
In the following, a brief explanation of the various mechanisms is presented. In general, the chain propagation step consists of the reaction between an adsorbed intermediate/hydrocarbon and a C1 intermediate (surface moiety which contains just one carbon atom) [111]. Based on the characteristics of C1 intermediates and chain initiation step (monomer formation), the primary pathways are divided into the two pathways: via oxygenated species (Figure 2.22) and via deoxygenated species (Figure 2.23) [46]. In the pathways via oxygenated surface species, the chain propagation mechanism consists of two separate steps. Carbon monoxide is initially dissociated (either H-assisted or directly) to form a monomer (surface partially hydrogenated CO or surface oxygen containing species). Following CO dissociation, another CO molecule reacts with the surface species to generate longer carbon chains [111]. There are pathways proposed via deoxygenated species which are based on the ‘carbide theory’ (Figure 2.24). Carbide theory basically presumes that surface carbide is formed via CO activation. The surface carbide then is hydrogenated to surface CHx species followed by formation of different hydrocarbons from these active surface species. The models based on
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 43 Chapter 2
carbide theory are known as ‘carbene mechanism’ in recent years [46]. The final step of all these mechanisms is the termination step. For termination step, there are generally three main reactions as H-addition, β-CO cleavage and CO insertion to form paraffin, olefins and alcohols respectively [111].
Figure 2.22. A general schematic presentation of CO hydrogenation chain growth mechanism via oxygenated surface species [111].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 44 Chapter 2
Figure 2.23. A general schematic presentation of CO hydrogenation chain growth mechanism via deoxygenated surface species (carbene mechanism) [111].
Figure 2.24. Schematic presentation of activation of carbon monoxide on a catalyst surface giving surface carbide (carbide theory) [46].
The first mechanism involving an oxygenated species for chain growth step was proposed by Pichler and Schulz [112]. Different oxygenated species such as CO,
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 45 Chapter 2
HCO and HCOH has been proposed as the C1 block, which is inserted into a surface hydrocarbon [92, 113, 114]. Two of the well-known CO hydrogenation mechanisms via oxygenated species are CO insertion and enol mechanisms
(Figure 2.25 and Figure 2.26).
Figure 2.25. Schematic view of CO insertion mechanism [113].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 46 Chapter 2
Figure 2.26. Schematic view of enol mechanism [113].
Similarly, different CHx intermediates (such as CH [115] and CH2 [116, 117]) have been proposed for the chain growth mechanisms involving deoxygenated surface species. It is believed that the nature of CHx intermediates and the hydrocarbon chains during polymerization step are strongly related [111]. There are three main proposed mechanisms for CO hydrogenation via deoxygenated species. These mechanisms are characterised by the nature of the growing hydrocarbon chain:
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 47 Chapter 2
alkyl (Figure 2.27) [116, 117], alkenyl (Figure 2.28) [118] and alkylidene (Figure
2.29) [119-121].
Figure 2.27. Proposed chain growth reaction and its schematic view involving alkyl as growing chain [46, 111].
Figure 2.28. Proposed chain growth reaction and its schematic view involving alkenyl as growing chain [46, 111].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 48 Chapter 2
Figure 2.29. Proposed chain growth reaction and its schematic view involving alkylidene as growing chain [46, 111].
According to carbene mechanisms, methane formation and chain growth reactions are the primary competing processes during CO hydrogenation. To enhance selectivity toward production of hydrocarbons with longer chains, it is necessary that rate of CHx formation from CO (chain initiation or monomer formation) and rate of C1 block insertion into the growing chain (chain propagation) are higher than rate of CHx hydrogenation (methane formation) and hydrocarbon termination (desorption step). It is reported that CO activation (or C
– O bond cleavage) is probably the rate limiting step [111]. In contrast, in mechanisms involving oxygenated species via CO insertion, the rate of C – O
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 49 Chapter 2
bond cleavage has to be fast after insertion of CO into the hydrocarbon chain
[111].
In addition to understanding the mechanism of formation of C2+ hydrocarbons from carbon monoxide hydrogenation, it is also necessary to review the mechanisms developed to explain methanation reaction. Many researchers have developed mechanisms for methanation of carbon oxides over nickel catalysts
[122-129]. One of these proposed pathways (Figure 2.30) [123] suggests that CO and H2 compete to adsorb on the same active sites. This suggestion was made based on the mutual interaction of CO and H2 reaction orders that was found in the experimental analysis.
Figure 2.30. Mechanism of CO methanation proposed by van Meerten et al. Symbol * represents type 1 (ordinary) reaction site [24, 123].
In an alternate mechanism (Figure 2.31) [126], it is suggested that CO and H2 compete for site for dissociative adsorption. They also found that the rate determining step is the reaction of surface C and H. Moreover, the formation of
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 50 Chapter 2
different forms of surface carbon were found as a function of feed composition, exposure time and temperature. It is also reported that the variety of surface carbons species possess different hydrogenation activities. Another research group employed in-situ Auger electron spectroscopy and thermal desorption coupled with H/D isotope labelling techniques to investigate mechanism of CO hydrogenation [128]. They found that methane is the only product is gas phase while stable adsorbed CDiH4-i (1 ≤ i ≤ 3) species were detected. They concluded that the rate determining step is formation of adsorbed CH species via hydrogenation of adsorbed carbon atom.
Figure 2.31. Mechanism of CO methanation proposed by Hayes et al. Symbol * represents type 1 (ordinary) reaction site [24, 126].
The mechanism of carbon monoxide hydrogenation under reaction conditions at high H2/CO has been also studied (Figure 2.32) [129]. The presence of two different types of active sites (type 1 and 2) were proposed. Type 1 sites are ordinary reaction sites for adsorption of all species except surface carbon (e.g.
H). Type 2 sites are 5-fold coordinated sites adsorbing surface carbon. It is
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 51 Chapter 2
suggested that CO almost completely covers the surface at high CO pressures, while at high H2/CO ratios both CO and H2 compete for adsorption sites. Based on this mechanism, type 1 sites are responsible for non-dissociative adsorption of CO and H2 while dissociative CO adsorption (rate determining step) takes place on type 2 sites. These findings were also confirmed by experimental studies performed by other groups [124].
Figure 2.32. Mechanism of CO methanation proposed by Sehested et al. Symbols * and # represent type 1 (ordinary) and type 2 (5-fold coordinated) reaction sites respectively [24, 129].
The reaction mechanism of CO2 methanation has been also extensively investigated [130-137]. The nature of intermediate species has been debated, and the type of intermediate compounds is still not completely resolved. There are two primary pathways for CO2 methanation (Figure 2.33): via carbon monoxide intermediates, or via intermediates other than CO. The mechanisms involve the initial formation of adsorbed CO, followed by subsequent methanation of CO intermediates are more accepted [24, 33, 78]. However, some researchers found that CO2 can be hydrogenated via oxygen containing species other than
CO (e.g. carbonate, formate and methoxy) [135-137].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 52 Chapter 2
Figure 2.33. Simplified different reaction mechanisms for CO2 methanation [33].
2.3.5 Summary
Fisher-Tropsch synthesis (FTS) and methanation are key technologies to synthesis hydrocarbons from carbonaceous resources by hydrogenation of carbon oxides. There are variety of reasons (such as economics and environmental concerns) for the historical importance of these catalytic processes. In its 100 years history, carbon oxides catalytic hydrogenation (i.e. methanation/FTS) has been an active area of research. Significant scientific resources and research activities have been undertaken to develop the process and elucidate the fundamental steps involved in the reaction. Studies focusing on catalyst development, process conditions, mechanistic insight, thermodynamics and many more different aspects has been published by different groups. In general, all group 8 – 10 transition metals were found active for this process.
Developing supported and/or unsupported catalysts using these metals as the active component were studied by different groups. Nickel is one of the intensively studied active component for carbon oxide hydrogenation due to its high activity and low price. Although nickel is not the preferred choice for
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 53 Chapter 2
producing long chain hydrocarbons, different studies focused on the modification of nickel-based catalysts has led to improvements in its activity and selectivity.
Moreover, one of the main issues with commercializing nickel-based catalysts is deactivation due to carbon deposition and formation of volatile metal carbonyls.
To gain insight into the reaction, different researchers have studied various mechanistic aspects of the catalytic hydrogenation of carbon oxides. The mechanism of the reaction is still not completely clear and is the subject of vigorous scientific debate. However, with respect to the nature of surface intermediates, it is found that some mechanisms are more widely accepted in the literature. It seems that formation of surface carbon followed by subsequent hydrogenation to methane is the most important intermediate reaction in the methanation of CO and CO2. In contrast, some reports suggest that oxygenated species are probably the main pathways to form hydrocarbons with longer carbon chain lengths.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 54 Chapter 2
2.4 References
[1] E. Saldivar-Guerra, E. Vivaldo-Lima, Handbook of Polymer Synthesis,
Characterization, and Processing, Wiley2013.
[2] J. Gallagher, M. McLinden, G. Morrison, N.I.o. Standards, Technology, NIST
Thermodynamic Properties of Refrigerants and Refrigerant Mixtures Database:
Version 2.0 : User's Guide, U.S. Department of Commerce, National Institute of
Standards and Technology, Standard Reference Data Program1991.
[3] H.M. Torres Galvis, K.P. De Jong, Catalysts for production of lower olefins from synthesis gas: A review, ACS Catalysis, 3 (2013) 2130-2149.
[4] H. Jahangiri, J. Bennett, P. Mahjoubi, K. Wilson, S. Gu, A review of advanced catalyst development for Fischer-Tropsch synthesis of hydrocarbons from biomass derived syn-gas, Catalysis Science & Technology, 4 (2014) 2210-2229.
[5] P. Tian, Y. Wei, M. Ye, Z. Liu, Methanol to Olefins (MTO): From Fundamentals to Commercialization, ACS Catalysis, 5 (2015) 1922-1938.
[6] R.A. Meyers, Handbook of Petroleum Refining Processes, McGraw-Hill
Education2003.
[7] R. Razzaq, C. Li, S. Zhang, Coke oven gas: Availability, properties, purification, and utilization in China, Fuel, 113 (2013) 287-299.
[8] P.B. Weisz, Basic Choices and Constraints on Long-Term Energy Supplies,
Physics Today, 57 (2004) 47-52.
[9] C. Song, Global challenges and strategies for control, conversion and utilization of CO2 for sustainable development involving energy, catalysis, adsorption and chemical processing, Catalysis Today, 115 (2006) 2-32.
[10] J.H. Lunsford, Catalytic conversion of methane to more useful chemicals and fuels: A challenge for the 21st century, Catal. Today, 63 (2000) 165-174.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 55 Chapter 2
[11] V. Shadravan, E. Kennedy, M. Stockenhuber, An experimental investigation on the effects of adding a transition metal to Ni/Al2O3 for catalytic hydrogenation of CO and CO2 in presence of light alkanes and alkenes, Catalysis Today,
(2017).
[12] B.L. Farrell, V.O. Igenegbai, S. Linic, A Viewpoint on Direct Methane
Conversion to Ethane and Ethylene Using Oxidative Coupling on Solid Catalysts,
ACS Catalysis, 6 (2016) 4340-4346.
[13] J.C.W. Kuo, C.T. Kresge, R.E. Palermo, Evaluation of direct methane conversion to higher hydrocarbons and oxygenates, Catalysis Today, 4 (1989)
463-470.
[14] D.Y.C. Leung, G. Caramanna, M.M. Maroto-Valer, An overview of current status of carbon dioxide capture and storage technologies, Renewable and
Sustainable Energy Reviews, 39 (2014) 426-443.
[15] https://esseacourses.strategies.org/module.php?module_id=170.
[16] M.A. Vannice, The Catalytic Synthesis of Hydrocarbons from Carbon
Monoxide and Hydrogen, Catalysis Reviews, 14 (1976) 153-191.
[17] K. Fang, D. Li, M. Lin, M. Xiang, W. Wei, Y. Sun, A short review of heterogeneous catalytic process for mixed alcohols synthesis via syngas,
Catalysis Today, 147 (2009) 133-138.
[18] V. Subramani, S.K. Gangwal, A Review of Recent Literature to Search for an
Efficient Catalytic Process for the Conversion of Syngas to Ethanol, Energy &
Fuels, 22 (2008) 814-839.
[19] J. Kopyscinski, T.J. Schildhauer, S.M.A. Biollaz, Production of synthetic natural gas (SNG) from coal and dry biomass - A technology review from 1950 to
2009, Fuel, 89 (2010) 1763-1783.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 56 Chapter 2
[20] B. Jager, R. Espinoza, Advances in low temperature Fischer-Tropsch synthesis, Catalysis Today, 23 (1995) 17-28.
[21] A.Y. Khodakov, W. Chu, P. Fongarland, Advances in the development of novel cobalt Fischer-Tropsch catalysts for synthesis of long-chain hydrocarbons and clean fuels, Chemical Reviews, 107 (2007) 1692-1744.
[22] NASA DATA Live Access Server - Advanced: Monthly Daylight Column
Carbon Monoxide (MOPITT) (10^13 mol/cm^2), (2012).
[23] NASA DATA Live Access Server - Advanced: Monthly Carbon Dioxide in
Troposphere (AIRS on AQUA) (ppmv), (2012).
[24] S. Rönsch, J. Schneider, S. Matthischke, M. Schlüter, M. Götz, J. Lefebvre,
P. Prabhakaran, S. Bajohr, Review on methanation – From fundamentals to current projects, Fuel, 166 (2016) 276-296.
[25] M.E. Dry, High quality diesel via the Fischer–Tropsch process – a review,
Journal of Chemical Technology & Biotechnology, 77 (2002) 43-50.
[26] G.A. Mills, F.W. Steffgen, Catalytic methanation, Catal. Rev., 8 (1973) 159-
210.
[27] B.C. Enger, A. Holmen, Nickel and Fischer-Tropsch Synthesis, Catal. Rev.-
Sci. Eng., 54 (2012) 437-488.
[28] M. Krämer, K. Stöwe, M. Duisberg, F. Müller, M. Reiser, S. Sticher, W.F.
Maier, The impact of dopants on the activity and selectivity of a Ni-based methanation catalyst, Applied Catalysis A: General, 369 (2009) 42-52.
[29] R.A. Dagle, A. Karim, G. Li, Y. Su, D.L. King, Syngas Conditioning, Fuel
Cells: Technologies for Fuel Processing2011, pp. 361-408.
[30] M.E. Dry, J.C. Hoogendoorn, Technology of the Fischer-Tropsch Process,
Catalysis Reviews, 23 (1981) 265-278.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 57 Chapter 2
[31] B. Hu, S. Frueh, H.F. Garces, L. Zhang, M. Aindow, C. Brooks, E. Kreidler,
S.L. Suib, Selective hydrogenation of CO2 and CO to useful light olefins over octahedral molecular sieve manganese oxide supported iron catalysts, Applied
Catalysis B: Environmental, 132-133 (2013) 54-61.
[32] F. Fischer, H. Tropsch, P. Dilthey, Reduction of carbon monoxide to methane in the presence of various metals, Brennst. Chem, 6 (1925) 265-271.
[33] P. Frontera, A. Macario, M. Ferraro, P. Antonucci, Supported Catalysts for
CO2 Methanation: A Review, Catalysts, 7 (2017) 59.
[34] P. Munnik, P.E. de Jongh, K.P. de Jong, Recent Developments in the
Synthesis of Supported Catalysts, Chemical Reviews, 115 (2015) 6687-6718.
[35] D.W. Lee, B.R. Yoo, Advanced metal oxide (supported) catalysts: Synthesis and applications, Journal of Industrial and Engineering Chemistry, 20 (2014)
3947-3959.
[36] S.L. Suib, K.C. McMahon, L.M. Tau, C.O. Bennett, Synthesis, characterization, and Fischer-Tropsch studies of iron-containing zeolites, Journal of Catalysis, 89 (1984) 20-34.
[37] L.V. Sineva, E.Y. Asalieva, V.Z. Mordkovich, The role of zeolite in the
Fischer–Tropsch synthesis over cobalt–zeolite catalysts, Russian Chemical
Reviews, 84 (2015) 1176.
[38] J. Gao, Q. Liu, F. Gu, B. Liu, Z. Zhong, F. Su, Recent advances in methanation catalysts for the production of synthetic natural gas, RSC Advances,
5 (2015) 22759-22776.
[39] I. Levin, D. Brandon, Metastable alumina polymorphs: Crystal structures and transition sequences, Journal of the American Ceramic Society, 81 (1998) 1995-
2012.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 58 Chapter 2
[40] G. Garbarino, P. Riani, L. Magistri, G. Busca, A study of the methanation of carbon dioxide on Ni/Al2O 3 catalysts at atmospheric pressure, Int. J. Hydrogen
Energ., 39 (2014) 11557-11565.
[41] J. Gao, C. Jia, J. Li, M. Zhang, F. Gu, G. Xu, Z. Zhong, F. Su, Ni/Al2O3 catalysts for CO methanation: Effect of Al2O3 supports calcined at different temperatures, J. Energ. Chem., 22 (2013) 919-927.
[42] D. Hu, J. Gao, Y. Ping, L. Jia, P. Gunawan, Z. Zhong, G. Xu, F. Gu, F. Su,
Enhanced investigation of CO methanation over Ni/Al 2O 3 catalysts for synthetic natural gas production, Industrial and Engineering Chemistry Research, 51
(2012) 4875-4886.
[43] P. Sabatier, J.B. Senderens, New methane synthesis, C. R. Acad. Sci. Paris,
134 (1902) 514-516.
[44] A. Steynberg, M. Dry, Fischer-Tropsch Technology, Elsevier Science2004.
[45] T. Bligaard, J.K. Nørskov, S. Dahl, J. Matthiesen, C.H. Christensen, J.
Sehested, The Brønsted–Evans–Polanyi relation and the volcano curve in heterogeneous catalysis, Journal of Catalysis, 224 (2004) 206-217.
[46] O.O. James, B. Chowdhury, M.A. Mesubi, S. Maity, Reflections on the chemistry of the Fischer-Tropsch synthesis, RSC Advances, 2 (2012) 7347-7366.
[47] J.K. Nørskov, T. Bligaard, A. Logadottir, S. Bahn, L.B. Hansen, M. Bollinger,
H. Bengaard, B. Hammer, Z. Sljivancanin, M. Mavrikakis, Y. Xu, S. Dahl, C.J.H.
Jacobsen, Universality in Heterogeneous Catalysis, Journal of Catalysis, 209
(2002) 275-278.
[48] P. Panagiotopoulou, D.I. Kondarides, X.E. Verykios, Selective methanation of CO over supported Ru catalysts, Applied Catalysis B: Environmental, 88
(2009) 470-478.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 59 Chapter 2
[49] B.S. Baker, J. Huebler, H.R. Linden, J. Meek, Process for selective removal by methanation of carbon monoxide from a mixture of gases containing carbon dioxide, in: U.S. Patent (Ed.), Company; Consolidated Natural Gas (PA)
Company; Southern California Gas (CA)
Company; Southern Counties Gas (CA), United States, 1971.
[50] R.A. Dagle, Y. Wang, G.G. Xia, J.J. Strohm, J. Holladay, D.R. Palo, Selective
CO methanation catalysts for fuel processing applications, Applied Catalysis A:
General, 326 (2007) 213-218.
[51] A. Rehmat, S.S. Randhava, Selective methanation of carbon monoxide,
Industrial and Engineering Chemistry Product Research and Development, 9
(1970) 512-515.
[52] Z. Li, W. Mi, J. Gong, Z. Lu, L. Xu, Q. Su, CO removal by two-stage methanation for polymer electrolyte fuel cell, Journal of Natural Gas Chemistry,
17 (2008) 359-364.
[53] C. Galletti, S. Specchia, G. Saracco, V. Specchia, CO-selective methanation over Ru-γAl2O3 catalysts in H2-rich gas for PEM FC applications, Chemical
Engineering Science, 65 (2010) 590-596.
[54] T. Abe, M. Tanizawa, K. Watanabe, A. Taguchi, CO2 methanation property of Ru nanoparticle-loaded TiO2 prepared by a polygonal barrel-sputtering method, Energy & Environmental Science, 2 (2009) 315-321.
[55] S. Takenaka, T. Shimizu, K. Otsuka, Complete removal of carbon monoxide in hydrogen-rich gas stream through methanation over supported metal catalysts,
International Journal of Hydrogen Energy, 29 (2004) 1065-1073.
[56] S. Eckle, Y. Denkwitz, R.J. Behm, Activity, selectivity, and adsorbed reaction intermediates/reaction side products in the selective methanation of CO in
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 60 Chapter 2
reformate gases on supported Ru catalysts, Journal of Catalysis, 269 (2010) 255-
268.
[57] E. Kok, J. Scott, N. Cant, D. Trimm, The impact of ruthenium, lanthanum and activation conditions on the methanation activity of alumina-supported cobalt catalysts, Catalysis Today, 164 (2011) 297-301.
[58] G. Prieto, A. Martínez, P. Concepción, R. Moreno-Tost, Cobalt particle size effects in Fischer–Tropsch synthesis: structural and in situ spectroscopic characterisation on reverse micelle-synthesised Co/ITQ-2 model catalysts,
Journal of Catalysis, 266 (2009) 129-144.
[59] S. Rojanapipatkul, B. Jongsomjit, Synthesis of cobalt on cobalt-aluminate via solvothermal method and its catalytic properties for carbon monoxide hydrogenation, Catalysis Communications, 10 (2008) 232-236.
[60] H. Zhu, R. Razzaq, L. Jiang, C. Li, Low-temperature methanation of CO in coke oven gas using single nanosized Co3O4 catalysts, Catalysis
Communications, 23 (2012) 43-47.
[61] B.G. Baker, N.J. Clark, Selective Fischer-Tropsch Catalysts Containing Iron and Lanthanide Oxides, in: P.G.P.A.J. B. Delmon, G. Poncelet (Eds.) Studies in
Surface Science and Catalysis, Elsevier1987, pp. 455-465.
[62] J.M. Basset, J. Bousquet, R. Mutin, Process for preparing catalysts for use in olefins conversion reactions, Google Patents, 1978.
[63] J. Barrault, C. Forquy, J.C. Menezo, R. Maurel, Selective hydrocondensation of CO to light olefins with alumina-supported iron catalysts, Reaction Kinetics and
Catalysis Letters, 15 (1980) 153-158.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 61 Chapter 2
[64] H.M. Torres Galvis, J.H. Bitter, C.B. Khare, M. Ruitenbeek, A.I. Dugulan, K.P.
De Jong, Supported Iron Nanoparticles as Catalysts for Sustainable Production of Lower Olefins, Science, 335 (2012) 835-838.
[65] L. Bruce, G. Hope, T.W. Turney, Light olefin production from CO/H2 over silica supported Fe/Mn/K catalysts derived from a bimetallic carbonyl anion,
[Fe2Mn(CO)12]−, Reaction Kinetics and Catalysis Letters, 20 (1982) 175-180.
[66] F. Stoop, K. van der Wiele, Formation of olefins from synthesis gas over silica-supported RuFe bimetallic catalysts, Applied Catalysis, 23 (1986) 35-47.
[67] M.E. Cooper, J. Frost, New iron/nickel alloy catalyst for Fischer-Tropsch synthesis, Applied Catalysis, 57 (1990) L5-L8.
[68] V.U.S. Rao, R.J. Gormley, Catalyst for converting synthesis gas to light olefins, Google Patents, 1982.
[69] J.W. Bae, S.H. Kang, Y.J. Lee, K.W. Jun, Method of direct synthesis of light hydrocarbons from natural gas, Google Patents, 2009.
[70] S.H. Kang, J.W. Bae, P.S.S. Prasad, K.W. Jun, Fischer-Tropsch synthesis using zeolite-supported iron catalysts for the production of light hydrocarbons,
Catalysis Letters, 125 (2008) 264-270.
[71] S.-H. Kang, J.W. Bae, K.-J. Woo, P.S. Sai Prasad, K.-W. Jun, ZSM-5 supported iron catalysts for Fischer–Tropsch production of light olefin, Fuel
Processing Technology, 91 (2010) 399-403.
[72] L. Xu, Q. Wang, Y. Xu, J. Huang, A new supported Fe-MnO catalyst for the production of light olefins from syngas. I. Effect of support on the catalytic performance, Catalysis Letters, 24 (1994) 177-185.
[73] M.L. Cubeiro, C.M. López, A. Colmenares, L. Teixeira, M.R. Goldwasser,
M.J. Pérez-Zurita, F. Machado, F. González-Jiménez, Use of aluminophosphate
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 62 Chapter 2
molecular sieves in CO hydrogenation, Applied Catalysis A: General, 167 (1998)
183-193.
[74] B.L. Gustafson, Process for the selective production of olefins from synthesis gas, Google Patents, 1985.
[75] J. Barrault, C. Renard, Selective hydrocondensation of carbon monoxide into light olefins with iron-manganese catalysts, Applied Catalysis, 14 (1985) 133-143.
[76] P. Panagiotopoulou, D.I. Kondarides, X.E. Verykios, Selective methanation of CO over supported noble metal catalysts: Effects of the nature of the metallic phase on catalytic performance, Applied Catalysis A: General, 344 (2008) 45-54.
[77] A.R. Vilchis-Nestor, M. Avalos-Borja, S.A. Gómez, J.A. Hernández, A.
Olivas, T.A. Zepeda, Alternative bio-reduction synthesis method for the preparation of Au(AgAu)/SiO2-Al2O3 catalysts: Oxidation and hydrogenation of
CO, Applied Catalysis B: Environmental, 90 (2009) 64-73.
[78] W. Wang, S. Wang, X. Ma, J. Gong, Recent advances in catalytic hydrogenation of carbon dioxide, Chem. Soc. Rev., 40 (2011) 3703-3727.
[79] J. Gao, C. Jia, M. Zhang, F. Gu, G. Xu, F. Su, Effect of nickel nanoparticle size in Ni/alpha-Al2O3 on CO methanation reaction for the production of synthetic natural gas, Catalysis Science & Technology, 3 (2013) 2009-2015.
[80] B. Sen, M.A. Vannice, The influence of platinum crystallite size on H2 and
CO heats of adsorption and CO hydrogenation, Journal of Catalysis, 130 (1991)
9-20.
[81] S. Hwang, J. Lee, U.G. Hong, J.G. Seo, J.C. Jung, D.J. Koh, H. Lim, C. Byun,
I.K. Song, Methane production from carbon monoxide and hydrogen over nickel– alumina xerogel catalyst: Effect of nickel content, Journal of Industrial and
Engineering Chemistry, 17 (2011) 154-157.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 63 Chapter 2
[82] C.H. Bartholomew, R.B. Pannell, J.L. Butler, Support and crystallite size effects in CO hydrogenation on nickel, Journal of Catalysis, 65 (1980) 335-347.
[83] T. Ishihara, K. Eguchi, H. Arai, Hydrogenation of carbon monoxide over SiO2- supported Fe-Co, Co-Ni and Ni-Fe bimetallic catalysts, App. Catal., 30 (1987)
225-238.
[84] T. Ishihara, N. Horiuchi, K. Eguchi, H. Arai, The effect of supports on the activity and selectivity of Co Ni alloy catalysts for CO hydrogenation, Journal of
Catalysis, 130 (1991) 202-211.
[85] T. Ishihara, N. Horiuchi, T. Inoue, K. Eguchi, Y. Takita, H. Arai, Effect of alloying on CO hydrogenation activity over SiO2-supported CO-Ni alloy catalysts,
Journal of Catalysis, 136 (1992) 232-241.
[86] K.I. Hadjiivanov, Identification of Neutral and Charged NxOy Surface Species by IR Spectroscopy, Catalysis Reviews, 42 (2000) 71-144.
[87] I. Czekaj, F. Loviat, F. Raimondi, J. Wambach, S. Biollaz, A. Wokaun,
Characterization of surface processes at the Ni-based catalyst during the methanation of biomass-derived synthesis gas: X-ray photoelectron spectroscopy (XPS), Applied Catalysis A: General, 329 (2007) 68-78.
[88] G. Garbarino, I. Valsamakis, P. Riani, G. Busca, On the consistency of results arising from different techniques concerning the nature of supported metal oxide (nano)particles. The case of NiO/Al2O3, Catalysis Communications, 51
(2014) 37-41.
[89] C.-W. Hu, J. Yao, H.-Q. Yang, Y. Chen, A.-M. Tian, On the Inhomogeneity of Low Nickel Loading Methanation Catalyst, Journal of Catalysis, 166 (1997) 1-
7.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 64 Chapter 2
[90] J.M. Rynkowski, T. Paryjczak, M. Lenik, On the nature of oxidic nickel phases in NiO/γ-Al2O3 catalysts, Applied Catalysis A, General, 106 (1993) 73-82.
[91] Z. Qin, J. Ren, M. Miao, Z. Li, J. Lin, K. Xie, The catalytic methanation of coke oven gas over Ni-Ce/Al2O3 catalysts prepared by microwave heating: Effect of amorphous NiO formation, Applied Catalysis B: Environmental, 164 (2015) 18-
30.
[92] M.P. Andersson, F. Abild-Pedersen, I.N. Remediakis, T. Bligaard, G. Jones,
J. Engbæk, O. Lytken, S. Horch, J.H. Nielsen, J. Sehested, J.R. Rostrup-Nielsen,
J.K. Nørskov, I. Chorkendorff, Structure sensitivity of the methanation reaction:
H2-induced CO dissociation on nickel surfaces, Journal of Catalysis, 255 (2008)
6-19.
[93] T. Shido, M. Lok, R. Prins, Characterization of highly dispersed Ni/Al2O3 catalysts by EXAFS analysis of higher shells, Topics in Catalysis, 8 (1999) 223-
236.
[94] J.R. Rostrup-Nielsen, K. Pedersen, J. Sehested, High temperature methanation: Sintering and structure sensitivity, Applied Catalysis A: General,
330 (2007) 134-138.
[95] C.H. Bartholomew, Carbon Deposition in Steam Reforming and Methanation,
Catalysis Reviews, 24 (1982) 67-112.
[96] J. Sehested, Four challenges for nickel steam-reforming catalysts, Catalysis
Today, 111 (2006) 103-110.
[97] J. Gao, Y. Wang, Y. Ping, D. Hu, G. Xu, F. Gu, F. Su, A thermodynamic analysis of methanation reactions of carbon oxides for the production of synthetic natural gas, RSC Advances, 2 (2012) 2358-2368.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 65 Chapter 2
[98] T. Inui, T. Hagiwara, Y. Takegami, Prevention of catalyst deactivation caused by coke formation in the methanation of carbon oxides, Fuel, 61 (1982) 537-541.
[99] J. Lu, B. Fu, M.C. Kung, G. Xiao, J.W. Elam, H.H. Kung, P.C. Stair, Coking- and Sintering-Resistant Palladium Catalysts Achieved Through Atomic Layer
Deposition, Science, 335 (2012) 1205.
[100] E. Nikolla, A. Holewinski, J. Schwank, S. Linic, Controlling carbon surface chemistry by alloying: Carbon tolerant reforming catalyst, Journal of the American
Chemical Society, 128 (2006) 11354-11355.
[101] J. Hepola, P. Simell, Sulphur poisoning of nickel-based hot gas cleaning catalysts in synthetic gasification gas: I. Effect of different process parameters,
Applied Catalysis B: Environmental, 14 (1997) 287-303.
[102] S. Albertazzi, F. Basile, J. Brandin, J. Einvall, G. Fornasari, C. Hulteberg,
M. Sanati, F. Trifirò, A. Vaccari, Effect of fly ash and H2S on a Ni-based catalyst for the upgrading of a biomass-generated gas, Biomass and Bioenergy, 32 (2008)
345-353.
[103] C.H. Bartholomew, P.K. Agrawal, J.R. Katzer, Sulfur Poisoning of Metals, in: D.D. Eley, H. Pines, P.B. Weisz (Eds.) Advances in Catalysis, Academic
Press1982, pp. 135-242.
[104] B. Wang, G. Ding, Y. Shang, J. Lv, H. Wang, E. Wang, Z. Li, X. Ma, S. Qin,
Q. Sun, Effects of MoO3 loading and calcination temperature on the activity of the sulphur-resistant methanation catalyst MoO3/γ-Al2O3, Applied Catalysis A:
General, 431-432 (2012) 144-150.
[105] P.R. Wentrcek, B.J. Wood, H. Wise, The role of surface carbon in catalytic methanation, Journal of Catalysis, 43 (1976) 363-366.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 66 Chapter 2
[106] Y. Borodko, G.A. Somorjai, Catalytic hydrogenation of carbon oxides – a
10-year perspective, Applied Catalysis A: General, 186 (1999) 355-362.
[107] S.-I. Fujita, N. Takezawa, Difference in the selectivity of CO and CO2 methanation reactions, Chemical Engineering Journal, 68 (1997) 63-68.
[108] G.P. Van Der Laan, A.A.C.M. Beenackers, Kinetics and Selectivity of the
Fischer–Tropsch Synthesis: A Literature Review, Catalysis Reviews, 41 (1999)
255-318.
[109] G.W. Norval, Notes on the issues of equilibrium in the Fischer–Tropsch synthesis, The Canadian Journal of Chemical Engineering, 86 (2008) 1062-1069.
[110] M.C. Bell, Thermodynamically Controlled Catalysis: Equilibria in Fischer-
Tropsch Synthesis, Canadian Metallurgical Quarterly, 34 (1995) 331-341.
[111] R.A. van Santen, A.J. Markvoort, I.A.W. Filot, M.M. Ghouri, E.J.M. Hensen,
Mechanism and microkinetics of the Fischer-Tropsch reaction, Physical
Chemistry Chemical Physics, 15 (2013) 17038-17063.
[112] H.S. Pichler, Hans; Hojabri, Fereidun, Synthesis of α-olefins from carbon monoxide and hydrogen, Brennstoff-Chemie, 45 (1964) 215 - 221.
[113] B.H. Davis, Fischer–Tropsch synthesis: current mechanism and futuristic needs, Fuel Processing Technology, 71 (2001) 157-166.
[114] Y.H. Zhao, K. Sun, X. Ma, J. Liu, D. Sun, H.Y. Su, W.X. Li, Carbon Chain
Growth by Formyl Insertion on Rhodium and Cobalt Catalysts in Syngas
Conversion, Angewandte Chemie International Edition, 50 (2011) 5335-5338.
[115] I.M. Ciobica, R.A. Van Santen, Carbon monoxide dissociation on planar and stepped Ru(0001) surfaces, Journal of Physical Chemistry B, 107 (2003)
3808-3812.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 67 Chapter 2
[116] R.C. Brady, R. Pettit, Mechanism of the Fischer-Tropsch reaction. The chain propagation step, Journal of the American Chemical Society, 103 (1981)
1287-1289.
[117] R.C. Brady, R. Pettit, Reactions of diazomethane on transition-metal surfaces and their relationship to the mechanism of the Fischer-Tropsch reaction,
Journal of the American Chemical Society, 102 (1980) 6181-6182.
[118] J. Gaube, H.F. Klein, Studies on the reaction mechanism of the Fischer–
Tropsch synthesis on iron and cobalt, Journal of Molecular Catalysis A: Chemical,
283 (2008) 60-68.
[119] P.M. Maitlis, R. Quyoum, H.C. Long, M.L. Turner, Towards a chemical understanding of the Fischer–Tropsch reaction: alkene formation, Applied
Catalysis A: General, 186 (1999) 363-374.
[120] B.E. Mann, M.L. Turner, R. Quyoum, N. Marsih, P.M. Maitlis, Demonstration by 13C NMR spectroscopy of regiospecific carbon-carbon coupling during
Fischer-Tropsch probe reactions [1], Journal of the American Chemical Society,
121 (1999) 6497-6498.
[121] P.M. Maitlis, V. Zanotti, The role of electrophilic species in the Fischer-
Tropsch reaction, Chemical Communications, (2009) 1619-1634.
[122] M. Araki, V. Ponec, Methanation of carbon monoxide on nickel and nickel- copper alloys, Journal of Catalysis, 44 (1976) 439-448.
[123] R.Z.C. van Meerten, J.G. Vollenbroek, M.H.J.M. de Croon, P.F.M.T. van
Nisselrooy, J.W.E. Coenen, The kinetics and mechanism of the methanation of carbon monoxide on a nickel-silica catalyst, Applied Catalysis, 3 (1982) 29-56.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 68 Chapter 2
[124] J.A. Dalmon, G.A. Martin, The kinetics and mechanism of carbon monoxide methanation over silica-supported nickel catalysts, Journal of Catalysis, 84
(1983) 45-54.
[125] J. Klose, M. Baerns, Kinetics of the methanation of carbon monoxide on an alumina-supported nickel catalyst, Journal of Catalysis, 85 (1984) 105-116.
[126] R.E. Hayes, W.J. Thomas, K.E. Hayes, A study of the nickel-catalyzed methanation reaction, Journal of Catalysis, 92 (1985) 312-326.
[127] S.i. Fujita, M. Nakamura, T. Doi, N. Takezawa, Mechanisms of methanation of carbon dioxide and carbon monoxide over nickel/alumina catalysts, App. Catal.
A, 104 (1993) 87-100.
[128] T. Kammler, J. Küppers, Methanation of carbon on Ni(100) surfaces at 120
K with gaseous H atoms, Chemical Physics Letters, 267 (1997) 391-396.
[129] J. Sehested, S. Dahl, J. Jacobsen, J.R. Rostrup-Nielsen, Methanation of
CO over nickel: Mechanism and kinetics at high H2/CO ratios, Journal of Physical
Chemistry B, 109 (2005) 2432-2438.
[130] G.D. Weatherbee, C.H. Bartholomew, Hydrogenation of CO2 on group VIII metals: II. Kinetics and mechanism of CO2 hydrogenation on nickel, Journal of
Catalysis, 77 (1982) 460-472.
