Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Methane decomposition and partial oxidation in a cyclic mode over supported Co and Ni catalysts Li, Jerry Kai Hsu 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-ubc_2006-0548.pdf [ 8.78MB ]
Metadata
JSON: 831-1.0058758.json
JSON-LD: 831-1.0058758-ld.json
RDF/XML (Pretty): 831-1.0058758-rdf.xml
RDF/JSON: 831-1.0058758-rdf.json
Turtle: 831-1.0058758-turtle.txt
N-Triples: 831-1.0058758-rdf-ntriples.txt
Original Record: 831-1.0058758-source.json
Full Text
831-1.0058758-fulltext.txt
Citation
831-1.0058758.ris

Full Text

Methane Decomposition and Partial Oxidation in a Cyclic Mode over Supported Co and Ni Catalysts by Jerry Kai Hsu Li B. Sc., Fu Jen University, 1994 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF APPLIED SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (CHEMICAL A N D BIOLOGICAL ENGINEERING) The University Of British Columbia September, 2006 © Jerry Kai Hsu L i , 2006 A b s t r a c t CH4 decomposition to produce C and H 2 , followed by partial oxidation of the C has been investigated in.a cyclic process over supported Co and N i catalysts. The catalysts are characterized by TPR, BET surface area, X R D and T E M . A tapered element oscillating microbalance (TEOM), used in the present study to monitor the reactions, operates as fixed bed micro-reactor that can measure the real time mass change of the catalyst during reaction. Product gases from the T E O M reactor were analyzed continuously by a quadrupole mass spectrometer. N i catalysts showed a better activity and stability compared to Co catalysts. At 773 K, the initial CH4 decomposition rate at 773 K of N i catalysts (1.23 x 10"5 mol/(g-s)) was faster than Co catalysts (9.44 x 10"6 mol/(g-s)). Steady H 2 production and initial CH4 decomposition rate were achieved for 6 cycles over N i catalysts, but complete deactivation was observed after the 2nd cycle over Co catalysts. A time delay of up to 2500 seconds occurred between the introduction of CH4 and the production of H 2 on Co catalysts but no such induction period occurred for N i catalysts. Catalysts can also be regenerated by carbon removal using C O 2 . The CO to H 2 ratio during C H 4 decomposition decreased compared to carbon removal with O 2 since formation of NiO was reduced. However, the rate of carbon removal by C O 2 was too slow for any practical application of the cyclic reaction. The H 2 production profile obtained during the first cycle of C H 4 decomposition on a freshly loaded N i catalyst was in qualitative agreement with that predicted by the deactivation model proposed by Zhang and Smith (2005). The H 2 production rate reached a maximum initially, decreased rapidly and then increased slowly before decreasing again. The statistical analysis of a fractional factorial design of experiments for the C H 4 decomposition, partial oxidation cyclic reaction indicated that higher H 2 production rate can be obtained by conducting the C H 4 decomposition at a higher temperature with a shorter duration. No significant effects were identified for selectivity to CO during oxidation and the overall quantity of CO produced during C H 4 decomposition. iii T a b l e o f C o n t e n t s Abstract ii Table of Contents iv List of Tables - . v i List of Figures vii Acknowledgements xii Chapter I Introduction 1 1.1 BACKGROUND l 1.2 MOTIVATION 2 1.3 OBJECTIVE OF THE PRESENT STUDY 4 Chapter II Literature Review 6 2.1 INTRODUCTION 6 2.2 C H 4 DECOMPOSITION ON SUPPORTED METAL CATALYST 8 2.2.1 Metal and Support Selection 8 2.2.2 Catalyst Promoters 11 2.3 CATALYST REGENERATION AND CYCLIC PROCESSES 14 2.4 SUMMARY OF LITERATURE REVIEW 18 Chapter III Experimental Methods and Analysis 20 3.1 INTRODUCTION 2 0 3.2 CATALYST PREPARATION 21 3.3 CATALYST CHARACTERIZATION 2 2 3.3.1 Temperature Programmed Reduction (TPR) 22 3.3.2 BET Surface Area 24 3.3.3 X-Ray Diffraction (XRD) 24 3.3.4 Transmission Electron Microscopy (TEM) and STEM EDX. 25 3.4 REACTOR AND ANALYSIS 2 5 3.4.1 TEOM 25 3.4.2 Experimental Set-up and Conditions 31 Chapter IV Supported Co Catalyst 35 4.1 INTRODUCTION 35 4.2 THE EFFECT OF M G O AS A PROMOTER ON THE DEGREE OF REDUCTION OF THE CO CATALYST . 36 4.3 CYCLIC PERFORMANCE OF THE SUPPORTED CO CATALYST 3 9 4.3.1 Hydrogen Production and Catalyst Deactivation 41 4.3.2 The Effects of Subsequent Partial Oxidation 46 4.3.3 CO, C02 Production, Carbon Removal Percentage and Selectivity to CO during Oxidation 51 4.3.4 CO to H2 Ratio and Induction Period during CH4 Decomposition 56 4.4 SUMMARY 6 0 Chapter V Supported Ni Catalyst 63 5.1 INTRODUCTION 63 5.2 THE DEGREE OF REDUCTION OF THE SUPPORTED N I CATALYST 64 5.3 CYCLIC PERFORMANCE OF THE SUPPORTED N I CATALYST 67 5.3.1 Hydrogen Production and Catalyst Deactivation 68 5.3.2 Induction Period 73 5.3.3 Possible Source of Solid Oxygen and CO, C02 Production during CH4 Decomposition 77 iv 5.3.4 Carbon Removal Percentage and Selectivity to CO during Oxidation 90 5.4 USING C 0 2 INSTEAD OF 0 2 TO REMOVE THE CARBON DEPOSIT 95 5.5 SUMMARY 102 Chapter VI Fractional Factorial Analysis on the Supported Ni Catalyst 104 6.1 INTRODUCTION 104 6.2 METHOD .' 105 6.3 METHANE DECOMPOSITION AND DEACTIVATION OF CATALYST 107 6.4 THE CHANGE OF N I PARTICLE SIZE AND THE ACTIVITY 118 6.5 STATISTICAL ANALYSIS 122 6.5.1 Hydrogen Production 123 6.5.2 CO Impurity during the CH4 Decomposition Step 125 6.5.3 CO Selectivity during Partial Oxidation 126 6.5.4 Carbon Removal Percentage 127 6.6 SUMMARY 129 Chapter VII Conclusions and Recommendations for Future Work 132 7.1 CONCLUSIONS 132 7.2 RECOMMENDATIONS FOR FUTURE WORK 134 References 137 Appendix A : Calibration for T C D Voltage 142 Appendix B: Sample Calculations 143 Appendix C: Calibration for Mass Flow Controller 149 Appendix D: Calibration for Prolab Mass Spectrometer 154 Appendix E: Error Analysis 156 Appendix F: Summary Results of Factorial Design Experiments 165 Appendix G: Brief Description of Experiment Operation 172 Appendix H: Activation Energy Measurement 176 v List of Tables Table III-1 Typical experimental parameters 33 Table IV-l Experimental parameters for supported Co catalyst 40 Table V-l Experimental parameters for supported Ni catalyst 67 Table V-2 Experimental parameters of Ni catalyst for comparison between deposited carbon removal by0 2 andC0 2 97 Table VI-1 The factors with low and high levels used in the factorial design 105 Table VI-2 A 25"1 fractional factorial design at two levels of the supported Ni catalyst 106 Table VI-3 Comparison of Ni particle size at different steps 118 Table VI-4 Results of BET surface area measurements 129 Table A-l Summary results of TPR of Cu 20 142 Table C-l Summary results of measured and setup flow rates for calibration of 5 mass flow controllers in different gases , 150 Table C-2 Summary results of calibration equation for mass flow controllers 153 Table E-l Summary of pure error for the measurements 156 Table G-l Summary of experiment parameters 175 vi L i s t o f F i g u r e s Figure 1-1 Conceptual dual-reactor design to perform stepwise partial oxidation 3 Figure 11-1 Current technology for hydrogen production 6 Figure III-1 Block diagram of the experimental set-up used for temperature programmed reduction. ..23 Figure 111-2 Block diagram of the major parts of the TEOM 26 Figure 1II-3 Picture of sample bed and one cent for relative size comparison 26 Figure III-4 The effect of temperature difference on mass reading at constant gas composition 28 Figure III-5 The effect of changing CH 4 concentration on mass reading at 823 K 29 Figure III-6 The effect of changing 0 2 concentration on mass reading at 823 K 30 Figure 1II-7 Block diagram of the experimental set-up used for in situ reduction, C H 4 decomposition and partial oxidation 32 Figure III-8 Flow chart of the sequence of CH 4 decomposition, partial oxidation and corresponding He flushing 34 Figure IV-1 The effect of MgO used to enhance the degree of reduction of cobalt oxide. The catalyst sample was heated at a rate of 10 K/min to 1007 K in 60 mL/min of the 10 % H2/90 % Ar reducing gas