[131] J.L. Falconer, A.E. Zaǧli, Adsorption and methanation of carbon dioxide on a nickel/silica catalyst, Journal of Catalysis, 62 (1980) 280-285.
[132] M. Marwood, R. Doepper, A. Renken, In-situ surface and gas phase analysis for kinetic studies under transient conditions The catalytic hydrogenation of CO2, Applied Catalysis A: General, 151 (1997) 223-246.
[133] W. Wei, G. Jinlong, Methanation of carbon dioxide: an overview, Frontiers of Chemical Science and Engineering, 5 (2011) 2-10.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 69 Chapter 2
[134] S. Akamaru, T. Shimazaki, M. Kubo, T. Abe, Density functional theory analysis of methanation reaction of CO2 on Ru nanoparticle supported on TiO2
(101), Applied Catalysis A: General, 470 (2014) 405-411.
[135] P.A.U. Aldana, F. Ocampo, K. Kobl, B. Louis, F. Thibault-Starzyk, M. Daturi,
P. Bazin, S. Thomas, A.C. Roger, Catalytic CO2 valorization into CH4 on Ni- based ceria-zirconia. Reaction mechanism by operando IR spectroscopy,
Catalysis Today, 215 (2013) 201-207.
[136] Q. Pan, J. Peng, T. Sun, S. Wang, S. Wang, Insight into the reaction route of CO2 methanation: Promotion effect of medium basic sites, Catalysis
Communications, 45 (2014) 74-78.
[137] Q. Pan, J. Peng, S. Wang, S. Wang, In situ FTIR spectroscopic study of the
CO2 methanation mechanism on Ni/Ce0.5Zr0.5O2, Catalysis Science &
Technology, 4 (2014) 502-509.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 70
Chapter 3
Materials and methods
Chapter 3
3.1 Catalyst preparation
3.1.1 Preparation method
One of the most widely adopted methodologies for the synthesis of supported catalysts is incipient wetness impregnation (dry impregnation). In this method, support pores are loaded with a solution comprised of soluble metal precursors in a suitable solvent. The volume of precursor’s solution should not exceed the total pore volume of the support. The catalyst support is then mixed with the precursor solution and the now impregnated catalyst is dried and calcined. In the application of the incipient wetness method, it is widely asserted that the drying step is more important than the specific interaction during the support impregnation [1]. One of the key advantages of the dry impregnation method is that the quantity of metallic precursor loaded on the support can be carefully controlled, when compared to other impregnation methods. However, achieving high levels of uniformity (metal dispersion) is a common challenge with the incipient wetness technique.
Figure 3.1. Pore filling during (a) impregnation and the effect of subsequent (b) drying [1].
In this case, a systematically controlled drying step may reduce metallic crystal size (enhancing dispersion). Figure 3.1 illustrates a simple schematic
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 72 Chapter 3
presentation of pore filling following a subsequent drying step during impregnation catalyst preparation. Ideally, no excess metal precursor solution remains following catalyst preparation by incipient wetness method [1-4].
3.1.2 Materials
3.1.2.1 Catalyst support
Sasol alumina spheres (1.8/210) were used as catalysts’ support in this study.
Table 3.1 shows the materials specification as provided by the manufacturer [5].
Table 3.1. Standard specifications of Sasol alumina spheres, used in this study. Spheres 1.8/210 Diameter [mm] 1.8 Crush strength [N] min. 50 Packed bulk density [g/L] 540-580 Surface area [m2/g] 200-220 Pore volume [cm3/g] min. 0.75
3.1.2.2 Metal precursors
Various nickel based catalysts were prepared and analysed in this study. Table
3.2 illustrates different precursors that has been used during catalyst preparation in this study.
Table 3.2. Different transition metal catalyst precursors. Formula Supplier Purity (%) Nickel Ni(NO3)2.6H2O Sigma-Aldrich 99.9999 Iron Fe(NO3)3.9H2O Sigma-Aldrich 98 Cobalt Co(NO3)2.6H2O Sigma-Aldrich 98 Copper Cu(NO3)2.2.5H2O Sigma-Aldrich 98 Chromium Cr(NO3)3.9H2O Sigma-Aldrich 99 Manganese Mn(NO3)2.4H2O Sigma-Aldrich 97 Zinc Zn(CH3COO)2.2H2O Chem-Supply 98 Ruthenium RuCl3.H2O Precious Metals Online 99 Rhodium RhCl3.H2O Precious Metals Online 99 Silver AgNO3 Ajax Chemicals 99.9 Cadmium Cd(NO3)2.4H2O Aldrich 98
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 73 Chapter 3
3.1.3 Preparation procedure
In this study, a series of nickel-based alumina supported catalysts were prepared by dry impregnation (incipient wetness) method. Crushed and calcined (500 °C)
Sasol alumina spheres (1.8/210) were used as the support for all catalysts. For each catalyst (either single- or bi-metallic), a metal salt solution was prepared using a predetermined amount of catalyst precursor dissolved in distilled water.
The volume of the solution was adjusted to a value close to but not exceeding the total pore volume of catalyst support. The impregnation step consisted of the drop-wise addition of the precursor solution to the alumina powder with continuous mixing in clean agate mortar and pestle. At the end of impregnation step, the catalyst slurry was collected and there was no evidence of any of the metal salt solution remaining. The slurry was then carefully dried in a two-step procedure, consisting initial drying at 80 °C for 12 hours followed by drying at 110
°C for 12 hours. The dried slurry was then calcined in static air at 500 °C for 2 hours. The final prepared catalyst was than sieved and sized (250 to 425 μm) for catalyst testing.
3.2 Catalytic measurements
3.2.1 Experimental apparatus
The catalytic measurement experiments were performed in a lab-scale apparatus which included gas control and monitoring instruments and an on-line analytical train. Figure 3.2 displays a schematic view of the experimental apparatus (Figure
3.3a). The two most important experimental parameters that were systematically adjusted and controlled during catalyst evaluation were the reaction temperature and gas flow rates. To control the reaction temperature, a three-zone tubular furnace (Labec: HTF 40/12/3, Figure 3.3d) was used.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 74 Chapter 3
Figure 3.2. Schematic view of the experimental apparatus.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 75 Chapter 3
Figure 3.3. Experimental apparatus used for catalyst performance experiments: a) overview of experimental setup, b) Shimadzu-GC, c) Alicat mass flow controllers, d) Labec furnace and e) Varian Micro-GC.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 76 Chapter 3
Temperature controllers of the furnace were programmable and provided a uniform, constant temperature along the furnace and the reactor tube. Mass flow controllers (Alicat: MC-D-I, Figure 3.3c) were used to control the flow rate of each of the gas supplied for reaction experiments. These mass flow controllers (±1% variation) were interfaced to a lab computer, such that inlet flows for each gas were adjusted and monitored using commercial computer software (Flow Vision).
3.2.2 Analytical instruments
During each experiment, two online gas chromatographs were used to analyse the reactor feed and product streams. One of the gas chromatographs was a
Micro-GC (Varian – 490, Figure 3.3e) and the other was a GC (Shimadzu-2014,
Figure 3.3b). The Micro-GC is equipped with two thermal conductivity detectors
(TCD) and two micro columns. In this GC, one of the TCDs is connected to a
PoraPLOT-Q column with helium as reference gas. The second TCD of the
Micro-GC is linked to a Molecular Sieve-5Å column and used argon as reference gas. The acquisition method adopted to collect the data points during the experiments is elucidated in Table 3.3.
Table 3.3. Method parameters of data acquisition by Micro-GC. TCD 1 TCD 2 Column Type MS-5Å PPQ Column Temperature 60 °C 90 °C Column Pressure 100 KPa 150 KPa Reference/Career Gas Argon Helium Injector Temperature 100 °C 100 °C Run Time 5 min 5 min
Another GC used to monitor the gas streams, which was equipped with three detectors including a thermal conductivity detector (TCD), a methanizer/flame ionization detector (mFID) and a flame ionization detector (FID). In this
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 77 Chapter 3
instrument, the TCD and mFID are connected in series to a dual packed column.
The dual packed column (SRI Instruments, CA, USA) and contained two columns
(MS-5Å and HayeSep-D) in parallel. In addition, a FID was linked with a RT-Q-
Bond capillary column (Restek, PA, USA) to separate and detect hydrocarbon species. The carrier and reference gas used was helium for all detectors and columns. In addition, both FID flame gases (air and hydrogen) and helium are filtered before reaching the GC. Figure 3.4 shows typical chromatograms of
Shimadzu-GC. The acquisition method adopted for Shimadzu-GC is summarised in Table 3.4.
Figure 3.4. Typical chromatograms (Top: mFID, Bottom FID).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 78 Chapter 3
Table 3.4. Method parameters of data acquisition by Shimadzu GC. TCD mFID FID Column Type Dual packed Dual packed Capillary Column Temperature 40 – 120 °C 40 – 120 °C 40 – 120 °C Detector Temperature 200 °C 395°C 395°C Reference/Career Gas Helium Helium Helium Career Gas Flow Rate 45 ml/min 45 ml/min 3.5 - 5 ml/min Injector Temperature 110 °C 110 °C 110 °C Run Time 15 min 15 min 15 min
The gas chromatography systems were calibrated with standard calibration mixtures (Coregas, NSW and BOC, NSW). The limit of detection (LOD) and limit of quantification (LOQ) were estimated using signal-to-noise (S/N) method. In this case, it is generally adopted that S/N ratios of 3:1 and 10:1 are the approximate requirements to determine LOD and LOQ respectively [6]. Table 3.5 shows the estimated detection limits for each components based on S/N method.
Table 3.5. Estimated detection limits for each components based on S/N method. Calibration Estimated Estimated Signal/Noise Concentration LOD LOQ CH4 2010 24612.4 0.24 0.81 C2H4 2049 44414.8 0.14 0.46 C2H6 2037 41928.4 0.14 0.48 C3H6 2041 43988.4 0.14 0.46 C3H8 1994 33741.6 0.18 0.59 iC4H8 2018 46975.2 0.13 0.43 nC4H10 2082 54333.6 0.12 0.38 CO (MS5A) 2003 19465.6 0.31 1.03 CO (HS-D) 10 725.4 0.04 0.14 CO2 10 418.4 0.07 0.24
3.2.3 Variable gas feed composition
One of the primary aims of the experiments in this project was to examine the performance of various catalysts during their reaction with a number of different
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 79 Chapter 3
feed compositions. Table 3.6 illustrates the gas used and their suppliers for all experiments. The gas cylinders were connected to the experimental setup through stainless steel lines.
Table 3.6. Gas specifications. Gas Purity/Contents Argon 99.99999 % Air Compressed Air Helium 99.99996 % Hydrogen 99.9999 % Carbon monoxide 99.995 % Carbon dioxide 99.995 % Methane 99.9995 % Ethane: 9.99 % Ethylene: 9.99 % C2-C3 hydrocarbon mixture Propane: 0.997 % Propylene: 1.0 % Helium: Balance Ethane: 10 % Ethane (diluted) Helium: Balance Ethylene: 10 % Ethylene (diluted) Helium: Balance
3.2.4 Reactor experiments procedure
The catalytic experiments were conducted in tubular, quarter-inch (OD) stainless tubing. The sieved catalysts particles (normally 250 – 425 μm) were weighted
(normally 250 mg) and mixed with quartz sand (normally 200 mg) to inhibit hotspot formation during experiments. The catalyst mixture was charged in the middle of the tubular reactor, adopting the same procedure for each run. To pack the catalyst, a plug of inert quartz wool (approximately 0.5 mm) was pushed inside the reactor, and then the reactor was filled with the catalyst on top of the quartz plug, followed by another layer of silica wool. The packed catalyst was initially calcined in air flow at 500 °C for 2 hours. The calcined and fresh catalyst
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 80 Chapter 3
was then reduced in a flow of H2/He (1/1) at 500 °C for two hours prior to each experimental run. To prevent reaction between hydrogen and oxygen, the catalyst was purged with helium for 30 min before the activation process commenced.
The reaction temperature was monitored by a thermocouple placed just outside the exit end of the catalyst bed, inside the reactor. The pressure drop in the reactor tube was measured with a pressure gauge located downstream of the reactor tube. All of the catalyst performance analysis experiments performed at atmospheric pressure.
3.2.5 Quantitative assessment of catalyst performance
Various calculations were performed to characterise the performance of each catalyst studied, such as species concentration (Ci), reactants conversion (Xr), product selectivity (Sp) and yield (Yp). As in the following chapters, there are some differences in the inlet compositions and experimental procedure, methods to calculate mentioned parameters are explained in a generalised way. Assuming the simple reaction below (Eq. 3.1) as the general basic model:
r → p1 + p2 Eq. 3.1
Then the concentration (Eq. 3.2), conversion (Eq. 3.3), selectivity (Eq. 3.4) and yield (Eq. 3.5) has been calculated by equations below:
Ci = Ai × RFi Eq. 3.2
Cr,in−Cr,out Xr = × 100 Eq. 3.3 Cr,in
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 81 Chapter 3
C − C pi,out pi,in Spi = × 100 Eq. 3.4 Cr,in−Cr,out
C − C pi,out pi,in Ypi = × 100 Eq. 3.5 Cr,in
Where Ai and RFi are respectively the peak area and response factor for each species, acquired by gas chromatographic analysis of feed- and product- streams data.
Carbon balance were also estimated to investigate overall process parameters such as presence/absence of carbon deposition. Equation below (Eq. 3.6) outlines the carbon balance calculation method.
TC Carbon balance = out × 100 Eq. 3.6 TCin
Where TCout and TCin are total moles of carbon in outlet and inlet streams
(detected by GC) respectively. It should be mentioned that for the catalyst performance analysing experiments in this study, the carbon balance was within a satisfactory level (98 ± 2 %). Therefore, it can be concluded that the carbon deposition level was negligible [7] during the reaction flow experiments in this study.
3.3 X-ray diffraction (XRD)
X-ray diffraction (XRD) has been used to study and determine crystallinity of solid materials [8]. In case of heterogeneous catalysis, used XRD is among the most widely used techniques [9]. The wavelength of X-rays falls in the range of 0.01 –
10 nanometre. Crystalline materials also have unit-cell parameters with same
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 82 Chapter 3
order of magnitude as X-rays wavelength range. The periodic arrangement of atoms (or the electron clouds) in crystalline materials can interact X-ray waves resulting in diffraction of the X-rays (Figure 3.5) [9].
Figure 3.5. Schematic stretching of Bragg’s law which can be derived from the triangle ABC [9].
In this case, diffracted electromagnetic waves are only detectable when constructive interference occurs, i.e. Bragg’s Law (Eq. 3.7) is satisfied. The
Bragg’s Law is basically a description for the constructive electrostatic scattering of electromagnetic waves in a certain direction [10].
2 푑′ sin θ = n λ Eq. 3.7
Where d’ is the spacing between atomic layers, θ is the incident angle, λ is the wavelength of X-rays and n is the scattering order. In addition to Bragg’s Law, which is geometrical explanation of constructive scattering, the constructive scattering conditions can be also explained by the Laue equations (Eq. 3.8 – 10).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 83 Chapter 3
a (cos φa − cos φa0) = h λ Eq. 3.8
a (cos φb − cos φb0) = k λ Eq. 3.9
a (cos φc − cos φc0) = l λ Eq. 3.10
Where φa, φb, φc are the incident waves angle, φa0, φb0, φc0 are the propagating waves angle and h, k, l are the miller indices (Figure 3.6). Miller indices can be used to describe diffraction peaks (reflections) as hkl vectors, which are representing planes, layers and directions [11]. In XRD, a diffracted X-rays beam from a sample is used to identify crystalline structure. In this case, a diffractometer consisting an X-ray source, a sample holder and a detector, which can be arranged in different ways, is used as XRD instrument [4, 12].
Figure 3.6. An example of Miller indices for cubic structure [8].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 84 Chapter 3
In this study, XRD characterisations were undertaken at The Electron Microscope and X-Ray Unit (EMX), University of Newcastle, NSW [13]. To obtain powder
XRD patterns for crystal phase investigation, the EMX Philips X’Pert MPD machine equipped with a copper anode (Kα = 1.54060 Å at 40 kV and 40 mA) was used. The diffraction patterns were analysed using two software, X’Pert
Highscore Plus [14] and Match [15]. The reference patterns were downloaded from Inorganic Crystal Structure Database (ICSD) [16] and Crystallography Open
Database (COD) [17] and compared to the experimental patterns for each sample. A typical spectrum of XRD analysis is displayed in Figure 3.7.
Figure 3.7. Typical data spectrum of X-ray diffraction instrument (Sample: Al2O3).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 85 Chapter 3
3.4 In-situ Fourier transform infrared spectroscopy (in-situ FTIR)
Another spectroscopic method used in the present investigation was in-situ
Fourier transform infrared spectroscopy (in-situ FTIR), which has been widely used for studying catalysts and catalytic reactions in infrared spectral range [18,
19]. In a Fourier transform infrared spectrometer, the collimated infrared beam is initially processed by passing through a Michelson interferometer [20]. The interferometer splits the IR beam equally into two sub-beams. The sub-beams then are reflected towards a movable and a fixed mirror separately. The beam directed toward the sample is a combination of reflected beams from two mirrors.
By scanning the movable mirror, the variation of retardation (difference between path lengths of the two interferometer arms) over time is obtained as an interferogram in the form of signal recorded by the detector. The signal output is
Fourier transformed, resulting in the IR spectra [21, 22]. The processed beam is focused on the sample and then the exiting beam is collected a the detector [20].
There are a number of setups with different configurations that are available for both ex-situ and in-situ characterization of solid catalysts. One of the most commonly used configurations is transmission mode (Figure 3.8), in which the infrared beam is focused through a self-supported sample located inside a reaction chamber [23].
Figure 3.8. A schematic view FTIR setup in transmission mode (TIR) [21].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 86 Chapter 3
In this study, a purpose built transmission mode in-situ FTIR apparatus was used
[24]. The apparatus consisted of a Bruker Tensor 27 FTIR system and an ultra- high vacuum stainless steel reaction chamber. The Tensor 27 is a single beam instrument equipped with a liquid nitrogen cooled mercury cadmium telluride
TM (LN2-MCT) detector and RockSolid interferometer. The acquisition software connected to the FTIR system was OPUS [25]. The reaction chamber (Figure
3.9a) consisted of a high-grade stainless steel equipped with a built-in sample holder, a heating wire and a thermocouple (Figure 3.9b).
Figure 3.9. a) The reaction chamber; and b) the sample holder of in-situ FTIR apparatus.
The sample’s temperature was controlled with a Eurotherm temperature controller connected to the heater and thermocouple. The ultra-high vacuum condition of the reaction cell was achieved by employing rotary and turbo molecular pumps (Pfeiffer Vacuum). The pressure of the system was monitored by a full-range and a ceramic pressure gauge (Pfeiffer Vacuum). The probe or
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 87 Chapter 3
reaction gases were injected to the cell by a high-accuracy dosing valve (Pfeiffer
Vacuum). Figure 3.10 shows the in-situ FTIR apparatus used in this study.
Figure 3.10. The in-situ Fourier transform infrared spectroscopy system.
In this study, nitric oxide was mainly used as the probe molecule for in-situ FTIR experiments. A typical spectrum acquired with the instrument is shown in Figure
3.11. In this case, different peaks arose following NO adsorption that can be
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 88 Chapter 3
separated into two main groups as reactive and non-reactive surface species
[26]. The surface species arising from non-reactive adsorption of NO are basically coordinated nitric oxide molecules (e.g. mononitrosyls). These coordinated nitric oxide molecules, bonded to the catalyst’s active sites, have properties which are similar to carbon monoxide on the active sites [26].
Figure 3.11. Typical analysed spectrum of in-situ FTIR instrument (Sample: Ni/Al2O3, adsorbate: NO).
3.5 Temperature programmed desorption (TPD)
Temperature programmed desorption (TPD) is a technique to control and measure the product of desorption/decomposition of an adsorbed substance from a solid material as a function of temperature. In the field of heterogeneous catalysis, the data obtained from TPD analysis can also be used to gain insight
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 89 Chapter 3
into identifying and characterising catalytically active sites (either for adsorption or reaction), binding energies, desorption kinetics and surface concentration [27].
One widely adopted TPD method is to perform temperature programmed desorption under ultra-high vacuum conditions, UHV-TPD, in which the surface of solids can be studied under ideal conditions [28]. In general, a number of different steps are involved in performing a TPD analysis. These steps can be include [29]:
- Pre-treatment of the solid material under UHV.
- Adsorption of the adsorbate (either a probe molecule or reactants).
- Evacuation of the adsorbate (physical removal) under UHV.
- Temperature programmed desorption of residual adsorbate under UHV.
- Detection of the thermally desorbed substances (qualitative and quantitative).
In the current study, temperature programmed analysis was performed employing a purpose built UHV-TPD system (Figure 3.12) [24]. The system consisted of a tubular furnace controlled by a Eurotherm temperature controller, a quartz reactor tube, a rotary pump (Pfeiffer Vacuum), two turbo-molecular pumps (Pfeiffer Vacuum) and a quadrupole mass spectrometer (Prisma – Pfeiffer
Vacuum). The temperature of the sample was monitored by a thermocouple connected to the temperature controller and system pressure was monitored by different pressure gauges (Full-range, Pirani and Capacitance – Pfeiffer
Vacuum). The adsorbate was injected to the sample compartment of the system
(quartz tube) to achieve the desired adsorption pressure. The sample and detection compartments where connected to each other by a dosing valve
(Pfeiffer Vacuum) in order to keep the mass spectrometer in the desired vacuum range.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 90 Chapter 3
Figure 3.12. The temperature programmed desorption (TPD) system.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 91 Chapter 3
Temperature programmed desorption was used to study different states of the active metal sites on various catalysts. The catalyst particles used for TPD analysis were similar to the ones used for catalytic performance experiments (in most cases 250 – 425 μm). Different adsorbate (e.g. CO and H2) were employed.
The TPD profiles were analysed and peak deconvolution were applied by using
Grams/AI software. Figure 3.13 illustrates a typical data acquired by UHV-TPD system used in this study. It is reported that the CO/CO2 desorption peaks from
TPD analysis can be assigned to sites with different properties (e.g. single or double chemisorption sites) [30, 31].
Figure 3.13. A typical spectrum of TPD analysis (Sample: CO desorption from
Ni/Al2O3).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 92 Chapter 3
3.6 Temperature programmed reduction (TPR)
Temperature programmed reduction (TPR) is also a commonly used technique in the field of heterogeneous catalysis. In this technique, the main purpose is studying the reduction of a solid catalyst in hydrogen flow (commonly H2 – Ar mixture) while the system’s temperature is changed in a predetermined way. The consumption of hydrogen as a function of temperature is recorded as TPR profile, which usually consists a number of peaks. Each peak describes a distinct reduction reaction of a component as a function of temperature. The properties of each peak such as peak’s centre and area represent the nature of reduced species and their concentration respectively [32].
In this study, a purpose-built temperature programmed reduction system was employed to study the reduction state of fresh catalysts (Figure 3.14).
Figure 3.14. The temperature programmed reduction (TPR) system.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 93 Chapter 3
The TPR apparatus consisted of a number of parts as follow: mass flow controllers, the furnace, the reactor, the detector and the data acquisition system.
Two high-accuracy mass flow controllers (Alicat) were used for both sample and reference lines. The gas cylinder used was a H2 (2 %) and Ar (98 %) mixture supplied by Coregas. A tubular furnace (Carbolite) equipped with a high accuracy controller (Eurotherm) was used to provide a consistent temperature increase during the analysis. The reactor (or sample line) was a quarter inch quartz tube.
The sample, supported by two layers of quartz wool on its both sides, was placed in the middle of the reactor tube. The sample bed consisted of catalyst particles with size range of 250 – 425 μm which is the same range used for catalyst testing analyses.
Figure 3.15. A typical spectrum for temperature programmed reduction (Sample:
Ni/Al2O3).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 94 Chapter 3
The system to analyse hydrogen consumption was a thermal conductivity detector (TCD) cell located inside a separate furnace (at 110 °C). The reference line was directly connected to the corresponding TCD channel, while the outlet stream of the reactor entered TCD cell after passing through a moisture trap. The trap consisted a U-shaped tube that was filled with Molecular-Sieve spheres and cooled with dry ice. The TCD cell was connected to the data recording system.
Temperature programmed reduction was used to investigate the reducibility of different catalysts in this study. In this case, the hydrogen consumption were plotted versus temperature and data was analysed. Figure 3.15 shows a typical output data from TPR instrument. The reducible sites on nickel-based alumina supported catalysts are basically classified into three main regions. The lower temperature range is associated with weakly bonded nickel species. The species reducible in the mid-range have stronger interaction between metal and the support. The high temperature range is related to stable nickel-aluminate species
[33-35].
3.7 Chemisorption (CO and H2)
One of the main objectives in the field of heterogeneous catalysis is gaining insight into the catalytic reactions on atomic scales. In this case, investigating properties of catalysts by the measurement of the extent of interaction between catalysts’ sites and a probe molecule, known as chemisorption, is one of the widely used methods [36]. Chemisorption is basically a phenomenon when a probe molecule forms a chemical bond with the solid (catalyst) surface [37]. There are various different techniques available to analyse the chemisorption of probed surfaces. In this case, volumetric chemisorption is one of the quantitative methods in which the amount of probe gas adsorbed on the catalyst’s surface is
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 95 Chapter 3
determined from changes of the pressure in a constant known volume. To calculate the number of adsorbed probe molecule moles on the catalyst’s surface during chemisorption commonly the ideal gas law is employed [36].
In this study, a purpose-built volumetric chemisorption apparatus (Figure 3.16) was used to determine quantitatively the capacity of different catalysts for chemisorption of probe molecules such as hydrogen and carbon monoxide.
Figure 3.16. The volumetric chemisorption system.
The apparatus was made of borosilicate glass, stainless steel and quartz parts.
The system had a number of different sections as below:
. Sample or reactor quartz tube.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 96 Chapter 3
. Constant volume borosilicate Schlenk flasks, connected to the system via
Schlenk manifold (GPE Scientific)
. Vacuum compartment; consisting a roughing and a turbo-molecular pump
(Pfeiffer Vacuum).
. Heating compartment; consisting a furnace, a temperature controller and
a thermocouple.
. Capacitance high accuracy pressure gauges (Pfeiffer Vacuum).
. Data acquisition system; recording pressure changes by time.
For each experiment, a known amount sample were located into the U-shape tube and the bed was supported with quartz wool on each side. The particle size of the sample bed were in the range of 250 – 425 μm which is the same range used for catalyst testing analyses.
3.8 Inductively coupled plasma optical emission spectroscopy
One of the most important and commonly used technique in the field of atomic spectroscopy is employing inductively coupled plasma (ICP) source combined by optical emission spectroscopy (OES). The technique covers a large number of elements at trace levels. In addition, it can be used to analyse different type of samples. Due to its versatility, ICP-OES has been used for a various applications as an almost ideal analytical technique. This technique has been widely used for elemental analysis of major, minor and trace amounts of species in solid materials
[38].
In general, an ICP-OES instrument consists of four main compartments as below:
- A sample introduction/injection systems.
- A plasma torch.
- A plasma power source and impedance matcher.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 97 Chapter 3
- An optical emission spectrometer.
The sample must be injected to the system in a form that can be vaporized and atomized [39]. In this case, solid samples should be dissolved in or digested with proper acidic solvents prior to the analysis [38]. The injected sample (commonly a small droplet of the solution) is ionised/atomized in the plasma torch (ICP). Due to the excitation by plasma, atoms and ions emit light. The optical emission spectrometer (OES) then analyses the emitted light. The wavelengths characteristic and the intensity of emitted light beam contain information about the element’s type and its concentration [39].
In this study, the elemental compositions of catalyst samples were analysed by a
Varian Radial 715-ES inductively coupled plasma optical emission spectrometer.
Liquid solution of solid samples were prepared by microwave assisted digestion of samples in a highly acidic mixture. The acidic solvent was prepared by mixing
4.5 ml of 65 % HNO3, 4.5 ml of 37 % HCl and 3 ml of 50 % HBF4. For each sample almost 100 mg of solid were mixed with the acid mixture. To complete the digestion procedure, solutions were transferred into a Milestone Start D microwave. The microwave assisted digestion took place at 220 °C for 30 minutes
[40]. Finally, samples’ solutions were diluted to required level (10 – 100 times) and transferred to the instrument’s auto-sampler for ICP-OES analysis.
3.9 Nitrogen adsorption manometry
Nitrogen adsorption manometry has been pre-eminently used as one of the available techniques for characterising porous materials. In this method, nitrogen adsorption isotherms are determined at the liquid nitrogen temperature (≂ 77 K).
This kind of approach was known as ‘BET volumetric method’ because the original method involved measuring the change of gas volume during N2
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 98 Chapter 3
adsorption. However, the term volumetric is not applicable for the most of modern instruments as determination of adsorbed nitrogen is evaluated by changes in the gas pressure rather than volume. One of the operational techniques to determine adsorption isotherms is the discontinuous point-by-point procedure. In this procedure, sufficient time is given to the system to reach equilibrium after introducing successive amount of adsorptive to the sample. Therefore, a series of data points form the adsorption isotherm. On the other hand, continuous procedure in based on the principle of quasi-equilibrium. In this procedure, enough time should be given to the system to provide continuous equilibrium isotherm [41].
Most of studies on adsorption equilibria involve classification of adsorption isotherms. According to the International Union of Pure and Applied Chemistry
(IUPAC) adsorption isotherms are classified in six groups as below (Figure 3.17)
[42].
Figure 3.17. Classification of adsorption isotherms (IUPAC) [43].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 99 Chapter 3
Type I. Microporous materials
Type II. Macroporous materials (strong adsorbate – adsorbent interaction)
Type III. Macroporous materials (weak adsorbate – adsorbent interaction)
Type IV. Mesoporous materials
Type V. Mesoporous materials (weak adsorbate – adsorbent interaction)
Type VI. Homogenous surface materials.
In this study, the nitrogen adsorption manometry analysis has been performed by a Micrometrics TriStar 3000 automated gas adsorption analyser. The analysis took place at liquid nitrogen temperature (77 K). Prior to each experiment, solid samples were degassed by heating up to 180 °C under vacuum conditions overnight. The porous structure properties such as average pore diameter and surface area were calculated based on Barett−Joyner−Halenda (BJH) [44] and
Brunauer−Emmett−Teller (BET) [45] methods. Figure 3.18 shows a typical nitrogen adsorption/desorption isotherm acquired by Micrometrics TriStar 3000 automated gas adsorption analyser.
Figure 3.18. Typical nitrogen adsorption/desorption isotherm (Sample: Ni/Al2O3).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 100 Chapter 3
3.10 Solid-state electrostatic potential calculations
The application of quantum mechanical ab initio simulation technique is growing rapidly because of lower computational cost, more available computer programs and the increased capability of computer codes. In this case, the number of computer programs, specialised to study periodic systems, has been also gaining importance in recent decades. In this study, the CRYSTAL17 program was used for computational studies on the electronic structure of crystals [46, 47].
In general, ab initio codes generate the wavefunction of the system, which is a mathematical tool to describe the behaviour and properties of a chemical system.
The system’s wavefunction is generally acquired as the eigenvector of the time- independent Schrodinger equation (Eq. 3.12) [48].
ĤΨ = EΨ Eq. 3.12
Where Ψ is the wave function of the quantum system, Ĥ is the electronic
Hamiltonian operator and E is the energy of the state. According to the third postulate of quantum mechanics, in any measurement of the associated with an operator (e.g. Ω̂), the only values that will ever be observed are the eigenvalues
(Eq. 3.13) [49].
⟨Ψ|Ω̂|Ψ⟩ 〈Ω〉 = Eq. 3.13 ⟨Ψ|Ψ⟩
Except the Hamiltonian operator, all the other operators depend on just one electron. Therefore, the measured properties associated with these operators all called monoelectronic properties, which can be calculated as follow (Eq. 3.14):
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 101 Chapter 3
푁 〈Ω〉 = ∑푖 ⟨ϕi(r, σ)|Ω|ϕi(r, σ)⟩ Eq. 3.14
Where ϕi is the one-electron wave function (orbital), N is the number of electrons, r and σ are the one electron’s space and spin coordinates respectively. Based on the Linear Combination of Atomic Orbitals (for molecular orbitals and crystalline structures) approximation (Eq. 3.15), the above expression can be written as a matrix product (Eq. 3.16) [50].
ϕi = ∑ ciχ Eq. 3.15
N/2 〈Ω〉 = 2 ∑i ⟨∑μi cμiχμ |Ω| ∑νi cνiχν⟩ = ∑μ ∑ν ΡμνΩμν Eq. 3.16
Where ci is the corresponding coefficient (weights of the contribution of n atomic orbital in the molecular orbital), χ represents the atomic orbital and μ, ν are local orbitals. The Ρμν matrix is density matrix which holds the wave functions information. The P matrix is available at the end of Self-Consistent Field (SCF) calculations by the CRYSTAL17 software [47]. The other matrix (Ωμν) represent the mono electronic properties on the basis of the basis-set functions [51].
In order to perform the electrostatic potential calculations and generate isodensity surfaces (electron density surfaces colour-coded by electrostatic potential), two monoelectronic properties must be acquired. These necessary monoelectronic properties are the electron density and the electrostatic potential. These properties are calculated in CRYSTAL17 code as explained in the following [47].
To evaluate the electron density (ρ) at point R, an operator called Dirac’s delta function is used (Eq. 3.17).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 102 Chapter 3
ρ̂(R) = δ (R − r) Eq. 3.17
In addition, to evaluate the electrostatic potential, its classic mathematical form is used (Eq. 3.18).
1 V̂(R) = Eq. 3.18 |r−R|
The computed monoelectronic properties of the system (i.e. electron density and electrostatic potential) are saved as two separate output files, containing 3- dimnetal information. These two files can be used by visualisation softwares such as VMD [52] or UCSF Chimera [53]. In this case, initially an isodensity surface is generated by employing the electron density at a desirable level. Finally, the isodensity surface can be coloured by the electrostatic potential values from the file containing electrostatic potential information.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 103 Chapter 3
3.11 References
[1] J. Ross, Heterogeneous CatalysisCatalysis - Fundamentals and Applications,
Elsevier.
[2] G. Ertl, H. Knözinger, J. Weitkamp, Preparation of Solid Catalysts, Wiley1999.
[3] K.P. de Jong, Synthesis of Solid Catalysts, Wiley2009.
[4] G. Sanchez Combita, Enhancing allyl alcohol selectivity in the heterogeneous catalytic conversion of glycerol, 2016.
[5] http://www.sasolgermany.de/index.php?id=alumina-home.
[6] A. Shrivastava, V. Gupta, Methods for the determination of limit of detection and limit of quantitation of the analytical methods, Chronicles of Young Scientists,
2 (2011) 21-25.
[7] A.N. Fatsikostas, X.E. Verykios, Reaction network of steam reforming of ethanol over Ni-based catalysts, Journal of Catalysis, 225 (2004) 439-452.
[8] N. Waeselmann, Structural transformations in complex perovskite-type relaxor and relaxor-based ferroelectrics at high pressures and temperatures, 2012.
[9] J.W. Niemantsverdriet, Spectroscopy in Catalysis: An Introduction,
Wiley2008.
[10] W.L. Bragg, The diffraction of short electromagnetic Waves by a Crystal,
Scientia, 23 (1929) 153.
[11] C. Kittel, Introduction to solid state physics, Wiley1986.
[12] Y. Waseda, E. Matsubara, K. Shinoda, X-Ray Diffraction Crystallography:
Introduction, Examples and Solved Problems, Springer Berlin Heidelberg2011.
[13] https://www.newcastle.edu.au/research-and-innovation/resources/central- scientific-services/emx.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 104 Chapter 3
[14] T. Degen, M. Sadki, E. Bron, U. König, G. Nénert, The HighScore suite,
Powder Diffraction, 29 (2014) S13-S18.
[15] Match! - Phase Identification from Powder Diffraction, Crystal Impact - Dr. H.
Putz & Dr. K. Brandenburg GbR, Kreuzherrenstr. 102, 53227 Bonn, Germany.
[16] https://www.fiz-karlsruhe.de/en/leistungen/kristallographie/icsd.html.
[17] http://www.crystallography.net/cod/index.php.
[18] N.-Y. Topsøe, In situ FTIR: A versatile tool for the study of industrial catalysts,
Catalysis Today, 113 (2006) 58-64.
[19] K. Hinrichs, In Situ Infrared Spectroscopy, in: D. Li (Ed.) Encyclopedia of
Microfluidics and Nanofluidics, Springer US, Boston, MA, 2013, pp. 1-4.
[20] J.A.d.H. P. R. Griffiths, Fourier Transform Infrared Spectroscopy, Vol. 83 aus der Reihe: Chemical Analysis—A Series of Monographs of Analytical Chemistry and Its Applications, Berichte der Bunsengesellschaft für physikalische Chemie,
90 (1986) 1240-1241.
[21] F. Zaera, New advances in the use of infrared absorption spectroscopy for the characterization of heterogeneous catalytic reactions, Chemical Society
Reviews, 43 (2014) 7624-7663.
[22] A.G. Marshall, F.R. Verdun, Fourier Transforms in NMR, Optical, and Mass
Spectrometry: A User's Handbook, Elsevier Science2016.
[23] F. Zaera, Infrared and molecular beam studies of chemical reactions on solid surfaces, International Reviews in Physical Chemistry, 21 (2002) 433-471.
[24] J.P.H. Li, M. Stockenhuber, A temperature programmed desorption study of the interaction of ethyl cyanoacetate and benzaldehyde on metal oxide surfaces,
Catalysis Today, 245 (2015) 108-115.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 105 Chapter 3
[25] https://www.bruker.com/products/infrared-near-infrared-and-raman- spectroscopy.html.
[26] K.I. Hadjiivanov, Identification of Neutral and Charged NxOy Surface Species by IR Spectroscopy, Catalysis Reviews, 42 (2000) 71-144.
[27] V. Rakić, L. Damjanović, Temperature-Programmed Desorption (TPD)
Methods, in: A. Auroux (Ed.) Calorimetry and Thermal Methods in Catalysis,
Springer Berlin Heidelberg, Berlin, Heidelberg, 2013, pp. 131-174.
[28] J.M. Kanervo, T.J. Keskitalo, R.I. Slioor, A.O.I. Krause, Temperature- programmed desorption as a tool to extract quantitative kinetic or energetic information for porous catalysts, Journal of Catalysis, 238 (2006) 382-393.
[29] R.J. Cvetanović, Y. Amenomiya, Application of a Temperature-Programmed
Desorption Technique to Catalyst Studies, in: D.D. Eley, H. Pines, P.B. Weisz
(Eds.) Advances in Catalysis, Academic Press1967, pp. 103-149.
[30] A. Tanksale, J.N. Beltramini, J.A. Dumesic, G.Q. Lu, Effect of Pt and Pd promoter on Ni supported catalysts-A TPR/TPO/TPD and microcalorimetry study,
Journal of Catalysis, 258 (2008) 366-377.
[31] G. Ehrlich, Adsorption and electrical conduction in thin films, The Journal of
Chemical Physics, 35 (1961) 2165-2167.
[32] A. Jones, Tempature-Programmed Reduction for Solid Materials
Characterization, Taylor & Francis1986.
[33] D. Hu, J. Gao, Y. Ping, L. Jia, P. Gunawan, Z. Zhong, G. Xu, F. Gu, F. Su,
Enhanced investigation of CO methanation over Ni/Al 2O 3 catalysts for synthetic natural gas production, Industrial and Engineering Chemistry Research, 51
(2012) 4875-4886.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 106 Chapter 3
[34] C. Guo, Y. Wu, H. Qin, J. Zhang, CO methanation over ZrO2/Al2O3 supported Ni catalysts: A comprehensive study, Fuel Process. Technol., 124
(2014) 61-69.
[35] J.M. Rynkowski, T. Paryjczak, M. Lenik, On the nature of oxidic nickel phases in NiO/γ-Al2O3 catalysts, Applied Catalysis A, General, 106 (1993) 73-82.
[36] J.L.G. Fierro, Chemisorption of Probe Molecules, Elsevier Science1990.
[37] J.R. Smith, Theory of Chemisorption, Springer Berlin Heidelberg2013.
[38] J. Cazes, Analytical Instrumentation Handbook, Third Edition, CRC
Press2004.
[39] C.R. Brundle, C.A. Evans, S. Wilson, Encyclopedia of Materials
Characterization: Surfaces, Interfaces, Thin Films, Butterworth-Heinemann1992.
[40] M. Ghoorah, B.Z. Dlugogorski, H.C. Oskierski, E.M. Kennedy, Study of thermally conditioned and weak acid-treated serpentinites for mineralisation of carbon dioxide, Minerals Engineering, 59 (2014) 17-30.
[41] K. Sing, The use of nitrogen adsorption for the characterisation of porous materials, Colloids and Surfaces A: Physicochemical and Engineering Aspects,
187-188 (2001) 3-9.
[42] M. Donohue, G.L. Aranovich, Classification of Gibbs adsorption isotherms,
1998.
[43] http://www.microtrac-bel.com/en/tech/bel/seminar02.html.
[44] E.P. Barrett, L.G. Joyner, P.P. Halenda, The Determination of Pore Volume and Area Distributions in Porous Substances. I. Computations from Nitrogen
Isotherms, Journal of the American Chemical Society, 73 (1951) 373-380.
[45] S. Brunauer, P.H. Emmett, E. Teller, Adsorption of Gases in Multimolecular
Layers, Journal of the American Chemical Society, 60 (1938) 309-319.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 107 Chapter 3
[46] R. Dovesi, R. Orlando, C. Roetti, C. Pisani, V. Saunders, The periodic
Hartree-Fock method and its implementation in the Crystal code, Physica Status
Solidi B Basic Research, 217 (2000) 63-88.
[47] R. Dovesi, R. Orlando, A. Erba, C.M. Zicovich-Wilson, B. Civalleri, S.
Casassa, L. Maschio, M. Ferrabone, M. De La Pierre, P. D'Arco, Y. Noël, M.
Causà, M. Rérat, B. Kirtman, CRYSTAL14: A program for the ab initio investigation of crystalline solids, International Journal of Quantum Chemistry,
114 (2014) 1287-1317.
[48] E. Schrödinger, An Undulatory Theory of the Mechanics of Atoms and
Molecules, Physical Review, 28 (1926) 1049-1070.
[49] R. Shankar, Principles of Quantum Mechanics, Springer US2012.
[50] R.A. Evarestov, Quantum Chemistry of Solids: The LCAO First Principles
Treatment of Crystals, Springer Berlin Heidelberg, Berlin, Heidelberg, 2007, pp.
105-146.
[51] M. Towler, An introductory guide to Gaussian basis sets in solid-state electronic structure calculations, (2000).
[52] W. Humphrey, A. Dalke, K. Schulten, VMD: visual molecular dynamics,
Journal of molecular graphics, 14 (1996) 33-38.
[53] E.F. Pettersen, T.D. Goddard, C.C. Huang, G.S. Couch, D.M. Greenblatt,
E.C. Meng, T.E. Ferrin, UCSF Chimera - A visualization system for exploratory research and analysis, Journal of Computational Chemistry, 25 (2004) 1605-
1612.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 108
Chapter 4
On the catalytic hydrogenation of carbon oxides in the presence of C1-C3 olefins and paraffins
Chapter 4
4.1 Abstract
Simultaneous hydrogenation of CO and CO2 was conducted over 10 wt%
Ni/Al2O3 catalyst, prepared by incipient wetness, in a fixed-bed reactor. The influence of the presence of C1-C3 paraffins and olefins on hydrogenation of the carbon oxides was systematically studied. Changes in the concentration of reactants and products were monitored in the temperature range up to 500 °C.
Catalysts were characterized by power X-ray diffraction and N2 adsorption/desorption techniques. The catalytic activities studies showed that the concentration of carbon oxides and olefins decreases at lower reaction temperatures (150 – 300 °C), while a considerable decrease in paraffin concentration was observed at higher temperatures (above 350 °C). For feed compositions containing COx and C1-C3, the consumption of carbon monoxide, carbon dioxide and C2/C3 paraffins was observed and their maximum conversion was attained over different temperature ranges, in the following order: CO < CO2
< C2/C3 paraffins. However, the concentration of feed olefins decreased significantly over the entire reaction temperature range studied.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 110 Chapter 4
4.2 Introduction
Notably in recent decades, the worldwide development of new and often alternative energy resources is continuously increasing due to population and industrial growth [1]. As a result, light hydrocarbons from fossil sources are in high demand for use as energy resources and as feedstocks for different chemical industries. However, using fossil hydrocarbon resources has many economic and environmental constraints such as high oil prices and the growing concern over global warming due to the elevated and increasing concentration of
CO2 in the atmosphere. Subsequently, researchers have been motivated to design and develop technologies to produce synthetic hydrocarbons from alternative feedstocks such as natural gas and biomass. For instance, the oxidative coupling of methane (OCM) has been studied intermittently over the past 30 years [2, 3]. In many studies on the conversion of natural gas to higher molecular weight compounds, carbon oxides are produced as unwanted by- products, and COx can be a limiting parameter in the commercialization of many of these new and otherwise commercially attractive processes [4]. Consequently, the removal, separation or conversion of carbon oxides present in a hydrocarbon product stream is critical to make these processes more commercially attractive.
In an effort to address this challenge, research has been conducted, focussing on the development of processes which reduce the level of carbon oxides in the final product stream. One approach, and the one presented in the present study, is to catalytically convert the carbon oxides in a second process step but without reducing the concentration of desirable products which are present in the raw product stream.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 111 Chapter 4
Under a wide range of reaction conditions, the catalytic hydrogenation of carbon oxides (Eq. 4.1 & 4.2) to produce hydrocarbons (mostly methane) is thermodynamically spontaneous.
0 CO + 3H2 → CH4 + H2O; ∆H298 K = −206.1 kJ⁄mol Eq. 4.1
0 퐶푂2 + 4퐻2 → 퐶퐻4 + 2퐻2푂; 훥퐻298 퐾 = −165 푘퐽⁄푚표푙 Eq. 4.2
The hydrogenation of carbon oxides was first reported by Sabatier and
Senderens at the beginning of the 20th century [5], and since then a number of important studies have been undertaken in an effort to improve the process. The hydrogenation of carbon oxides has been widely considered for different applications including removal of trace amounts of CO from H2–rich feed gas, in ammonia synthesis plants, for the purification of the reformate gas for use in a fuel cell, Fischer–Tropsch synthesis, and the production of substitute natural gas
(SNG) from coal, biomass or other carbon oxides resources [6-9].
Various catalysts, usually based on group 8-10 transition metals (Ni [10, 11], Ru
[12, 13], Rh [14], Pd [15], Pt [16], Fe [17], and Co [1]) on several different support materials (Al2O3 [10], SiO2 [18], ZrO2 [19], CeO2 [20], La2O3 [21], MgO [22], TiO2
[14], carbon materials [23] and zeolites [12]) have been investigated for carbon oxide hydrogenation. It is reported by Mills and Steffgen that the intrinsic activity of carbon oxide methanation (in case of conversion of COx to methane) over various metals can be characterised in the following order: Ru > Ir > Rh > Ni >
Co > Os > Pt > Fe > Mo > Pd > Ag [24]. Among all these metals, supported nickel catalysts (especially on alumina) has been extensively studied, since nickel
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 112 Chapter 4
catalysts have high activity for hydrogenation of carbon oxides. Nickel-based catalysts have been applied in different industrial processes for the hydrogenation of carbon oxides and unsaturated hydrocarbons [25]. Most commercial catalyst manufacturers market alumina-supported nickel catalysts for large scale industrial plants [26].
To understand the multitude of complex reactions in catalytic processes such as carbon oxides hydrogenation (especially where the reaction takes place in presence of light hydrocarbons) an assessment of the thermodynamic properties of the system is useful. The equilibrium state of the system can be calculated based on the Gibbs free energy minimization [27-29].
The main focus of the current study is exploring the influence of light hydrocarbons on carbon oxides hydrogenation over Ni/Al2O3 in a fixed bed reactor. In this case, we have measured the change in carbon oxides concentration during reaction over alumina-supported nickel catalysts, when reacting in the presence of light hydrocarbons including ethane, ethylene, propane and propylene in the feed. An assessment of the catalysts’ performance in various feed streams has been performed in temperature range from 150 °C to 500 °C.
4.3 Experimental
4.3.1 Catalyst Preparation
An alumina supported nickel catalyst was prepared by incipient wetness method.
The alumina used in the synthesis of the catalysts was originally porous, high surface area alumina spheres (Sasol, Alumina Spheres 1.8/210). Nickel (II) nitrate hexahydrate (99.999% trace metals basis, Aldrich) was used as precursor.
Typically, 4.97 g nickel (II) nitrate hexahydrate was dissolved in distilled water to
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 113 Chapter 4
achieve 6 ml of nickel nitrate aqueous solution to synthesise 10 wt% nickel catalyst. The solution was added drop-wise to 9 g of crushed and calcined alumina powder while grinding in an agate mortar and pestle until the incipient wetness point was reached. The slurry was dried overnight at 130 °C. The dried green colour slurry was ground and then calcined at 500 °C for 2 hours in static air. Finally, the calcined 10 wt% Ni/Al2O3 catalyst powder was shaped and sized by laboratory sieves. The particles with size between 250 µm to 425 µm were collected.
4.3.2 Catalyst Evaluation
Catalytic experiments were performed in a fixed bed tubular reactor. The reactor was heated to the reaction temperature with a three-zone furnace and the temperature was monitored by a thermocouple placed just after the catalyst bed inside the reactor. Gas flow rates were controlled by mass flow controllers (Alicat,
MC series). The sized catalyst particles were packed in the middle of the tubular reactor in the same manner for each run. To charge the reactor with catalyst, a layer of inert silica wool was pushed inside the reactor, and then the reactor was filled with about 240 milligram of 250-425 µm 10 wt% Ni/Al2O3, followed by another layer of quartz wool. Prior to each run, in-situ catalyst calcination and reduction were performed. The packed catalyst was firstly calcined in 100 ml/min air at 500 °C for 2 hours, followed by in-situ catalyst reduction in a mixture of 50 ml/min He and 50 ml/min H2 at 500 °C for 2 hours. To remove any oxygen remaining in the reactor, the catalyst was purged in 50 ml/min helium for 30 min before starting the reduction process. The total gas hourly space velocity of around 20,000 hr-1 was adopted for all catalyst screening experiments. The
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 114 Chapter 4
pressure drop in the reactor tube was measured with a pressure gauge located before the reactor tube. In all runs, the pressure drop was negligible.
In order to study the influence of light hydrocarbons on carbon oxides hydrogenation, the alumina supported nickel catalyst was studied with feed streams of varying compositions. Each feed stream was comprised of a mixture of different gases for each set of experiments. The reactant gases (Coregas,
NSW) were connected to the experimental setup via stainless steel quarter inch tubes. Carbon monoxide (99.995 %), hydrogen (99.9999 %), helium (99.99996
%), methane (99.9995 %), carbon dioxide (99.995 %) and C2/C3 (9.99 % ethane,
9.99 % ethylene, 0.997 % propane and 1.0% propylene in helium balance) were used. Nine sets of experiments (Table 4.1) were undertaken to determine the effect of C2/C3 olefins and paraffins on the reactions taking place between carbon oxides and hydrogen. Temperature range for the first set was from 50 °C to 500
°C while the experiments for sets 2-9 were carried out between 150 °C to 500 °C.
Table 4.1. Feed compositions for each set of experiments. Partial Pressure (KPa) Set H2 CO CO2 CH4 C2H6 C2H4 C3H8 C3H6 He 1 85.11 2.03 0.68 10.13 0.34 0.34 0.03 0.03 2.67 2 85.11 0.00 0.00 0.00 0.34 0.34 0.03 0.03 15.50 3 85.11 2.03 0.00 0.00 0.00 0.00 0.00 0.00 14.19 4 85.11 2.03 0.00 0.00 0.34 0.34 0.03 0.03 13.47 5 85.11 0.00 0.68 0.00 0.00 0.00 0.00 0.00 15.54 6 85.11 0.00 0.68 0.00 0.34 0.34 0.03 0.03 14.82 7 85.11 2.03 0.68 0.00 0.00 0.00 0.00 0.00 13.51 8 85.11 2.03 0.68 0.00 0.34 0.34 0.03 0.03 12.80 9 85.11 2.03 0.68 10.13 0.34 0.34 0.03 0.03 2.67
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 115 Chapter 4
4.3.3 Catalyst Characterization
The porous structure of the prepared catalyst and its support were analysed by
Micrometrics TriStar 3000 automated gas adsorption analyser. The analysis gas was nitrogen and the analysis bath temperature was kept at 77 K. Prior to undertaking porosity and surface area measurements, the solid samples were degassed at 180 °C overnight under vacuum conditions. The average pore width and specific surface area of the samples were calculated using
Barett−Joyner−Halenda (BJH) and Brunauer−Emmett−Teller (BET) methods respectively. A Philips X’Pert MPD with copper anode (Kα = 1.54060 Å at 40 kV and 40 mA) were used to take X-ray diffraction (XRD) patterns for further crystal phase investigations.
4.4 Thermodynamic Analysis
Effect of temperature (T) on the chemical reaction’s equilibrium constant (K) is defined by Van ‘t Hoff equation (Eq. 4.3) [30]:
d ln K ∆H° = Eq. 4.3 d T RT2
Where ∆H° is the standard enthalpy of the reaction and R is the universal gas constant. Assuming that ∆H° is independent of temperature, the equilibrium constant at an arbitrary temperature (T) is related to the standard enthalpy of
° reaction (∆H ) and equilibrium constant (K0) at standard temperature (T0 = 298.15
K) by equation below (Eq. 4.4, derived from Van ‘t Hoff equation):
° ∆H T0 ln K = ln K0 + (1 − ) Eq. 4.4 RT0 T
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 116 Chapter 4
K0 can be calculated by Eq. 4.5:
− ∆G° ln K0 = Eq. 4.5 RT0
Where ∆G° is the standard Gibbs free energy of reaction, which is related to standard enthalpy (∆H°) and entropy (∆S°) of reaction by equation 4.6. Moreover, standard enthalpy (∆H°) and entropy (∆S°) of reaction were calculated by equations 4.7 and 4.8 respectively.
∆G° = ∆H° − T ∆S° Eq. 4.6
° ° ° ∆H = ∑ ∆Hf(products) − ∑ ∆Hf(reactants) Eq. 4.7
∆S° = ∑ S°(products) − ∑ S°(reactants) Eq. 4.8
° ° Where ∆Hf is the standard enthalpy of formation and S is the standard entropy of species. Table 4.2 shows the possible species in the system studied in this chapter. The data in Table 4.2 contains values used in the thermodynamic calculations to evaluate the effect of temperature on equilibrium constants of possible reactions in the system.
The chemical equilibrium concentration of all species at different temperatures were calculated by employing Gibbs free energy minimization method. At the equilibrium state, the change of total Gibbs free energy (Gt) of the system at given
T and P does not change while differential variations occur (Eq. 4.9) [27].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 117 Chapter 4
t (d G )T,P = 0 Eq. 4.9
Commercial software (COSILAB, [31]) was used for chemical equilibrium calculations with consideration of the species in Table 4.2. The software uses a built-in solver called EQUI to perform the Gibbs-function minimization by employing element-potential method [32].
Table 4.2. Standard properties of possible components in the system. The bracket is the reference used to source the thermodynamic for each species. ° ° Component ∆Hf (kJ⁄mol) S (J⁄mol. K) CO (g) -110.5 [33] 197.66 [33] CO2 (g) -393.52 [33] 213.79 [33] CH4 (g) -74.6 [34] 186.25 [33] H2 (g) 0 130.68 [33] -Ch2- (g) 386.39 [33] 193.93 [33] H2O (g) -241.83 [33] 188.84 [33] C (s) 0 5.83 [33] C2H4 (g) 52.47 [33] 219.32 [35] C2H6 (g) -84.01 [34] 229.49 [36] C3H6 (g) 20.41 [37] 266.94 [36] C3H8 (g) -104.71 [38] 267.77 [36]
4.5 Results and Discussion
4.5.1 Thermodynamic Analysis
Theoretical thermodynamic calculations were performed to study the effect of temperature on equilibrium constant and species concentration at chemical equilibrium state. Table 4.3 shows the possible reactions and their standard properties [28, 39-42].
Figure 4.1 illustrates the natural logarithm of equilibrium constants (K) of the possible reactions (Table 4.3) occurring in carbon oxides hydrogenation in presence of light hydrocarbons as a function of temperature. The equilibrium
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 118 Chapter 4
constants at different temperature were calculated by Van ‘t Hoff equation (Eq.
4.3).
Table 4.3. Possible chemical reactions and their standard properties. ∆H° ∆S° ∆G° Reaction (kJ⁄mol) (J⁄mol. K) (kJ⁄mol)
R1 CO + 3H2 → CH4 + H2O -206.1 [28] -214.6 -142.1
R2 CO2 + 4H2 → CH4 + 2H2O -165.0 [28] -172.6 -113.6
R3 2CO + 2H2 → CH4 + CO2 -247.3 [28] -256.6 -170.8
R4 2CO → C + CO2 -172.4 [28] -175.7 -120.0
R5 CO + H2O → H2 + CO2 -41.2 [28] -42.0 -28.7
R6 CH4 → C + 2H2 74.8 [28] 80.9 50.7
R7 CO + H2 → C + H2O -131.3 [28] -133.7 -91.5
R8 CO2 + 2H2 → C + 2H2O -90.1 [28] -91.6 62.8
R9 CO + 2H2 → (−CH2 −) + CO2 -152.0 [39] -76.2 -129.3
R10 C2H4 + H2 → C2H6 -136.0 [33] -120.5 -100.1
R11 C2H6 + H2 → 2CH4 -65.2 [33] 12.3 -68.9
R12 C3H6 + H2 → C3H8 -125.1 [33] -129.8 -86.4
R13 C3H8 + 2H2 → 3CH4 -119.1 [33] 29.6 -127.9
R14 C3H8 + H2 → CH4 + C2H6 -53.9 [33] 17.3 -59.0
All reactions except methane cracking (R6) are exothermic. The suppression of exothermic reactions by increasing temperature is clear while lnK of methane cracking (endothermic) increases. It can be seen that due to the high K values of reactions consuming CO (e.g. reactions 1, 3, 4, 7 and 9), complete conversion of carbon monoxide in lower temperature ranges is more likely. On the other hand, reactions involving CO2 consumption (e.g. 2 and 8) have lower equilibrium constants than those of CO. Moreover, carbon dioxide can be produced by reactions 3 – 5 [28]. In case of carbon formation, it is seen that the equilibrium
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 119 Chapter 4
constant for Boudouard reaction (R4) is relatively higher than those of other reactions producing carbon (i.e. reactions 6, 7 and 8).
Figure 4.1. Calculated equilibrium constants for possible reactions involved in the hydrogenation of carbon oxides in presence of light hydrocarbons.
In complex systems involving many different known and/or unknown reactions, using Gibbs free energy minimization method is common to calculate the chemical equilibrium state properties [27]. The method is based on the product distribution in a system to obtain minimum free energy. Thus, transport, kinetic and individual reactions data is not considered in the Gibbs free energy minimization method [28]. Moreover, the method can be also used for condensed
[43] and non-reacting [44] species of the system.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 120 Chapter 4
In this study, based on the Gibbs free energy minimization technique, chemical equilibrium of the co-hydrogenation of CO and CO2 in presence of C1-C3 hydrocarbons were studied. The system was investigated under two groups of conditions, which formation of solid carbon was included or excluded from the product species. In both groups (with or without solid carbon), the equilibrium concentrations were calculated for different H2 content (in feed composition) and as a function of temperature.
Five different feed compositions were used to perform chemical equilibrium calculations. The feed compositions (Table 4.4) were adjusted based on the ratio of molar fraction of hydrogen (푦퐻) to sum of COx and hydrocarbons except CH4
(∑ 푦푟).
Table 4.4. Feed compositions used in thermodynamic analysis. Molar Fraction (%) STO H50 H100 H200 HRICH H2 9.33 14.00 18.67 37.33 86.00 CO 2.00 2.00 2.00 2.00 2.00 CO2 0.67 0.67 0.67 0.67 0.67 CH4 8.00 8.00 8.00 8.00 8.00 C2H4 0.33 0.33 0.33 0.33 0.33 C2H6 0.33 0.33 0.33 0.33 0.33 C3H6 0.03 0.03 0.03 0.03 0.03 C3H8 0.03 0.03 0.03 0.03 0.03 He 79.27 74.60 69.93 51.27 2.60 Total 100 100 100 100 100 푦 퐻⁄ 1.0 1.5 2.0 4.0 9.1 ∑ 푦푟
Figure 4.2 shows the equilibrium concentrations of CO, CO2, CH4 and C2-C3 for the calculations without solid carbon. It is seen that increasing temperature resulted in increasing levels of CO, CO2 and C2-C3 concentrations. In contrast, methane equilibrium concentration decreased by increasing temperature. This
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 121 Chapter 4
observation can be explained by exothermic/endothermic characteristics of the possible reactions during the process. In this case, all possible reactions which consume CO, CO2 and C2-C3 are exothermic and are favoured at lower reaction temperatures. Moreover, methane producing reactions such R1 – R3 are also exothermic and methane production is decreased by temperature. It is also found that increasing hydrogen content in feed composition resulted in equilibrium concentrations of COx and C2-C3 to values equal or near zero, especially at lower temperature. In contrast, the temperature for decreasing methane concentration shifted to higher values by increasing H2 in feed composition.
Figure 4.2. Equilibrium concentrations of CO, CO2, CH4 and C2-C3 calculated without
solid carbon for different H2 content.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 122 Chapter 4
Figure 4.3 shows the equilibrium concentrations calculated based on considering solid carbon as a reaction product. It is found that the equilibrium concentration of all species (CO, CO2 and C2-C3) except methane increased with increasing temperature. Based on the reactions in Table 4.3, the only endothermic reaction is methane cracking (R6) which can be enhanced by increasing temperature to consume methane. Moreover, all other reactions (R1-R14 except R6) are exothermic, which are suppressed by temperature.
Figure 4.3. Equilibrium concentrations of CO, CO2, CH4 and C2-C3 calculated with solid
carbon for different H2 content.
The effect of feed H2 concentration is minor on the equilibrium CO concentration for feed compositions having 푦퐻⁄∑ 푦푟 ≤ 2. Similar to calculations without solid
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 123 Chapter 4
carbon, CO equilibrium concentration reached zero at lower temperatures compared to CO2. Methane equilibrium concentration increased by increasing H2 in feed composition at same temperature. On the other hand, increasing hydrogen content resulted in reducing the equilibrium concentration for COx and
C2-C3.
Figure 4.4. Produced solid carbon at chemical equilibrium state for different H2 content.
The amount of solid carbon produced under equilibrium conditions at different temperatures for various feed compositions is shown in Figure 4.4. It is observed that increasing temperature resulted in producing more solid carbon at equilibrium state. This can be explained by the endothermic methane cracking
(R6) or possibly other cracking of other hydrocarbons [45]. The addition of more hydrogen to the feed suppressed solid carbon formation. For feed compositions having 푦퐻⁄∑ 푦푟 > 2 no solid carbon was produced at equilibrium state.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 124 Chapter 4
4.5.2 Catalyst Characterization
According to IUPAC classification of adsorption/desorption isotherms, it is found that isotherm profiles for the catalyst and its support (Figure 4.5) are type IV. The hysteresis loop, exhibiting a type IV isotherm, is associated with capillary condensations in the mesopores [46]. The BET surface area, BJH adsorption average pore width and BJH adsorption cumulative volume of pores for the catalyst are smaller than that of its support (Table 4.5). It is reported that the addition of nickel to alumina decreases the specific surface area, pore size and pore volume of the support [10].
Figure 4.5. Nitrogen adsorption/desorption isotherm linear plot for the a) Ni/Al2O3 and
b) Al2O3.
Table 4.5. N2 adsorption/desorption results for the catalyst and its support. Catalyst Support BET surface area (m2/gr) 48.4 53.9 BJH adsorption average pore width (Å) 113 126 BJH adsorption cumulative volume of pores (cm3/gr) 0.155 0.185
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 125 Chapter 4
The XRD patterns for the 10 wt% Ni/Al2O3 catalyst and the support are shown in the Figure 4.6. The catalyst pattern showed the presence of NiO in the unreduced fresh catalyst. In this case, reflections at 37.1°, 43.3°, 62.9°, 75.3° and 79.3° have been identified. The first four reflections correspond to NiO 111, 200, 220 and
311 planes respectively [47].
Figure 4.6. XRD patterns Ni/Al2O3 and Al2O3 (Blue arrows: Al2O3, green arrows: NiO).
4.5.3 Catalyst activity assessment
4.5.3.1 Gas phase reactions/reactions on internal surface of the reactor
(Set 1)
It is important to rule out the occurrence of homogeneous gas phase reactions or reactions on the reactor internal surface in order to understand the role of the catalyst. In a temperature range from 50 °C to 500 °C, blank experiments were
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 126 Chapter 4
carried out for a feed stream containing all C1-C3 hydrocarbons, carbon oxides and hydrogen. The concentration of the most of the components were almost constant in this temperature range. However, the ethane concentration increased slightly at higher temperatures while the corresponding ethylene concentration decreased (Figure 4.7). It seems that ethylene hydrogenation in either the gas phase or over the reactor’s internal surface is taking place, to some limited extent, under these conditions. One of the reasons for this observation could be catalytic activity of the reactor’s internal surface, which is made of stainless steel. The maximum level of ethylene conversion was approximately 7% at 400 °C.
Figure 4.7. Concentration of ethane and ethylene versus catalyst bed temperature for set 1. The inlet values are showed as dashed lines with the same line colour for each component.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 127 Chapter 4
4.5.3.2 C2/C3 olefins and paraffins (Set 2)
The change in the concentration of C2/C3 alkenes and alkanes have been studied over Ni/Al2O3 in a hydrogen rich feed and a temperature range from 150 °C to
500 °C. No significant changes in the overall concentration of C2/C3 hydrocarbons were observed at temperatures below 260 °C. The concentration of ethylene and propylene dropped dramatically. Even at low temperatures (150 °C) almost all of the inlet olefins were consumed. Not surprisingly, under these conditions unsaturated C2 and C3 olefins are converted to saturated C2 and C3 paraffins due to catalytic hydrogenation [42, 48].
Figure 4.8. Concentration of C2H6 and C3H8 versus catalyst bed temperature for set 2. The inlet values are showed as dashed lines with the same line colour for each component.
The conversion of propane and ethane started at 260 °C and reached a maximum at approximately 350 °C and 400 °C respectively (Figure 4.8). Simultaneously to
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 128 Chapter 4
the consumption of C2H6 and C3H8, the concentration of methane increased in the outlet stream. This suggest that the conversion of alkanes at higher temperature range (above 260 °C) is a result of the catalytic hydrocracking of ethane and propane to methane, although carbon formation could also be occurring [41].
4.5.3.3 Carbon monoxide hydrogenation (Set 3)
Carbon monoxide hydrogenation on Ni/Al2O3 has been investigated. Methane was the major product during CO hydrogenation (Figure 4.9), although higher hydrocarbons, mainly ethane and propane, were also produced in a temperature range between 150 °C and 315 °C. The concentration of C2+ hydrocarbons peaked around 230 °C. Formation of both saturated and unsaturated light hydrocarbons has been reported, when conducted under specific reaction conditions, during the hydrogenation of carbon monoxide over alumina supported nickel catalysts [25, 49, 50].
Figure 4.9. Concentration of CH4, CO and C2-C3 versus catalyst bed temperature for set 3. The inlet values are showed as dashed lines with the same line colour for each component.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 129 Chapter 4
Almost all of the inlet carbon monoxide was converted at 270 °C. However, trace amount of carbon monoxide was detected in the outlet stream at 500 °C.
Detecting carbon monoxide in the product stream at high temperatures (500 °C) is possibly because that carbon monoxide hydrogenation is highly exothermic and increasing temperature may decrease CO consumption [49].
4.5.3.4 Effect of C2/C3 hydrocarbon addition on CO hydrogenation (Set 4)
Carbon monoxide hydrogenation over Ni/Al2O3 in the presence of C2/C3 hydrocarbons was studied (Figure 4.10).
Figure 4.10. Concentrations of CH4, CO and C2-C3 versus catalyst bed temperature for set 4. The inlet values are showed as dashed lines with the same line colour for each component.
It is found that the concentration of carbon monoxide, ethane and propane decreased to below detection limit in separate temperature ranges. In contrast, the concentration of methane increased, in two temperature windows. In the first
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 130 Chapter 4
window (between 150 °C and 255 °C), methane formation occurred coincident with the conditions where CO consumption was observed, and the second window (between 255 °C and 400 °C) was evident when the concentration of
C2/C3 hydrocarbons started to decrease. The maximum level of CO conversion
(almost 100 %) was at 255 °C, while the total concentration of C2/C3 hydrocarbons (based on number of carbon atoms) increased by 7% compared to the feed at 230 °C. The concentration of C2/C3 hydrocarbons then decreased from 230 °C and became lower than its inlet value at temperatures higher than
270 °C. C2/C3 hydrocarbons conversion reached 100% at 400 °C.
4.5.3.5 Carbon dioxide hydrogenation (Set 5)
CO2 hydrogenation over Ni/Al2O3 had been studied (Figure 4.11), where conversion started at 150 °C and CO2 concentration was below detection limits at temperatures above 400 °C.
Figure 4.11. Concentrations of CH4 and CO2 versus catalyst bed temperature for set 5. The inlet values are showed as dotted lines with the same line colour for each component.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 131 Chapter 4
The primary product detected was methane while carbon monoxide was not detected in the product stream and higher hydrocarbon formation was negligible.
This is consistent with the literature, where higher hydrocarbons were not observed during CO2 hydrogenation over nickel catalysts [51].
4.5.3.6 Effect of C2/C3 hydrocarbons addition on CO2 hydrogenation (Set 6)
Hydrogenation of CO2 over Ni/Al2O3 in presence of C2/C3 hydrocarbons has been investigated (Figure 4.12). Similar to CO2 hydrogenation in absence of C2/C3 hydrocarbons, conversion of CO2 started at 150 °C and reached 100% at 400 °C.
Light hydrocarbons consumption was observed over a similar temperature range to that observed in previous experiments.
Figure 4.12. Concentrations of CH4, CO2 and C in C2-C3 versus catalyst bed temperature for set 6. The inlet values are showed as dotted lines with the same line colour for each component.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 132 Chapter 4
The C2-C3 hydrocarbon concentration started to decrease at around 270 °C and reached 100 % at 400 °C. Comparing the results from the current set (CO2 hydrogenation in presence of C2-C3 hydrocarbons) with set 2 (C2-C3 hydrocarbons with hydrogen) and set 5 (hydrogenation of neat CO2), it can be confirmed that the reaction of both CO2 and light hydrocarbons with hydrogen did not have any notable influence on each other when the feed stream contains both
CO2 and C2-C3 hydrocarbons.
4.5.3.7 Carbon monoxide and carbon dioxide co-hydrogenation (Set 7)
To understand the efficacy of the catalyst for reducing carbon oxides, co- hydrogenation of CO and CO2 over Ni/Al2O3 has been undertaken (Figure 4.13).
The concentration of carbon monoxide decreased in a very similar trend compared to the hydrogenation of neat CO. In contrast, the conversion of CO2 started at a higher temperatures compared to the hydrogenation of neat CO2. The
CO2 concentration decreased at temperatures above 260 °C and reached 100 % conversion at 400 °C. The concentration of light hydrocarbons increased with decreasing CO concentration. Similar to sets 3 and 4, the C2-C3 hydrocarbons concentration peaked at approximately 230 °C. It has been reported that hydrogenation of carbon dioxide occurs at a higher rate and is more selective toward methane formation than that of hydrogenation of carbon monoxide over nickel catalysts [50, 52, 53]. On the other hand, it has been reported that the rate of carbon monoxide hydrogenation co-fed with carbon dioxide is higher than that of hydrogenation of neat CO [53]. It is suggested that CO hydrogenation inhibits
CO2 hydrogenation until all carbon monoxide is consumed during simultaneous hydrogenation of carbon oxides [51]. Although hydrogenation of carbon dioxide is thermodynamically more favourable, CO hydrogenation is preferential at low
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 133 Chapter 4
temperatures during simultaneous hydrogenation of CO and CO2 over Ni/Al2O3.
The reason for this is most likely related to the stronger adsorption of carbon monoxide with metal catalysts surfaces compared to carbon dioxide [53].
Figure 4.13. Concentrations of CH4, CO, CO2 and C2-C3 versus catalyst bed temperature for set 7. The inlet values are showed as dotted lines with the same line colour for each component.
4.5.3.8 Effect of C2/C3 hydrocarbons addition on CO and CO2
hydrogenation (Set 8)
Hydrogenation of carbon monoxide and carbon dioxide in presence of C2/C3 hydrocarbons over the Ni/Al2O3 catalyst has been studied (Figure 4.14). Carbon monoxide, carbon dioxide and C2/C3 hydrocarbons conversion was observed over distinct temperature ranges. Carbon oxides conversion started and reached their maximum (100 %) at a temperature range similar to that of CO and CO2 co-
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 134 Chapter 4
hydrogenation in the absence of hydrocarbons in feed stream. The concentration of C2/C3 hydrocarbons maximised at 230 °C, with approximately 8 % increase in the total C2/C3 concentration (based on number of carbon atoms) compared to their inlet concentration. Light hydrocarbons (C2+) were reduced to below their inlet concentration at temperatures above 270 °C. The concentration of light hydrocarbons was undetectable at higher temperatures (above 400 °C). Methane was the only hydrocarbon detected in the product stream.
Figure 4.14. Concentrations of CH4, CO, CO2 and C2-C3 versus catalyst bed temperature for set 8. The inlet values are showed as dotted lines with the same line colour for each component.
Similar to the observations from comparing set 3 (hydrogenation of just CO) and
4 (hydrogenation of CO in presence of C2-C3 hydrocarbons), it is found that the addition of light hydrocarbons slightly increased the maximum overall yield of production of light hydrocarbons. In this case, it is found that the maximum yield
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 135 Chapter 4
of production for light hydrocarbons increased from approximately 3 % for set 7 to roughly 6 % for set 8.
4.5.3.9 Effect of C1-C3 hydrocarbons on CO and CO2 hydrogenation (Set 9)
Co-hydrogenation of carbon monoxide and carbon dioxide in presence of C2/C3 hydrocarbons and methane had been investigated (Figure 4.15). Similar to all other previous sets of experiments in this study, the concentration of CO was reduced at temperatures above 150 °C, while the concentration of C2/C3 increased from 150 °C and reached a maximum at 230 °C. The concentration changes with changing reaction temperature for carbon dioxide was similar to that of set 8 (co-hydrogenation of CO and CO2 in the presence of C2-C3 hydrocarbons).
Figure 4.15. Concentrations of CH4, CO, CO2 and C2-C3 versus catalyst bed temperature for set 9. The inlet values are showed as dotted lines with the same line colour for each component.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 136 Chapter 4
Similar to the results from set 8, trace quantities of unsaturated C2 and C3 olefins were detected at low temperatures (below 210 °C) in the product stream. The concentration of C2/C3 hydrocarbons fell below its inlet value at temperatures above 275 °C until no C2-C3 hydrocarbon was detected (above 400 °C).