stream 37 Figure IV-2 XRD pattern comparison of the 12 wt. % Co-MgO/Al 20 3 catalyst precursor after calcination at 723 K in air and reduction at 1007 K in 10 % H2/90 % Ar reducing gas 38 Figure IV-3 Measurement of H 2 , CH 4 , CO, C0 2 and the mass of the supported Co catalyst during C H 4 decomposition (5 vol. % CH 4 , 773 K) in the 1st cycle of experiment Co-1 41 Figure IV-4 Initial CH 4 decomposition rate vs cycle numbers for supported Co catalyst for experiment Co-1, Co-2 and Co-3. All 3 experiments were performed at 773 K with 5 vol. % CH 4 in the C H 4 decomposition step and 5 vol. % 0 2 in the partial oxidation step. The only difference among the 3 experiments was the duration of each step 43 Figure IV-5 H 2 production during CH 4 decomposition for supported Co catalyst for experiment Co-1, Co-2 and Co-3. All 3 experiments were performed at 773 K with 5 vol. % CH 4 for the C H 4 decomposition step and 5 vol. % 0 2 for the partial oxidation step. The only difference among these 3 experiments was the duration of each step 44 Figure IV-6 H 2 production during C H 4 decomposition in the 1st cycle for experiment Co-1 and Co-3. Same experimental parameters was applied for both experiments during CH 4 decomposition 45 Figure IV-7 Water peak integrated area during in situ reduction for experiment Co-1, Co-2 and Co-3 46 Figure 1V-8 Measurements of CO, C0 2 , 0 2 and the mass of the catalyst during partial oxidation (5 vol. % 0 2, 773 K) in the 1st cycle for experiment Co-1 47 Figure IV-9 Initial oxidation rate of carbon deposit vs cycle numbers for supported Co catalyst of experiment Co-1, Co-2 and Co-3. All 3 experiments were performed at 773 K with 5 vol. vii % CH 4 in the CH 4 decomposition step and 5 vol. % 0 2 in the partial oxidation step. The only difference among the 3 experiments was the duration of each step 48 Figure IV-10 Comparison of H 2 , CH 4 , CO, C0 2 and the mass of catalyst during CH 4 decomposition (5 vol. % CH 4 , 773 K) in the 1st and 2nd cycle of experiment Co-1 (Different Y scales were applied for H 2 -CH 4 and CO-C0 2) 49 Figure IV-11 Measurement of CO, C0 2 , 0 2 and the mass of the catalyst during the partial oxidation (5 vol. % 0 2, 773 K) in the 1st cycle for experiment Co-1 with the expanded time scale 51 Figure IV-12 CO and C 0 2 production during partial oxidation on supported Co catalyst vs cycle number for experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % CH 4 for CH 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of partial oxidation 52 Figure IV-13 The correlation between the CO and C0 2 produced during partial oxidation and H 2 produced during the preceding CH 4 decomposition for experiment Co-1 and Co-3 53 Figure IV-14 Carbon removal fraction vs cycle number of experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % C H 4 for CH 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of partial oxidation 55 Figure IV-15 Selectivity to CO during partial oxidation of experiment Co-1 and Cb-3. Both experiments were performed at 773 K with 5 vol. % CH 4 for CH 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of partial oxidation 56 Figure IV-16 CO/H 2 during CH 4 decomposition of experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % CH 4 for C H 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of partial oxidation 57 Figure IV-17 Amount of CO produced during CH 4 decomposition vs cycle number for experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % CH 4 for CH 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of each step 58 Figure IV-18 Induction period during CH 4 decomposition vs cycle number for experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % CH 4 for C H 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of each step .60 Figure V-l Comparison of TCD voltage signal during TPR for 12 wt. % Ni-Ce0 2/MgO/Al 20 3 and Ce0 2/MgO/Al 20 3 65 Figure V-2 XRD pattern comparison for 12 wt. % Ni-Ce0 2/MgO/Al 20 3 after calcination and reduction 66 Figure V-3 Initial C H 4 decomposition rate vs cycle number for the supported Ni catalyst of experiment Ni-4, Ni-5, Ni-6, and Ni-7 69 Figure V-4 Comparison of H 2 molar flow rate during the 1st cycle of CH 4 decomposition of experiment Ni-4, Ni-5, Ni-6 and Ni-7 70 viii Figure V-5 H 2 production during CH 4 decomposition vs cycle number for the supported Ni catalyst of experiment Ni-4, Ni-5, Ni-6, and Ni-7 71 Figure V-6 Comparison of H 2 , CH 4 , CO, C0 2 , water and mass during CH 4 decomposition for the first 3 cycles of experiment Ni-4 74 Figure V-7 Molar flow rate of H 2 and CH 4 during CH 4 decomposition of experiment Ni-7 75 Figure V-8 Molar flow rate of H 2 and CH 4 during CH 4 decomposition of experiment Co-1. Induction period is required during CH 4 decomposition from the 2nd cycle for supported Co catalyst 76 Figure V-9 XRD pattern comparison for 12 wt. % Ni-Ce0 2/MgO/Al 20 3 after partial oxidation (10 % 0 2, 773 K) and CH 4 decomposition (5 % CH 4 , 773 K) 77 Figure V-10 The comparison of adjusted mass change of reduced 12 wt. % Ni-Ce0 2/MgO/Al 20 3 and 12 wt. % Ni-MgO/Al 20 3 during the TPO : 80 Figure V- l 1 The linear fit to show the correlation between the overall increase in mass during TPO and the corresponding increase in mass due to the oxidation of Ni according to the Ni content of the catalyst 81 Figure V-l 2 The linear fit to show the correlation between the Ni content of the catalyst and the initial mass increasing due to oxygen uptake 83 Figure V-l 3 Comparison of oxygen (moles) obtained between NiO and CO, C 0 2 during CH 4 decomposition for each cycle of experiment Ni-4, Ni-5, Ni-6 and Ni-7 84 Figure V-14 Carbon removal percentage vs cycle number of experiment Ni-4, Ni-5, Ni-6 and Ni-7....86 Figure V-l 5 Correlations between the amount of oxygen obtained from CO and C 0 2 during CH 4 decomposition and the accumulated deposited carbon of Ni-4, Ni-6 and Ni-7 88 Figure V-l6 Correlations between the initial CH 4 decomposition rate and the accumulated deposited carbon of Ni-4, Ni-6 and Ni-7 89 Figure V-l 7 Selectivity to CO during partial oxidation of experiment Ni-4, Ni-5, Ni-6 and Ni-7 91 Figure V-18 Comparison of molar flow rate of CO, C 0 2 and 0 2 during partial oxidation of cycle 1, 3, and 5 of experiment Ni-5 and Ni-6 92 Figure V-l 9 Comparison of initial oxidation rate of deposited carbon during partial oxidation of each cycle of experiment Ni-4, Ni-5, Ni-6 and Ni-7 94 Figure V-20 Comparison of molar flow rate of CO, C0 2 and 0 2 during partial oxidation of cycle 1, 2, and 3 of experiment Ni-7 95 Figure V-21 Comparison of CO to H 2 ratio (in percentage) during CH 4 decomposition from the 2nd to 6th cycle of experiments Ni-4, Ni-5, Ni-6 and Ni-7 96 Figure V-22 XRD pattern comparison of the Ni catalyst after carbon removal by 0 2 and C 0 2 98 Figure V-23 Comparison of CO to H 2 ratio (in percentage) during C H 4 decomposition from the 2nd to 5th cycle of experiment Ni-8-02 and Ni-9-C02 99 ix Figure V-24 Comparison of carbon removal percentage of each cycle of experiment Ni-8-02 and Ni-9-C 0 2 100 Figure V-25 Comparison of initial CH 4 decomposition rate of experiment Ni-8-02 and Ni-9-C02 101 Figure V-26 Comparison of carbon removal rate of experiment Ni-8-02 and Ni-9-C02 101 Figure VI-1 H 2 molar flow rate indicating an unique feature during the 1st CH 4 decomposition (5 % CH 4 , 723 K and 823 K) 109 Figure VI-2 Filamentous carbon formation on supported Ni catalyst after CH 4 decomposition (5 % CH 4 , 773 K) 110 Figure VI-3 Fringes observed on filamentous carbon on supported Ni catalyst after CH 4 decomposition (5 % C H 4 , 773 K) 110 Figure VI-4 H 2 molar flow rate during CH 4 decomposition to show the activity of experiment FD-4 (5 % CH 4 , 723 K, 50 minutes) 111 Figure VI-5 H 2 molar flow rate during CH 4 decomposition to show the activity of experiment FD-3 (5 % CH 4 , 823 K, 50 minutes) 111 Figure VI-6 Flow chart of the sequence of CH 4 decomposition and partial oxidation (He flush steps are not included) for 6 cycles 113 Figure Vl-7 (a) Carbon recovery percentage during oxidation, (b) Initial CH 4 