4.6 Conclusions
Simultaneous hydrogenation of carbon monoxide and carbon dioxide in the presence of light hydrocarbons such as methane, ethylene, ethane, propylene and propane has been investigated. The inlet species were consumed and converted in different temperature ranges. Olefins were converted under all reaction conditions at lower temperatures (below 150 °C) due to the hydrogenation reaction which resulted in the formation of saturated hydrocarbons. The concentration of carbon monoxide started to decrease at 150
°C and reached 100% conversion at around 250 °C. It was found that addition of
CO2 and C1-C3 hydrocarbons has little influence on the CO concentration trend.
The temperature for CO2 conversion under the conditions studied was influenced by the presence or absence of carbon monoxide in the feed stream. Decreasing the CO2 concentration was observed to take place at 150 °C when CO was not present in the feed stream, while the lowest temperature observed for CO2 conversion was increased by 90 °C following the addition of CO to the feed stream. The concentration of light hydrocarbons (C2+) maximised at approximately 230 °C for all sets containing CO in the feed stream. For the feed streams containing C2/C3 hydrocarbons, the concentration of C2/C3 hydrocarbons decreased below its inlet value at temperatures above 275 °C. Addition of either carbon monoxide or carbon dioxide to the inlet stream had negligible effect on
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 137 Chapter 4
the onset temperature for conversion or on the maximum conversion of C2/C3 hydrocarbons.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 138 Chapter 4
4.7 References
[1] R. Razzaq, C. Li, S. Zhang, Coke oven gas: Availability, properties, purification, and utilization in China, Fuel, 113 (2013) 287-299.
[2] B.L. Farrell, V.O. Igenegbai, S. Linic, A viewpoint on direct methane conversion to ethane and ethylene using oxidative coupling on solid catalysts,
ACS Publications, 2016.
[3] J.H. Lunsford, The Catalytic Oxidative Coupling of Methane, Angewandte
Chemie International Edition in English, 34 (1995) 970-980.
[4] J.H. Lunsford, Catalytic conversion of methane to more useful chemicals and fuels: A challenge for the 21st century, Catal. Today, 63 (2000) 165-174.
[5] P. Sabatier, J.B. Senderens, New methane synthesis, C. R. Acad. Sci. Paris,
134 (1902) 514-516.
[6] K.O. Xavier, R. Sreekala, K.K.A. Rashid, K.K.M. Yusuff, B. Sen, Doping effects of cerium oxide on Ni/Al2O3 catalysts for methanation, Catal. Today, 49 (1999)
17-21.
[7] E.D. Park, D. Lee, H.C. Lee, Recent progress in selective CO removal in a
H2-rich stream, Catal. Today, 139 (2009) 280-290.
[8] D. Tian, Z. Liu, D. Li, H. Shi, W. Pan, Y. Cheng, Bimetallic Ni-Fe total- methanation catalyst for the production of substitute natural gas under high pressure, Fuel, 104 (2013) 224-229.
[9] W. Wang, S. Wang, X. Ma, J. Gong, Recent advances in catalytic hydrogenation of carbon dioxide, Chem. Soc. Rev., 40 (2011) 3703-3727.
[10] D. Hu, J. Gao, Y. Ping, L. Jia, P. Gunawan, Z. Zhong, G. Xu, F. Gu, F. Su,
Enhanced investigation of CO methanation over Ni/Al 2O 3 catalysts for synthetic
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 139 Chapter 4
natural gas production, Industrial and Engineering Chemistry Research, 51
(2012) 4875-4886.
[11] D.C.D. Da Silva, S. Letichevsky, L.E.P. Borges, L.G. Appel, The Ni/ZrO 2 catalyst and the methanation of CO and CO 2, Int. J. Hydrogen Energ., 37 (2012)
8923-8928.
[12] A.M. Abdel-Mageed, S. Eckle, H.G. Anfang, R.J. Behm, Selective CO methanation in CO2-rich H2 atmospheres over a Ru/zeolite catalyst: The influence of catalyst calcination, J. Catal., 298 (2013) 148-160.
[13] S. Eckle, H.G. Anfang, R.J. Behm, Reaction intermediates and side products in the methanation of CO and CO2 over supported Ru catalysts in H2-rich reformate gases, J. Phys. Chem. C, 115 (2011) 1361-1367.
[14] A. Karelovic, P. Ruiz, Mechanistic study of low temperature CO2 methanation over Rh/TiO2 catalysts, J. Catal., 301 (2013) 141-153.
[15] J.N. Park, E.W. McFarland, A highly dispersed Pd-Mg/SiO2 catalyst active for methanation of CO2, J. Catal., 266 (2009) 92-97.
[16] K.P. Yu, W.Y. Yu, M.C. Kuo, Y.C. Liou, S.H. Chien, Pt/titania-nanotube: A potential catalyst for CO2 adsorption and hydrogenation, App. Catal. B, 84 (2008)
112-118.
[17] A.L. Kustov, A.M. Frey, K.E. Larsen, T. Johannessen, J.K. Nørskov, C.H.
Christensen, CO methanation over supported bimetallic Ni-Fe catalysts: From computational studies towards catalyst optimization, App. Catal. A, 320 (2007)
98-104.
[18] J.L. Falconer, A.E. Zaǧli, Adsorption and methanation of carbon dioxide on a nickel/silica catalyst, Journal of Catalysis, 62 (1980) 280-285.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 140 Chapter 4
[19] Q. Liu, X. Dong, X. Mo, W. Lin, Selective catalytic methanation of CO in hydrogen-rich gases over Ni/ZrO2 catalyst, J. Nat. Gas Chem., 17 (2008) 268-
272.
[20] S. Sharma, Z. Hu, P. Zhang, E.W. McFarland, H. Metiu, CO2 methanation on Ru-doped ceria, J. Catal., 278 (2011) 297-309.
[21] H. Song, J. Yang, J. Zhao, L. Chou, Methanation of carbon dioxide over a highly dispersed Ni/La2O3 catalyst, Chinese J. Catal., 31 (2010) 21-23.
[22] N. Takezawa, H. Terunuma, M. Shimokawabe, H. Kobayashib, Methanation of carbon dioxide: preparation of Ni/MgO catalysts and their performance, App.
Catal., 23 (1986) 291-298.
[23] F.W. Chang, M.T. Tsay, S.P. Liang, Hydrogenation of CO2 over nickel catalysts supported on rice husk ash prepared by ion exchange, App. Catal. A,
209 (2001) 217-227.
[24] G.A. Mills, F.W. Steffgen, Catalytic methanation, Catal. Rev., 8 (1973) 159-
210.
[25] Y.J. Huang, J.A. Schwarz, The effect of catalyst preparation on catalytic activity: I. The catalytic activity of Ni/Al2O3 catalysts prepared by wet impregnation, App. Catal., 30 (1987) 239-253.
[26] G. Garbarino, P. Riani, L. Magistri, G. Busca, A study of the methanation of carbon dioxide on Ni/Al2O 3 catalysts at atmospheric pressure, Int. J. Hydrogen
Energ., 39 (2014) 11557-11565.
[27] J.M. Smith, H.C. Van Ness, M.M. Abbott, Introduction to chemical engineering thermodynamics, McGraw-Hill2005.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 141 Chapter 4
[28] J. Gao, Y. Wang, Y. Ping, D. Hu, G. Xu, F. Gu, F. Su, A thermodynamic analysis of methanation reactions of carbon oxides for the production of synthetic natural gas, RSC Advances, 2 (2012) 2358-2368.
[29] N.A. Khan, Selective oxidation of methane with nitrous oxide over cobalt catalyst.
[30] M.J.H. van't Hoff, Etudes de dynamique chimique, Recueil des Travaux
Chimiques des Pays-Bas, 3 (1884) 333-336.
[31] COSILAB 4.0.4 - Rotexo GmbH & Co. KG.
[32] W.C. Reynolds, The Element Potential Method for Chemical Equilibrium
Analysis : Implementation in the Interactive Program STANJAN, Version 3,
Technical Rept., (1986).
[33] M.W.J. Chase, NIST-JANAF Thermochemical Tables, American Inst. of
Physics1998.
[34] J.A. Manion, Evaluated Enthalpies of Formation of the Stable Closed Shell
C1 and C2 Chlorinated Hydrocarbons, Journal of Physical and Chemical
Reference Data, 31 (2002) 123-172.
[35] Y. Takahashi, E.F. Westrum, Glassy Carbon low-temperature thermodynamic properties, The Journal of Chemical Thermodynamics, 2 (1970)
847-854.
[36] E.S. Domalski, E.D. Hearing, Estimation of the Thermodynamic Properties of Hydrocarbons at 298.15 K, Journal of Physical and Chemical Reference Data,
17 (1988) 1637-1678.
[37] S. Furuyama, D.M. Golden, S.W. Benson, Thermochemistry of the gas phase equilibria i-C3H7I C3H6 + HI, n-C3H7I i-C3H7I, and C3H6 + 2HI C3H8 + I2,
The Journal of Chemical Thermodynamics, 1 (1969) 363-375.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 142 Chapter 4
[38] D.A. Pittam, G. Pilcher, Measurements of heats of combustion by flame calorimetry. Part 8.-Methane, ethane, propane, n-butane and 2-methylpropane,
Journal of the Chemical Society, Faraday Transactions 1: Physical Chemistry in
Condensed Phases, 68 (1972) 2224-2229.
[39] A. Jess, C. Kern, Modeling of Multi-Tubular Reactors for Fischer-Tropsch
Synthesis, Chemical Engineering & Technology, 32 (2009) 1164-1175.
[40] A.O. Odunsi, T.S. O'Donovan, D.A. Reay, Temperature stabilisation in
Fischer–Tropsch reactors using phase change material (PCM), Applied Thermal
Engineering, 93 (2016) 1377-1393.
[41] J.R. Anderson, B.G. Baker, The Hydrocracking of Saturated Hydrocarbons
Over Evaporated Metal Films, Proceedings of the Royal Society of London.
Series A, Mathematical and Physical Sciences, 271 (1963) 402-423.
[42] A.S. Crampton, M.D. Rötzer, F.F. Schweinberger, B. Yoon, U. Landman, U.
Heiz, Ethylene hydrogenation on supported Ni, Pd and Pt nanoparticles: Catalyst activity, deactivation and the d-band model, Journal of Catalysis, 333 (2016) 51-
58.
[43] M.K. Nikoo, N.A.S. Amin, Thermodynamic analysis of carbon dioxide reforming of methane in view of solid carbon formation, Fuel Processing
Technology, 92 (2011) 678-691.
[44] G.A. Nahar, S.S. Madhani, Thermodynamics of hydrogen production by the steam reforming of butanol: Analysis of inorganic gases and light hydrocarbons,
International Journal of Hydrogen Energy, 35 (2010) 98-109.
[45] J. Kopyscinski, T.J. Schildhauer, S.M.A. Biollaz, Production of synthetic natural gas (SNG) from coal and dry biomass – A technology review from 1950 to 2009, Fuel, 89 (2010) 1763-1783.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 143 Chapter 4
[46] N. Srisawad, W. Chaitree, O. Mekasuwandumrong, A. Shotipruk, B.
Jongsomjit, J. Panpranot, CO 2 hydrogenation over Co/Al 2O 3 catalysts prepared via a solid-state reaction of fine gibbsite and cobalt precursors, Reaction
Kinetics, Mechanisms and Catalysis, 107 (2012) 179-188.
[47] F. Ocampo, B. Louis, A.C. Roger, Methanation of carbon dioxide over nickel- based Ce0.72Zr0.28O2 mixed oxide catalysts prepared by sol-gel method, App.
Catal. A, 369 (2009) 90-96.
[48] J. Horiuti, K. Miyahara, Hydrogenation of Ethylene on Metallic Catalysts,
1967.
[49] J. Gao, C. Jia, J. Li, M. Zhang, F. Gu, G. Xu, Z. Zhong, F. Su, Ni/Al2O3 catalysts for CO methanation: Effect of Al2O3 supports calcined at different temperatures, J. Energ. Chem., 22 (2013) 919-927.
[50] S.i. Fujita, M. Nakamura, T. Doi, N. Takezawa, Mechanisms of methanation of carbon dioxide and carbon monoxide over nickel/alumina catalysts, App. Catal.
A, 104 (1993) 87-100.
[51] D.E. Peebles, D.W. Goodman, J.M. White, Methanation of carbon dioxide on
Ni(100) and the effects of surface modifiers, J. Phys. Chem., 87 (1983) 4378-
4387.
[52] K. B. Kester, E. Zagli, J. L. Falconer, Methanation of carbon monoxide and carbon dioxide on Ni/Al2O3 catalysts: effects of nickel loading, App. Catal., 22
(1986) 311-319.
[53] H. Habazaki, M. Yamasaki, B.P. Zhang, A. Kawashima, S. Kohno, T. Takai,
K. Hashimoto, Co-methanation of carbon monoxide and carbon dioxide on supported nickel and cobalt catalysts prepared from amorphous alloys, App.
Catal. A, 172 (1998) 131-140.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 144 Chapter 4
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 145
Chapter 5
An experimental investigation on the effect of adding a transition metal to Ni/Al2O3 for the catalytic hydrogenation of CO and CO2 in presence of light alkanes and alkenes
Chapter 5
5.1 Abstract
The effect of transition metals on the activity and selectivity of a Ni/Al2O3 catalyst for carbon oxides hydrogenation was studied. It is found that each transition metal has different and distinct promoting or inhibiting influence. Nitric oxide was used as a probe molecule to study the electronic structure of a primary and two bi- metallic catalysts. Based on a series of catalyst activity and selectivity experiments, promoted bimetallic catalysts which enhanced and inhibited the activity and selectivity of the primary Ni/Al2O3 catalyst were analysed by NO-FTIR experiments. The presence of multiple sites, with different electronic properties, were observed. It is concluded that the addition of transition metals to Ni/Al2O3 markedly changed the electronic structure of the primary catalyst. Hydrogen and carbon monoxide chemisorption experiments showed that the ratio of CO/H2 chemisorbed species was affected by addition of Mn and Cu to Ni/Al2O3.
Temperature programmed desorption of H2 and CO confirmed the presence of different active sites on Ni/Al2O3.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 147 Chapter 5
5.2 Introduction
The worldwide drive for the development of alternative energy resources is increasing due to high rates of population and industrial growth. Carbon oxides,
(CO and CO2 or COx) are often generated as undesirable by-products of higher hydrocarbon synthesis, especially during the partial oxidation or oxidative coupling of methane [1]. A significant concentration of carbon oxides in a hydrocarbon stream can inhibit commercialization of synthetic hydrocarbon processes [2]. Extensive research has been conducted, mainly focussing on the design of processes which increase overall efficiency and in turn reduce carbon oxides emission. Since the hydrogenation of carbon oxides was reported by
Sabatier and Senderens at the beginning of the last century [3], many studies have resulted in improvements to the process [4]. Two of the most important and commercialized processes for COx conversion are Fischer-Tropsch Synthesis
(FTS) to produce long chain hydrocarbons and Methanation to produce synthetic natural gas.
Nickel is a highly active transition metal for CO and CO2 hydrogenation. However, nickel has not been commercialized for the FTS process due to its lighter product range being shifted towards methanation, and the formation of volatile carbonyls
[5]. Although nickel has not been considered as a suitable catalyst or promoter for FTS [4], there have been many research studies relevant to Ni and FTS [6].
On the other hand, numerous studies have also been undertaken, with focus on methanation of CO and CO2 over nickel catalysts to produce synthetic natural gas [7-9].
Nickel is highly active for complete conversion of carbon oxides (especially CO) and capable of producing higher hydrocarbons [2, 10-12]. This is due to the high
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 148 Chapter 5
rate of CO hydrogenation and the formation of hydrogenolysis products which are observed during COx hydrogenation reactions over nickel catalysts. The rate of formation of surface oxides, resulting in oxygenated species synthesis, is relatively low over nickel catalysts [13]. Moreover, the relatively low cost of nickel, compared to the most of metals that are active COx hydrogenation [14], suggests that nickel is a suitable choice for complete removal of COx and producing light hydrocarbons. It is hypothesised that the structural and electronic changes of the nickel catalyst have a direct influence on the performance of the catalyst, in terms of both activity and selectivity during carbon oxides hydrogenation [15-18]. For instance, it has been shown that bimetallic alloys of Ni and Co have distinctly different properties when compared to single metal Ni and Co catalysts for CO hydrogenation [15]. The addition of transition metals, which are suggested to modify the electronic properties of nickel catalysts for carbon oxides hydrogenation, has been studied [15, 19-21].
In this work, we present the performance of a series of bi-metallic catalysts for hydrogenation of CO and CO2 in presence of light hydrocarbons. The primary focus of this study is to explore the influence of adding a second transition metal to Ni/Al2O3 on the activity; and its influence on the selectivity towards enhancing the yield of light hydrocarbons. Enhancing the yield of production for light hydrocarbons is one of the key objectives of this thesis; thus the effects of different catalysts on this value is highlighted in this chapter.
5.3 Experimental
5.3.1 Catalyst testing
A series of Ni-M/Al2O3 catalysts (Ni(NO3)2.6H2O, Sigma-Aldrich 99.999%) were prepared using the incipient wetness method, where “M” designates an added
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 149 Chapter 5
transition metal. The following transition metals were used as the second metal:
Fe (Fe(NO3)3.9H2O, Sigma-Alrich 98%), Co (Co(NO3)2.6H2O, Sigma-Aldrich
98%), Cu (Cu(NO3)2.2.5H2O, Sigma-Aldrich 98%), Cr (Cr(NO3)3.9H2O, Sigma-
Aldrich 99%), Mn (Mn(NO3)2.4H2O, Sigma-Aldrich 97%), Zn
(Zn(CH3COO)2.2H2O, Chem-Supply 98%), Ru (RuCl3.H2O, Precious Metals
Online 99%), Rh (RhCl3.H2O, Precious Metals Online 99%), Ag (AgNO3, Ajax
Chemicals 99.9%) and Cd (Cd(NO3)2.4H2O, Aldrich 98%). Ground and calcined
(at 500 °C) alumina spheres (Sasol, Alumina Spheres 1.8/210) are used as the support. The loading of Ni and the second transition metal in the catalysts were fixed at 6 wt% for each metal, except for the single metal Ni/Al2O3 base catalyst in which the Ni loading is 12 wt%. For catalyst preparation, a solution of the metal salt precursors in distilled water was prepared for each catalyst. The solution was added dropwise to the support, with continuous mixing. The slurry was dried at
80 °C for 12 hr, followed by a second 12 hr drying period at 110 °C. After drying, the catalysts were calcined at 500 °C in static air. Prior to each run, the catalysts were reduced in hydrogen flow at 500 °C for 2 hr. Catalyst testing experiments were performed in a fixed bed tubular reactor in a temperature range from 150
°C to 500 °C. The feed stream composition is shown in Table 5.1.
Table 5.1. Feed stream composition. Partial pressure of feed species (KPa) H2 CO CO2 CH4 C2H6 C2H4 C3H8 C3H6 He 87.14 2.03 0.68 8.11 0.34 0.34 0.03 0.03 2.67
The feed stream contained a mixture of carbon oxides and light hydrocarbons.
With this feed gas mixture, the effect of the COx hydrogenation catalyst on the inlet hydrocarbons was determined.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 150 Chapter 5
The aim of chapter 4 was more of a feasibility study about possible advantages of using catalytic hydrogenation of COx in the presence of light hydrocarbons.
Therefore, step-wise study of the process was performed by using different feed streams containing different components. After finding that the process in advantageous, this chapter is mainly focused on the catalyst development.
Therefore, the main feed stream (containing all COx and light hydrocarbons) were used to analyse the performance of different catalysts.
5.3.2 Nitric oxide adsorption Fourier Transform Infra-Red Spectroscopy
(NO-FTIR)
Adsorbed NO spectra were recorded at 100 °C with a Bruker Tensor27 FTIR spectrometer. Wafers with approximate weight of 15 mg were prepared by using
13 mm dies. The pressed wafers were positioned into the sample holder and placed inside the in-situ FTIR cell equipped with KBr windows. Prior to nitric oxide adsorption, the wafers were heated to 500 °C at 5 °C/min, and remained at 500
°C for 30 min under vacuum conditions. Subsequently, the catalyst samples were reduced at 500 °C by injecting hydrogen into the cell. The reduction process consisted of 3 cycles of the following two steps: 10 mbar H2 for 10 min, then applying vacuum to the sample for 10 min. The samples were then allowed to cool to 100 °C for NO-adsorption. NO-adsorption was conducted at a pressure range between 1.0⨉10-4 mbar to 10 mbar and spectra were collected with 4 cm−1 resolution. For difference spectra, the spectrum of the clean, activated and reduced sample at 100 °C was subtracted. To confirm the absence of cell contamination, the spectrum of gas phase nitric oxide was also collected in the empty FTIR cell.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 151 Chapter 5
5.3.3 Temperature programmed desorption of H2 and CO
Catalysts samples were reduced in hydrogen at 500 °C for 2 hours prior to TPD analysis. The temperature programmed desorption experiments were conducted in ultra-high vacuum after the activation of the 100 mg of each sample at 500 °C for 1 hr. The desorbed species were analysed by a Pfeiffer Prisma quadrupole mass spectrometer. Carbon monoxide or hydrogen adsorbed at 35 °C on fresh samples. The TPD analysis was performed with heating the saturated samples up to 500 °C following evacuation of residual gas from the system.
5.3.4 Hydrogen and carbon monoxide chemisorption
Volumetric chemisorption analysis involved approximately 400 mg of catalyst sample which were calcined and reduced prior to each experiment. Prior to undertaking the chemisorption experiments, the apparatus and sample tube were evacuated using foreground and turbo-molecular pumps. The chemisorption of hydrogen and carbon monoxide were performed separately at 35 °C over a pressure range between 30 and 90 mbar. The total number of moles of gas adsorbed on the catalyst was calculated using the ideal gas law.
5.4 Results and Discussion
5.4.1 Catalyst activity
The temperature-controlled catalyst activity experiments were carried out over the temperature range from 150 °C to 500 °C, with reaction taking place at atmospheric pressure. Figure 5.1 and Figure 5.2 show the conversion of carbon oxide as a function of reaction temperature for each catalyst.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 152 Chapter 5
Figure 5.1. Carbon monoxide conversion versus temperature over different catalysts.
The onset of conversion of CO and CO2 was detected at different temperatures over different bi-metallic catalysts. The majority of CO and CO2 converted to methane, and the maximum conversion achieved for both CO and CO2 was affected following the addition of a second transition metal to nickel.
Figure 5.2. Carbon dioxide conversion versus temperature over different catalysts.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 153 Chapter 5
The results disclose that the selectivity of light hydrocarbons formation is strongly dependant on the bi-metallic catalyst examined. Some transition metals (such as
Mn, Fe and Rh) enhance the selectivity of C2-C4 production, as shown in Figure
5.3. On the other hand, other transition metals (such as Cu and Ag) when added to nickel decrease the activity for COx hydrogenation and do not improve C2-C4 production yield. The changes of catalyst activity and selectivity by adding a second metal have been attributed to a variety of reasons, most notably a change in the number of metallic sites available on the catalyst for reaction, the formation of a suitable (enhanced catalytic activity) alloy (i.e., formation of new sites) and an enhancement in the concentration of adsorbed carbonate species on the surface of the catalyst [21].
Figure 5.3. Maximum yield of C2-C4 hydrocarbon produced over different catalysts.
Figure 5.4 shows the changes in concentration of C2-C4 hydrocarbons as a function of temperature over different catalysts. It is clear that there is an optimal
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 154 Chapter 5
temperature window for each catalyst, whereby the concentration of C2-C4 increases to a value significantly above their inlet concentration. Increasing the catalyst bed temperature beyond this critical point appears to lead to the consumption of C2-C4 hydrocarbons, above which the concentration of these species decreases significantly. With the C2-C4 hydrocarbons concentration decreasing, there is a commensurate increase in the concentration of methane in the product stream. This suggests that at higher temperatures, the cracking of light hydrocarbons to methane is the dominant route. Most of the catalysts, most notably Co, Cr, Mn, Zn, Ru, and Rh, result in the complete removal of inlet C2-C4 hydrocarbons at temperatures below 500 °C.
Figure 5.4. Concentration changes of C2 – C4 hydrocarbons over temperature by
hydrogenation of CO and CO2 in presence of light hydrocarbons.
5.4.2 NO-FTIR
To gain insight into the surface properties of the primary and bi-metallic catalysts, nitric oxide was used as a probe molecule for infrared spectroscopy analysis. The
In-situ NO-FTIR technique is a widely used methodology to investigate the properties of active sites on the surface of heterogeneous catalysts [22].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 155 Chapter 5
Information about the electronic state and properties of the active metals can be elucidated from NO-FTIR analysis [4, 15]. Moreover, coordinated NO species bonded to active transition metal sites in a manner similar to that of CO [23]. In the current investigation, two bi-metallic catalysts were selected for study: Ni-
Cu/Al2O3 and Ni-Mn/Al2O3. Both of these catalysts displayed a markedly different influence on the performance of Ni/Al2O3. The addition of copper to the nickel catalyst significantly shifts the optimum temperature range for CO and CO2 hydrogenation to higher temperatures. In addition, the C2-C4 production yield was negligible over Ni-Cu/Al2O3 and decreased when compared to the activity of primary Ni/Al2O3 catalyst. On the other hand, adding managanese to Ni/Al2O3 did not change the CO and CO2 conversion temperature range. Moreover, the maximum production yield of C2-C4 almost doubled over Ni-Mn/Al2O3 compared to the selectivity of the primary Ni/Al2O3 catalyst.
The species detected following NO adsorption based on previous studies are commonly divided into two main groups, the species attributed to reactive and nonreactive NO adsorption. The nonreactive adsorbed species are surface monitrosyl and dinitrosyl complexes. Depending on the adsorption conditions
(especially the adsorption temperature and NO pressure), reactive NO adsorption can take place under conditions where NO acts as an oxidizing/reducing agent or disproportionates. The common species arising due to reactive NO adsorption
+ - are surface NO , NOx (x = 2, 3), N2O and NO2 [23].
Spectra collected during NO adsorption experiments for Ni/Al2O3, Ni-Cu/Al2O3 and Ni-Mn/Al2O3 catalysts are shown in Figure 5.5, Figure 5.6 and Figure 5.7.
The bands around 1850 cm-1 can be assigned to weakly adsorbed NO surface complexes (i.e., nonreactive adsorption species). These bands essentially
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 156 Chapter 5
disappeared following evacuation at the adsorption temperature. On the other hand, bands arising at wavenumbers around 1100cm-1, 1300cm-1 and 1600cm-1 can be assigned to the presence of reactive NO adsorption surface species. The intenstity of these bands increased considerably with increasing NO adsorption pressure. Moreover, a relatively high temperature of evacution was required to engender their desorption. These bands are most likely due to different types of
- -1 NO2 and NO2 surface species [23, 24]. The bands arising around 2200 cm can be assigned as adsorbed N2O [23].
Figure 5.5. IR spectra for NO adsorption (left) over Ni/Al2O3 and temprature program desorption (right).
Figure 5.6. IR spectra for NO adsorption (left) over Ni-Cu/Al2O3 and temprature program desorption (right).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 157 Chapter 5
Figure 5.7. IR spectra for NO adsorption (left) over Ni-Mn/Al2O3 and temprature program desorption (right).
In an effort to characterise the transition metal oxide sites, the focus of the current investigation is on the species arising due to nonreactive NO adorption, especially surface metal – mononitrosyls. The bonding orbitals of the gas phase nitric oxide molecule are ocupied by three electron pairs. The NO molecule has also one unpaired electron located on an antibonding orbital. Therefore, the nitrogen – oxygen bond order in NO molecule is 2.5 and its gas phase stretching vibration frequency is 1876 cm-1 [23]. The electron pair ocupying the weakly antibonding orbital (5σ) engenders the NO molecule as an electron donor or a weak Lewis base. Similar to carbon monoxide, NO coordination via a nitrogen atom to a Lewis acid is possible by a partial charge transfer from the 5σ orbital to metal site. On the other hand, π-back bonding, which decreases the N – O stretching mode frequency, can also be formed. The partial charge transfer is generally denoted as the coordination of NO molecule by NOδ+ or NOδ- [23].
The interaction between nitric oxide and all transition metals gives rise to the formation of metal – nitrosyl complexes, mononitrosyls and dinitrosyls. The metal
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 158 Chapter 5
– mononitrosyl complexes are formed by the coordination of NO molecules by
NOδ+ or NOδ-. In the case of NOδ+ coordination, NO acts as an electron donor to an electron accepting site, which results in a linear geometry of M – NO. For NOδ-
, NO acts as an electron acceptor from an electron donating site, which gives rise to a bent geometry of M – NO [25]. Evidence of nitrosyl complexes adsorption have been found in a wide spectral frequency range from 1700 cm-1 to 1970 cm-
1. It is expected that N – O stretching modes (generally linear mononitrosyl) have a frequency above gas phase NO species when only a σ bond is formed. In contrast, the frequency of N – O stretching modes (linear/bent mononitrosyl and dinitrosyl) can fall below 1876 cm-1 when π-back donation is also possible [23].
Using peak fitting software, four infrared bands are identified in the wavenumber
-1 -1 range between 1750 cm to 1910 cm for each Ni/Al2O3, Ni-Cu/Al2O3 and Ni-
-1 Mn/Al2O3 at NO pressure of 1.0⨉10 mbar (Figure 5.8, Figure 5.9 and Figure
5.10). By comparing the changes of these four bands at different NO coverage, the neareset two bands with wavenumbers below the gaseous NO frequency were assigned as metal – mononitrosyl species. This interpretation was made because of the low probabilty of dinitrosyl formation when compared to mononitrosyl at low NO coverage [26, 27]. Moreover, at high levels of NO coverage, the bands around 1800 cm-1 shift (slightly) to higher wavenumbers.
High coverage of nitric oxide can change the oxidation state of the metal and it is reported that metals in their highest oxidation state are more likely suitable to form just linear mononitrosyl species [28, 29].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 159 Chapter 5
-1 Figure 5.8. Identified IR bands around 1850 cm for Ni/Al2O3.
-1 Figure 5.9. Identified IR bands around 1850 cm for Ni-Cu/Al2O3.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 160 Chapter 5
-1 Figure 5.10. Identified IR bands around 1850 cm for Ni-Mn/Al2O3.
The promoting influence of the second transition metal on the nickel catalyst electronic properties is investigated by calculating the ratio of linear to bent metal–mononitrosyl bands for each Ni/Al2O3, Ni-Cu/Al2O3 and Ni-Mn/Al2O3 at
1.0x10-1 mbar of nitric oxide (Table 5.2).
Table 5.2. caclulated linear-/bent-type metal – mononitrosyl for Ni/Al2O3, Ni-Cu/Al2O3
and Ni-Mn/Al2O3 catalysts.
Ni Ni-Cu Ni-Mn Linear Bent Linear Bent Linear Bent Wavenumber 1860.2 1834.5 1850.0 1808.0 1865.4 1845.0 Area 0.47 0.06 3.71 0.15 0.23 0.34 Linear/Bent 7.70 24.89 0.67
As explained previously, metal–mononitrosyl complexes with linear geometry are formed via partial charge transfer from NO to an electron electron accepting site on the metal and the bent geometry is formed by charge transfer from the active
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 161 Chapter 5
site to an NO molecule. Thus, the absorbance of bent and linear metal – mononitrosyl groups can reflect the number of electron-donating and electron accepting sites of the catalyst [15].
According to the Table 5.2, the linear-/bent-type metal – mononitrosyl ratio changed markedly following the addition of Cu and Mn to Ni/Al2O3. Ni-Cu catalyst has a significantly higher linear/bent metal – mononitrosyl ratio compare to Ni catalyst. In contrast, the linear/bent ratio decreased following the addition of Mn to the nickel catalyst and its value is slightly below 1.0. This suggests that Mn addition is most likely balanced by the number of electron accepting and electron donating sites of the Ni/Al2O3 catalyst, while copper addition led to a significant increase of electron donating sites.
5.4.3 Temperature programmed desorption of H2 and CO
To investigate the different states of the active metal sites of the catalysts, temperature programmed desorption (TPD) experiments were performed. Two bimetallic catalysts, Ni-Cu/Al2O3 and Ni-Mn/Al2O3, together with the single metal
Ni/Al2O3 were selected for TPD experiments. Hydrogen, carbon monoxide and nitric oxide were used as the adsorbent gases and the TPD profiles were collected separately for H2, CO and NO desorption from the fresh samples. It is reported that over nickel-based catalysts, a quantity of H2 migrates to the subsurface layer or to the support, and therefore, the amount of desorbed hydrogen exceeds the monolayer chemisorbed hydrogen [30, 31]. In this case, low temperature H2-TPD is used to determine desorbed hydrogen from the monolayer hydrogen chemisorbed on metal particles. Based on peak fitting of the
TPD profiles (Figure 5.11a) for low temperature H2-TPD of Ni/Al2O3, Ni-Cu/Al2O3 and Ni-Mn/Al2O3 two peaks were identified for each sample. The first and second
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 162 Chapter 5
peaks are centred at approximately 100 °C and 170 °C respectively. As summarised in Table 5.3, it is found that the majority of chemisorbed hydrogen on Ni/Al2O3 and Ni-Mn/Al2O3 desorbed at the higher temperatures. In contrast,
H2 mostly desorbed at lower temperature from Ni-Cu/Al2O3.
Table 5.3. Calculated percentage of the area for each fitted peak on H2-TPD profiles of
Ni/Al2O3, Ni-Cu/Al2O3 and Ni-Mn/Al2O3 catalysts. Peak1 (%) Peak2 (%) Peak 2/Peak 1 Ni 17.34 82.66 4.77 Ni-Mn 24.00 76.00 3.17 Ni-Cu 54.05 45.95 0.85
From the CO-TPD profiles, two main peaks were observed for Ni-Cu/Al2O3 and three main peaks for Ni/Al2O3 (Figure 5.11b) and Ni-Mn/Al2O3. Carbon dioxide desorption was also monitored during the CO-TPD experiment. The number of fitted peaks for desorbed CO2 profiles are similar to the CO profiles. The mechanism of carbon monoxide desorption is complex and still not well understood in many cases [32]. With respect to CO desorption from the samples, a portion of adsorbed carbon monoxide is converted to CO2 which is most probably due to a combination of water gas shift and Boudouard reaction, dominated by the former mechanism [33, 34]. Based on peak fitting of CO desorption and CO2 emission profiles (Figure 5.11c for Ni/Al2O3), three peaks were identified for each profile.
As suggested by other researchers [33], it is possible to categorise the fitted peaks in three broad regions for both CO desorption (α, β1, β2) and CO2 emission
(α’ , β’1, β’2). It is reported that α region peaks are most probably due to desorption from a single site chemisorption while the peaks from β regions are from double
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 163 Chapter 5
site chemisorption sites[35]. The presence of a mixture of single and double sites on nickel catalysts was previously reported [33, 36].
Figure 5.11. Deconvolution of temperature programmed desorption profiles for Ni/Al2O3
– H2 (a), CO (b) and CO2 (c).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 164 Chapter 5
Table 5.4 and Table 5.5 disclose the peak area of each region for CO and CO2 desorption respectively. It is found that the addition of Cu and Mn to Ni/Al2O3 clearly affected the dominant adsorption sites. For both Ni/Al2O3 and Ni-Mn/Al2O3, the β region desorption is dominant. In contrast, the addition of Cu to nickel catalyst significantly increased the single site adsorption.
Table 5.4. Calculated percentage of the area for each fitted peak on CO-TPD profiles
of Ni/Al2O3, Ni-Cu/Al2O3 and Ni-Mn/Al2O3 catalysts.
α (%) β1 (%) β2 (%) β/α Ni 16.24 34.35 49.40 0.19 Ni-Mn 32.90 16.73 50.36 0.49 Ni-Cu 82.39 6.98 10.63 4.68
Table 5.5. Calculated percentage of the area for each fitted peak on CO2-TPD profiles
of Ni/Al2O3, Ni-Cu/Al2O3 and Ni-Mn/Al2O3 catalysts.
α' (%) β'1 (%) β’2 (%) β’/α’ Ni 13.19 47.19 39.63 0.15 Ni-Mn 34.68 27.02 38.30 0.53 Ni-Cu 79.74 10.31 9.95 3.94
It has been reported that strongly adsorbed CO molecules (β-CO) are precursors for surface carbide formation. In addition, the interaction between the partially hydrogenated carbon species (formed from surface carbides) and weakly chemisorbed CO (α-CO) results in the formation of higher hydrocarbons [37]. This argument can be used to explain the lower activity and selectivity of Ni-Cu/Al2O3 compared to Ni/Al2O3. In addition, the temperature programmed desorption results showed that by adding Mn to Ni/Al2O3, the ratio of α-CO/ β-CO adsorption sites increased slightly and resulted in higher selectivity toward production of light hydrocarbons.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 165 Chapter 5
5.4.4 Hydrogen and carbon monoxide chemisorption
In order to study the effect of copper and manganese addition on CO and H2 adsorption capacity of Ni/Al2O3, the amount of CO and H2 chemisorbed on different catalysts was measured (Table 5.6).
Table 5.6. Chemisorbed CO and H (μmol gas/g catalyst) over Ni/Al2O3, Ni-Cu/Al2O3
and Ni-Mn/Al2O3.
Ni Ni-Cu Ni-Mn H2 277.8 120.3 103.3 CO 296.6 441.3 122.1 CO/H 0.53 1.83 0.59
It is found that CO/H2 ratio increased almost four times by addition of Cu to Ni. In contrast, the addition of Mn to Ni slightly increased the CO/H2 ratio.
According to a widely accepted mechanism for carbon oxide hydrogenation, the so called carbide mechanism, the catalytic hydrogenation of carbon oxides
(especially CO) follows three main steps: chain initiation, chain propagation and finally desorption. The chain initiation step is basically cleavage of carbon – oxygen bonds and formation of surface active carbon species followed by hydrogenation of these species to different CHx species. The chain growth step has been debated extensively by many studies and several different mechanism are proposed [38]. For chain initiation steps, several studies have been focussed on the cleavage of carbon – oxygen bonds on different catalysts [39]. It is suggested that the presence of neighbouring sites with different electronic characteristics (e.g. electron-donating and electron-accepting sites) can enhance carbon-oxygen bond cleavage and hydrogenation rate [39-41]. Similarly, the
δ+ same surface sites structure may increase H2 adsorption and polarisation to H and Hδ- (Figure 5.12) [39].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 166 Chapter 5
Figure 5.12. Diagramatic representation of the interaction of CO and H2 with neighboring sites that have different electronic properties.
Copper and manganese addition to Ni/Al2O3 both resulted in changing the number of electron donating and electron accepting sites. Figure 5.13 shows a digramathic reperesentation of the effects of adding copper and manganese on the site structure of Ni/Al2O3.
Figure 5.13. Diagramathic representation of the effects of adding Cu and Mn on the site
structure of Ni/Al2O3.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 167 Chapter 5
Linear M – CO adsorption is the most preferred type for carbon monoxide interaction with catalyst sites [42]. We can interpret the electron accepting sites as carbon accepting sites and electron donating sites as oxygen accepting sites.
In this case, by addition of copper to Ni/Al2O3 the number carbon accepting sites increased extensively compared to oxygen accepting sites of the catalyst. Thus, carbon monoxide adsorbed primarily in a linear configuration and carbon – oxygen bond cleavage requires higher temperatures. On the other hand, following the addition of manganese to Ni/Al2O3, the ratio of carbon to oxygen accepting sites became closer to 1.0, wherby C – O bond cleavage becomes quite facile. Moreover, by having a balanced number of electron accepting and electron donating sites in vicinty of each other, polarization of H2 becomes more facile. Finally, by having more activated carbon and hydrogen species on the surface of the catalyst, the probability for interaction of these species increases and the rate of chain propagation step is enhanced (resulted in higher yield of
C2-C4 hydrocarbons).
5.5 Conclusion
Catalyst activity measurements for a series of nickel- based bi-metallic catalysts have shown that the addition of a transition metal to Ni/Al2O3 can influence the activity of the catalyst for hydrogenating carbon oxides and change the selectivity toward C2-C4 production. The characteristics of partially charged active sites of the catalysts were probed by infrared spectroscopy of NO adsorption. The nitric oxide FTIR experiments show that by adding a second transition metal to Ni/Al2O3 the electronic properties of the catalyst are altered. This is interpreted as a change in the number and ratio of active sites on the surface of the catalyst. The
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 168 Chapter 5
ratio of adsorbed linear and bent type NO were changed by adding a second transition metal to nickel. This is consistent with the electronic structure of the catalyst changing and is evidence of the presence of different adsorption sites.
Addition of copper resulted in increasing the ratio of C-accepting/ O-accepting sites, which is probably increased the chance of linear CO adsorption that needs higher temperature for C – O cleavage. By adding manganese to Ni/Al2O3 catalyst the ratio of electron accepting to donating sites balanced on the catalyst surface.
Thus, most probably the number active carbon and hydrogen species increased on the surface, which is resulted in higher yield toward production of C2-C4 hydrocarbons. From the results of CO-TPD and H2-TPD experiments, it is also found that the addition of manganese to nickel enhanced the selectivity toward light hydrocarbons formation by optimizing the ration of α- and β-CO adsorption sites. While copper addition dramatically increased the ratio of α-CO/ β-CO, resulted in low COx hydrogenation activity.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 169 Chapter 5
5.6 References
[1] R. Razzaq, C. Li, S. Zhang, Coke oven gas: Availability, properties, purification, and utilization in China, Fuel, 113 (2013) 287-299.
[2] J.H. Lunsford, Catalytic conversion of methane to more useful chemicals and fuels: A challenge for the 21st century, Catal. Today, 63 (2000) 165-174.
[3] P. Sabatier, J.B. Senderens, New methane synthesis, C. R. Acad. Sci. Paris,
134 (1902) 514-516.
[4] B.C. Enger, A. Holmen, Nickel and Fischer-Tropsch Synthesis, Catal. Rev.-
Sci. Eng., 54 (2012) 437-488.
[5] M.E. Dry, FT catalysts, Studies in Surface Science and Catalysis, 2004, pp.
533-600.
[6] E. Rytter, T.H. Skagseth, S. Eri, A.O. Sjåstad, Cobalt fischer-tropsch catalysts using nickel promoter as a rhenium substitute to suppress deactivation, Industrial and Engineering Chemistry Research, 49 (2010) 4140-4148.
[7] G.A. Mills, F.W. Steffgen, Catalytic methanation, Catal. Rev., 8 (1973) 159-
210.
[8] M.A. Vannice, The Catalytic Synthesis of Hydrocarbons from Carbon
Monoxide and Hydrogen, Catalysis Reviews, 14 (1976) 153-191.
[9] J. Kopyscinski, T.J. Schildhauer, S.M.A. Biollaz, Production of synthetic natural gas (SNG) from coal and dry biomass - A technology review from 1950 to
2009, Fuel, 89 (2010) 1763-1783.
[10] М.М. Zyryanova, P.V. Snytnikov, R.V. Gulyaev, Y.I. Amosov, A.I. Boronin,
V.A. Sobyanin, Performance of Ni/CeO2 catalysts for selective CO methanation in hydrogen-rich gas, Chemical Engineering Journal, 238 (2014) 189-197.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 170 Chapter 5
[11] S. Tada, R. Kikuchi, A. Takagaki, T. Sugawara, S.T. Oyama, K. Urasaki, S.
Satokawa, Study of RuNi/TiO2 catalysts for selective CO methanation, Applied
Catalysis B: Environmental, 140–141 (2013) 258-264.
[12] K.O. Xavier, R. Sreekala, K.K.A. Rashid, K.K.M. Yusuff, B. Sen, Doping effects of cerium oxide on Ni/Al2O3 catalysts for methanation, Catal. Today, 49
(1999) 17-21.
[13] G.A. Hadjigeorghiou, J.T. Richardson, Fischer-Tropsch selectivity of
Ni/Al2O3 catalysts, Applied Catalysis, 21 (1986) 11-35.
[14] R.A. Dagle, A. Karim, G. Li, Y. Su, D.L. King, Syngas Conditioning, Fuel
Cells: Technologies for Fuel Processing2011, pp. 361-408.
[15] T. Ishihara, N. Horiuchi, T. Inoue, K. Eguchi, Y. Takita, H. Arai, Effect of alloying on CO hydrogenation activity over SiO2-supported CO-Ni alloy catalysts,
Journal of Catalysis, 136 (1992) 232-241.
[16] M.A. Vannice, R.L. Garten, Metal-support effects on the activity and selectivity of Ni catalysts in CO H2 synthesis reactions, Journal of Catalysis, 56
(1979) 236-248.
[17] C.C. Kao, S.C. Tsai, Y.W. Chung, Surface electronic properties and CO hydrogenation activity of nickel deposited on rutile TiO2(100) as a model supported catalyst, Journal of Catalysis, 73 (1982) 136-146.
[18] C.H. Bartholomew, R.B. Pannell, J.L. Butler, D.G. Mustard, Nickel-support interactions: Their effects on particle morphology, adsorption, and activity selectivity properties, Industrial & Engineering Chemistry Product Research and
Development, 20 (1981) 296-300.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 171 Chapter 5
[19] T. Ishihara, K. Eguchi, H. Arai, Hydrogenation of carbon monoxide over SiO2- supported Fe-Co, Co-Ni and Ni-Fe bimetallic catalysts, App. Catal., 30 (1987)
225-238.
[20] H. Lu, X. Yang, G. Gao, J. Wang, C. Han, X. Liang, C. Li, Y. Li, W. Zhang,
X. Chen, Metal (Fe, Co, Ce or La) doped nickel catalyst supported on ZrO2 modified mesoporous clays for CO and CO2 methanation, Fuel, 183 (2016) 335-
344.
[21] D. Pandey, G. Deo, Promotional effects in alumina and silica supported bimetallic Ni–Fe catalysts during CO2 hydrogenation, J. Mol. Catal. A: Chem.,
382 (2014) 23-30.
[22] S.K. Bhargava, D.B. Akolekar, Adsorption of NO and CO over transition- metal-incorporated mesoporous catalytic materials, Journal of Colloid and
Interface Science, 281 (2005) 171-178.
[23] K.I. Hadjiivanov, Identification of Neutral and Charged NxOy Surface Species by IR Spectroscopy, Catalysis Reviews, 42 (2000) 71-144.
[24] T. Tanaka, T. Okuhara, M. Misono, Intermediacy of organic nitro and nitrite surface species in selective reduction of nitrogen monoxide by propene in the presence of excess oxygen over silica-supported platinum, Applied Catalysis B,
Environmental, 4 (1994) L1-L9.
[25] R.H. Crabtree, General Properties of Organometallic Complexes, The
Organometallic Chemistry of the Transition Metals, John Wiley & Sons, Inc.2005, pp. 29-52.
[26] P. Atanasova, A.L. Agudo, Infrared studies of nitric oxide adsorption on reduced and sulfided P-Ni-Mo/Al2O3 catalysts, Applied Catalysis B,
Environmental, 5 (1995) 329-341.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 172 Chapter 5
[27] J.B. Peri, Infrared studies of Ni held at low concentrations on alumina supports, Journal of Catalysis, 86 (1984) 84-94.
[28] G. Spoto, S. Bordiga, D. Scarano, A. Zecchina, Well defined CuI(NO),
- - CuI(NO)2 and CuII(NO)X (X = O and/or NO2 ) complexes in CuI-ZSMS prepared by interaction of H-ZSM5 with gaseous CuCl, Catalysis Letters, 13 (1992) 39-44.
[29] A.W. Aylor, S.C. Larsen, J.A. Reimer, A.T. Bell, An Infrared Study of NO
Decomposition over Cu-ZSM-5, Journal of Catalysis, 157 (1995) 592-602.
[30] M. Mihet, M.D. Lazar, Methanation of CO2 on Ni/γ-Al2O3: Influence of Pt,
Pd or Rh promotion, Catalysis Today, (2016).
[31] S. Velu, S.K. Gangwal, Synthesis of alumina supported nickel nanoparticle catalysts and evaluation of nickel metal dispersions by temperature programmed desorption, Solid State Ionics, 177 (2006) 803-811.
[32] B.M.W. Trapnell, D.O. Hayward, Chemisorption, Butterworths1964.
[33] A. Tanksale, J.N. Beltramini, J.A. Dumesic, G.Q. Lu, Effect of Pt and Pd promoter on Ni supported catalysts-A TPR/TPO/TPD and microcalorimetry study,
Journal of Catalysis, 258 (2008) 366-377.
[34] S.D. Jackson, B.M. Glanville, J. Willis, G.D. McLellan, G. Webb, R.B. Moyes,
S. Simpson, P.B. Wells, R. Whyman, Supported Metal Catalysts: Preparation,
Characterization, and Function. II. Carbon Monoxide and Dioxygen Adsorption on Platinum Catalysts, Journal of Catalysis, 139 (1993) 207-220.
[35] G. Ehrlich, Adsorption and electrical conduction in thin films, The Journal of
Chemical Physics, 35 (1961) 2165-2167.
[36] R.P. Eischens, W.A. Pliskin, The Infrared Spectra of Adsorbed Molecules,
Advances in Catalysis, 1958, pp. 1-56.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 173 Chapter 5
[37] R.D. Kelley, S. Semancik, On the mechanism of fischer-tropsch synthesis on a single crystal nickel catalyst, Journal of Catalysis, 84 (1983) 248-251.
[38] O.O. James, B. Chowdhury, M.A. Mesubi, S. Maity, Reflections on the chemistry of the Fischer-Tropsch synthesis, RSC Advances, 2 (2012) 7347-7366.
[39] P.M. Maitlis, V. Zanotti, The role of electrophilic species in the Fischer-
Tropsch reaction, Chemical Communications, (2009) 1619-1634.
[40] A.T. Bell, The influence of metal oxides on the activity and selectivity of transition metal catalysts, Journal of Molecular Catalysis. A, Chemical, 100
(1995) 1-11.
[41] M.A. Vannice, Hydrogenation of co and carbonyl functional groups, Catalysis
Today, 12 (1992) 255-267.
[42] J.M. Fukuto, S.J. Carrington, D.J. Tantillo, J.G. Harrison, L.J. Ignarro, B.A.
Freeman, A. Chen, D.A. Wink, Small Molecule Signaling Agents: The Integrated
Chemistry and Biochemistry of Nitrogen Oxides, Oxides of Carbon, Dioxygen,
Hydrogen Sulfide, and Their Derived Species, Chemical Research in Toxicology,
25 (2012) 769-793.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 174
Chapter 6
Effect of manganese on the selective catalytic hydrogenation of COx in the presence of light hydrocarbons over Ni/Al2O3: An experimental and computational study
Chapter 6
6.1 Abstract
The promoting effect of manganese on Ni/Al2O3 catalyst for the hydrogenation of carbon oxides, in presence of light hydrocarbons, was studied. Ni/Al2O3 displayed a high activity for the complete conversion of CO and CO2 to methane and C2+ hydrocarbons. Moreover, over a discrete and relatively narrow temperature range, the net concentration of light C2+ hydrocarbons was elevated, with the exit stream containing a higher concentration of C2+ species than was present in the feed stream and the product stream being virtually free of carbon oxides. It is found that the addition of manganese can enhance the selectivity toward the production of light hydrocarbons.
A series of Ni-Mn/Al2O3 catalysts, prepared with different Ni/Mn ratios, was studied. Various characterisation techniques such as XRD, CO and H2 chemisorption, in situ NO-FTIR and TPR were performed to gain insight into how the addition of Mn to the primary catalyst enhances the yield of light hydrocarbons. It is concluded that manganese influences the electronic structure of the catalyst. The electrostatic properties of crystalline nickel and nickel- manganese particles were studied by computational methods. In this case, the
KS-DFT calculation were performed at the PW91/pob-TZVP level of the theory.
The effects of Mn addition on the electronic properties of Ni/Al2O3 catalyst has been also observed by computational calculations.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 176 Chapter 6
6.2 Introduction
For decades, the catalytic hydrogenation of carbon oxides has been widely studied for a wide variety of applications. The worldwide demand for development of new energy resources has reanimated research into CO and CO2 hydrogenation [1]. Two of the most important and well-known processes for conversion of carbon oxides are Fischer-Tropsch Synthesis (FTS) [2] and carbon oxide methanation [3]. FTS is an exothermic polymerization reaction (Eq. 6.1) which produces a variety of hydrocarbons, most significantly parrafins, olefins and alcohols [4].
CO + 2H2 → −CH2 − + H2O; ∆H = −40 kcal/mol Eq. 6.1
The FTS process has received renewed interest for producing hydrocarbons from both CO and CO2 [5]. Carbon oxide methanation reactions (Eq. 6.2,3) has also been used for various applications, such as hydrogen purification for use in ammonia synthesis and fuel cells, producing Synthetic Natural Gas (SNG) and chemical storage of electricity [1, 6].
CO + 3H2 → CH4 + H2O; ∆H = −49 Kcal/mol Eq. 6.2
CO2 + 4H2 → CH4 + 2H2O; ∆H = −39 Kcal/mol Eq. 6.3
Significant quantities of carbon monoxide and carbon dioxide, present as undesirable by-products, can inhibit commercialization of different processes especially those involving the catalytic synthesis of hydrocarbons [7, 8]. For these
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 177 Chapter 6
applications, removing CO and CO2 from a stream containing hydrocarbons is essential. Under these circumstances, it is necessary that carbon oxide hydrogenation does not reduce the initial concentration of hydrocarbons in the feed stream.
The group VIII transition metals have been extensively studied for carbon oxide hydrogenation and are generally proven to be highly active catalysts. Among this group, nickel has been widely investigated for both FTS and methanation processes because of its high activity and relatively low cost. However, nickel has not been used as FTS catalyst because of its high selectivity toward methane formation [2, 9-11].
Improving the catalytic properties of nickel-based catalysts for COx hydrogenation has been the aim of much research in this area. In these studies, using a second metal as a promoter to form bi-metallic nano particles, has been established as a technique to enhance the catalytic performance of Ni catalysts. It is found that the structural and electronic properties of single metal particles are significantly changed following the addition of a second metal to form bi-metallic nano particles. These changes have been shown to engender a noticeable influence on the performance of the catalyst, in terms of both activity and selectivity [11-
17].
Manganese has been used to improve both nickel-based and other COx hydrogenation catalysts. Differing effects on the catalyst (following the addition of manganese) have been found. For example, the structural and electronic properties of Ni/Al2O3 was changed by adding Mn to the primary catalyst.
Moreover, it has been shown that by adding manganese to nickel catalysts, carbon deposition and the dispersion of the active sites are decreased and
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 178 Chapter 6
increased respectively. It is also reported that manganese addition can improve the reducibility by weakening the interaction between nickel and the support. [18-
22]. Based on these studies and the observations from chapter 5, manganese was decided as one of the best options to prepare high-performance nickel-based bi-metallic catalyst to overcome the key objectives of this thesis. Thus, an in- depth analysis of the effects of adding Mn to Ni/Al2O3 was performed.
In this work, we present an assessment of the catalytic performance of a series of Ni-Mn/Al2O3 catalysts for hydrogenation of CO and CO2 where the carbon oxides are present in a gas stream together with a significant concentration of light hydrocarbons. The primary focus of this study is to explore the influence of manganese addition to Ni/Al2O3 on the activity for complete conversion of COx and selectivity toward enhancing the yield of production for light hydrocarbons and catalyst’s properties.
6.3 Experimental methods
6.3.1 Catalyst testing
A series of Ni-Mn/Al2O3 catalysts were prepared by the incipient wetness method.
The total metal loading for each catalyst was adjusted to 12 wt%. Table 6.1 illustrates the labels and calculated metal contents for each sample.
Table 6.1. The labels and calculated metal contents of catalysts. Catalyst Label Nickel (wt%) Manganese (wt%) Ni/Mn Ni:Mn 12:0 12.0 0.0 NA Ni:Mn 10:2 10.0 2.0 5.0 Ni:Mn 8:4 8.0 4.0 2.0 Ni:Mn 6:6 6.0 6.0 1.0 Ni:Mn 4:8 4.0 8.0 0.5 Ni:Mn 2:10 2.0 10.0 0.2 Ni:Mn 0:12 0.0 12.0 0.0
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 179 Chapter 6
Hydrated metal nitrates for Ni (Ni(NO3)2.6H2O, Sigma-Aldrich 99.999%) and Mn
(Mn(NO3)2.4H2O, Sigma-Aldrich 97%) were used as precursors. Alumina spheres
(Sasol, Alumina Spheres 1.8/210) were ground and calcined at 500 °C and used as catalyst supports. To prepare each catalyst, a predetermined amount of metal precursors to achieve desired metal loading and Ni/Mn ratio was dissolved in distilled water. The solution then added drop-wise slowly to the support powder with continuous mixing. The slurry was dried in two steps: initially at 80 °C for 12 hr and at 110 °C for 12 hr. The dried slurry was then transferred to furnace for calcination in static air at 500 °C for. The catalysts were sieved and particles between 250 to 425 micrometres were collected. For each run, 250 mg of the sized catalyst particles with 200 mg of quartz sand were mixed (to avoid formation of any hotspots). A tubular fixed-bed reactor was used to perform the catalyst experiments. Prior to each run, the fresh catalyst was reduced in hydrogen flow at 500 °C or 2 hr in situ. Temperature programmed reaction (TPR) analysis performed at atmospheric pressure in a range between 150 °C and 500 °C.
Feed stream contained CO, CO2 and H2 with light alkanes and alkenes (CH4,
C2H4, C2H6, C3H6 and C3H8). Four different feed compositions were used to analyse the catalysts performance. The compositions of each feed were adjusted based on the ratio of H2 to all other reactive components i.e. CO, CO2, C2H4,
C2H6, C3H6 and C3H8. In this case, each feed labelled as the value of H2/Other ratio (Table 6.2).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 180 Chapter 6
Table 6.2. Composition of each feed stream Partial Pressure (kPa) H2/Other: 1/1 H2/Other: 1.5/1 H2/Other: 2/1 H2/Other: 4/1 H2 9.46 14.2 18.9 37.8 CO 2.03 2.03 2.03 2.03 CO2 0.68 0.68 0.68 0.68 CH4 8.11 8.11 8.11 8.11 C2H4 0.34 0.34 0.34 0.34 C2H6 0.34 0.34 0.34 0.34 C3H6 0.03 0.03 0.03 0.03 C3H8 0.03 0.03 0.03 0.03 He 80.3 75.6 70.9 52.0 Total 101 101 101 101
Conversion, selectivity and product yield are defined by the equations below (Eq.
6.4-6). The conversion was calculated for CO and CO2. The selectivity and yield were calculated for light hydrocarbons (C2-C4). It should be noted that the selectivity and yield were calculated based on the total mole number of carbon in
C2+ hydrocarbons.
nCO or CO2,in−nCO or CO2,out Conversion (%): XCO or CO2 = × 100 Eq. 6.4 nCO or CO2,in
nC2−C4,out − nC2−C4,in Selectivity (%): SC2−C4 = × 100 Eq. 6.5 nCO,in−nCO,out
n − n ( ) C2−C4,out C2−C4,in Yield % : YC2−C4 = × 100 Eq. 6.6 nCO,in
Experiments for CO and CO2 hydrogenation in the absence of hydrocarbons in the feed stream were performed. It is found that from CO2 hydrogenation in this condition the major product is methane and the concentration of higher
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 181 Chapter 6
hydrocarbons was negligible. Therefore, the selectivity and yield of C2-C4 hydrocarbons was calculated based on CO conversion.
The specific reaction rate and activation energy were determined for Ni-Mn/Al2O3 catalysts under differential reaction conditions (below 10 % conversion). In this case, the reaction of hydrogen with CO, CO2 and C2H6 has been analysed separately. The activation energies were calculated using Arrhenius equation
[23].
6.3.2 Inductively Coupled Plasma – Optical Emission Spectroscopy (ICP-
OES)
The elemental composition of the catalysts for Ni and Mn was determined by an
ICP-OES spectrometer spectroscopy (Varian Radial 715-ES). The sample preparation consisted of digestion and dilution steps. Microwave-assisted digestion of each sample obtained in a mixture of HNO3, HCl and HBF4 as solvent
[24]. The solutions then diluted to a suitable extent and analysed by the spectrometer.
6.3.3 X-ray Diffraction analysis (XRD)
A Philips X’Pert MPD with copper anode (Kα = 1.54060 Å at 40 kV and 40 mA) was used to obtain powder XRD patterns for crystal phase investigation.
Obtained data was analysed using X’Pert Highscore Plus [25] and Match [26] programs. The reference patterns were downloaded from Inorganic Crystal
Structure Database (ICSD) and Crystallography Open Database (COD) and compared to the experimental patterns for each sample.
6.3.4 Temperature Programmed Reduction (TPR)
TPR analysis was performed in a purpose-built apparatus with a mixture of H2 (2
%) and Ar (98 %). Prior to each run, fresh samples were thermally-treated at 400
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 182 Chapter 6
°C for 30 min. For the TPR experiments, a temperature ramp was 10 °C/min was imposed on the catalyst sample and flow rate for each sample/reference lines was fixed at 50 ml/min. The moles of hydrogen consumed was analysed with a thermal conductivity detector (TCD). A similar mass of sample (225 mg) were used for each analysis, so no normalisation was needed during post-run analysis.
6.3.5 Nitric oxide adsorption Fourier Transform Infra-Red Spectroscopy
(NO-FTIR)
Nitric oxide adsorption in-situ experiments were carried out in an ultra-high vacuum cell. Spectra during all key steps (activation/reduction, adsorption and desorption) were recorded using a Bruker Tensor27 FTIR spectrometer. Wafers
(≂15 mg) were pressed using 13 mm dies. The wafers were heated up to 500 °C at a rate of 5 °C/minute and maintained at that temperature for 30 minutes under vacuum. The sample was then reduced in situ at 500 °C by injecting 10 mbar H2 into the cell and remaining at these conditions for 10 minutes followed by evacuation of the cell (and catalyst) for 10 minutes. The reduction procedure consisted of three cycles of this H2 injection/evacuation procedure. Nitric oxide adsorption spectra were collected at 50 °C at pressures ranging from 1.0⨉10-4 mbar to 10 mbar. After the completion of adsorption step, samples were heated to 500 °C with 5 °C/min and spectra were collected to study desorption of NO from the samples. For adsorption/desorption difference spectra, the spectrum of a “clean”, activated and reduced catalyst sample at 50 °C was subtracted. The gas phase NO spectrum was also recorded in the empty cell to confirm the absence of gas phase contaminants.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 183 Chapter 6
6.3.6 H2 and CO chemisorption
A volumetric chemisorption apparatus was used to determine quantitatively the capacity of each catalyst for H2 and CO chemisorption. For each sample, 250 mg of catalyst was calcined and reduced (in situ) in the apparatus for chemisorption analysis. The apparatus and samples were outgassed at 1.0⨉10-4 mbar using foreground and turbo pumps prior to each experiment. Hydrogen and carbon monoxide chemisorption were performed separately at 40 °C, over a pressure range between 30 and 90 mbar. Based on the volume of chemisorbed H2 and
CO, the metal particle dispersion (D), specific surface area (S) and average particle size (d) were calculated using below equations (Eq. 6.7-9) [27].
Sf × n [mol]× Fw [g/mol] Dispersion: D (%) = gas × 100 Eq. 6.7 Wt% × WSample [g]
2 m2 Sf × ngas [mol] × σm [m ⁄atom] × NA [atom/mol] Specific Surface Area: S ( ⁄g) = Eq. 6.8 Wt% × WSample [g]
6000 Particle Size: d (nm) = 2 3 Eq. 6.9 S [m ⁄g] × ρm [g⁄cm ]
Where ngas is the number of moles of gas adsorbed on the metal particles, Fw is the formula weight of the particle, Sf is the stoichiometric factor, wt% is the weight percent of the metal particle in the sample, Wsample is the total weight of sample,
σm is the atomic cross sectional area of metal, NA is the Avogadro number and
ρm is the metal particle density.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 184 Chapter 6
6.4 Computational method: Solid-state electrostatic potential
The electrostatic potential V(r) has been widely used in the study of the electronic structure of different systems and/or materials. It is a 3-dimentional local property that has a particular value at any point in the space (except the position of nuclei).
In contrast to other descriptors of atomic electronic properties (e.g. atomic partial charge), the electrostatic potential V(r) can be calculated experimentally or computationally based on its physical definition (Eq. 6.10) [28].
′ ZA ρ(r )dr V(r) = ∑A − ∫ ′ Eq. 6.10 |RA− r| |r −r|
Where ZA is the charge of the nucleus of atom A which is located RA and ρ(r) is the electron density.
For intermolecular interactions, the electrostatic potential on surface Vs(r) is commonly computed on the ρ(r) = 0.001 au surface contour. The Vs(r) at ρ(r) =
0.001 au surface is a suitable value because it corresponds roughly to the van der Waals radii [29-31]. This definition of Vs(r) is useful to study long-range interactions such as non-covalent or the early stage covalent interactions [28].
Two mono-electronic properties, the electron density and the electrostatic potential, are needed to be evaluated for computing isodensity surfaces. These properties can be calculated based on the wavefunction of the system [32].
The ab initio CRYSTAL17 [33] code was used for generating wavefunction and computing the monoelectronic properties (the electron density and the electrostatic potential). The cubic particles of nickel and nickel-manganese crystals with two different sizes were studied. The particles with smaller size contained only one conventional unit cell of each crystal while the larger particles
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 185 Chapter 6
contained 64 unit cells. The crystallographic geometries were obtained from
ICSD for both nickel [34] and nickel-manganese [35].
As a well-known method for calculating the electronic structure of materials,
Kohn-Sham Density Functional Theory [36] was used. In this case, the surface electrostatic potential Vs(r) were obtained on the ρ(r) = 0.001 at the PW91/pob-
TZVP level of the theory [37, 38] The graphical representations were performed with the UCSF Chimera package [39].
6.5 Results and discussion
6.5.1 Catalyst testing
The temperature programmed catalyst experiments were conducted at atmospheric pressure over the temperature range between 150 °C to 500 °C.
Initially, Ni/Al2O3 catalysts were examined, with different feed compositions as outlined in Table 6.2. The primary reason for this aspect of the research was to study the effect of excess H2 on catalyst activity (in case of higher conversion of
CO and CO2 at lower temperatures) and the selectivity towards the production of light hydrocarbons (C2+). The change in the level of conversion with temperature for different feed compositions over Ni/Al2O3 catalyst are shown in Figure 6.1 and
Figure 6.2 for CO and CO2 respectively. The conversion of carbon monoxide and carbon dioxide was first detected at different temperatures for all experiments.
Carbon monoxide conversion commenced and reached its maximum conversion level over a temperature range which was significantly lower than that of CO2.
Others have also reported that during co-hydrogenation of CO and CO2 the conversion of CO commences at a lower temperature than CO2 hydrogenation
[40-42].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 186 Chapter 6
Figure 6.1. CO conversion for different feed compositions over Ni/Al2O3 (Ni:Mn 12:0) catalyst.
At higher temperatures, conversion of both CO and CO2 decreased with increasing temperature. One explanation for this observation is both CO and CO2 hydrogenation reactions are highly exothermic and increasing temperature results in reducing the net level of conversion [43]. It is clear from these data that increasing the amount of excess hydrogen enhances the activity of CO and CO2 hydrogenation. Increasing the H2 ratio from 1 to 4 resulted in decrease in temperature of approximately 40 °C for maximum CO and CO2 conversion.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 187 Chapter 6
Figure 6.2. CO2 conversion for different feed compositions over Ni/Al2O3 (Ni:Mn 12:0) catalyst.
The main product of CO and CO2 hydrogenation over nickel catalysts is methane.
However, the formation of light hydrocarbons also was observed, and the yield of these hydrocarbons, for each feed composition studied, is shown in Figure 6.3.
The formation of C2+ hydrocarbons commences at low temperatures coinciding with the temperature at which CO conversion is detected. The yield of C2+ hydrocarbons then reached a maximum for each run and subsequently decreased with increasing temperature. With continuing increase in reaction temperature, the light hydrocarbons in the feed stream were converted, mainly to methane. Additional hydrogen in the feed stream enhanced the C2+ production yield, and the feed with the highest concentration of hydrogen resulted in the highest yield of C2+ hydrocarbons. Moreover, the starting temperature for C2+
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 188 Chapter 6
consumption increased by approximately 10 °C for hydrogen ratio of 4:1 compared to 1:1.
Figure 6.3. Yield of production of C2-C4 hydrocarbons for different feed compositions
over Ni/Al2O3 (Ni:Mn 12:0) catalyst.
To analyse the performance of catalysts with different Ni/Mn ratio, experiments were performed with the feed composition with the highest activity or level of COx conversion and the highest yield of C2-C4 hydrocarbons. The results disclose that the starting temperature for both CO and CO2 conversion are dependent on the
Ni/Mn ratio of the catalysts examined (Figure 6.4 and Figure 6.5). For instance, the Ni/Mn ratios of 5 and 2 increased the activity for CO and CO2 conversion compared to the single metallic Ni/Al2O3 benchmark catalyst. By decreasing the
Ni/Mn ratio below 2, it is found that the activity was significantly reduced as the manganese content increased. The single metallic Mn/Al2O3 catalyst did not
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 189 Chapter 6
show any activity for CO conversion. It is found that above 300 °C, carbon dioxide conversion was detected and CO production was observed over the same temperature range. This is most probably due to the dry reforming of methane
(DRM) or reverse water gas shift (RWGS) reactions. The activity of manganese containing catalysts for both DRM and RWGS has been reported by other researchers [44, 45].
Figure 6.4. CO conversion for catalysts with varying Ni/Mn ratio in the feed stream with
H2/Other=4 feed composition.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 190 Chapter 6
Figure 6.5. CO2 conversion for catalysts with varying Ni/Mn ratios in the feed stream
with H2/Other=4 feed composition.
The yield of light hydrocarbons also changed and is dependent on the nickel and manganese concentration in the catalysts (Figure 6.6). Similar to CO and CO2 conversion, the catalysts with a Ni/Mn ratio of 5 and 2 showed elevated levels of
C2-C4 yield compared to Ni/Al2O3 catalysts. Conversely, the yield of C2-C4 decreased for catalysts with Ni/Mn ratio lower than 2.
From a catalyst performance perspective, the existence of a temperature window, in which CO and CO2 conversion is essentially complete (100%) and the C2-C4 hydrocarbons concentration is elevated when compared to that in the feed stream, is intriguing. By increasing the reaction temperature above that temperature window, light hydrocarbons are consumed. The increase of methane
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 191 Chapter 6
concentration at higher temperature suggests that cracking of light hydrocarbons to form methane is taking place.
Figure 6.6. Yield of production of C2-C4 hydrocarbons for catalysts with varying Ni/Mn
ratios in the feed stream with H2/Other=4 feed composition.
To further understand the influence of manganese addition, estimates for the activation energy for the reaction of CO, CO2 and C2H6 with hydrogen over
Ni/Al2O3 and Ni-Mn/Al2O3 (with Ni/Mn = 2 and 0.5) catalysts was obtained. In this study, the activation energies were calculated from the specific reaction rates in differential regime using the Arrhenius equation. The calculated values are summarised in Table 6.3.
For CO hydrogenation (Figure 6.7), the activation energies followed the order of
Ni:Mn 4:8 > Ni:Mn 12:0 > Ni:Mn 8:4. The calculated values of this study are in
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 192 Chapter 6
good agreement with the activation energies reported for CO hydrogenation over nickel-based catalysts (approximately between 80 and 120 kJ/mol) [46-50].
Table 6.3. Calculated activation energies for the reaction of CO, CO2 and C2H6 with H2 over Ni:Mn 12:0, Ni:Mn 8:4 and Ni:Mn 4:8 catalysts. Catalyst Reactant Activation Energy (kJ/mol) Ni:Mn 12:0 98.3 Ni:Mn 8:4 CO 93.3 Ni:Mn 4:8 113 Ni:Mn 12:0 80.3 Ni:Mn 8:4 CO2 75.7 Ni:Mn 4:8 99.4 Ni:Mn 12:0 151 Ni:Mn 8:4 C2H6 161 Ni:Mn 4:8 169
Figure 6.7. Arrhenius plots for CO hydrogenation over Ni/Al2O3 and Ni-Mn/Al2O3 (Ni/Mn = 2 and 0.5) catalysts.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 193 Chapter 6
It is found that by adding a moderate loading of manganese (Ni:Mn 8:4) to the
Ni/Al2O3 catalyst the activation barrier for CO hydrogenation slightly decreased.
However, adding more quantities of manganese (Ni:Mn 4:8) resulted in an increase in the activation energy.
The CO2 hydrogenation activation energy followed the same order as CO hydrogenation over different catalysts (Ni:Mn 4:8 > Ni:Mn 12:0 > Ni:Mn 8:4).
Similar to CO hydrogenation, the activation energy for CO2 hydrogenation (Figure
6.8) decreased by adding a moderate amount of manganese (Ni/Mn = 2) to
Ni/Al2O3 and increased by further increasing of the added manganese (Ni/Mn =
0.5). The values calculated in this study are also close to the reported activation energy of CO2 hydrogenation over nickel catalysts [47, 50, 51].
Figure 6.8. Arrhenius plots for CO2 hydrogenation over Ni/Al2O3 and Ni-Mn/Al2O3 (Ni/Mn = 2 and 0.5) catalysts.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 194 Chapter 6
Figure 6.9 shows the Arrhenius plots for the reaction between C2H6 and H2
(ethane hydrocracking). The calculated values for ethane hydrocracking in this study are of the same order of magnitude compared to the reported values in the literature [52-54]. It is found that the activation barrier for ethane hydrocracking increased by adding manganese to the primary Ni/Al2O3 catalyst. In this case, the activation energy for ethane hydrocracking can be arranged in the following order
Ni:Mn 4:8 > Ni:Mn 8:4 > Ni:Mn 12:0.
Figure 6.9. Arrhenius plots for CO2 hydrogenation over Ni/Al2O3 and Ni-Mn/Al2O3 (Ni/Mn = 2 and 0.5) catalysts.
Differences in the activity and selectivity for bi-metallic catalysts have been explained in a number of ways, such as changes in the number of active metal
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 195 Chapter 6
sites and the formation of catalytically active sites with new characteristics [55], which are discussed in the following sections.
6.5.2 Inductively Coupled Plasma – Optical Emission Spectroscopy (ICP-
OES)
Table 6.4 shows the elemental analysis (Ni and Mn) obtained for each catalyst by ICP-OES technique. The results are reasonably close to the amount of nickel and manganese initially impregnated. The ratio of Ni/Mn, is similar to the values invoked during catalyst preparation via incipient wetness.