decomposition rate, (c) H 2 activity profiles during CH 4 decomposition for all 6 cycles of experiment FD-11 (5 % CH 4 , 723 K for 50 minutes, 2.5 % 0 2, 723 K for 3 minutes), (d) H 2 activity profiles during CH 4 decomposition for all 6 cycles during the first 6 minutes 115 Figure VI-8 (a) Carbon recovery percentage during oxidation, (b) Initial CH 4 decomposition rate, (c) H 2 activity profiles during CH 4 decomposition for all 6 cycles of experiment FD-9 (5 % CH 4 , 823 K for 50 minutes, 2.5 % 0 2, 823 K for 3 minutes), (d) H 2 activity profiles during CH 4 decomposition for all 6 cycles during the first 6 minutes 117 Figure VI-9 TEM images show the change of Ni particle size after reduction, after C H 4 decomposition and after oxidation 119 Figure VI-10 H 2 production as a function of cycle number of experiment FD-4 (5 % CH 4 at 723 K for 50 minutes and 2.5 % 0 2 at 823 K for 15 minutes) and FD-10 (5 % CH 4 at 823 K for 50 minutes and 20 % 0 2 at 723 K for 3 minutes) 122 Figure VI-11 Initial CH 4 decomposition rate as a function of cycle number, (a) Decomposition of CH 4 at 823 K for 3 minutes, (b) Decomposition of C H 4 at 723 K for 3 minutes, (c) Decomposition of CH 4 at 823 K for 50 minutes, (d) Decomposition of CH 4 at 723 K for 50 minutes 123 Figure VI-12 H 2 production profile during the first CH 4 decomposition for FD-10 (decomposition of CH 4 at 823 K) and FD-4 (decomposition of CH 4 at 723 K). The duration for both experiment are 50 minutes 124 Figure VI-13 H 2 production rate as a function of cycle number, (a) Decomposition of CH 4 at 723 K for 50 minutes, (b) Decomposition of CH 4 at 823 K for 3 minutes 125 x Figure VI-14 Normal probability plot with CO/H 2 as the response. 5 main effects and 10 two-factor interaction effects are all insignificant 126 Figure VI-15 Plots generated from statistical analysis when carbon removal percentage was chosen to be the response, (a) normal probability plot, (b) main effects plot and (c) interaction effects plot , 128 Figure A-l TCD voltage and temperature vs time for TPR of Cu 20 142 Figure C-l Calibration curve for MFC 1 (He), MFC2 (5 % CH4/Ar) and MFC2 (CH4) 151 Figure C-2 Calibration curve for MFC2 (4.82 % H 2 , 10.4 %CH 4 , 2.01%C2H4, 3.75%C2H6/Ar), MFC3 (H2)and MFC4 (02) 152 Figure C-3 Calibration curve for MFC5 (He) 153 Figure D-l Calibration curves for CH 4 , C0 2 , CO and H 2 of Prolab mass spectrometer 155 Figure F-1 Summary results of total H 2 production during CH 4 decomposition for FD series 165 Figure F-2 Summary results of total CO production during CH 4 decomposition for FD series 166 Figure F-3 Summary results of total C0 2 production during CH 4 decomposition for FD series 167 Figure F-4 Summary results of total CO production during partial oxidation for FD series 168 Figure F-5 Summary results of total C0 2 production during partial oxidation for FD series 169 Figure F-6 Summary results of average H 2 production rate during CH 4 decomposition for FD series 170 Figure F-7 Summary results of initial C H 4 decomposition rate for FD series 171 Figure H-l Activation energy measurement of C H 4 decomposition on Co catalysts 176 Figure H-2 Activation energy measurement of CH 4 decomposition on Ni catalysts 177 xi Acknowledgements I would like to thank Dr. Kevin Smith, my supervisor, for giving me the opportunity to participate in this project. I first met him in May 2001 while I was working as a research scientist in a nanotechnology company. I often consulted with Dr. Smith and obtained valuable knowledge and support. My respect for Dr. Smith's expertise intensified during these interactions, eventually leading me to enroll for graduate studies under his supervision at UBC. Both Dr. Smith's knowledge and research skills have advanced me. It is really a great experience to work for him. His tremendous patience is highly appreciated, especially while writing my thesis. I would also like to thank NSERC for financial support. Dr. C. C. Y u deserves credit for initial experiment setup and integration. I would also like to express my deepest appreciation to my mother and father, Chao Chih Yu and Kwang Chi L i . With their endless love and support, I have been able to concentrate on my studies without having to worry about trivial things. There is no doubt that they will continue to support me if I intend to further my studies. But now it is my turn to contribute and spend more time with them. Jessica Liu, my girl friend, as a fantastic source of cheer and care throughout my entire studies even though she was mostly overseas during the past four years. With her remote encouragement, the stress of research was relieved. x i i Chapter I Introduction 1.1 Background The demand for hydrogen is growing due to an increased production of ammonia, syngas, and methanol. More importantly, petroleum refinery industries, especially those processing Canadian heavy oils require more H 2 to reduce the sulfur levels of fuels to meet strict environmental regulatory requirements. Proton exchange membrane (PEM) fuel cells also use high purity H 2 and O 2 to produce electrical power through electrochemical processes and increased use of fuel cells is expected to further increase demand for H 2 . However, H 2 is not a resource. Production of high purity H 2 at low cost is a key requirement for further development of the H 2 economy. . Current technologies for H 2 production include steam reforming of methane (SRM), partial oxidation of methane (POX), and C O 2 reforming of methane. S R M is an endothermic reaction in which C H 4 reacts with steam to produce CO and/or C O 2 and H 2 . SRM is normally carried out at high temperature (900-1100 °C) and high pressure (20-30 atm) and produces the greatest number of moles of H 2 per mole of C H 4 . POX involves reacting C H 4 with pure O 2 to produce H 2 and CO with a ratio 1.6-1.8 to 1. This ratio is suitable for the Fischer-Tropsch process, which converts syngas to long-chain hydrocarbons. However, expensive air separation is required to provide pure O 2 for POX. Besides, the exothermic POX reaction needs to be controlled very carefully to prevent thermal runaway. C O 2 reforming of C H 4 is an endothermic reaction that also produces syngas. It is attractive since this process converts two major greenhouse gases into more 1 valuable H 2 and CO. Nevertheless, the deactivation caused by excessive carbon formation on the surface of the catalyst means that C O 2 reforming of C H 4 has not been commercialized. Catalytic decomposition of C H 4 into C and H 2 is an alternative route to H 2 production. Since only H 2 and carbon are formed in the decomposition process, purification of H 2 is not required. High purity H 2 without CO contamination can be directly used for P E M fuel cells. However, large amounts of carbon deposit are produced during this endothermic reaction. 1.2 Motivation Among the technologies for H 2 production, stepwise partial oxidation has recently drawn attention. The stepwise partial oxidation involves two steps. The first step is the catalytic decomposition of C H 4 to produce H 2 and carbon deposit. The second step regenerates the catalyst by introducing O 2 or steam to remove the deposited carbon and control the selectivity toward CO rather than C O 2 . Theoretically, the second exothermic reaction step provides enough energy for the first endothermic reaction step of the two-step process. By repeating the stepwise partial oxidation, i.e., CH4 decomposition and partial oxidation in a cyclic mode, H 2 can be produced and deposited carbon can be utilized to provide energy and valuable CO gas. The catalyst can be regenerated and reused. Following this concept, Figure 1-1 shows a conceptual dual-reactor system that would perform such a stepwise partial oxidation of CH4. 