Table 6.4. Metal contents of each catalyst based on ICP-OES analysis. Catalyst Nickel (wt%) Manganese (wt%) Ni/Mn Ni:Mn 12:0 11.6% 0.0% N/A Ni:Mn 10:2 9.5% 1.9% 5.0 Ni:Mn 8:4 7.7% 3.8% 2.0 Ni:Mn 6:6 5.7% 5.8% 1.0 Ni:Mn 4:8 3.8% 7.8% 0.5 Ni:Mn 2:10 2.3% 9.5% 0.2 Ni:Mn 0:12 0.0% 11.5% 0.0
6.5.3 X-ray Diffraction analysis (XRD)
The X-ray powder diffraction patterns of calcined samples were collected for crystal and structural analysis of the catalysts (Figure 6.10). The following reflections; 19°, 32°, 37°, 39°, 61°, 67° and 85° were detected on the alumina support [56]. It is found that, the single metal Ni/Al2O3 catalyst contains NiO and
NiAl2O4 particles, most notably the distinct NiO reflections were observed at 37°,
43°, 63°, 76° and 80° [57]. For NiAl2O4, reflections at 37°, 45°, 59° and 66° have been reported [58]. However, these reflections were masked by the reflections of other materials such as Al2O3, NiO and NiMnO3 which have higher intensities. It is clear that the intensity of the reflection at approximately 64° decreased as a
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 196 Chapter 6
result of a lower concentration of NiAl2O4 in samples that have lower nickel content (especially for single metal Mn/Al2O3).
Figure 6.10. XRD patterns for the alumina support and all catalysts. The symbols ◆,
●, ▲, ■ and ★ were used to mark the identified reflections for Al2O3, NiO, NiMnO3,
Mn2O3 and MnO2 phases respectively.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 197 Chapter 6
By increasing the manganese content in the samples, NiO reflections also became less intensive and new reflections were observed. For samples with
Ni/Mn ratios of 2 (Ni:Mn 8:4), 1 (Ni:Mn 6:6) and 0.5 (Ni:Mn 4:8), reflections corresponding to NiMnO3 phase emerged. The reflections attributed to NiMnO3 are at 24°, 34°, 37°, 42°, 50° and 56° [59]. The formation of NiMnO3 particles on bi-metallic Ni-Mn/Al2O3 catalyst has been reported previously [60]. Two oxide forms of manganese were also found for the samples with higher manganese contents (Ni:Mn 4:8, Ni:Mn 2:10 and Ni:Mn 0:12). The presence of Mn2O3 was identified with two low intensity reflections at 33° and 38° for the catalyst with a
Ni/Mn ratio of 0.5. For the other two samples with higher Mn content the reflections at 23°, 33°, 38°, 51° and 55° were identified as Mn2O3 [61].
Manganese (IV) oxide (MnO2) observed for the sample with Ni/Mn ratio of 0.2 and the single metal Mn/Al2O3. The reflections at 29°, 38°, 41°, 43°, 57°, 60°, 64°,
73° and 86° correspond to MnO2 particles on the samples [62].
6.5.4 Temperature Programmed Reduction (TPR)
The reducibility of metal sites for catalysts was investigated by temperature programmed reduction. Figure 6.11 presents the TPR profiles for single metal
Ni/Al2O3 and Mn/Al2O3, together with bimetallic Ni-Mn/Al2O3 samples having
Ni/Mn ratios of 2, 1 and 0.5. The consumption of hydrogen commenced at temperature above 300 °C for all the samples. In general, the reducible nickel species on alumina supported catalysts are classified into four types: α, β1, β2, and γ [63-65]. The α-type NiO (also known as free NiO weakly-bonded to alumina) species are usually reducible in low temperature regions. The mid-range or β-type NiO species interacts more strongly with the support and are reduced at a higher temperature range than α-type NiO. The β-type nickel oxide are
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 198 Chapter 6
classified into two subgroups; β1-type and β2-type. The β1-type and β2-type are known as Ni-rich and Al-rich mixed oxide phase respectively. In this case, the Ni- rich phase is more reducible than the Al-rich phase. The Ni containing species reduced at high temperature range is γ-type NiO. It is reported that the γ-type NiO is the stable nickel aluminate (NiAl2O4) [66, 67].
The TPR profile for single metal Mn/Al2O3 sample shows peaks in 300 °C to 500
°C range. It is reported that alumina-supported manganese oxides are usually reduced via a two-step reduction process (Eq. 6.11) [60, 68-70]:
MnO2 or Mn2O3 → Mn3O4 → MnO Eq. 6.11
For bimetallic Ni-Mn catalysts, the TPR profiles changed significantly with varying
Ni/Mn ratio. It is found that by decreasing the Ni/Mn ratio, detection of hydrogen consumption shifted to lower temperatures. There is no distinct peak attributable to manganese oxide reduction for all of the Ni-Mn/Al2O3 sample examined during the TPR analysis, which is constant with the XRD results of these sample. The
TPR profiles for bi-metallic Ni-Mn/Al2O3 catalysts showed that the concentration of Ni-containing species in β and γ region decreased following the addition of additional manganese to the catalysts. Moreover, a new peak emerged, located at more reducible α-type region in the TPR spectra. The presence of more reducible Ni containing species for Ni-Mn/Al2O3 samples can be attributed to the formation of NiMnO3 complex. These observations from TPR analysis of Ni-
Mn/Al2O3 catalysts are in good agreement with the results reported in literature
[20, 22, 60].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 199 Chapter 6
Figure 6.11. Temperature programmed reduction profiles or samples with different Ni and Mn contents.
6.5.5 Nitric oxide adsorption Fourier Transform Infra-Red Spectroscopy
(NO-FTIR)
To understand the catalytic properties of the samples with different nickel and manganese contents, samples were analysed by in-situ infrared spectroscopy
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 200 Chapter 6
using nitric oxide as a probe molecule (NO-FTIR). Characteristic information of the catalysts can be investigated by NO-FTIR technique [11, 12, 71]. The coordinated nitric oxide molecules, bonded to the catalyst’s active sites, have properties which are similar to carbon monoxide on the active sites [72]. This makes the NO-FTIR technique even more suitable for analysis of the catalysts for processes such as COx hydrogenation and involving CO activation. Nitric oxide FTIR analysis were performed on Ni/Al2O3, Mn/Al2O3 and bimetallic Ni-
Mn/Al2O3 with Ni/Mn ratio of 2, 1 and 0.5. All of the samples displayed different catalytic performance for hydrogenation of COx in presence of light hydrocarbons.
The results of the catalysts screening experiments showed that addition of manganese changed the activity of Ni/Al2O3 for conversion of CO and CO2.
Moreover, the yield of production for C2-C4 hydrocarbons was also varied by changing the Ni/Mn ratio.
Following the adsorption of NO, several peaks arose at different wavenumbers in the catalyst spectra. These surface species are generally divided into two main categories as reactive and nonreactive NO adsorption surface species. The reactive adsorption (depending on the adsorptive sample and adsorption conditions) occurs when nitric oxide acts as reducing/oxidizing agent or
+ - disproportionates. Compounds such as NO , NOx (x = 2, 3), N2O and NO2 can be formed during the reactive adsorption of NO. On the other hand, NO surface species such as mononitrosyl and dinitrosyl complexes are the products of nitric oxide nonreactive adsorption [72]. In the current investigation, the focus is on the nonreactive NO adsorption.
Figure 6.12, Figure 6.13, Figure 6.14, Figure 6.15 and Figure 6.16 demonstrate the NO-FTIR spectra for the samples studied. Sections ‘a’ and ‘b’ show the
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 201 Chapter 6
adsorption and desorption spectra respectively. For the samples that contained nonreactive NO adsorption, zoomed-in windows representing the region for mononitrosyl adsorption are highlighted. The peaks, present in the spectra at around 1850 cm-1 wavenumbers (below the gaseous NO frequency) can be assigned as mononitrosyls arising from the nonreactive adsorption of NO on the
Ni containing samples [73-76]. The disappearance of these peaks following the evacuation of the FTIR cell also confirms the assignment of the weakly bound peak at 1850 cm-1. It should be noted that, compared to mononitrosyls, the formation of dinitrosyls are less likely at low NO coverage [77, 78]. Other peaks appearing at wavenumbers around 1300cm-1 and 1600cm-1 can arise as a result of the reactive adsorption of NO molecules on the sample. The peaks are most likely due to the formation of charged and neutral NxOy species such as NO2 and
- NO2 surface compounds [72, 79]. These peaks were significantly increased in size by introducing more NO to the sample. In addition, applying a higher temperature was required to desorb these species. Peaks attributable to nonreactive NO adsorption were absent on single metal Mn/Al2O3 samples. It is found that at high levels of NO coverage, the bands assigned as metal– mononitrosyl species are shifted slightly to higher wavenumbers. The explanation for this slight shift is attributed to changes in the oxidation state of the metal to higher values. It is reported that on metals at their highest oxidation state formation of linear-type NO is more likely than the bent-type [80, 81].
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 202 Chapter 6
Figure 6.12. In-situ FTIR spectra for a. NO adsorption over single metal Ni/Al2O3 (Ni:Mn 12:0), followed by b. temperature programmed desorption.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 203 Chapter 6
Figure 6.13. In-situ FTIR spectra for a. NO adsorption over bi-metallic Ni-Mn/Al2O3 (Ni:Mn 8:4), followed by b. temperature programmed desorption.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 204 Chapter 6
Figure 6.14. In-situ FTIR spectra for a. NO adsorption over bi-metallic Ni-Mn/Al2O3 (Ni:Mn 6:6), followed by b. temperature programmed desorption.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 205 Chapter 6
Figure 6.15. In-situ FTIR spectra for a. NO adsorption over bi-metallic Ni-Mn/Al2O3 (Ni:Mn 4:8), followed by b. temperature programmed desorption.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 206 Chapter 6
Figure 6.16. In-situ FTIR spectra for a. NO adsorption over bi-metallic Ni-Mn/Al2O3 (Ni:Mn 0:12), followed by b. temperature programmed desorption.
The focus of the current study is on the species formed during the nonreactive adsorption of NO, especially metal – mononitrosyls. The nitric oxide molecule has three electron pairs and one unpaired electron in its bonding and antibonding orbitals respectively. Thus the bond order of N – O is 2.5 with a stretching vibration at a frequency of 1876 cm-1 in the gas phase [82]. Nitric oxide can act as an electron donor (weak Lewis base) and electron acceptor (weak Lewis acid).
Therefore, the coordination of NO molecule by partial charge transfer from weakly antibonding (5σ) orbital to a metal site via nitrogen atom forms is possible. In
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 207 Chapter 6
contrast, partial charge transfer from the metal site to nitric oxide molecule or π- back bonding may occur. These partial charge transfers are known as NO molecule coordination and are denoted as NOδ+ or NOδ-. For simplicity, the partial charge symbol (δ) has been removed in many publications [72].
The metal – mononitrosyl compounds is often formed during the adsorption of
NO on transition metals. The formation of these species is the result of a partial charge transfer between the active sites and the NO molecule which leads to the coordination of NO by NOδ+ or NOδ-. Nitric oxide, in form NOδ+ coordination, is basically an electron donor to an electron accepting site. In contrast, NO, in the form of NOδ- coordination, accepts an electron from an electron donator site. The linear and bent geometries of coordinated nitric oxide molecule are the results of
NOδ+ and NOδ- respectively [83].
To compare the ratio of Linear/Bent metal – mononitrosyl species on samples with different Ni and Mn contents, the peaks assigned as mononitrosyls were deconvoluted. In this case, adsorption spectra, collected at same NO pressure
(1.0⨉10-1 mbar), were chosen. Figure 6.17 displays the deconvoluted peaks for the nickel containing samples.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 208 Chapter 6
Figure 6.17. Peak deconvolution for metal – mononitrosyl species formed on nickel containing samples during in-situ NO-FTIR analysis.
Figure 6.18 shows the changes of Linear/Bent ratio of metal – mononitrosyl compounds on the samples. Comparing this ratio gives an insight about the influence of adding different levels of manganese to Ni/Al2O3 catalyst on the electronic properties of catalyst. In fact, the Linear/Bent ratio represents the ratio of electron accepting and electron donating sites on each catalyst [12]. According to Figure 6.18, decreasing the Ni/Mn ratio resulted in reducing the value of
Linear/Bent ratio. However, for samples with Ni/Mn ratio of 2 and 1, the number
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 209 Chapter 6
of sites reliable for formation of linear and bent metal – mononitrosyl compounds increased.
Figure 6.18. The calculated value of Linear/Bent ratio of metal – mononitrosyl compounds for catalysts with different Ni/Mn ratio.
6.5.6 H2 and CO chemisorption
In order to investigate the influence of differing nickel and manganese content on the catalyst properties, the quantity of chemisorbed gas, mole ratio of chemisorbed CO to Hydrogen, dispersion, specific surface area and particle size were estimated. It was found that when reducing the Ni/Mn ratio, the amount of chemisorbed gas (both CO and H2) decreased (Figure 6.19). The chemisorption of CO decreased significantly for catalysts with elevated manganese content. In this case, the amount of CO chemisorbed on single metal Mn/Al2O3 was negligible. On the other hand, H2 chemisorption occurred on all samples. The
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 210 Chapter 6
dispersion, specific surface area and particle size were calculated for all samples based on the H2 consumption (Table 6.5).
Table 6.5. Dispersion, specific surface area and particle size calculated for metal
particles based on H2 chemisorption. Catalyst Dispersion (%) Specific Area (m2/g) Particle Size (nm) Ni:Mn 12:0 23.1 165.3 4.1 Ni:Mn 10:2 18.1 134.3 5.2 Ni:Mn 8:4 15.2 117.0 6.1 Ni:Mn 6:6 10.6 85.3 8.6 Ni:Mn 4:8 6.2 51.6 14.7 Ni:Mn 2:10 3.5 30.2 25.8 Ni:Mn 0:12 2.2 20.0 40.5
The data presented in Table 6.5 suggests that when decreasing the Ni content in samples, the dispersion and specific area decreased. In contrast, the average particle size increased with decreasing the Ni/Mn ratio. Changing the metallic composition of samples showed more influence on the samples with Ni/Mn ratio below 1. For example, the average particle size increased about 4 nm for Ni/Mn=1 compared to single metal Ni/Al2O3 but the particle size increment is almost 36 nm for Ni/Mn=0. By comparing the XRD patterns and H2-Chemisorption data it was found that adding low or medium amounts of manganese (Ni/Mn<1) led to formation of moderately dispersed and relatively small NiO or NiMnO3 particles.
In contrast, a higher manganese content (Ni/Mn>1) resulted in a considerable reduction in the level of metal dispersion and an increase in particle size of the metal.
The chemisorption of carbon monoxide is considered to be negligible for single metal Mn/Al2O3 catalyst. In this case, it was assumed that CO adsorption occurs on Ni containing samples. Thus the data obtained from CO chemisorption was
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 211 Chapter 6
used to study the dispersion, specific surface area and particle size of nickel particles (Table 6.6). Similar to the values calculated from H2 chemisorption for overall metal particles, the dispersion and specific surface area decreased while the metal particle size increased with decreasing Ni/Mn ratio. However, it was found that by increasing the Mn content in the catalysts, the properties of the nickel particles narrowed to a smaller particle size range compared to those calculated for overall metal particles on each sample.
Table 6.6. Dispersion, specific surface area and particle size calculated for nickel particles based on CO chemisorption. Catalyst Dispersion (%) Specific Area (m2/g) Particle Size (nm) Ni:Mn 12:0 10.8 77.2 8.7 Ni:Mn 10:2 10.9 77.7 8.7 Ni:Mn 8:4 7.3 52.5 12.8 Ni:Mn 6:6 7.5 53.7 12.5 Ni:Mn 4:8 6.5 46.6 14.5 Ni:Mn 2:10 7.1 51.1 13.2 Ni:Mn 0:12 N/A N/A N/A
Catalysts’ activity and selectivity with different active metal (single metal and alloys) for COx hydrogenation has been investigated by CO and H2 chemisorption. It has been reported that the ratio of chemisorbed CO/H2 correlates with activity and selectivity toward C2+ hydrocarbons [12, 13, 84].
Figure 6.19 illustrates the changes in the ratio of chemisorbed CO/H2 over catalysts with different nickel and manganese contents. It was found that the addition of manganese changed the CO/H2 ratio. By increasing the Mn content
(Ni/Mn<5), the CO/H2 ratio reduced for most samples. The decrease in the CO/H2 ratio suggests that CO adsorption occurs more in the bridge type configuration in preference over the linear configuration. The correlation between a change in
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 212 Chapter 6
CO/H2 and C2+ selectivity due to intimate metal-metal interactions has been shown by other researchers [85]. The two samples with highest manganese contents (Ni/Mn=0.2 and 0) and with lowest activity for CO and CO2 hydrogenation also had the lowest CO/H2 ratio. It is not possible to clearly explain the changes in activity and selectivity by only studying the CO/H2 ratio. This is possibly due to the presence of different particle types (crystals) which was confirmed by both XRD and TPR patterns.
Figure 6.19. Chemisorbed carbon monoxide, hydrogen and CO/H ratio on each catalyst.
6.5.7 Solid-state electrostatic potential
The results of catalyst evaluation displayed improved activity and selectivity for
Ni-Mn/Al2O3 (Ni/Mn>1) bi-metallic catalysts compared to single metal nickel catalyst. The presence of a bi-metallic oxide for catalysts with Ni/Mn ratio more
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 213 Chapter 6
than one was found from XRD patterns. Furthermore, the TPR profile showed a single reduction peak at low temperature range compared to single metal
Mn/Al2O3 catalyst. The NO-FTIR investigation indicated the changes in the number of sites responsible for linear or bent type adsorption of nitric oxide by adding manganese to the Ni/Al2O3 catalyst. The changes observed by NO-FTIR analysis can be interpreted as the changes in the electronic structure of the catalyst [16, 18].
Solid state surface electrostatic potential Vs(r) were computed, based on KS-DFT
(Kohn-Sham Density Functional Theory) to investigate the electronic structure of the catalysts. Choosing an approximation of the universal exchange-correlation is critical to achieve high accuracy DFT calculations. In this case, many different functionals have been suggested [37]. The simplest functional is Local Density
Approximation [36]. However, Generalised Gradient Approximations (GGA) such as PW91 and PBE [86] are among the most widely used functionals. The PW91 functional has been used for metallic systems in other studies [87]. Furthermore, the modified basis-sets (pob-TZVP) has been shown to improve performance when compared to the standard basis-sets for solid state systems (especially metallic systems) [38].
The surface electrostatic potential Vs(r) at 0.001 au isodensity level were calculated for nickel and nickel-manganese particles containing only one conventional unit cell of each crystal (Figure 6.20).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 214 Chapter 6
Figure 6.20. Surface electrostatic potential Vs(r) on 0.001 au isodensity for nickel (left) and nickel-manganese (right) particles with the size of one crystallographic unit cell.
For a neutral atom, the value of V(r) is spherically symmetric and positive at all points with a local maxima at the position of nuclei. However, in systems containing more than one atom (molecule, crystals, and etc.), the electron density is redistributed toward the more electronegative atoms [28]. The redistribution of the electron density in multi-atomic systems leads to the formation of minima
(Vs,min) and maxima (Vs,max) of the surface electrostatic potential Vs(r). These surface minimum/maximum values are due to the presence of negative and positive areas or positive areas with different magnitude [88]. The Vs,min and Vs,max or areas with different electrostatic values on the surface of the catalyst can be interpreted as different catalytic sites. In this case, the sites with Vs,min values can act as electron donors (basic sites) and the ones with Vs,max can act as electron
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 215 Chapter 6
acceptors (acidic sites). These sites can attract other species to form different types of noncovalent bonding (e.g. hydrogen bonding, dihydrogen bonding, sigma-hole bonding, pi-hole bonding, halogen bonding and etc.) [89].
In molecular structures, the Vs,max regions which has a lower electron density are defined as sigma-holes. These regions are generated by the formation of σ molecular orbitals [90]. The σ-hole concept has been recently extended to the solid-state metallic systems. The formation of sigma-holes in gold and platinum nano particles has been explained by the overlapping of 6s1 valence orbitals to form sigma-orbitals [88]. In our investigation, explaining the Vs,max regions is much more complicated with catalysts containing nickel and manganese. This complexity is due to the different electron configuration of Ni ([Ar] 3d8 4s2) and
Mn ([Ar] 3d5 4s2) compared to Au ([Xe] 4f14 5d10 6s1) and Pt ([Xe] 4f14 5d9 6s1).
According to the metal-metal bonding orbital theory, the bond’s strength depends on different factors such as number of available electrons and the properties
(radial and angular) of the valence orbitals. The bonding orbitals for metal-based systems are known as σ, π and δ (for valance d and f orbitals). However, it is believed that δ-bonding is weaker compared to σ and π bonds, and thus the overall metal-metal bond is dominated by σ and π symmetries [91]. In this case, the regions of positive electrostatic potential for Ni and NiMn particles can be explained (at least to some extent) by the similar concepts for Au and Pt. One important observation is the obvious difference of the electrostatic potential maps for NiMn particle compared to single metal Ni particle.
The surface electrostatic potential Vs(r) at 0.001 au isodensity level were also computed for Ni and NiMn with a larger particle size containing 64 conventional unit cell of each crystal (Figure 6.21).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 216 Chapter 6
Figure 6.21. Surface electrostatic potential Vs(r) on 0.001 au isodensity for nickel (left) and nickel-manganese (right) particles with the size of 64 crystallographic unit cell.
The presence of Vs,max regions at the corners (or in general the charge distribution) of the larger clusters is probably due to the same reason as smaller clusters. Similarly, the colour-coded maps for larger Ni and NiMn particles are considerably different. This can be used as an explanation to confirm the changes in the electronic structure of the Ni/Al2O3 catalyst by the addition of manganese.
Different mechanisms have been proposed for carbon oxide hydrogenation reactions and a carbide-based mechanism is among the widely accepted ones.
According to the carbide mechanism, COx hydrogenation initiates via the dissociation of carbon – oxygen bonds. The cleavage of C – O bond generates active surface carbon species which can be followed by their hydrogenation
(known as chain growth step) to different CxHy species [92]. It has been reported
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 217 Chapter 6
that the electronic properties of catalysts may affect both chain initiation (C – O cleavage) and chain growth (CxHy formation) steps [18, 93-95]. The changes in the electronic structure of catalysts can also influence the polarised adsorption of hydrogen (Hδ+ and Hδ-) [28, 93]. Carbon monoxide, as a Lewis base, is known to have two electrostatically negative regions around carbon and oxygen atoms
(carbon is more negative). However, it is also suggested that the region around the carbon oxygen bound has a positive electrostatic potential value. Thus, the
CO molecule can react via side-on and end-on interactions with electrostatically negative and positive surfaces respectively [96]. This can explain the fact that CO preferably adsorbs as linear type configuration on metal sites from the C-end [97].
In our study, the catalysts with different Ni and Mn contents showed different CO and H2 adsorption capacities. From the CO and H2 chemisorption, it was found that the addition of manganese most likely leads to the adsorption of CO in bridge type configuration, which can be correlated to changes in the electronic structure.
The changes in CO/H2 chemisorption has been shown to affect the chain growth step during the COx hydrogenation. In addition, the NO-FTIR analysis showed the presence of sites with different electronic properties structure that are responsible for linear and bent adsorption configurations of NO. These sites with electron accepting (linear NO) and electron donating (bent NO) can be interpreted as carbon accepting and oxygen accepting sites [18, 72]. It is found that adding manganese to Ni/Al2O3 resulted in changing the ratio of Linear/Bent sites (i.e. C- accepting/O-accepting). These changes in the electronic structure of the catalyst were also confirmed by the surface electrostatic potential maps. Therefore, the enhancement of CO hydrogenation activity and C2+ hydrocarbons selectivity by adding manganese to Ni/Al2O3 can be explained by decreasing C-accepting/O-
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 218 Chapter 6
accepting ratio. In this case, the presence of more O-accepting sites in vicinity of
C-accepting sites can decrease the energy barrier for C – O bond cleavage and
CO hydrogenation.
For CO2 hydrogenation over nickel catalysts, one of the proposed mechanisms is hydrogenation via initial formation of surface CO species following by C – O bond dissociation and surface carbon hydrogenation [50]. It is also reported that modifying the electronic structure of nickel-based catalysts affects the CO2 hydrogenation activity. In this case, it is found that adding promoters that donate electron density (such as K) to the primary catalyst (i.e. forming more sites with negative electronic properties) enhances the COx hydrogenation activity, while adding promoters that decreases the electron density (such as S) decreased COx hydrogenation activity [98-101]. In this study, it is found that adding manganese changed the electronic structure of Ni/Al2O3, which is confirmed by some techniques such as in-situ NO-FTIR and electrostatic potential calculations. It is shown that adding moderate amounts of manganese (e.g. Ni/Mn = 2) enhanced the CO2 hydrogenation activity which can be explained by increasing the number of electron donating sites (i.e. oxygen accepting).
The addition of manganese also affected the hydrocracking activity (e.g. ethane hydrocracking). It is reported that hydrocracking activity over nickel catalysts depends on the C – C bond rupture [52, 54]. In this case, the C – C bond cleavage happens more readily on catalysts with stronger carbon-metal bonding or more carbon accepting sites [53]. In this paper, it is found that adding manganese to
Ni/Al2O3 decreased the ratio of C-accepting/O-accepting sites by modifying the electronic structure of the primary catalyst (which are confirmed by techniques such as in-situ NO-FTIR and surface electrostatic potential calculations). These
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 219 Chapter 6
changes resulted in the increasing of the energy barrier for C2H6 hydrocracking and C – C bond rupture. This discussion can be used for further explanation of the enhanced selectivity toward higher hydrocarbons production over some of the
Ni-Mn/Al2O3 (e.g. Ni/Mn = 2) catalysts compared to Ni/Al2O3 catalyst. In this case, the addition of manganese increased the O-accepting/C-accepting ratio which is resulted in higher activity for CO hydrogenation (more C – O bond cleavage) and formation of more surface CHx species. In addition, decreasing the C- accepting/O-accepting ratio inhibited the C – C bond rupture, and thus less C2+ hydrocarbons were cracked to form methane.
It should be noted that adding manganese above a certain amount (Ni/Mn = 1) can significantly decrease both CO/CO2 hydrogenation activity and C2+ hydrocarbons selectivity. In this case, comparison of XRD patterns of the catalysts suggests that the addition of manganese to the Ni/Al2O3 catalyst and adjusting the Ni/Mn ratio, a new form of bi-metallic oxide (NiMnO3) was formed.
The formation of manganese oxides enhanced by decreasing the Ni/Mn ratio below 1, which resulted in loss of activity for hydrogenation of CO and CO2.
Consequently, there is an optimum value for Ni/Mn ratio to achieve the optimum catalytic performance. Figure 6.22 shows a diagrammatic summary of the effects of adding manganese to Ni/Al2O3 on the hydrogenation of CO and CO2 in the presence of light hydrocarbons.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 220 Chapter 6
Figure 6.22. Diagrammatic representation of the effects of adding manganese to
Ni/Al2O3 on the hydrogenation of CO and CO2 in the presence of light hydrocarbons.
6.6 Conclusions
In summary, it is showed that the addition of manganese to Ni/Al2O3 catalyst significantly altered its catalytic performance for hydrogenation of CO and CO2 in the presence of light hydrocarbons. Adding manganese to Ni/Al2O3 enhanced the selectivity of C2-C4 production. However, there is an optimum amount of Mn added to the primary catalyst which enhanced the catalyst activity and selectivity.
The more hydrogen amount in the feed stream improved the catalyst activity for
COx hydrogenation and selectivity toward C2-C4 production. Increasing the catalyst bed temperature beyond this narrow temperature window leads to a dramatic decrease in the concentration of C2-C4 hydrocarbons, with a concomitant increase in methane concentration in the product stream. This
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 221 Chapter 6
suggests that possibly the cracking of higher hydrocarbons to methane is the dominant route at higher temperatures. It is found that the Mn addition changed the CO/H2 chemisorption ratio. According to investigation of the catalysts’ electronic properties with different Ni and Mn contents, changes in catalytic activity (for COx hydrogenation) and selectivity (for light hydrocarbons formation) can be interpreted as being due to the effect of different electronic structure of the catalysts with variety of Ni/Mn ratios.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 222 Chapter 6
6.7 References
[1] S. Rönsch, J. Schneider, S. Matthischke, M. Schlüter, M. Götz, J. Lefebvre,
P. Prabhakaran, S. Bajohr, Review on methanation – From fundamentals to current projects, Fuel, 166 (2016) 276-296.
[2] M.E. Dry, J.C. Hoogendoorn, Technology of the Fischer-Tropsch Process,
Catalysis Reviews, 23 (1981) 265-278.
[3] P. Sabatier, J.B. Senderens, New methane synthesis, C. R. Acad. Sci. Paris,
134 (1902) 514-516.
[4] A.K. Dalai, B.H. Davis, Fischer–Tropsch synthesis: A review of water effects on the performances of unsupported and supported Co catalysts, Applied
Catalysis A: General, 348 (2008) 1-15.
[5] B. Hu, S. Frueh, H.F. Garces, L. Zhang, M. Aindow, C. Brooks, E. Kreidler,
S.L. Suib, Selective hydrogenation of CO2 and CO to useful light olefins over octahedral molecular sieve manganese oxide supported iron catalysts, Applied
Catalysis B: Environmental, 132-133 (2013) 54-61.
[6] J. Gao, Q. Liu, F. Gu, B. Liu, Z. Zhong, F. Su, Recent advances in methanation catalysts for the production of synthetic natural gas, RSC Advances, 5 (2015)
22759-22776.
[7] R. Razzaq, C. Li, S. Zhang, Coke oven gas: Availability, properties, purification, and utilization in China, Fuel, 113 (2013) 287-299.
[8] J.H. Lunsford, Catalytic conversion of methane to more useful chemicals and fuels: A challenge for the 21st century, Catal. Today, 63 (2000) 165-174.
[9] M.E. Dry, FT catalysts, Studies in Surface Science and Catalysis, 2004, pp.
533-600.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 223 Chapter 6
[10] G.A. Mills, F.W. Steffgen, Catalytic methanation, Catal. Rev., 8 (1973) 159-
210.
[11] B.C. Enger, A. Holmen, Nickel and Fischer-Tropsch Synthesis, Catal. Rev.-
Sci. Eng., 54 (2012) 437-488.
[12] T. Ishihara, N. Horiuchi, T. Inoue, K. Eguchi, Y. Takita, H. Arai, Effect of alloying on CO hydrogenation activity over SiO2-supported CO-Ni alloy catalysts,
Journal of Catalysis, 136 (1992) 232-241.
[13] M.A. Vannice, R.L. Garten, Metal-support effects on the activity and selectivity of Ni catalysts in COH2 synthesis reactions, Journal of Catalysis, 56
(1979) 236-248.
[14] C.C. Kao, S.C. Tsai, Y.W. Chung, Surface electronic properties and CO hydrogenation activity of nickel deposited on rutile TiO2(100) as a model supported catalyst, Journal of Catalysis, 73 (1982) 136-146.
[15] C.H. Bartholomew, R.B. Pannell, J.L. Butler, D.G. Mustard, Nickel-support interactions: Their effects on particle morphology, adsorption, and activity selectivity properties, Industrial & Engineering Chemistry Product Research and
Development, 20 (1981) 296-300.
[16] T. Ishihara, K. Eguchi, H. Arai, Hydrogenation of carbon monoxide over SiO2- supported Fe-Co, Co-Ni and Ni-Fe bimetallic catalysts, App. Catal., 30 (1987)
225-238.
[17] G.L. Bezemer, P.B. Radstake, U. Falke, H. Oosterbeek, H.P.C.E. Kuipers,
A.J. van Dillen, K.P. de Jong, Investigation of promoter effects of manganese oxide on carbon nanofiber-supported cobalt catalysts for Fischer–Tropsch synthesis, Journal of Catalysis, 237 (2006) 152-161.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 224 Chapter 6
[18] V. Shadravan, E. Kennedy, M. Stockenhuber, An experimental investigation on the effects of adding a transition metal to Ni/Al2O3 for catalytic hydrogenation of CO and CO2 in presence of light alkanes and alkenes, Catalysis Today,
(2017).
[19] S.-H. Seok, S.H. Han, J.S. Lee, The role of MnO in Ni/MnO-Al2O3 catalysts for carbon dioxide reforming of methane, Applied Catalysis A: General, 215
(2001) 31-38.
[20] K. Zhao, Z. Li, L. Bian, CO2 methanation and co-methanation of CO and
CO2 over Mn-promoted Ni/Al2O3 catalysts, Frontiers of Chemical Science and
Engineering, 10 (2016) 273-280.
[21] N. Lohitharn, J.G. Goodwin, Impact of Cr, Mn and Zr addition on Fe Fischer–
Tropsch synthesis catalysis: Investigation at the active site level using SSITKA,
Journal of Catalysis, 257 (2008) 142-151.
[22] A. Zhao, W. Ying, H. Zhang, M. Hongfang, D. Fang, Ni/Al2O3 catalysts for syngas methanation: Effect of Mn promoter, Journal of Natural Gas Chemistry,
21 (2012) 170-177.
[23] S. Arrhenius, Über die Dissociationswärme und den Einfluss der Temperatur auf den Dissociationsgrad der Elektrolyte, Zeitschrift für Physikalische Chemie,
1889, pp. 96.
[24] M. Ghoorah, B.Z. Dlugogorski, H.C. Oskierski, E.M. Kennedy, Study of thermally conditioned and weak acid-treated serpentinites for mineralisation of carbon dioxide, Minerals Engineering, 59 (2014) 17-30.
[25] T. Degen, M. Sadki, E. Bron, U. König, G. Nénert, The HighScore suite,
Powder Diffraction, 29 (2014) S13-S18.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 225 Chapter 6
[26] Match! - Phase Identification from Powder Diffraction, Crystal Impact - Dr. H.
Putz & Dr. K. Brandenburg GbR, Kreuzherrenstr. 102, 53227 Bonn, Germany.
[27] S. Velu, S.K. Gangwal, Synthesis of alumina supported nickel nanoparticle catalysts and evaluation of nickel metal dispersions by temperature programmed desorption, Solid State Ionics, 177 (2006) 803-811.
[28] J.S. Murray, P. Politzer, The electrostatic potential: An overview, Wiley
Interdisciplinary Reviews: Computational Molecular Science, 1 (2011) 153-163.
[29] J.S. Murray, P. Politzer, Molecular surfaces, van der Waals Radii and electrostatic potentials in relation to noncovalent interactions, Croatica Chemica
Acta, 82 (2009) 267-275.
[30] F.A. Bulat, A. Toro-Labbé, T. Brinck, J.S. Murray, P. Politzer, Quantitative analysis of molecular surfaces: areas, volumes, electrostatic potentials and average local ionization energies, Journal of Molecular Modeling, 16 (2010)
1679-1691.
[31] T. Brinck, The use of the electrostatic potential for analysis and prediction of intermolecular interactions, in: C. Párkányi (Ed.) Theoretical and Computational
Chemistry, Elsevier1998, pp. 51-93.
[32] E. Jimenez-Izal, F. Chiatti, M. Corno, A. Rimola, P. Ugliengo, Glycine
Adsorption at Nonstoichiometric (010) Hydroxyapatite Surfaces: A B3LYP Study,
The Journal of Physical Chemistry C, 116 (2012) 14561-14567.
[33] R. Dovesi, R. Orlando, A. Erba, C.M. Zicovich-Wilson, B. Civalleri, S.
Casassa, L. Maschio, M. Ferrabone, M. De La Pierre, P. D'Arco, Y. Noël, M.
Causà, M. Rérat, B. Kirtman, CRYSTAL14: A program for the ab initio investigation of crystalline solids, International Journal of Quantum Chemistry,
114 (2014) 1287-1317.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 226 Chapter 6
[34] S.N. Achary, D. Errandonea, D. Santamaria-Perez, O. Gomis, S.J. Patwe,
F.J. Manjón, P.R. Hernandez, A. Muñoz, A.K. Tyagi, Experimental and
Theoretical Investigations on Structural and Vibrational Properties of Melilite-
Type Sr2ZnGe2O7 at High Pressure and Delineation of a High-Pressure
Monoclinic Phase, Inorganic Chemistry, 54 (2015) 6594-6605.
[35] V.E. Egorushkin, S.N. Kulkov, S.E. Kulkova, Electronic structure and the theory of phase transformations in NiMn, Physica B+C, 123 (1983) 61-68.
[36] P. Hohenberg, W. Kohn, Inhomogeneous electron gas, Physical Review, 136
(1964) B864-B871.
[37] A.E. Mattsson, R. Armiento, P.A. Schultz, T.R. Mattsson, Nonequivalence of the generalized gradient approximations PBE and PW91, Physical Review B -
Condensed Matter and Materials Physics, 73 (2006).
[38] M.F. Peintinger, D.V. Oliveira, T. Bredow, Consistent Gaussian basis sets of triple-zeta valence with polarization quality for solid-state calculations, Journal of
Computational Chemistry, 34 (2013) 451-459.
[39] E.F. Pettersen, T.D. Goddard, C.C. Huang, G.S. Couch, D.M. Greenblatt,
E.C. Meng, T.E. Ferrin, UCSF Chimera - A visualization system for exploratory research and analysis, Journal of Computational Chemistry, 25 (2004) 1605-
1612.
[40] H. Habazaki, M. Yamasaki, B.P. Zhang, A. Kawashima, S. Kohno, T. Takai,
K. Hashimoto, Co-methanation of carbon monoxide and carbon dioxide on supported nickel and cobalt catalysts prepared from amorphous alloys, App.
Catal. A, 172 (1998) 131-140.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 227 Chapter 6
[41] K. B. Kester, E. Zagli, J. L. Falconer, Methanation of carbon monoxide and carbon dioxide on Ni/Al2O3 catalysts: effects of nickel loading, App. Catal., 22
(1986) 311-319.
[42] S.i. Fujita, M. Nakamura, T. Doi, N. Takezawa, Mechanisms of methanation of carbon dioxide and carbon monoxide over nickel/alumina catalysts, App. Catal.
A, 104 (1993) 87-100.
[43] J. Gao, Y. Wang, Y. Ping, D. Hu, G. Xu, F. Gu, F. Su, A thermodynamic analysis of methanation reactions of carbon oxides for the production of synthetic natural gas, RSC Advances, 2 (2012) 2358-2368.