2 H 2 •4 Regenerated catalyst C O • C H 4 Decomposition Partial Oxidation C H 4 Catalyst with carbon deposit on surface Figure 1-1 Conceptual dual-reactor design to perform stepwise partial oxidation There are some additional benefits of the proposed reaction sequence. Since there is no C H 4 and O 2 co-fed, further oxidation of H 2 to H 2 O is avoided. Furthermore, in principle the catalyst is regenerated during the oxidation step so that the requirement to replace the catalyst can be reduced. The concept shown in Figure 1-1 allows for flexibility in application; either high purity H 2 for P E M fuel cells or synthesis gas for gas-to-liquid (GTL) transformation. Although C H 4 is used as a feedstock in the present study, other hydrocarbons could also be used, including hydrocarbons derived from biomass. Note that the hydrogen can not be considered a true "green" fuel unless the feedstock is derived from a sustainable source. 3 There are a number of undesired reactions that may cause challenges for the concept outlined herein. The spillover of carbon ( C H 4 —* C(s> + 2 H 2 ) from the metal catalyst to the support may result in the presence of other components in the H 2 . For example, the spillover carbon may react with solid oxygen from the support to produce CO or C O 2 . Also, the deposited carbon may not be completely removed during the partial oxidation step resulting in deactivation of the catalyst. Controlling the partial oxidation will also be important to minimize oxidation of the metal catalyst and to ensure high selectivity to CO during the partial oxidation step. A kinetic model of C H 4 decomposition on supported Co catalysts was reported recently by Zhang and Smith (2005). The catalyst activity and deactivation associated with filamentous and encapsulating carbon formation were presented. The model predicts a H 2 production profile that rapidly reaches a maximum initially, followed by a rapid decrease and then a slow increase. In that study, however, gas chromatography was used for analysis. Because of the low sampling rate associated with gas chromatography, some of the model predictions could not be verified. 1.3 Objective of the Present Study The goal of the present work is to investigate the stability of supported metal (Co and Ni) catalysts during the stepwise H 2 production by decomposition of C H 4 ( C H 4 —•> C + 2 H 2 ) and partial oxidation of carbon ((n+m)C + (Vin + m)02 —• nCO + OTCO2). Deactivation of the catalyst, CO and C O 2 distribution and carbon removal efficiency are the main response variables of interest. Moreover, the effects of operating conditions are also 4 examined in order to obtain the optimal conditions for H 2 production and CO/ C O 2 selectivity control. The quadrupole mass spectrometer used in the present study provides a continuous product gas analysis. Therefore, the data collected during C H 4 decomposition can be used to verify the features of the kinetic model proposed previously by Zhang and Smith (2005). The main objectives of the present study are: 1. To demonstrate the production of H 2 by the stepwise partial oxidation of C H 4 ; 2. To investigate the cyclic ability of the supported metal catalyst; 3. To evaluate the H 2 production rate and to quantify the presence of impurities during C H 4 decomposition; 4. To determine the optimal operating conditions for maximum H 2 production rate and CO selectivity during the partial oxidation step; 5. To verify the main features of the kinetic model of deactivation of supported metal catalysts during C H 4 decomposition reported previously by Zhang and Smith (2005). 5 Chapter II Literature Review 2.1 Introduction The hydrogen economy is developing fast due to an increasing demand for H 2 in modern world economics. According to Armor (2005), forecasts suggest that future H 2 needs exceed the world's current production by a factor of 14. Issues such as storage, delivery and production of hydrogen, are receiving more attention in order to boost world production. Among the different techniques to produce H 2 shown in Figure II-1, reforming of natural gas ( C H 4 ) to produce H 2 is a mature technology for near-term purposes. Current Technology for Hydrogen Production Thermal Processes Electrolytic Processes Photolytic Processes Reforming of Natural Gas Gasification of Coal Photobiological Water Splitting Photoelectrochemical Water Splitting! Gasification of Biomass { Reforming of Renewable Liquid Fuels High-Temperature Water Splitting Information adapted from U. S. Department of Energy http://www.eere. energy.gov/hydrogenandfuelcells/production/current_technology. html Figure II-l Current technology for hydrogen production 6 There are two processes for reforming of natural gas. Steam methane reforming (SRM)) is the most important technology for H 2 production currently. In SRM, methane reacts with steam at 900-1100 °C and 20-30 atm using an alkali-promoted N i catalyst supported on alumina to produce CO and H 2 . Subsequently, CO and steam react via the water-gas shift reaction to produce extra hydrogen. The reactions of S R M are listed below. C H 4 + H 2 0 -»• CO + 3 H 2 AH°298 = 205.9 kJ/mole (2.1) CO + H 2 O —> C O 2 + H-2 AH°298=-41.2 kJ/mole (2.2) Reaction (2.1) is highly endothermic and therefore high energy input is required. According to the U.S. Department of Energy, about 9 5 % of the H 2 produced today in the United States is made via SRM. The partial oxidation of methane (POX) is a process that reacts methane with oxygen and in which the amount of O 2 is less than the stoichiometric quantity required to produce CO and H 2 . The reaction of POX is listed below. C H 4 + V2O2 -> CO + 2 H 2 AH°298 = - 35 .9 kJ/mole (2.3) This exothermic reaction is a much faster process than SRM. However, less hydrogen per methane is produced compared to SRM. Normally, an oxygen plant is required to provide pure O 2 in order to avoid an expensive N 2 separation process downstream. 7 Thermocatalytic decomposition of C H 4 into C and H 2 is another attractive alternative for H 2 production. The major advantage of catalytic decomposition of C H 4 is that H 2 is the only gaseous product according to the following reaction: C H 4 — C(s) + 2H 2 AH°298 = 75.6 kJ/mole (2.4) In principle, the need for a H 2 purification step is reduced since the carbon is a solid deposit on the catalyst and no CO is produced. Since the majority of the deposited carbon is nano-structured, it can be further utilized in many applications that require nano-carbon materials. Alternatively, the carbon deposit can be reacted with O 2 to produce heat, part of which could be utilized to drive the decomposition reaction ( C(s) + O 2 — C O 2 AH°298= -3 93.5 kJ/mole). 2.2 CH4 Decomposition on Supported Metal Catalysts The decomposition of CH4 on metal catalysts has been studied extensively. Topics such as CH4 homologation (Boskovic and Smith, 1997; Zadeh and Smith, 1999) and kinetics of CH4 decomposition on supported Co catalysts (Zadeh and Smith, 1998; Zhang and Smith, 2002; Zhang and Smith, 2005) have been discussed in the literature. Catalytic decomposition of methane to hydrogen has drawn more attention in the past few years. 2.2.1 M e t a l a n d S u p p o r t Select ion C H 4 dissociation on different transition metals has been widely studied. Solymosi et al. (1994) showed that the decomposition of C H 4 to produce H 2 occurred over supported Pd catalysts at 473 K. Matsui et al. (2000) reported that dissociative adsorption of C H 4 occurs at temperatures above 373 K on Ru and Rh catalysts. 