[44] P. Littlewood, X. Xie, M. Bernicke, A. Thomas, R. Schomäcker,
Ni0.05Mn0.95O catalysts for the dry reforming of methane, Catalysis Today, 242
(2015) 111-118.
[45] F.M. Gottschalk, G.J. Hutchings, Manganese oxide water—gas shift catalysts initial optimization studies, Applied Catalysis, 51 (1989) 127-139.
[46] Q. Liu, J. Gao, F. Gu, X. Lu, Y. Liu, H. Li, Z. Zhong, B. Liu, G. Xu, F. Su, One- pot synthesis of ordered mesoporous Ni–V–Al catalysts for CO methanation,
Journal of Catalysis, 326 (2015) 127-138.
[47] X. Lu, F. Gu, Q. Liu, J. Gao, Y. Liu, H. Li, L. Jia, G. Xu, Z. Zhong, F. Su, VOx promoted Ni catalysts supported on the modified bentonite for CO and CO2 methanation, Fuel Processing Technology, 135 (2015) 34-46.
[48] J. Sehested, S. Dahl, J. Jacobsen, J.R. Rostrup-Nielsen, Methanation of CO over nickel: Mechanism and kinetics at high H2/CO ratios, Journal of Physical
Chemistry B, 109 (2005) 2432-2438.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 228 Chapter 6
[49] Q. Liu, F. Gu, X. Lu, Y. Liu, H. Li, Z. Zhong, G. Xu, F. Su, Enhanced catalytic performances of Ni/Al2O3 catalyst via addition of V2O3 for CO methanation,
Applied Catalysis A: General, 488 (2014) 37-47.
[50] D.E. Peebles, D.W. Goodman, J.M. White, Methanation of carbon dioxide on
Ni(100) and the effects of surface modifiers, J. Phys. Chem., 87 (1983) 4378-
4387.
[51] R.A. Hubble, J.Y. Lim, J.S. Dennis, Kinetic studies of CO2 methanation over a Ni/[gamma]-Al2O3 catalyst, Faraday Discussions, 192 (2016) 529-544.
[52] J.R. Anderson, B.G. Baker, The Hydrocracking of Saturated Hydrocarbons
Over Evaporated Metal Films, Proceedings of the Royal Society of London.
Series A, Mathematical and Physical Sciences, 271 (1963) 402-423.
[53] A.V. Zeigarnik, R.E. Valdés-Pérez, O.N. Myatkovskaya, C−C Bond Scission in Ethane Hydrogenolysis, The Journal of Physical Chemistry B, 104 (2000)
10578-10587.
[54] D.W. Goodman, Ethane hydrogenolysis over single crystals of nickel: Direct detection of structure sensitivity, Surface Science, 123 (1982) L679-L685.
[55] D. Pandey, G. Deo, Promotional effects in alumina and silica supported bimetallic Ni–Fe catalysts during CO2 hydrogenation, J. Mol. Catal. A: Chem.,
382 (2014) 23-30.
[56] R.-S. Zhou, R.L. Snyder, Structures and transformation mechanisms of the
[eta], [gamma] and [theta] transition aluminas, Acta Crystallographica Section B,
47 (1991) 617-630.
[57] R.W. Cairns, E. Ott, X-Ray Studies of the System Nickel—Oxygen—Water.
I. Nickelous Oxide and Hydroxide1, Journal of the American Chemical Society,
55 (1933) 527-533.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 229 Chapter 6
[58] J.N. Roelofsen, R.C. Peterson, M. Raudsepp, Structural variation in nickel aluminate spinel (NiAl2O4), American Mineralogist, 77 (1992) 522-528.
[59] W.H. Cloud, Crystal structure and ferrimagnetism in NiMnO3 and CoMnO3,
Physical Review, 111 (1958) 1046-1049.
[60] L. Huang, F. Zhang, R. Chen, A.T. Hsu, Manganese-promoted nickel/alumina catalysts for hydrogen production via auto-thermal reforming of ethanol, International Journal of Hydrogen Energy, 37 (2012) 15908-15913.
[61] R. Norrestam, Alpha-manganese(III) oxide - a C-type sesquioxide of orthorhombicsymmetry, Acta Chemica Scandinavica, 21 (1967) 2871-2884.
[62] A.I.Z. Yu. D. Kondrashev, The structure of the modifications of manganese oxide, Izvestiya Akademii Nauk SSSR, Seriya Fizicheskaya, 15 (1951) 179-186.
[63] D. Hu, J. Gao, Y. Ping, L. Jia, P. Gunawan, Z. Zhong, G. Xu, F. Gu, F. Su,
Enhanced investigation of CO methanation over Ni/Al 2O 3 catalysts for synthetic natural gas production, Industrial and Engineering Chemistry Research, 51
(2012) 4875-4886.
[64] C. Guo, Y. Wu, H. Qin, J. Zhang, CO methanation over ZrO2/Al2O3 supported Ni catalysts: A comprehensive study, Fuel Process. Technol., 124
(2014) 61-69.
[65] J.M. Rynkowski, T. Paryjczak, M. Lenik, On the nature of oxidic nickel phases in NiO/γ-Al2O3 catalysts, Applied Catalysis A, General, 106 (1993) 73-82.
[66] J. Yang, X. Wang, L. Li, K. Shen, X. Lu, W. Ding, Catalytic conversion of tar from hot coke oven gas using 1-methylnaphthalene as a tar model compound,
Applied Catalysis B: Environmental, 96 (2010) 232-237.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 230 Chapter 6
[67] A. Zhao, W. Ying, H. Zhang, H. Ma, D. Fang, Ni–Al2O3 catalysts prepared by solution combustion method for syngas methanation, Catalysis
Communications, 17 (2012) 34-38.
[68] G.S. Pozan, Effect of support on the catalytic activity of manganese oxide catalyts for toluene combustion, Journal of Hazardous Materials, 221-222 (2012)
124-130.
[69] J.I. Gutiérrez-Ortiz, R. López-Fonseca, U. Aurrekoetxea, J.R. González-
Velasco, Low-temperature deep oxidation of dichloromethane and trichloroethylene by H-ZSM-5-supported manganese oxide catalysts, Journal of
Catalysis, 218 (2003) 148-154.
[70] Q. Tang, X. Huang, C. Wu, P. Zhao, Y. Chen, Y. Yang, Structure and catalytic properties of K-doped manganese oxide supported on alumina, Journal of
Molecular Catalysis A: Chemical, 306 (2009) 48-53.
[71] S.K. Bhargava, D.B. Akolekar, Adsorption of NO and CO over transition- metal-incorporated mesoporous catalytic materials, Journal of Colloid and
Interface Science, 281 (2005) 171-178.
[72] K.I. Hadjiivanov, Identification of Neutral and Charged NxOy Surface Species by IR Spectroscopy, Catalysis Reviews, 42 (2000) 71-144.
[73] P.G. Harrison, E.W. Thornton, Tin oxide surfaces. Part 9.-Infrared study of the adsorption of CO, NO and CO + NO mixtures on tin(IV) oxide gels containing ion-exchanged CrIII, MnII, FeIII, CoII, NiII and CuII, Journal of the Chemical
Society, Faraday Transactions 1: Physical Chemistry in Condensed Phases, 74
(1978) 2703-2713.
[74] E.E. Platero, B. Fubini, A. Zecchina, Nitric oxide adsorption on NiO: An infrared and calorimetric study, Surface Science, 179 (1987) 404-424.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 231 Chapter 6
[75] A. Zecchina, D. Scarano, S. Bordiga, G. Ricchiardi, G. Spoto, F. Geobaldo,
IR studies of CO and NO adsorbed on well characterized oxide single microcrystals, Catalysis Today, 27 (1996) 403-435.
[76] M. Mihaylov, K. Hadjiivanov, N. Abadjieva, D. Klissurski, L. Mintchev,
Characterization of Zirconia-supported Nickel Catalysts Prepared by Multiple Ion- exchange, in: B. Delmon, P.A. Jacobs, R. Maggi, J.A. Martens, P. Grange, G.
Poncelet (Eds.) Studies in Surface Science and Catalysis, Elsevier1998, pp. 295-
304.
[77] P. Atanasova, A.L. Agudo, Infrared studies of nitric oxide adsorption on reduced and sulfided P-Ni-Mo/Al2O3 catalysts, Applied Catalysis B,
Environmental, 5 (1995) 329-341.
[78] J.B. Peri, Infrared studies of Ni held at low concentrations on alumina supports, Journal of Catalysis, 86 (1984) 84-94.
[79] T. Tanaka, T. Okuhara, M. Misono, Intermediacy of organic nitro and nitrite surface species in selective reduction of nitrogen monoxide by propene in the presence of excess oxygen over silica-supported platinum, Applied Catalysis B,
Environmental, 4 (1994) L1-L9.
[80] G. Spoto, S. Bordiga, D. Scarano, A. Zecchina, Well defined CuI(NO),
- - CuI(NO)2 and CuII(NO)X (X = O and/or NO2 ) complexes in CuI-ZSMS prepared by interaction of H-ZSM5 with gaseous CuCl, Catalysis Letters, 13 (1992) 39-44.
[81] A.W. Aylor, S.C. Larsen, J.A. Reimer, A.T. Bell, An Infrared Study of NO
Decomposition over Cu-ZSM-5, Journal of Catalysis, 157 (1995) 592-602.
[82] K. Nakamoto, Theory of Normal Vibrations, Infrared and Raman Spectra of
Inorganic and Coordination Compounds, John Wiley & Sons, Inc.2008, pp. 1-147.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 232 Chapter 6
[83] R.H. Crabtree, General Properties of Organometallic Complexes, The
Organometallic Chemistry of the Transition Metals, John Wiley & Sons, Inc.2005, pp. 29-52.
[84] I. Rodriguez-Ramos, A. Guerrero-Ruiz, J.L.G. Fierro, Effect of oxide promoters on the surface characteristics of carbon-supported Co and Ru catalysts, Applied Surface Science, 40 (1989) 239-247.
[85] P.C. Das, N.C. Pradhan, A.K. Dalai, N.N. Bakhshi, Carbon monoxide hydrogenation over various titania-supported Ru–Ni bimetallic catalysts, Fuel
Processing Technology, 85 (2004) 1487-1501.
[86] J.P. Perdew, J.A. Chevary, S.H. Vosko, K.A. Jackson, M.R. Pederson, D.J.
Singh, C. Fiolhais, Atoms, molecules, solids, and surfaces: Applications of the generalized gradient approximation for exchange and correlation, Physical
Review B, 46 (1992) 6671-6687.
[87] M. Neef, K. Doll, CO adsorption on the Cu(111) surface: A density functional study, Surface Science, 600 (2006) 1085-1092.
[88] J.H. Stenlid, T. Brinck, Extending the σ-Hole Concept to Metals: An
Electrostatic Interpretation of the Effects of Nanostructure in Gold and Platinum
Catalysis, Journal of the American Chemical Society, 139 (2017) 11012-11015.
[89] H. Wang, W. Wang, W.J. Jin, σ-Hole Bond vs π-Hole Bond: A Comparison
Based on Halogen Bond, Chemical Reviews, 116 (2016) 5072-5104.
[90] J.S. Murray, P. Lane, P. Politzer, Expansion of the σ-hole concept, Journal of Molecular Modeling, 15 (2009) 723-729.
[91] J.E. McGrady, Introduction and General Survey of Metal-Metal Bonds,
Molecular Metal-Metal Bonds: Compounds, Synthesis, Properties, (2015) 1-22.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 233 Chapter 6
[92] O.O. James, B. Chowdhury, M.A. Mesubi, S. Maity, Reflections on the chemistry of the Fischer-Tropsch synthesis, RSC Advances, 2 (2012) 7347-7366.
[93] P.M. Maitlis, V. Zanotti, The role of electrophilic species in the Fischer-
Tropsch reaction, Chemical Communications, (2009) 1619-1634.
[94] A.T. Bell, The influence of metal oxides on the activity and selectivity of transition metal catalysts, Journal of Molecular Catalysis. A, Chemical, 100
(1995) 1-11.
[95] M.A. Vannice, Hydrogenation of co and carbonyl functional groups, Catalysis
Today, 12 (1992) 255-267.
[96] H. Kim, V.D. Doan, W.J. Cho, R. Valero, Z. Aliakbar Tehrani, J.M.L.
Madridejos, K.S. Kim, Intriguing Electrostatic Potential of CO: Negative Bond- ends and Positive Bond-cylindrical-surface, Scientific Reports, 5 (2015) 16307.
[97] J.M. Fukuto, S.J. Carrington, D.J. Tantillo, J.G. Harrison, L.J. Ignarro, B.A.
Freeman, A. Chen, D.A. Wink, Small Molecule Signaling Agents: The Integrated
Chemistry and Biochemistry of Nitrogen Oxides, Oxides of Carbon, Dioxygen,
Hydrogen Sulfide, and Their Derived Species, Chemical Research in Toxicology,
25 (2012) 769-793.
[98] J. Bayer, K.C. Stein, L.J.E. Hofer, R.B. Anderson, Effect of preadsorbed sulfur compounds on chemisorption of CO and CO2 on iron catalysts, Journal of
Catalysis, 3 (1964) 145-155.
[99] J. Benziger, R.J. Madix, The effects of carbon, oxygen, sulfur and potassium adlayers on CO and H2 adsorption on Fe(100), Surface Science, 94 (1980) 119-
153.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 234 Chapter 6
[100] E.L. Garfunkel, J.E. Crowell, G.A. Somorjai, The strong influence of potassium on the adsorption of carbon monoxide on platinum surfaces: a TDS and HREELS study, The Journal of Physical Chemistry, 86 (1982) 310-313.
[101] M.P. Kiskinova, CO adsorption on alkali-metal covered Ni(100), Surface
Science, 111 (1981) 584-594.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 235
Chapter 7
Selective conversion of CO and CO2 from a refinery offgas by catalytic hydrogenation over nickel-based catalysts
Chapter 7
7.1 Abstract
Complete removal of trace amounts of carbon monoxide and carbon dioxide present in an ethane offgas (ExxonMobil refinery, Altona, VIC) via catalytic hydrogenation (over Ni-Mn/Al2O3) was studied. The addition of hydrogen gas into the feed reduced the concentration of CO and CO2 to below the detection limits
(40 ppb CO, 70 ppb CO2). Pre-treatment of the saturate gas plant ethane (SGPE) offgas using molecular sieves to remove water vapour from the feed gas stream did not affect COx hydrogenation at temperatures below 300 °C. However, pre- treatment resulted in a significant reduction in CO and CO2 concentrations at temperatures above 300 °C. The results also disclosed significant catalyst deactivation using SGPE mixture during experiments over the Ni-Mn/Al2O3 catalyst. In this case, it is found that the used catalysts can be regenerated by hydrogen treatment at 500 °C.
The effect of ethane and ethylene in the feed gas stream on catalytic hydrogenation of low concentration CO and CO2 has also been investigated.
Ethane addition did not influence the hydrogenation of COx at 180 °C while it inhibited the hydrogenation reaction at 320 °C. On the other hand, ethylene addition inhibited CO and CO2 hydrogenation at both 180 °C and 320 °C.
Moreover, ethylene addition caused deactivation effects which is most probably due to the coke formation.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 237 Chapter 7
7.2 Introduction
It has been estimated that 40 % of total energy demand and 97 % of the transportation fuels consumed in the United States is supplied by petroleum refineries [1]. A wide range of materials such as gasoline, jet fuel, diesel and other hydrocarbons (used as energy source or feedstock for other industries) are produced in oil refineries. In the last few decades, the refining industry has become progressively complex due to the new environmental regulations, higher levels of competition and lower profit margins [1].
Global demand of transportation fuels, especially gasoline and diesel fuels, is continuously increasing. One of the key processes in oil refinery industry is Fluid
Catalytic Cracking (FCC), used to synthesise lighter (lower molecular weight) products from crude oil. It is predicted that over the next few decades, the FCC process will continue to play a key role in the fuel producing industry as it can be also be adapted to produce biofuels [2-4]. The catalytic cracker is an important unit operation in all refineries. The FCC process was first commercialised in 1942.
The process has been improved in many different ways continuously since then, especially with respect to its mechanical reliability, feedstock compatibility and environmental performance [5, 6] It is reported that more than 400 FCC units are currently in operation [7].
In the modern refineries, the fluid catalytic cracking process is the most important method to convert crude hydrocarbon feedstocks (e.g. gas oils, cracked gas oils, de-asphalted gas oils and vacuum resids [8, 9]) to lighter and more valuable products such as light fuel oil and feedstocks for petrochemical plants. The FCC process was originally designed for cracking of vacuum gas oil
(vacuum/atmospheric distillation units’ overhead). However, most modern FCC
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 238 Chapter 7
units are also using feedstocks with higher boiling points such as distillation residues (vacuum distillation bottoms) [10, 11].
The FCC feedstocks generally contain hydrocarbons with boiling points in a range between 315 °C – 650 °C. Cracked naphtha is a highly valued FCC product, as it is a primary constituent of gasoline fuel. For example, about 35 % of the United
States gasoline pool is from FCC gasoline. Moreover, oxygenates (e.g. MTBE,
DIPE, ETBE and TAME) produced by alkylation of FCC light gases further highlight the FCC products contribute to the gasoline pool [12].
Fluid catalytic cracking (FCC) process consists of three main steps as below:
I. Reaction.
II. Product separation.
III. Regeneration.
Fluid catalytic cracking units typically operate at high process temperature range
(750 K and 800 K) and pressures close to atmospheric. The majority of reactions in the fluid catalytic cracking process take place over a multi-functional catalyst containing USHY zeolite, an active alumina matrix, an inert kaolin matrix and a binder [13]. Various additives has been explored to improve the FCC process, for example increasing the overall yield and/or decreasing pollutant emissions [5,
14]. The most common additives are used to promote combustion in the regenerator unit [15], boosting light olefins and octane yield [14, 16], reducing
SOx and NOx [17, 18] and trapping metal containing species [19].
Similar to most industrial unit operations, fluid catalytic cracking (FCC) process design and operation is strongly influenced by environmental regulations. In this case, changes to the process aimed at decreasing air-pollutant emissions and modifying the product stream’s composition to meet the new and often stringent
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 239 Chapter 7
regulations. Criteria pollutants and air toxic components levels are specified by air emission regulations. Carbon monoxide (CO), sulfur oxides (SOx), nitrogen oxides (NOx), ozone and particulates are defined as criteria pollutants [5].
A typical petroleum refinery consists of several unit operation, designed to synthesise various products (gasoline, diesel, jet fuel and heating oil) from a raw crude oil feed. The first unit operation in refinery processes is called a crude unit, in which the raw crude oil is processed in distillation towers to form intermediate products such as gas oil, diesel, kerosene and naphtha (Figure 7.1). The bottom stream of the atmospheric tower contains the heaviest portion of the crude oil. As this portion cannot be processed in the atmospheric tower it is transported to the vacuum tower after heating. The vacuum tower then splits (distils) the heaviest portion of the crude oil into gas oil and residue. The vacuum tower residue can be further processed in other units such as residue cracker, visbreaker, deasphalting unit and delayed coker, or it can be sold as road asphalt or fuel oil.
Finally, the feed stream for FCC unit (gas oil) is mainly from atmospheric/vacuum towers and the delay coker. In some refineries, the atmospheric/vacuum residue is blended into the FCC feedstock to be cracked in FCC process. The FCC feedstock can be hydrotreated (partial/full hydrotreating) or unhydrotreated. The
FCC process is broken down into separate sections for clarity [6]:
• Feed preheat
• Feed nozzles—riser
• Catalyst separation
• Stripping section
• Regenerator—heat/catalyst recovery
• Partial versus complete combustion
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 240 Chapter 7
• Regenerated catalyst standpipe/slide valve
• Flue gas heat and pressure recovery schemes
• Catalyst handling facilities
• Main fractionator
• Gas plant
• Treating facilities
Figure 7.1. The process flow diagram of a typical crude oil refinery.
The FCC unit converts gas oil to more valuable products. Generally, the primary objective of fluid catalytic cracking is maximizing the gasoline and LPG yield of production. The common products from an FCC unit are listed below [6]:
• Dry gas (hydrogen, methane, ethane, and ethylene)
• LPG (propane, propylene, isobutane, normal butane and butylene)
• Gasoline
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 241 Chapter 7
• Light cycle oil (LCO)
• Heavy cycle oil (HCO)
• Decanted (or slurry) oil
• Combustion coke
This chapter examines the treatment of an FCC gas plant ethane-rich offgas sampled directly from a commercial process, as feed stream. A typical refinery may have numerous gas plants, operating in order to process the light gas streams from different process units. However, it is also possible that a gas plant handles a combination of overhead vapour gas stream, sourced from different units including fluid catalytic cracking, hydro-cracking, reforming units, coker or distillation tower [7].
The FCC gas plant can be broken down into the units below [6]:
• Primary absorber
• Sponge oil or secondary absorber
• Stripper or de-ethaniser
• Debutanizer
• Gasoline splitter
• Water wash system (WGCs and HPS)
The gas plant of fluid catalytic cracker main objective is separating the unestablished gasoline and light gases into the below products.
• Fuel gas
• C3’s and C4’s
• Gasoline.
The debutaniser overhead contains C3 and C4 hydrocarbons including propane, propylene, n-butane, i-butane and butylene. The C3’s and C4’s are commonly
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 242 Chapter 7
separated using splitting towers for further processing of C4’s in alkylation units.
Treatment facilities to remove sulfur and other contaminants are also used in many FCC units. The FCC gas plants process flow commences at a wet gas compressor (WGC) unit, which often consists of a double stage centrifugal compressor. However, in some FCC units single-stage wet gas compressors are employed. The liquefied hydrocarbon stream from WGCs is directed to either the striper column or to the high-pressure separator (HPS) unit. The liquid stream from HPS is pumped to the stripper and the gas stream is sent to the primary absorber. The vapour streams from the stripper and the primary absorber are recycled to the HPS [6].
As the vapour from primary absorber contains small quantities of gasoline, a secondary absorber (sponge oil) is used to recover the gasoline. The gas plant offgas is the lean gas leaving the secondary absorber or sponge oil tower. The offgas is sent to the treating facilities for removing H2S and other acidic components. The treated FCC gas plant offgas then enters the refinery fuel gas system [6].
The fuel gas stream originates from the top of secondary absorber or sponge oil tower in the FCC gas plant offgas. It primarily contains C2/lighter hydrocarbons,
H2 and some C3+ species. The offgas also contains contaminants such as H2S,
CO, CO2, O2, N2, and water vapour [6, 7] In some refineries, the gas plant may be considered as either an unsaturated or saturate gas plant. The principal of these gas plants has many similarities. Most gas plants are working with absorption-fractionation systems (as described above for FCC gas plant) [20].
In most refineries, the offgas streams are usually blended into refinery fuel gas system. However, depending on the offgas composition some refineries recover
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 243 Chapter 7
H2 and other valuable products (e.g. light hydrocarbons) present in the offgas stream by applying different methods, such as cryogenic recovery, membrane separation or pressure-swing absorption [7] The offgas is an undesirable by- product of the fluid catalytic cracking gas plant unit (or, in general, the whole refinery). Excessive yields of offgas has many disadvantages such as limiting the unit feed rate and increasing carbon oxides emission levels [6] Therefore, further processing or sending the offgas to secondary plants (e.g. petrochemical plants, polymerization plants and etc.) can be helpful to minimise the disadvantages of offgas production and improve the overall yield of the unit and/or refinery [7, 20].
Re-using an FCC offgas stream (rich in ethane) as feedstock for light alkane cracker units (ethane cracker) to produce light olefins (ethylene) for polymerization purposes is an attractive option for this gas stream. However, the offgas needs to be treated to remove contaminants (e.g. COx) which are present in the gas. To utilise the ethane-rich offgas as feed stock in the ethane crackers, it is critical to remove COx species from the stream, especially carbon monoxide to sub-ppm levels, as it has very serious deleterious effect on the ethane cracking process and in particular on the polymer product.
Carbon oxides removal (especially CO) from a hydrocarbon-rich stream can be challenging because the treatment process may also lead to cracking the concentration of (desirable) hydrocarbons present in the feed stream. We have shown in earlier studies (reported in chapters 4 to 6), that it is possible to reduce
COx in the presence of light hydrocarbons by catalytic hydrogenation under conditions where the conversion of light hydrocarbons is negligible. To explore the application of the catalytic process we have developed, gas samples were obtained from two major industries, ExxonMobil and Qenos, in Australia.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 244 Chapter 7
Qenos (Altona, Victoria, Australia) is a polymer production company producing a wide range of polymer materials from gaseous feedstocks (mainly ethane).
Therefore, using a readily available and relatively low cost ethane feedstock to produce ethylene for further polymerisation processes (e.g. polyethylene production) is a very attractive commercial opportunity for Qenos. The
ExxonMobil refinery (also located in Altona) has offgas streams from different units that are rich in ethane and are potentially suitable feedstocks to use by
Qenos. One of the ethane-rich offgas streams is the saturate gas plant ethane
(SGPE) offgas. However, the SGPE contains trace amounts of COx which require removal to sub-ppm levels prior to consumption in the Qenos ethylene production unit.
The main objective outlined in this chapter is to catalytically convert the carbon oxides (especially CO) from the offgas mixtures while minimising any reduction in the concentration of hydrocarbons present in the SGPE. The structure of this chapter is briefly outlined below:
Addition of hydrogen: the feasibility of complete conversion of COx from
SGPE mixture is studied with/without addition of extra hydrogen.
Feed stream drying conditions: the effects of pre-treating the feed stream
(just SGPE) by using molecular sieve filters were investigated.
Used catalyst’s performance: the catalytic performance of a used catalyst
sample (in SGPE stream) was analysed, catalyst regeneration was
performed and results were compared to the fresh catalyst.
Effects of hydrocarbons on low concentration COx hydrogenation: low
concentration (5 ppm) CO/CO2 feed (with roughly equal amount of
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 245 Chapter 7
hydrogen present in SGPE) was used to study the effects of hydrocarbons
addition on the hydrogenation process.
7.3 Experimental
The bi-metallic nickel-manganese catalyst was prepared by the incipient wetness method. The detailed preparation procedure of Ni-Mn/Al2O3 catalyst is explained in chapter 6. The catalyst particles (250 – 425 μm) were mixed with quartz sand and placed in a tubular fixed bed reactor. The catalysts were reduced prior to each run as explained in chapter 6. The temperature programmed experiments performed at atmospheric pressure in a temperature range from 150 °C to 350
°C.
The Saturated Gas Plant Ethane Offgas is labelled as SGPE in this study. The composition of SGPE is showed in the Table 7.1.
Table 7.1. The composition of Saturate Gas Plant Ethane Offgas mixture supplied by Mobil Refining Australia. Content in the Gas mixture H2 2.3 % CH4 7.1 % C2H6 59.5 % C3H8 26.4 % C4 4.7 % C2H4 66 ppm C3H6 619 ppm C4+ Trace CO 3 ppm CO2 55 ppm
7.3.1 Addition of hydrogen
Four different feed compositions were analysed in this section to investigate the effect of hydrogen addition. Feed streams contained hydrogen and the SGPE.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 246 Chapter 7
The composition of each feed stream were adjusted based on the contents of H2 and SGPE in the feed stream. Table 7.2 displays the amount of H2 and SGPE in each feed stream.
Table 7.2. Feed stream composition with different amounts of hydrogen added to the SGPE.
Feed Label H2 Content SGPE Content Overall SGPE-00H 0 % 100 % 100 % SGPE-10H 10 % 90 % 100 % SGPE-25H 25 % 75 % 100 % SGPE-40H 40 % 60 % 100 %
7.3.2 Feed stream drying conditions
The SGPE containing feed stream has been pre-treated before entering the reactor. Two different pre-treatment methods were used to filter the SGPE feed stream. MS-3A (Molecular Sieve UOP Type 3A) was used as the adsorbent in the both methods. The particle size of adsorbent and the adsorption temperature were different for the two pre-treatment procedures. The detail of each pre- treatment methods are presented in Table 7.3.
Table 7.3. The conditions and materials used for each pre-treatment methods. Adsorption Adsorbent Feed Label Adsorbent Temperature (°C) Particle Size SGPE-00H N/A N/A N/A SGPE-00H-D1 MS-3A 25 Beads (2 mm) SGPE-00H-D2 MS-3A 0 250 – 425 μm
The feed stream contained just SGPE mixture with no additional hydrogen. The catalyst was the same Ni-Mn/Al2O3 catalyst used in the section 7.2.1. The experiment procedure was also similar to the section 7.2.1.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 247 Chapter 7
7.3.3 Used catalyst’s performance
The used catalysts from SGPE experiments were reused in a hydrogen-rich feed stream containing CO, CO2, H2, CH4, C2H4, C2H6, C3H6 and C3H8 (Table 7.4).
The detail of gas supplies for the hydrogen-rich feed stream are given in chapter
4. The used catalysts were collected from similar temperature programmed experiments with the feed stream containing just SGPE (no additional H2).
Table 7.4. The hydrogen-rich feed stream composition for analysing the performance of
used Ni-Mn/Al2O3 catalysts. Partial Pressure Species (KPa) H2 85.11 CO 2.03 CO2 0.68 CH4 10.13 C2H6 0.34 C2H4 0.34 C3H8 0.03 C3H6 0.03 He 2.67
Three sets of experiments performed with the fresh and used Ni-Mn/Al2O3 catalyst. Table 7.5 shows the detail of each set of experiment.
Table 7.5. The experiment details and pre-run conditions for fresh and used Ni-
Mn/Al2O3 catalysts. Set Label Catalyst Pre-run Conditions Activated in H2 stream for CF Fresh Ni-Mn/Al2O3 2 hr at 500 °C Used Ni-Mn/Al2O3 in UCF N/A SGPE Used Ni-Mn/Al2O3 in Activated in H2 stream for UCF-Act SGPE 2 hr at 500 °C
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 248 Chapter 7
A 5-step switch experiment also performed with the simulated feed stream (Table
7.4) and SGPE. The information for each step is explained below in Table 7.6.
Table 7.6. Details of each step for the switch experiment with the simulated feed stream (Table 7.4) and SGPE. Temperature Step Feed Composition Time (min) (°C) I Simulated feed (Table 7.4) 350 100 II SGPE (Table 7.1) 350 180 III Simulated feed (Table 7.4) 350 60 IV Hydrogen 500 180 V Simulated feed (Table 7.4) 350 40
All experiments in this section were performed in a tubular fixed bed reactor in atmospheric pressure. Except the switch experiment, all other runs were temperature programmed at atmospheric in a temperature range between 150
°C and 350 °C.
7.3.4 Effects of hydrocarbons on low concentration COx hydrogenation
A series of switch experiments have been carried out. In this case, four different feed streams have been used. All of the feed streams were in atmospheric pressure and the total flow rates were also similar. The only difference between each feed stream was feed composition (Table 7.7). The following gas cylinders were supplied by Coregas: hydrogen (99.9999 %), helium (99.99996 %), C2/C3
(9.99 % ethane, 9.99 % ethylene, 0.997 % propane and 1.0% propylene in helium balance), C2H4 (10 % ethylene in helium balance) and C2H6 (10 % ethane in helium balance). A low concentration mixture of CO and CO2 (10 ppm CO and
10 ppm CO2 in helium balance) was supplied by BOC.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 249 Chapter 7
Table 7.7. Feed compositions for each feed stream in this section. Feed Species Label H2 CO CO2 C2H6 C2H4 C3H8 C3H6 He Feed A 2.7 % 5 ppm 5 ppm N/A N/A N/A N/A 97.3 % Feed B 2.7 % 5 ppm 5 ppm 4.9 % 4.9 % 0.5 % 0.5 % 86.5 % Feed C 2.7 % 5 ppm 5 ppm 4.9 % N/A N/A N/A 92.4 % Feed D 2.7 % 5 ppm 5 ppm N/A 4.9 % N/A N/A 92.4 %
Five sets of switch experiments performed by using the feed compositions explained in Table 7.7. All experiments were performed at atmospheric pressure while the temperature of each set could be different. The details of each set of switch experiments are displayed in Table 7.8.
Table 7.8. Details of each step for each switch experiment with the low concentration
COx feed compositions (Table 7.7). Set (Label) Step Feed Composition Temperature (°C) Time (min) I Feed A 75 1 (C2/C3-LT) II Feed B 180 150 III Feed A 800 I Feed A 75 2 (C2H6-LT) II Feed C 180 150 III N/A N/A I Feed A 75 3 (C2H4-LT) II Feed D 180 150 III Feed A 450 I Feed A 75 4 (C2H6-HT) II Feed C 320 150 III Feed A 100 I Feed A 75 5 (C2H4-HT) II Feed D 320 150 III Feed A 350
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 250 Chapter 7
7.4 Results and discussion
7.4.1 Addition of hydrogen
The possibility of removing carbon monoxide and carbon dioxide from the SGPE mixture (Table 7.1) has been investigated over Ni-Mn/Al2O3 catalyst. Initially, the effect of adding excess hydrogen to the SGPE mixture has been investigated with different feed compositions as shown in Table 7.2. The change in the concentration over temperature for CO and CO2 are shown in Figure 7.2 and
Figure 7.3 respectively. Carbon monoxide concentration dropped below its inlet value at low temperature range (below 210 °C) for the feed stream containing just
SGPE mixture (SGPE-00H). The concentration of carbon monoxide continuously increased by increasing temperature. For the same feed stream (SGPE-00H), the concentration of carbon dioxide also increased at temperature range above 300
°C while it was almost constant below 300 °C. Addition of hydrogen showed remarkable differences in the concentration of both CO and CO2 during the temperature programmed analysis. For instance, carbon monoxide concentration dropped below the detection limit in a temperature range for all feed streams containing additional H2. The temperature range in which CO concentration was not detectable became wider by adding more hydrogen to the SGPE mixture
(Figure 7.2). Moreover, increasing the amount of hydrogen in the inlet stream resulted in lower CO production (less CO concentration increase compared to the baseline) over temperature. Similarly, carbon dioxide consumption enhanced by adding hydrogen to the SGPE mixture. The concentration of CO2 also reduced to value below the detection limit for feed streams containing 25 % and 40 % additional hydrogen (Figure 7.3). Addition of more H2 also inhibited the production of CO2 at higher temperatures (approximately above 300 °C). It is found from this
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 251 Chapter 7
data that increasing the amount of additional hydrogen resulted in reducing CO and CO2 concentration to values below the detection limit for COx in a wider temperature range. Increasing the additional H2 content in the feed stream (up to
40 %) reduced the concentration of produced CO and CO2 at higher temperatures
(350 °C).
Figure 7.2. CO concentration for feed streams containing SGPE and different additional
H2 contents over Ni-Mn/Al2O3 catalyst: (a) Full scale, (b) Zoomed-in.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 252 Chapter 7
Figure 7.3. CO2 concentration for feed compositions containing SGPE and different
additional H2 contents over Ni-Mn/Al2O3 catalyst.
7.4.2 Feed stream drying conditions
It is possible that a refinery offgas stream contains trace contaminants such as oxygen, ammonia, nitriles, sulfur species (e.g. H2S, COS and etc.), chlorides, carbon monoxide, carbon dioxide, water, arsenic, mercury and etc. Therefore, removing some or all of these contaminants is believed to be beneficial [21]. It is found from the results of previous section that addition of hydrogen to SGPE improves the catalytic hydrogenation of COx for complete removal of CO and CO2 even at low temperature (about 210 °C). From a process point of view, it is more favourable to treat the offgas stream without adding any extra hydrogen to the feed stream. In this case, the influence of using a molecular sieve absorber (to remove oxygenated contaminants such as water) for feed stream on the catalytic
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 253 Chapter 7
hydrogenation of carbon oxides was studied. The Mol-Sieve 3Å was selected for this purpose because MS-3Å showed good selectivity to separate water from hydrocarbons [22]. As explained in Table 7.3, two different drying conditions were applied. The reason for using MS-3Å with smaller particle size and at cooler temperature was minimizing the void space in the filtering tube and increasing the absorption capacity of MS-3Å respectively [23]. Figure 7.4 and Figure 7.5 show the changes in CO and CO2 concentration versus temperature over Ni-
Mn/Al2O3 for the SGPE feed stream with different drying conditions.
Figure 7.4. CO concentration for feed streams containing just SGPE with different
drying conditions over Ni-Mn/Al2O3: (a) Full scale, (b) Zoomed-in.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 254 Chapter 7
From Figure 7.4(b) it is clear that the concentration of carbon monoxide reduced below its inlet concentration at temperatures below 210 °C for all experiments.
CO concentration increased by increasing temperature. On the other hand, CO2 concentration remained constant at temperatures below 300 °C. The concentration of CO2 increased with increasing temperature (above 300 °C) for the untreated SGPE, while it showed a slight decrement at the same temperature range for pre-treated (dried) SGPE.
Figure 7.5. CO2 concentration for feed streams containing just SGPE with different
drying conditions over Ni-Mn/Al2O3 catalyst.
Applying different drying conditions did not show remarkable changes on removing CO and CO2 from saturate gas plant offgas. For example, carbon monoxide and carbon dioxide concentration changes at low temperature (below
300 °C) are fairly similar for all experiments. However, using a molecular sieve
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 255 Chapter 7
for pre-treating feed streams influenced the concentration of carbon oxides in the product stream at higher temperature range (above 300 °C). In this case, the carbon monoxide concentration at 350 °C for dried feed streams is almost half of the untreated one. Moreover, CO2 concentration is above its inlet value at 350 °C for untreated SGPE while it dropped below its inlet concentration at the same temperature for pre-treated feed streams.