8 Liao and Zhang (1998) investigated the dissociation of C H 4 on a number of transition metals (Ru, Rh, Ir, N i , Pd, Pt, Cu, Ag, Au) and indicated that Rh and N i were the most efficient according to the calculated total dissociation enthalpy. Boskovic and Smith (1997) reported that C H 4 decomposition was initiated at 553 K on reduced Co/SiC>2 and at 623 K on reduced C 0 / A I 2 O 3 . Zhang and Amiridis (1998) investigated the catalytic cracking of C H 4 over silica-supported N i catalysts at 823 K. H 2 was the only gaseous product detected. Zadeh and Smith (1998) performed CH4 decomposition at 723 K on supported Co catalysts prepared by incipient wetness impregnation with high surface area silica gel as a support. They verified that the extent of metal support interaction increased by decreasing the Co loading. A catalyst with 12 wt. % Co loading achieved highest extent of reduction (TPR in H 2 , 91 mol. %) and reasonable C H 4 decomposition activity (20.8 % conversion o f 5 % C H 4 in Arat450 °C). Wang and Ruckenstein (2002) investigated the carbon formation during C H 4 decomposition at 900 °C over Co/MgO catalysts. However, their focus was the carbonaceous deposits instead of the activity of C H 4 decomposition. Aiello et al. (2000) performed the catalytic cracking of C H 4 over a 15 wt. % Ni/SiC>2 at 923 K. Choudhary et al. (2001) also performed C H 4 decomposition at a range of 723 -823 K on various Ni-supported catalysts to produce H 2 . Piao et al. (2002) performed C H 4 decomposition on alumina supported N i aerogel catalyst between 450 and 700 °C. The 9 catalysts prepared by using a sol-gel technique with supercritical drying can achieve a high surface area (294-375 m2/g). Zhang and Smith (2002) reported the effects of Co dispersion on the kinetics of CH4 decomposition on Co/SiC>2. In the range of 5 - 13 % Co dispersion, the initial decomposition activity increased with decreasing metal dispersion and the deactivation rate decreased with decreasing metal dispersion. The study also indicated that 12 wt. % Co catalyst, reduced at 923 K, yielded 5 % Co dispersion and this catalyst showed the highest initial decomposition activity and slowest deactivation rate among the catalysts examined. Chen et al. (2004) used a coprecipitated Ni-Cu-alumina catalyst for direct decomposition of CH4 at 773-1013 K to produce COx-free H_. The doping with Cu enhanced the catalyst stability. Takenaka and co-workers (Takenaka et al. 2002, 2003, and 2004; Otsuka et al. 2003 and 2004) reported C H 4 decomposition over N i and Co on different supports (AI2O3, SiC>2, TiO_, and carbon fibers). Even though the activity varied depending on the support, Ni-based and Co-based catalysts showed a significant activity for CEL, decomposition at a temperature range 773 - 823 K. Noble metals appear to be able to dissociate CH4 at a relatively low temperature. However, other metals such as N i and Co with lower cost are typically considered a more practical choice even though a higher CH4 dissociation temperature is required. 10 Alumina (AI2O3) and silica (Si02) are commonly used as a support for catalysts due to their high surface area and thermal stability. Aparicio et al. (1997) performed the decomposition of C H 4 at 623 K on different metal catalysts (Co, N i , Ru, Rh, Ir, Pt) supported on AI2O3 and Si02- They reported that alumina supported metals had greater ability to dehydrogenate CH4 than the silica supported catalysts. Boskovic and Smith (1997) examined the effect of catalyst support on the activity of Co catalysts. At the same reaction temperature and reaction time, the Co catalyst supported on AI2O3 achieved higher initial CH4 decomposition rate and conversion than the Co on SiO_ catalysts. Takenaka et al. (2004) also investigated the catalytic C H 4 decomposition over Co catalyst on different supports (MgO, AI2O3, SiCh, and Ti02). The Co supported on AI2O3 and MgO showed a superior catalytic performance than TiC>2 and SiO_ for C H 4 decomposition at 773 K. Murata et al. (2005) also found that C H 4 decomposition over AI2O3 supported N i catalysts was more active and stable than Ni/SiC>2. Based on the studies summarized herein, N i and Co were chosen as the active metal with alumina as the support for the present studies. 2.2.2 Catalyst Promoters In general, alumina supported transition metal (Ni or Co) catalysts are often prepared by impregnating alumina with corresponding metal nitrate, followed by drying to remove water and calcining at temperature to decompose the metal nitrate to the metal oxide. Chen and Zhang (1992) suggested that the decomposition of the nitrate and formation of amorphous metal oxide on the surface was followed by diffusion of the metal ions into 11 the surface lattice of alumina, resulting in a strong metal support interaction (MSI) and irreducible metal ions. The interaction between the metal oxide precursor and the support has been considered a key point for the activity of supported catalysts. Chernavskii et al. (2000) indicated that MgO was used to modify the AI2O3 surface in order to reduce the interaction markedly between cobalt oxide and the support. Hence, by using MgO as a promoter to modify the alumina, the interaction between metal oxide precursor and the support can be reduced and the reducibility of active metal phase can be enhanced. Ce02 is an oxidation catalyst with high oxygen storage ability. The high oxygen storage capacity is due to the high reducibility of Ce 4 + . Loong and Ozawa (2000) reported that oxygen release and intake associated with the conversion between the 3+ and 4+ oxidation state of the Ce ions provide the oxygen storage capacity. Furthermore, the O2" associated with Ce02 is very mobile. Otsuka et al. (1998) presented the direct partial oxidation of C H 4 by Ce02 to produce a synthesis gas with H2/CO ratio of 2. The authors claimed that CO must be produced by the reaction of the carbon with the lattice oxygen of Ce02. Pantu et al. (2000) performed partial oxidation of CH4 over Pt and Ru supported on Ce02-Zr02 in the absence of gaseous oxygen. The oxygen supplied for the oxidation of 12 C H 4 was derived from the reduction of CeCh. Results indicated that a high selectivity (100 %) to CO was achieved. Zhu et al. (2001) investigated the catalytic partial oxidation of CH4 to synthesis gas over Ni/Ce02. Data showed that a high CO selectivity (84 %) was observed when the reaction was performed at 650°O. They also concluded that the oxygen from Ce02 can transfer to the N i interface and oxidize any carbon species produced by C H 4 dissociation on Ni . Roh et al. (2001) performed the catalytic partial oxidation of C H 4 over Ce02 directly. Ce02 catalyst was reduced first before introducing the CHVO2 reactant gas mixture. With a suitable ratio of CH4/O2, Ce02 exhibited oscillatory catalytic behavior and consumed 100 % of O2 with high CO selectivity (75 - 77 %). They proposed a mechanism in which the oxygen vacancies in the reduced Ce02 were supplied with gaseous reactant O2, and then activated adsorbed oxygen, followed by release of the active oxygen species to react with C H 4 to produced H2 and CO. Dong et al. (2002) studied the partial oxidation of C H 4 by a sequential pulse experiment over Ni/Ce02. They indicated that C H 4 dissociated on N i and the resultant carbon species quickly migrated to the interface of Ni-Ce02, and then reacted with lattice oxygen of C e 0 2 to form CO. Burno-Lopez et al. (2005) studied the effect of Ce02 in catalyzed soot oxidation using labeled oxygen. The data showed that the gas-phase labeled oxygen replaced the non-13 labeled lattice oxygen of CeCh, creating highly active oxygen. This highly active non-labeled oxygen reacted with soot to generate CO and CO2. Data showed that the selectivity to CO increased as the reaction temperature increased but CO2 was still the main carbon-containing product. Several other researchers have indicated that the addition of Ce02 helps the stability of catalysts. X u et al. (1999) indicated that adding Ce02 and MgO as promoters for Ni/Al203 during CO2 reforming of CH4 can increase N i dispersion and improve the resistance to carbon deposition. Hence, the stability of the catalyst can be improved. Jiang et al. (2003) also reported that CH4 decomposition over N i / a- AI2O3 was promoted by Ce02. Ce02 acted as an electronic promoter and could inhibit aggregation of the N i particles. Therefore, MgO and Ce02 were both chosen as promoters in the present study. MgO was used to enhance the reducibility of the metal oxide precursor. Gaseous oxygen can be stored in Ce02 as lattice oxygen and high selectivity to CO can be obtained by the reaction between carbon and lattice oxygen. Hence, selectivity to CO should be enhanced by adding Ce02 to the catalyst. 2.3 Catalyst Regeneration and Cyclic Processes In order to restore the activity of the catalyst after C H 4 decomposition, the catalyst needs to be regenerated by removing the carbon deposit in order to recover the active metal sites. Poirier and Sapundzhiev (1997) proposed a concept whereby natural gas is decomposed over a catalyst to produce hydrogen and carbon is deposited on the catalyst. 14 The deactivated catalyst can be regenerated by burning the deposited carbon with air. However, several studies indicated that the deposited carbon can be removed by burning with O 2 or gasification with C O 2 or steam, as well. Zhang and Amiridis (1998) performed the regeneration of a supported N i catalyst which had been used for CH4 decomposition. The catalysts were regenerated by removing carbon filaments by either oxidation in air or steam gasification. The authors claimed that both methods appeared to be able to fully restore the activity of the catalyst and the oxidation process was faster than the steam gasification. Choudhary and Goodman (1999) demonstrated a stepwise process for the production of CO-free hydrogen on supported N i catalyst at 650 K. The surface carbon produced by the catalytic decomposition of CH4 in the first step was quantitatively removed by steam in the second step. The catalyst was maintained at 648 K under flowing H 2 between cycles and a total of 13 cycles were completed. Hence, the catalyst was reduced during the H 2 flush period before the start of each CH4 decomposition step. Choudhary et al. (2001) examined H 2 production via catalytic decomposition of C H 4 over Ni-supported catalyst. After the CH4 decomposition at 723 K for 60 minutes, the catalyst was regenerated in air at 723 K for 30 minutes, followed by reduction with 50 % H 2 . There was no drop in activity of the catalyst through 12 cycles. However, similar experiments performed at slightly higher temperature (773 K) showed a gradual decrease in activity with every cycle. Unlike most of researchers who have claimed that CO-free 15 H_2 is produced during catalytic decomposition of CH4, these authors reported the formation of low levels of CO (ppm level) due to the interaction of surface carbon with the support. Fathi et al. (2000) also reported that the regeneration of the used catalysts after the partial oxidation of CH4 with O2 gave complete removal of carbon as CO with high selectivity. Regeneration with CO2 gave incomplete carbon removal. Aiello et al. (2000) investigated the deactivation and regeneration of Ni /Si02 for H 2 production via direct cracking of CH4. CH4 was decomposed at 923 K for 3 hours followed by steam regeneration at 923 K for up to 6 hours for 10 successive cycles. Data showed that the conversion of C H 4 was stable as a function of time on stream. Otsuka et al. (2004) investigated the catalytic performance of N i catalysts on different supports for the decomposition of CH4 followed by the oxidation of the deposited carbon nanofibers with O2. Based on their estimation, no energy input is required for the two-step process in which CH4 decomposition is followed by the oxidation of the deposited carbon with O2 and CO2. For N i (10 wt. %)/Ai203 catalyst, their results indicated that the conversion rate of CH4 improved remarkably from the first to the third cycle. After the third cycle, the results were reproducible and a total of 5 cycles was completed. They also claimed that the catalysts did not play any role in the reaction for the oxidation of carbon nanofibers with oxygen. 16 Takenaka et al. (2003 and 2004) performed the C H 4 decomposition and the subsequent gasification of deposited carbon with C O 2 over supported N i catalysts. Their data indicated that the specific H 2 yield (H2/N1) increased with the repeated cycles over N i (5 wt. %) /A l203 . They related the catalytic activity for the C H 4 decomposition to the N i metal particle size by investigating the structural change of N i catalysts on different supports (AI2O3, TiC»2, and SiC^). For N i / A ^ C ^ during the repeated reactions, the N i species were gradually reduced, forming fine Ni metal particles (< 20 nm), and these fine metal particles aggregated into larger ones (diameter 60 to 100 nm) which were the most effective for C H 4 decomposition. Most of the studies listed above focused on complete carbon removal by different methods and presented data that suggested that the activity of the catalyst can be fully recovered or enhanced. However, information about the stability of the catalysts during a cyclic process of C H 4 decomposition and partial oxidation is limited. Also, the interaction between the C H 4 decomposition step and the regeneration step remains unclear. The CO content in the H 2 stream during C H 4 decomposition is also a crucial issue but most of researchers have claimed that CO-free H 2 was produced. Plug flow, fixed bed reactors have been used by many researchers to perform activity measurements of C H 4 decomposition and gas chromatography (GC) was widely used as an analytical method for product gas analysis in these studies. Activity measurements of C H 4 decomposition were often presented as conversion of C H 4 versus time on stream instead of H 2 production versus time on stream. Some researchers used 17 thermogravimetric (TGA) methods to monitor the mass change of the catalyst due to the carbon deposition during C H 4 decomposition. Ideally, the reactivity should be determined by simultaneously monitoring the product gas composition and the mass change. However, simultaneous presentation of mass change of the catalysts due to carbon deposition (or removal) and product gas analysis has never been reported in any of these studies. Furthermore, the time required to complete one product gas analysis by GC normally takes several minutes. Activity data during the first few minutes of operation are, therefore, not available by GC. An analytical instrument providing very short analysis intervals, such as a qiiadrupole mass spectrometer, must be considered in order to acquire the complete activity profile. 2.4 Summary Ni and Co were selected for the present study of CH4 dissociation since they both show a significant activity for C H 4 decomposition to carbon and H 2 between 723 and 823 K. Alumina is chosen as the support since alumina supported catalysts have proven to be more active and stable during C H 4 decomposition. By adding MgO to modify the surface of the alumina support, the interaction between the metal oxide precursor and the alumina support is expected to be reduced. Hence, the degree of reduction of the metal catalysts, which is normally related to activity, can be enhanced. 18 CeC>2 is used as a promoter due to its high oxygen storage ability in which Ce02 can adsorb and convert the gaseous oxygen into lattice oxygen. As a result, selectivity to CO could be enhanced. Among several methods to regenerate the metal catalyst which have been investigated, oxidation of deposited carbon seems a fast process that can achieve complete regeneration. However, information about the interaction between the CH4 decomposition step and the regeneration step and stability of the catalysts during the cyclic process of CH4 decomposition and partial oxidation remains unclear and needs further investigation. 