The difference in carbon oxides concertation with and without molecular sieve absorbent for treating the feed streams can be explained by considering the various possible side reactions. As mentioned before, the refinery offgas possibly contains trace amount of contaminants other than CO and CO2. Considering O2 and H2O as two of the most common contaminants, a wide range of catalytic side- reactions such as steam reforming (SR) [24, 25], dry reforming (DR) [26-28], water gas shift (WGS) [29], reverse water gas shift (RWGS) [30] and partial oxidation (POX) [31] are possible (Eq. 7.1-5).
CnH2n+2 + nH2O → nCO + (2n + 1)H2 Eq. 7.1
CnH2n+2 + nCO2 → 2nCO + (n + 1)H2 Eq. 7.2
CO + H2O → CO2 + H2 Eq. 7.3
CO2 + H2 → CO + H2O Eq. 7.4
n C H + O → nCO + (n + 1)H Eq. 7.5 n 2n+2 2 2 2
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 256 Chapter 7
By considering the above side-reactions we can explain the differences in CO and CO2 concentrations especially in the higher temperature range (above 300
°C). In this case, the trace amounts of H2O and O2 in the untreated feed stream
(SGPE-00H) resulted in increment of CO and CO2 concentrations at temperatures above 300 °C. It is known that higher temperature is more favourable for reforming reactions, which are endothermic [26]. By applying the drying conditions for feed stream, resulting in partial or complete removal of trace
H2O from the SGPE mixture, partial or complete inhibition of the steam reforming reaction occurred. Therefore, the concentration of CO and CO2 in the outlet stream at 350 °C remarkably decreased for dried feed streams. Figure 7.6 shows a schematic illustration of the relation between the presence of H2O in the feed stream and COx concentrations in the product stream.
Figure 7.6. Schematic illustration of CO and CO2 formation via catalytic steam reforming of hydrocarbons.
7.4.3 Used catalyst’s performance
From the data in two previous sections, it was found that CO and CO2 could not be reduced below detection limit for the feed stream containing SGPE only. In
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 257 Chapter 7
this case, it is likely that high concentration of hydrocarbons (in total more than
96 %) or presence of some possible trace contaminants in the saturate gas plant ethane offgas cause partial or full deactivation of the catalyst [21]. Initially, the fresh and used Ni-Mn/Al2O3 were analysed by a surrogate feed stream containing hydrogen, carbon oxides and light hydrocarbons (Table 7.4). The change in concentrations over temperature for carbon monoxide and carbon dioxide are displayed in Figure 7.7 and Figure 7.8 respectively.
Figure 7.7. CO concentration for simulated feed stream containing H2, COx and
hydrocarbons over fresh and used Ni-Mn/Al2O3 catalyst.
Over the fresh catalyst, the concentration of carbon monoxide increased by temperature until no CO was detected at around 265 °C. For the same experiment, the concentration of CO2 started to decrease at around 245 °C and dropped to an undetectable value at 300 °C. By using the fresh catalyst, it is
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 258 Chapter 7
confirmed that the catalyst is able to completely remove COx in this reaction conditions.
Figure 7.8. CO2 concentration for simulated feed stream containing H2, COx and
hydrocarbons over fresh and used Ni-Mn/Al2O3 catalyst.
Two used catalysts were collected after reactor experiments with just SGPE at same process conditions. Both used catalysts were used under the flow of SGPE for about 10 hours. One of these used catalysts were tested without any pre- treatment while the other one were treated in hydrogen flow at 500 °C for 2 hours.
It is observed that the CO concentration started to decrease at 100 °C higher than that of fresh catalyst and it only reduced to roughly half of its inlet value even at
350 °C. Moreover, CO2 concentration remained constant until high temperatures
(330 °C). Therefore, using a highly active catalyst for COx hydrogenation (i.e. Ni-
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 259 Chapter 7
Mn/Al2O3) in a SGPE feed stream caused significant reduction of its ability to convert CO and CO2.
By pre-treating the similar used catalysts, interesting results were obtained. In this case, it is found that using hydrogen pre-treatment recovered the used catalyst. Comparing CO and CO2 concentration graphs for the pre-treated used catalyst with the fresh one confirms that the used catalyst was almost completely regenerated.
The deactivating effect of saturate gas plant ethane off gas on Ni-Mn/Al2O3 catalysts and the possibility to regenerate the catalyst with hydrogen flow at 500
°C has been also investigated by performing a switch experiment (Figure 7.9) as explained in Table 7.6.
Figure 7.9. CO and CO2 concentration for the switch experiment (Table 7.6) with
simulated feed, SGPE and H2 streams (dashed lines: inlet concentrations).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 260 Chapter 7
In this case, the simulated feed stream was used in first step and temperature was set to 350 °C to ensure the complete removal of CO and CO2. Then, the flow was switched to the SGPE stream for 3 hours and switched back to the simulated feed stream. It was found that a significant unreacted CO and CO2 were detected in the outlet stream by leaving the catalyst under the SGPE stream. In the next step, the simulated feed stream was switched to another stream containing just hydrogen. The furnace temperature was increased from 350 °C to 500 °C. The catalyst remained at 500 °C under hydrogen flow for toughly two hours and cooled to 350 °C again. Finally, the feed stream switched back again to the simulated stream. The concentration of carbon monoxide and carbon dioxide were reduced to an undetectable level after catalyst regeneration in H2 flow.
7.4.4 Effects of hydrocarbons on low concentration COx hydrogenation
To investigate the influence of light hydrocarbons on the catalytic hydrogenation of low concentration carbon monoxide and carbon dioxide, a series of switch experiments with Ni-Mn/Al2O3 have been performed (Table 7.8). In this case, four different feed streams containing hydrogen, C2-C3 light hydrocarbons mixture, ethane, ethylene and low concentration COx mixture have been used (Table 7.7).
In this section, the different feed compositions and experiments’ steps are mentioned with the labels explained in Table 7.7 and Table 7.8. It should be noticed that the total flow rate of each feed stream were equal.
7.4.4.1 Set 1 (C2/C3-LT)
Initially, an experiment was performed at 180 °C with using light hydrocarbons mixture, hydrogen and low concentration COx (Figure 7.10). In the first step, feed composition A (CO, CO2 and H2 in He) were used. In this step (C2/C3-LT: I), CO concentration reduced to an undetectable value due to the catalytic
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 261 Chapter 7
hydrogenation. The feed stream then switched to feed B (CO, CO2, H2 and light hydrocarbons mixture in He). It was found that the CO concentration came back to its inlet value by adding the C2-C3 hydrocarbon mixture to the inlet stream
(C2/C3-LT: II). The feed stream switched back again to feed A for the third and final step (C2/C3-LT: III). The result show that CO concentration started to decrease gradually after removing the C2-C3 hydrocarbons mixture from the inlet stream. The concentration of carbon monoxide finally reduced to an undetectable level again after being almost 11 hours under feed A stream at 180 °C.
Figure 7.10. CO concentration for the C2/C3-LT experiment (Table 7.8) with feed streams A and B (Table 7.7). Dashed line shows inlet concentration.
7.4.4.2 Set 2 (C2H6-LT)
The second set of experiment was performed with using feed stream A and C to investigate the influence of ethane addition on CO concentration at 180 °C
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 262 Chapter 7
(Figure 7.11). The first step of this set (C2H6-LT: I) is similar to the first of the previous set (C2/C3-LT: I). Therefore, it was also observed that the CO concentration was decreased to an undetectable level during the first step. After the first step, the feed stream switched from feed A to feed C (CO, CO2, C2H6 and H2 in He). During this step (C2H6-LT: II), the concentration of CO remained similar to the first step. Therefore, it was concluded that ethane addition did not influence the catalytic hydrogenation of CO over Ni-Mn/Al2O3 at 180 °C.
Figure 7.11. CO concentration for the C2H6-LT experiment (Table 7.8) with feed streams A and C (Table 7.7). Dashed line shows inlet concentration.
7.4.4.3 Set 3 (C2H4-LT)
The final experiment at 180 °C was performed with feed A and feed D to study the effect of ethylene addition on CO concentration (Figure 7.12). Similar to set 1 and 2, inlet carbon monoxide was completely converted and no CO was detected
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 263 Chapter 7
in the outlet stream during this step (C2H4-LT: I). In the second step (C2H4-LT: 2), ethylene was added to the feed stream by switching to feed D from feed A. It is found that the concentration of carbon monoxide increased by adding ethylene to the feed stream. By switching back to feed A in the final step (C2H4-LT: III) and removing ethylene from the feed stream, CO concentration reduced with time.
After approximately 7 hours during the last step, no carbon monoxide was detected in the outlet stream.
Figure 7.12. CO concentration for the C2H4-LT experiment (Table 7.8) with feed streams A and D (Table 7.7). Dashed line shows inlet concentration.
7.4.4.4 Set 4 (C2H6-HT)
The influence of ethane addition has also been investigated at higher temperature
(320 °C). Figure 7.13 illustrates the concentration of CO and CO2 over all steps of set 4 (C2H6-HT). Both CO and CO2 were fully converted at 320 °C in the first
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 264 Chapter 7
of this set (C2H6-HT: I). By switching back the feed stream from feed A to feed C
(C2H6-HT: II), both CO and CO2 concentration returned back to their inlet values.
It is found that unlike set 2 when addition of ethane did not affect the hydrogenation reaction at 180 °C, no CO and CO2 were converted in the presence of ethane at 320 °C. In the third step (C2H6-HT: III), ethane was removed from the feed stream by switching back to feed A. It was observed that by removing ethane from feed stream, CO and CO2 concentration suddenly dropped back to undetectable level again.
Figure 7.13. CO and CO2 concentration for the C2H6-HT experiment (Table 7.8) with feed streams A and C (Table 7.7). Dashed line shows inlet concentration.
7.4.4.5 Set 5 (C2H4-HT)
In this set, the effect of adding ethylene on CO hydrogenation was studied (Figure
7.14). The first step (C2H4-HT: I) is similar to first step set four. Therefore, the
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 265 Chapter 7
concentration of CO and CO decreased below detection limit. Ethylene was added to the feed stream by switching from feed A to feed D. Similar to switch experiment at 180 °C, it is found that the concentration of CO and CO2 came back to their inlet values at 320 °C due to ethylene addition. In the final step (C2H4-HT:
III), ethylene has been removed by switching back to feed A. In this case, the results show that the concentration of CO and CO2 gradually decreased and became close to the undetectable level after almost 2 hours.
Figure 7.14. CO and CO2 concentration for the C2H4-HT experiment (Table 7.8) with feed streams A and D (Table 7.7). Dashed line shows inlet concentration.
From the results of switch experiments in this section, it is found that the addition of light hydrocarbons have different effects on catalytic hydrogenation of low concentration CO and CO2. These effects are related to the type of light hydrocarbons (saturated or unsaturated) and process conditions (i.e.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 266 Chapter 7
temperature). For instance, addition of ethane did not influence hydrogenation of
CO at 180 °C while it inhibited COx hydrogenation at 320 °C. The hydrogenation of CO and CO2 was recovered shortly after switching off the ethane flow at 320
°C. The inhibition of CO and CO2 hydrogenation at higher temperature is most probably because of the dominant hydro-cracking of ethane to methane and consequent reduction of hydrogen concentration (Eq. 7.6) [32, 33].
C2H6 + H2 → 2CH4 Eq. 7.6
On the other hand, ethylene addition is found to inhibit CO and CO2 hydrogenation at both 180 °C and 320 °C. The COx hydrogenation was recovered gradually after removing ethylene from the feed stream. The recovery time was longer for lower temperature (7 hours at 180 °C) compared to higher temperature
(2 hours at 320 °C). The inhibition of CO and CO2 hydrogenation by ethylene addition can be explained by the dominant olefin saturation reaction (Eq. 7.7) [34,
35]. Moreover, CO and CO2 hydrogenation did not recover shortly after removing ethylene from the feed stream (unlike ethane experiments). This suggests that addition of ethylene probably had some deactivation effects on the catalyst. This deactivation is can be related to ethylene catalytic decomposition (Eq. 7.8) [36] or coking (Eq. 7.9) [37]. The catalyst then gradually regenerated by hydrogen in feed A [38].
C2H4 + H2 → C2H6 Eq. 7.7
C2H4 → CH4 + C Eq. 7.8
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 267 Chapter 7
C2H4 → 2H2 + 2C Eq. 7.9
7.5 Conclusion
The addition of hydrogen was observed to enhance the complete removal of CO and CO2 over Ni-Mn/Al2O3 from saturate gas plant ethane (SGPE) offgas, provided by ExxonMobil (Altona). In this case, feed streams containing 25 % and
40 % hydrogen showed complete removal of both CO and CO2. Influence of pre- treating the feed stream with just SGPE has been studied as well. It is found that removing possible contaminants (i.e. water) by molecular sieves from the feed stream changed the CO and CO2 concentration in the higher temperature range
(above 300 °C). In contrast, no significant changes were found on CO and CO2 concentration in the lower temperature range (below 300 °C) by pre-treating the feed stream with molecular sieving. The results confirm the saturate gas plant ethane offgas can considerably deactivate the Ni-Mn/Al2O3 catalyst. This deactivation can be explained by the following possibilities: the presence of trace contaminants in the saturate gas plant ethane offgas mixture and/or high concentration of light hydrocarbons in the feed stream. In this case, it is found that presence of ethane and ethylene can inhibit the hydrogenation of low concentration CO and CO2. Ethane addition inhibited COx hydrogenation at higher temperature (320 °C). While ethylene addition showed inhibition effects on
CO and CO2 hydrogenation both at lower (180 °C) and higher temperatures (320
°C). The inhibition of trace COx hydrogenation by ethane and ethylene can be explained by hydro-cracking and olefins hydrogenation reactions respectively.
Moreover, it is also observed that ethylene addition can cause catalyst deactivation which is most probably due to coke formation.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 268 Chapter 7
7.6 References
[1] N.K. Shah, Z. Li, M.G. Ierapetritou, Petroleum Refining Operations: Key
Issues, Advances, and Opportunities, Industrial & Engineering Chemistry
Research, 50 (2011) 1161-1170.
[2] I. Graça, F.R. Ribeiro, H.S. Cerqueira, Y.L. Lam, M.B.B. de Almeida, Catalytic cracking of mixtures of model bio-oil compounds and gasoil, Applied Catalysis B:
Environmental, 90 (2009) 556-563.
[3] M.E. Domine, A.C. van Veen, Y. Schuurman, C. Mirodatos, Coprocessing of
Oxygenated Biomass Compounds and Hydrocarbons for the Production of
Sustainable Fuel, ChemSusChem, 1 (2008) 179-181.
[4] A. Corma, G.W. Huber, L. Sauvanaud, P. O'Connor, Processing biomass- derived oxygenates in the oil refinery: Catalytic cracking (FCC) reaction pathways and role of catalyst, Journal of Catalysis, 247 (2007) 307-327.
[5] W.C. Cheng, G. Kim, A.W. Peters, X. Zhao, K. Rajagopalan, M.S. Ziebarth,
C.J. Pereira, Environmental Fluid Catalytic Cracking Technology, Catalysis
Reviews, 40 (1998) 39-79.
[6] R. Sadeghbeigi, Fluid Catalytic Cracking Handbook (Third Edition),
Butterworth-Heinemann, Oxford, 2012.
[7] R.A. Meyers, Handbook of Petroleum Refining Processes, McGraw-Hill
Education2003.
[8] M.L. Fernández, A. Lacalle, J. Bilbao, J.M. Arandes, G. de la Puente, U.
Sedran, Recycling Hydrocarbon Cuts into FCC Units, Energy & Fuels, 16 (2002)
615-621.
[9] A. Corma, F. Melo, L. Sauvanaud, F.J. Ortega, Different process schemes for converting light straight run and fluid catalytic cracking naphthas in a FCC unit for
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 269 Chapter 7
maximum propylene production, Applied Catalysis A: General, 265 (2004) 195-
206.
[10] K. Bryden, U. Singh, M. Berg, S. Brandt, R. Schiller, W.-C. Cheng, Fluid
Catalytic Cracking (FCC): Catalysts and Additives, Kirk-Othmer Encyclopedia of
Chemical Technology, John Wiley & Sons, Inc.2000.
[11] E.T. Habib, X. Zhao, G. Yaluris, W.-C. Cheng, L.T. Book, J.-P. Gilson,
Advances in Fluid Catalytic Cracking, 2005.
[12] C.I.C. Pinheiro, J.L. Fernandes, L. Domingues, A.J.S. Chambel, I. Graça,
N.M.C. Oliveira, H.S. Cerqueira, F.R. Ribeiro, Fluid Catalytic Cracking (FCC)
Process Modeling, Simulation, and Control, Industrial & Engineering Chemistry
Research, 51 (2012) 1-29.
[13] J. Biswas, I.E. Maxwell, Recent process- and catalyst-related developments in fluid catalytic cracking, Applied Catalysis, 63 (1990) 197-258.
[14] D. Wallenstein, R.H. Harding, The dependence of ZSM-5 additive performance on the hydrogen-transfer activity of the REUSY base catalyst in fluid catalytic cracking, Applied Catalysis A: General, 214 (2001) 11-29.
[15] M.C.N.A. de Carvalho, E. Morgado, H.S. Cerqueira, N.S. de Resende, M.
Schmal, Behavior of Fresh and Deactivated Combustion Promoter Additives,
Industrial & Engineering Chemistry Research, 43 (2004) 3133-3136.
[16] T. Blasco, A. Corma, J. Martínez-Triguero, Hydrothermal stabilization of
ZSM-5 catalytic-cracking additives by phosphorus addition, Journal of Catalysis,
237 (2006) 267-277.
[17] D.M. Stockwell, C.P. Kelkar, Reduction of NOx emissions from FCCU regenerators with additives, in: M. Occelli (Ed.) Studies in Surface Science and
Catalysis, Elsevier2004, pp. 177-188.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 270 Chapter 7
[18] R.E. Roncolatto, M.J.B. Cardoso, Y.L. Lam, M. Schmal, FCC SOx Additives
Deactivation, Industrial & Engineering Chemistry Research, 45 (2006) 2646-
2650.
[19] E.T. Habib, X. Zhao, G. Yaluris, W.C. Cheng, L.T. Boock, J.P. Gilson,
Advances in Fluid Catalytic Cracking, Zeolites for Cleaner Technologies,
PUBLISHED BY IMPERIAL COLLEGE PRESS AND DISTRIBUTED BY WORLD
SCIENTIFIC PUBLISHING CO.2002, pp. 105-130.
[20] Petroleum Refining Corrosion http://hghouston.com/resources/special- corrosion-topics/refining.
[21] R.N.a.J.R. Sheng-Yi Chuang, Contaminants key to refinery offgas treatment unit design, Oil & Gas Journal, 106 (2008).
[22] H.A.A. Farag, M.M. Ezzat, H. Amer, A.W. Nashed, Natural gas dehydration by desiccant materials, Alexandria Engineering Journal, 50 (2011) 431-439.
[23] R. Lin, A. Ladshaw, Y. Nan, J. Liu, S. Yiacoumi, C. Tsouris, D.W. DePaoli,
L.L. Tavlarides, Isotherms for Water Adsorption on Molecular Sieve 3A: Influence of Cation Composition, Industrial & Engineering Chemistry Research, 54 (2015)
10442-10448.
[24] C. Gillan, M. Fowles, S. French, S.D. Jackson, Ethane Steam Reforming over a Platinum/Alumina Catalyst: Effect of Sulfur Poisoning, Industrial &
Engineering Chemistry Research, 52 (2013) 13350-13356.
[25] М.М. Zyryanova, P.V. Snytnikov, A.B. Shigarov, V.D. Belyaev, V.A. Kirillov,
V.A. Sobyanin, Low temperature catalytic steam reforming of propane–methane mixture into methane-rich gas: Experiment and macrokinetic modeling, Fuel, 135
(2014) 76-82.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 271 Chapter 7
[26] H. Eltejaei, H. Reza Bozorgzadeh, J. Towfighi, M. Reza Omidkhah, M.
Rezaei, R. Zanganeh, A. Zamaniyan, A. Zarrin Ghalam, Methane dry reforming on Ni/Ce0.75Zr0.25O2–MgAl2O4 and Ni/Ce0.75Zr0.25O2–γ-alumina: Effects of support composition and water addition, International Journal of Hydrogen
Energy, 37 (2012) 4107-4118.
[27] L.B. Råberg, M.B. Jensen, U. Olsbye, C. Daniel, S. Haag, C. Mirodatos, A.O.
Sjåstad, Propane dry reforming to synthesis gas over Ni-based catalysts:
Influence of support and operating parameters on catalyst activity and stability,
Journal of Catalysis, 249 (2007) 250-260.
[28] B. Yan, X. Yang, S. Yao, J. Wan, M. Myint, E. Gomez, Z. Xie, S. Kattel, W.
Xu, J.G. Chen, Dry Reforming of Ethane and Butane with CO2 over PtNi/CeO2
Bimetallic Catalysts, ACS Catalysis, 6 (2016) 7283-7292.
[29] T. L. LeValley, A. Richard, M. Fan, The progress in water gas shift and steam reforming hydrogen production technologies – A review, 2014.
[30] Y.A. Daza, J.N. Kuhn, CO2 conversion by reverse water gas shift catalysis: comparison of catalysts, mechanisms and their consequences for CO2 conversion to liquid fuels, RSC Advances, 6 (2016) 49675-49691.
[31] B. Christian Enger, R. Lødeng, A. Holmen, A review of catalytic partial oxidation of methane to synthesis gas with emphasis on reaction mechanisms over transition metal catalysts, Applied Catalysis A: General, 346 (2008) 1-27.
[32] J.R. Anderson, B.G. Baker, The Hydrocracking of Saturated Hydrocarbons
Over Evaporated Metal Films, Proceedings of the Royal Society of London.
Series A, Mathematical and Physical Sciences, 271 (1963) 402-423.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 272 Chapter 7
[33] A. Cimino, M. Boudart, H. Taylor, Ethane Hydrogenation-Cracking on Iron
Catalysts with and without Alkali, The Journal of Physical Chemistry, 58 (1954)
796-800.
[34] J. Horiuti, K. Miyahara, Hydrogenation of Ethylene on Metallic Catalysts,
1967.
[35] A.S. Crampton, M.D. Rötzer, F.F. Schweinberger, B. Yoon, U. Landman, U.
Heiz, Ethylene hydrogenation on supported Ni, Pd and Pt nanoparticles: Catalyst activity, deactivation and the d-band model, Journal of Catalysis, 333 (2016) 51-
58.
[36] D.W. McKee, The Chemisorption and Catalytic Decomposition of Ethylene on Nickel, Journal of the American Chemical Society, 84 (1962) 1109-1113.
[37] G.C. Reyniers, G.F. Froment, F.-D. Kopinke, G. Zimmermann, Coke
Formation in the Thermal Cracking of Hydrocarbons. 4. Modeling of Coke
Formation in Naphtha Cracking, Industrial & Engineering Chemistry Research,
33 (1994) 2584-2590.
[38] I. Regeneration of precoked catalysts, P. Marecot*, S. Peyrovi, D. Bahloul,
J. Barbier, Regeneration by hydrogen treatment of bifunctional catalysts deactivated by coke deposition, Applied Catalysis, 66 (1990) 181-190.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 273
Chapter 8
Conclusions and recommendations
Chapter 8
8.1 Conclusions
The catalytic hydrogenation of carbon monoxide and carbon dioxide in the presence of light hydrocarbons (i.e. CH4, C2-C3 alkanes and alkenes) on alumina supported nickel-based catalysts has been studied. Experimental and computational techniques were used to investigate different aspects of the process. Concluding remarks of these investigations are highlighted below:
Based on thermodynamic calculations, it is found that due to the highly
exothermic characteristic of the hydrogenation and hydrocracking
reactions involved in the system, increasing reaction temperature reduced
the equilibrium conversion of the feed. On the other hand, increasing the
concentration of hydrogen in the inlet mixture resulted in enhancing
reactions equilibrium conversion at a constant temperature. Moreover, by
considering solid carbon as a reaction product in the calculations, it is
noted that higher temperatures favour the formation of solid carbon.
However, the amount of solid carbon produced (for temperatures up to
500 °C) for hydrogen-rich mixtures was negligible.
Experiments were conducted on COx hydrogenation in the presence of
light hydrocarbons over the benchmark Ni/Al2O3 catalyst. In this case, the
catalyst was studied under different feed compositions (e.g.
simultaneous/separate hydrogenation of CO and CO2 either in the
presence or absence of light hydrocarbons) and reaction conditions. The
inlet species were consumed and converted over different temperature
ranges. The results also showed that there is a temperature window
(approximately between 200 °C to 250 °C), whereby the concentration of
C2-C4 increases above their inlet concentration. The amount of C2-C4
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 275 Chapter 8
hydrocarbons produced from CO2 hydrogenation was negligible.
Hydrogenation of CO2 took place at higher temperature (about 90 °C) in
the presence of CO.
Catalyst activity measurements for a series of Ni-M/Al2O3 bi-metallic
catalysts (M: Fe, Co, Cu, Cr, Mn, Zn, Ru, Rh, Ag and Cu) have shown that
the addition of a transition metal to Ni/Al2O3 can change the activity of the
catalyst for hydrogenation of carbon oxides and the selectivity toward C2-
C4 production. In this case, it is found that each transition metal has
different and distinct promoting or inhibiting influence. For example,
manganese addition resulted in enhancement of the catalyst performance
while copper addition reduced both COx hydrogenation activity and light
hydrocarbons selectivity.
Studying the characteristics of active sites for Mn, Cu-Ni/Al2O3 catalyst
revealed that the composition of active sites with different electronic
properties was changed. Nitric oxide adsorption in-situ FTIR results
showed that by adding copper the ratio of electron-accepting/electron-
donating sites increased while manganese addition conversely affected
the ratio. By interpreting electron-accepting and electron-donating sites as
carbon-accepting and oxygen-accepting sites respectively, it is concluded
that adding manganese resulted in easier C – O bond cleavage while
copper addition increased the linear type CO adsorption and reduced the
activity. This discussion was also confirmed by chemisorption and TPD
(with CO and H2) studies. In this case, formation of more surface carbon
species and activated hydrogen is possibly the main reason for enhancing
the light hydrocarbons selectivity by Mn addition.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 276 Chapter 8
Ni-Mn/Al2O3 catalysts with different nickel and manganese contents were
studied. The results of catalyst evaluation displayed improved activity and
selectivity for Ni-Mn/Al2O3 (Ni/Mn<1) bimetallic catalysts compared to
single metal nickel catalyst. While exceeding Ni/Mn=1 ratio resulted in
reducing the catalytic performance. Formation of alloys containing Ni and
Mn was confirmed by employing different characterisation techniques (e.g.
XRD and TPR). The single metal Mn/Al2O3 catalyst did not show any COx
hydrogenation activity while dry reforming was observed by increasing
reactor temperature (above 300 °C).
To achieve insight on the promoting influence of manganese, the
electronic structure of the primary Ni/Al2O3 and bimetallic Ni-Mn/Al2O3
catalysts were investigated using in-situ FTIR technique and KS-DFT
method. The observations confirmed that by adding manganese the
electronic structure of the catalyst was changed. In this case, Mn addition
resulted in increasing electron-donating/-accepting sites ratio. This
resulted in decreasing the energy barrier (activation energy) for both CO
and CO2 hydrogenation over the catalyst with optimum Ni and Mn content
(e.g. Ni/Mn = 2), most possibly by reducing the required energy for C – O
bond dissociation and formation of active surface carbon species.
Moreover, it is found that decreasing the ratio C-/O-accepting sites
increased the activation energy for saturated hydrocarbon (e.g. ethane)
cracking (possibly by enhancing the energy barrier for C – C bond rupture).
A bimetallic Ni-Mn/Al2O3 catalyst has been used for converting trace
amounts of COx from saturate gas plant ethane (SGPE) offgas of
ExxonMobil, Altona refinery. It is found that adding hydrogen to the mixture
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 277 Chapter 8
is necessary for complete removal of CO and CO2. Moreover, increasing
reactor temperature resulted in increasing the amounts of CO and CO2 in
the product stream which is probably due to the presence of oxygen
containing contaminates such as H2O. In this case, feed pre-treatment was
attempted by using Mol-Sieve filters to reduce the concentration of water.
It is shown that applying pre-treatment conditions reduced the amount of
produced COx by increasing temperature.
By analysing the catalyst performance of fresh and used (in SGPE feed
stream), it is found that the industrial gas mixture has some deactivation
effects on Ni-Mn/Al2O3 catalyst. However, the used catalyst can be
reactivated by treatment in hydrogen at 500 °C. To understand this
phenomenon, a series of switch experiments were performed using low
concentration COx mixture, an alkane (i.e. ethane) and an alkene (i.e.
ethylene). It is observed that exposure of the catalyst to ethylene
(especially when ethylene content is more than hydrogen in the feed
stream) can inhibit COx hydrogenation. This inhabitation effect was found
to be more effective at lower temperature (180 °C).
8.2 Recommendations
Suggestions to continue studies on the topics presented in this thesis are provided in this section. The recommendations covers different aspects of the process such as process/catalyst development and advanced fundamental studies.
It was shown that modifying the catalyst’s electronic structure by adding a
transition metal to form bimetallic alloys can significantly affect its
performance for catalytic hydrogenation of COx in the presence of light
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 278 Chapter 8
hydrocarbons. Using a promoter (e.g. K, P, Na, Ca) to further improve the
catalysts performance might be interesting, as some of these additives
were proved to be effective in studies on single-metal catalysts [1-3].
One of the challenges in this study was the undesired conversion of light
hydrocarbons over the active catalysts for COx hydrogenation. Attempts to
design catalysts that restrict active sites into isolated cavities might be
useful to inhibit undesired reactions. In this case, designing hollow-shell
catalysts (e.g. core-shell and yolk-shell) and using supports such as
zeolites are interesting options [3-5].
Studying on effects of different process conditions (e.g. pressure) and
designing different reactor types (e.g. membrane) to minimise the contact
of light hydrocarbons with the catalyst’s active sites could be useful for
further process development.
To achieve a better insight into the reaction mechanisms advanced
experimental and theoretical investigations are necessary. Some of the
techniques that should be considered in future studies are as follow:
isotope labelling experiments, kinetics/reactor modelling and ab-initio
studies involving surface reactions [6-9].
Extended research on the electronic structure of modified catalysts can be
also helpful for further rational catalyst design and synthesis. X-ray photo-
electron spectroscopy (XPS) techniques coupled with computational solid-
state quantum chemical calculations can be used to achieve this goal [10-
12].
Studying the catalyst deactivation is also necessary as a main objective of
future works. In this case, long term time-on-stream experiments, studying
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 279 Chapter 8
the effects of different feed stream contaminants (e.g. H2O, H2S and NOx)
and detailed investigation on the structural and chemical changes
occurring in used catalysts should be performed.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 280 Chapter 8
8.3 References
[1] B.C. Enger, A. Holmen, Nickel and Fischer-Tropsch Synthesis, Catal. Rev.-
Sci. Eng., 54 (2012) 437-488.
[2] Y. Cheng, J. Lin, K. Xu, H. Wang, X. Yao, Y. Pei, S. Yan, M. Qiao, B. Zong,
Fischer–Tropsch Synthesis to Lower Olefins over Potassium-Promoted Reduced
Graphene Oxide Supported Iron Catalysts, ACS Catalysis, 6 (2016) 389-399.
[3] X. Zhou, J. Ji, D. Wang, X. Duan, G. Qian, D. Chen, X. Zhou, Hierarchical structured [small alpha]-Al2O3 supported S-promoted Fe catalysts for direct conversion of syngas to lower olefins, Chemical Communications, 51 (2015)
8853-8856.
[4] Z. Li, M. Li, Z. Bian, Y. Kathiraser, S. Kawi, Design of highly stable and selective core/yolk–shell nanocatalysts—A review, Applied Catalysis B:
Environmental, 188 (2016) 324-341.
[5] D. Farrusseng, A. Tuel, Perspectives on zeolite-encapsulated metal nanoparticles and their applications in catalysis, New Journal of Chemistry, 40
(2016) 3933-3949.
[6] V. Frøseth, S. Storsæter, Ø. Borg, E.A. Blekkan, M. Rønning, A. Holmen,
Steady state isotopic transient kinetic analysis (SSITKA) of CO hydrogenation on different Co catalysts, Applied Catalysis A: General, 289 (2005) 10-15.
[7] S.L. Shannon, J.G. Goodwin Jr, Characterization of catalytic surfaces by isotopic-transient kinetics during steady-state reaction, Chemical reviews, 95
(1995) 677-695.
[8] N. Kapur, J. Hyun, B. Shan, J.B. Nicholas, K. Cho, Ab Initio Study of CO
Hydrogenation to Oxygenates on Reduced Rh Terraces and Stepped Surfaces,
The Journal of Physical Chemistry C, 114 (2010) 10171-10182.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 281 Chapter 8
[9] M. Gajdoš, A. Eichler, J. Hafner, CO adsorption on close-packed transition and noble metal surfaces: trends from ab initio calculations, Journal of Physics:
Condensed Matter, 16 (2004) 1141.
[10] J.H. Stenlid, T. Brinck, Extending the σ-Hole Concept to Metals: An
Electrostatic Interpretation of the Effects of Nanostructure in Gold and Platinum
Catalysis, Journal of the American Chemical Society, 139 (2017) 11012-11015.
[11] A.M. Venezia, X-ray photoelectron spectroscopy (XPS) for catalysts characterization, Catalysis Today, 77 (2003) 359-370.
[12] J.-C. Dupin, D. Gonbeau, P. Vinatier, A. Levasseur, Systematic XPS studies of metal oxides, hydroxides and peroxides, Physical Chemistry Chemical
Physics, 2 (2000) 1319-1324.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 282 Appendix A
Appendix A
This appendix contains data on catalytic experiments using Ni-Mn/Al2O3 catalyst and Unsaturated Fluid Catalytic Cracking (UFCC) offgas supplied by ExxonMobil,
Altona. Table Appa. 1 shows the composition of UFCC mixture.
Table Appa. 1. The composition of UFCC Offgas supplied by Mobil Refining Australia. Content in the Gas mixture H2 5.16 % CH4 13.34 % C2H4 23.84 % C2H6 25.32 % C3H6 18.01 % C3H8 3.48 % C4 2.93 % C4+ 4.69 % CO 0.23 % CO2 2.99 % H2S 60 – 70 ppm
Section 1. Pure UFCC, effect of H2 addition and pre-treatment
Four different feed compositions were analysed in this section to investigate the
COx hydrogenation of Ni-Mn/Al2O3 catalyst for pure UFCC as well as effect of hydrogen addition and pre-treatment of feed stream. Table Appa. 2 shows the feed compositions.
Table Appa. 2. Feed stream composition with different amounts of hydrogen added to the UFCC.
Feed Label Pre-treatment H2 Content UFCC Content Overall UFCC-00H N/A 0 % 100 % 100 % UFCC-10H N/A 10 % 90 % 100 % UFCC-25H N/A 25 % 75 % 100 % UFCC-25H-D MS-3A @ 0 °C 25 % 75 % 100 %
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 283 Appendix A
The change in the concentration over temperature for CO and CO2 are shown in
Figure Appa. 1 and Figure Appa. 2 respectively.
Figure Appa. 1. CO concentration for feed streams containing UFCC and different
additional H2 contents over Ni-Mn/Al2O3 catalyst (dashed lines show inlet concentrations).
Figure Appa. 2. CO2 concentration for feed streams containing UFCC and different
additional H2 contents over Ni-Mn/Al2O3 catalyst (dashed lines show inlet concentrations).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 284 Appendix A
Section 2. Used and reactivated catalyst
The used catalysts from UFCC experiments were reused in a hydrogen-rich feed stream containing CO, CO2, H2 and He (Table Appa. 3).
Table Appa. 3. The hydrogen-rich feed stream composition for analysing the
performance of used Ni-Mn/Al2O3 catalysts. Species Partial Pressure (KPa) H2 85.1 CO 2.03 CO2 0.68 He 11.5
Two sets of experiments performed with the used and reactivated Ni-Mn/Al2O3 catalyst. Table Appa. 4 shows the detail of each set of experiment.
Table Appa. 4. The experiment details and pre-run conditions for used and reactivated
Ni-Mn/Al2O3 catalysts. Set Label Catalyst Pre-run Conditions Used Ni-Mn/Al2O3 in UC N/A UFCC Used Ni-Mn/Al2O3 in Activated in H2 stream for UC-Act UFCC 2 hr at 500 °C
The change in the concentration over temperature for CO and CO2 on used and reactivated catalysts are shown in Figure Appa. 3 and Figure Appa. 4 respectively.
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 285 Appendix A
Figure Appa. 3. CO concentration for hydrogen-rich feed stream over used and
reactivated Ni-Mn/Al2O3 catalyst (dashed lines show inlet concentrations).
Figure Appa. 4. CO2 concentration for hydrogen-rich feed stream over used and
reactivated Ni-Mn/Al2O3 catalyst (dashed lines show inlet concentrations).
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 286 Appendix A
Highlights
Due to the deactivation effects of sulfur compounds (e.g. H2S) in UFCC
mixture, the Ni-Mn/Al2O3 catalyst did not show any hydrogenation activity
in UFCC feed.
Addition of hydrogen and pre-treatment of feed stream did not improve the
catalyst hydrogenation activity in UFCC feed.
The deactivated catalyst was reduced in hydrogen flow at 500 °C. The
reduction of used catalyst did not improve the used catalyst activity for CO
and CO2 hydrogenation in H2S-free stream.
By increasing temperature (above 300 °C), the concentration of CO2
decreased while CO concentration increased, which can be explained by
the reactions showed below (Eq. Appa.1 and 2).
Hydrocarbon + CO2 → CO + H2 Eq. Appa.1
H2 + CO2 → H2O + H2 Eq. Appa.2
Catalytic hydrogenation of CO and CO2 in the presence of light hydrocarbons 287