19 Chapter III Experimental Methods and Analysis 3.1 Introduction In this chapter, the details of the experimental methods used for catalyst preparation, catalyst characterization and evaluation of the catalyst cyclic performance are presented. Section 3.2 illustrates preparation of the supported Co and N i catalysts by stepwise incipient wetness impregnation. MgO is used as promoter to minimize the interaction between AI2O3 and the metal oxide precursor. Ce02 is also used as promoter due to its high oxygen mobility as O ". Methods of catalyst characterization including temperature programmed reduction (TPR), BET surface area, X-ray diffraction (XRD) and transmission electron microscopy (TEM) are described in Section 3.3. Results of catalyst characterization are presented in the corresponding section. The tapered element oscillating microbalance (TEOM) used in the present study is a fixed bed micro-reactor which can measure the real time mass change of the catalyst during reaction. Section 3.4.1 provides theoretical details and operating principles of this instrument. The mass change of the catalyst sample can be measured at the same temperature and constant gas composition. Section 3.4.2 illustrates the experimental set-up and typical operating procedures used to evaluate the cyclic performance of the catalyst during CH4 decomposition followed by partial oxidation of the deposited carbon. The calcined catalyst was first reduced in situ to convert metal oxide to metal before the 20 cyclic performance evaluation. In the present work, a single cycle refers to the decomposition reaction of CH4 (CH 4 — C + 2H2) followed by partial oxidation of the carbon ((n+m)C + (Vin + m)02 —> nCO + mCOj) . The number of cycles in the present work was between 4 and 6. The product gases were analyzed by mass spectrometry. By combining the T E O M and mass spectrometry, mass change of the catalyst and the product gas analysis wereobtained continuously. 3.2 Catalyst Preparation The supported Co and N i catalysts were prepared by stepwise incipient wetness impregnation. Pre-dried (373 K for 8 hours) y-alumina (Sasol Germany GmbH, +95 %) with a surface area of 211 m Ig and pore volume of 0.814 mL/g was used as the support. The y-alumina support was impregnated with an aqueous solution of Mg(N03)2 '6H20 (BDH Chemical Ltd., 99 %) at a concentration of 1.10 M[Mg]. After impregnation, the sample was dried in air at 373 K for 2 hours, and then calcined in air at 1073 K for 8 hours. The modified alumina support (MgO/A^Os) was obtained after calcination. The MgO loading was 6.50 wt. % (MgO/(MgO+Al 20 3)). The second impregnation was carried out on the modified alumina support with an ethanol (Aldrich, 95 %) solution of Ce(N03)3«6H20 (Aldrich, 99 %) in a concentration of 0.75 M[Ce]. After impregnation, the sample was dried in air at 373 K for 8 hours and then calcined in air at 748 K for 3 hours. The modified alumina support with Ce02 as the promoter (Ce02/MgO/Al203) was obtained after calcination. The Ce02 loading was 14.4 wt. % (Ce02/(Ce02+MgO-Al 20 3)). 21 The Co and N i catalyst was achieved by a final impregnation. The modified alumina support with CeC>2, was impregnated with an aqueous solution of Co(N03)2 ,6H 20 (Acros Organics, ACS Reagent) or N i ( N 0 3 ) 2 ' 6 H 2 0 (Aldrich, 99.999 %). The molarity was 1.85 M for [Co] and 1.79 M for [Ni]. The metal loading was 12 wt. % for both Co and N i on the CeOi/MgO/Al203 support. Different drying techniques were applied after impregnation. For the supported Co catalysts, the catalysts were vacuum-dried at 378 K for 37 hours and then calcined at 723 K for 10 minutes. For supported N i catalysts, the catalysts were dried in air at 383 K for 37 hours and then calcined at 723 K for 10 minutes. The metal catalysts were also prepared on a support without Ce0 2 (i.e., MgO/AbOa) for comparison. 3.3 Catalyst Characterization 3.3.1 Temperature Programmed Reduction (TPR) TPR was performed in a stainless steel micro-reactor 50 mm in length and 7.7 mm inside diameter. The flow diagram of the reactor and the on-line analytical equipment is shown in Figure III-1. For each experiment, approximately 0.30 g of catalyst was loaded into the isothermal zone of the reactor. The catalyst was heated from room temperature to 623 K at a rate of 10 K/min and held at 623 K for 60 minutes under 60 mL/min Ar (Praxair, UHP, 99.999 %) flow. The purpose of this pre-heat was to desorb physically adsorbed water. At the same time, Ar was used to stabilize the TCD signal. 22 Reference Line Gas Outlet 1 Gas Outlet 2 Gas Outlet 3 TCD: thermal conductivity detector GC: gas chromatography Reactor Figure III-l Block diagram of the experimental set-up used for temperature programmed reduction The catalyst was then cooled to room temperature and the gas stream switched to 10 % H2/9O % Ar (Praxair, UHP, 99.999 %) reducing gas. The dry catalyst sample was heated at a rate of 10 K/min to 1007 K in 60 mL/min of the reducing gas. A 4A molecular sieve trap, placed on the down stream side of the reactor, removed produced water. The effluent gas was introduced to the TCD to quantify the H 2 consumption. The 4A molecular sieve trap was removed and regenerated by heating at 383 K for 12 hours after each experiment. By plotting the TCD signal against time, H 2 consumption was determined by the integrated area under the TCD curve. The degree of reduction was calculated from the ratio of the actual H 2 consumption to the theoretical H 2 consumption needed to 23 completely convert the metal oxide to metal. The measured H 2 consumption was calibrated using completely reducible Cu 2 0 (Rocky Mt. Research Inc. 99.999 %) as a standard. The calibration data for Cu 2 0 are shown in Appendix A and a sample calculation is given in Appendix B. 3.3.2 B E T Surface Area The BET surface area of the catalyst was measured using a Flowsorb II 2300 Micromeritics analyzer. The surface area measurement was accomplished with a simplified single point procedure by N 2 adsorption-desorption at 77 K . The sample was first degassed at 398 K for approximately 3 hours prior to measurement. A 30 mole % N 2 and 70 mole % He mixture at a flow of 15 mL (STP) /min was used to measure the N 2 adsoption. The instrument was calibrated by introducing l m L of N 2 gas to the detector. A sample calculation is given in Appendix B. 3.3.3 X-Ray Diffraction (XRD) X R D patterns of the prepared catalysts were obtained with a Rigaku Multiflex diffractometer using Cu K a radiation (A=1.5406 A), a scan range of 20 from 20° to 80° with a step of 0.04°. Diffraction patterns were compared with the inorganic crystal structure database (ICSD) for bulk phase identification. Particle size of the metal or metal oxide was obtained from the X R D data using the Scherrer Equation (3.1): d„= XKam" C 3- 1) /?cos0 where dp is the metal crystallite size, A; X is the X-ray wavelength, 1.5406 A; Kconst is a constant taken as 0.89; B is the full width at half height of the diffraction peak in radians; and 0 is the diffraction angle in radians. 24 3.3.4 Transmission Electron Microscopy ( T E M ) and S T E M E D X The formation of carbon deposits was examined by T E M using a FEI TECNAI G 2 electron microscope with an acceleration voltage of 200 kV. The magnification was in the range of 7,000 - 400,000x's. T E M specimens were prepared by grinding the catalyst to a fine powder before dispersing the catalysts in ethanol (Aldrich, 95 %) and applying one drop of the suspension onto a copper grid. The 200 mesh copper grid was a Formvar coated grid, stabilized with evaporated carbon film. Scanning Transmission Electron Microscopy (STEM) with Energy-dispersive X-ray (EDX) analysis was also used to determine chemical composition of the catalyst at selected locations. 3.4 Reactor and Analysis 3.4.1 T E O M A tapered element oscillating microbalance (TEOM, Rupprecht & Patashnick Co., Inc. TEOM series 1500 Pulse Mass Analyzer) was used in the present study. The T E O M is configured as a fixed bed micro-reactor in which the real time mass change of solid placed in the reactor can be determined. Typically one measures the catalyst mass change in situ during a gas-solid catalytic reaction. The major components of the T E O M are a hollow tapered tube (also called tapered element, TE) made of a quartz glass, a lower heating zone and two optical detectors located on opposite sides of the tapered tube. The sample bed is located at the bottom of the tapered tube with a volume of 0.2 cm 3. Figure III-2 shows the block diagram of the major components of the T E O M and a picture of the 25 sample bed and part of the tapered tube is shown in Figure III-3. Instrument operation and data acquisition are done using LabVIEW control software. Purge gas Purge gas Protection tube Protection tube Lower heat zone §§ • Sample bed Figure III-2 Block diagram of the major parts of the TEOM Tapered tube Quartz wool Figure III-3 Picture of sample bed and one cent for relative size comparison 26 The reactant and carrier gases pass through the tapered tube and sample bed where the reaction takes place. The purge gases pass down between the tapered tube and the protection tube to prevent diffusion of the product gases back into the system. The tapered tube vibrates at its natural frequency of oscillation with constant amplitude. This frequency is measured by two optical detectors (transmitter and receiver). The mass change of the sample can be determined by the following mass change equation, AM = Kox[\—U ( 3 > 2 ) f , 2 f o 2 where: AM = Mass change of the sample Kq = Spring constant of the tapered element fo = The natural oscillating frequency at initial time "0" fi = The natural oscillating frequency at a later time "1" The mass reported in the data plots of the present report is the sum of the initial mass and the mass change as determined by Equation 3.2. Normally, the initial mass is set as zero for each experiment. Consequently, the mass at a specific time does not necessarily reflect the true mass of the sample. In contrast, the mass difference measured between two time intervals is the true mass change during the period. According to the operating principle of the TEOM, the temperature and gas composition also affect the measurement of the mass change. Blank experiments were performed at different temperatures and gas compositions with an empty sample bed in order to verify these effects. The temperature can affect the mass change because the spring constant J^ o 27 is a function of temperature. Figure III-4 shows the mass change due to temperature variation with constant gas composition (100 % He). The mass increased from around 0.2 to 1.4 mg when temperature increased from 723 to 823 K. In order to prevent the incorrect mass measurement caused by a temperature variation, a period of inert gas flushing is employed to allow temperature adjustment. Temperature was adjusted to the desired temperature for the next step (either CH4 decomposition or partial oxidation) during the inert gas flush period. By avoiding the temperature variation, the mass measurement caused by the reaction between the catalyst and reactant gases (either C H 4 or O2) was obtained. 1.8x10* 1.6x10"JH 1.4x10-1.2x10-1.0x10" </> 03 Mass Change due to temperature variation Reactor Temperature 900 -2.0x10 I i i • | 700 2400 2600 2800 3000 3200 3400 3600 3800 4000 4200 4400 4600 Time(sec) Figure III-4 The effect of temperature difference on mass reading at constant gas composition 28 The effect of changing gas composition depends on the gas molecular mass and concentration. In general, the oscillating frequency of the tapered tube decreases as the density of the gas mixture passing through the tapered tube increases. The decreasing oscillating frequency (fi) generates a positive AM from Equation (3.2). Therefore, the mass appears to increase as the density of the gases passing through the tapered tube increases. Figure III-5 shows that there is no significant mass change caused by switching from pure He (200 seem He) to 5 % CH 4 /He (10 seem C H 4 + 190 seem He). 5 % CH 4 /He was widely used in the present study as the reactant/carrier gas during the C H 4 decomposition step. It can be concluded that there is no significant mass change caused by gas composition variation during the C H 4 decomposition step. 2.0x10"3-i 1.9x10'3-1.8x10"3-1.7x10"3-1.6x10"3-05 1.5x10"3-W Ma 1.4x10"3-1.3x10"3-1.2x10"3-1.1x10'3-1.0x10'3-Mass change due to gas composition variation He flow rate C H , flow rate 4 4000 4200 4400 4600 4800 5000 5200 5400 5600 5800 6000 Time(sec) Figure HI-5 The effect of changing C H 4 concentration on mass reading at 823 K 29 Figure III-6 shows the effect of changing O 2 composition at a constant temperature. When the gas was switched from pure He (200 seem He) to 2.5 % 02/He (5 seem O 2 +195 seem He), no significant mass change was observed. But a mass increase from around 1.05 to 1.15 mg was observed when the gas was switched from He (200 seem He) to 20 % 02/He (40 seem O 2 + I 6 O seem He). However, a 0.1 mg change is not considered significant given that the typical mass change is 10 mg for partial oxidation with 20 % O 2 concentration. Since 20 % O 2 concentration was the maximum O 2 concentration used in the present study, it is reasonable to claim that the mass change due to O 2 concentration variation is negligible during the partial oxidation step of the experimental cycle. 1.40x10"3-r 1.35X10"3-1.30x10"3-1.25x10"3-1.20x10"3-1.15X10"3-(/> 1.10X10"3-ro 1.05x10"3-1.00x10"3-9.50x10^-9.00x10^-8.50x10"4-8.00X10"4--• Mass change due to 0 2 concentration variation He flow rate 0 2 flow rate 5500 6000 6500 7000 7500 8000 8500 9000 9500 1000010500 Time(sec) Figure III-6 The effect of changing 0 2 concentration on mass reading at 823 K 30 It is important to note that the measured mass change of catalyst at the same temperature and gas composition is the real mass change unbiased by the temperature and gas composition. However, the mass change measured represents the net mass change of catalyst. If the catalyst sample is simultaneously gaining and losing mass (for example, by simultaneous carbon deposition and carbon removal reactions), only the net mass change is determined by the TEOM. 3.4.2 Experimental Set-up and Conditions CH4 decomposition and partial oxidation were measured in the T E O M , monitoring catalyst mass change while also using a bench-top Mass Spectrometer (VG ProLab) for continuous product gas analysis. A l l the experiments were carried out at atmospheric pressure. The total flow rate passing through sample bed was 200 seem and the purge gas was He at 400 seem. Gas flow rates were controlled by calibrated Brooks mass flow controllers. The calibration data of the Brooks mass flow controllers is shown in Appendix C. The flow diagram including the T E O M and Mass Spectrometer (MS) is shown in Figure III-7. A l l the gases used in the present study for cyclic performance evaluation were UHP, 99.999 % grade. Approximately 0.10 - 0.15 g of calcined supported metal catalyst was loaded into the TEOM. The calcined catalyst was reduced in situ by introducing 40 vol. % H 2 at 823 K for approximately 2 hrs. The metal oxide was reduced to metal according to the degree of reduction. Table III-1 shows the typical experimental conditions. 31 Vent TEOM: tapered element oscillating microbalance MFC: mass flow controller MS: Mass Spectrometer Figure III-7 Block diagram of the experimental set-up used for in situ reduction, C H 4 decomposition and partial oxidation After the catalyst was reduced in situ by H 2 , He was introduced to the T E O M for 15 minutes to flush any residual H 2 . The temperature was adjusted to the designated temperature for CH4 decomposition during this He flush period. Therefore, the stable designated temperature of CH4 decomposition was established during the He flush period. As described in Section 3.4.1, a mass change can be introduced by temperature variations. In order to avoid this error and ensure correct mass change measurement during CH4 decomposition, it is essential to establish a stable designated temperature during the He flush period before the CH4 decomposition was initiated. 32 Table III-l Typical experimental parameters Steps Experimental Parameters Reactant Gases a > b ' 0 Reactor Temperature (K) Duration (min) H 2 Cone, (vol. %) C H 4 Cone, (vol. %) O 2 Cone, (vol. %) in situ Reduction Step 40% • • — — 823 100-150 C H 4 Decomposition Step ' .../:|;::|i.:f':''-'-: 5% 723-823 3-45 Partial Oxidation Step 2.5-20 % 723-823 3-45 a: Carrier gas is He for all the experiments b: The total flow rate passing through sample bed was 200 seem c: The purge gas was He at 400 seem Once C H 4 was introduced, it decomposed on the reduced, supported metal catalyst. The mass change of the catalyst was measured by the T E O M and the effluent gases were analyzed by the mass spectrometer (MS). The MS monitored and recorded the intensity of mass peaks corresponding to H 2 , He, C H 4 , H 2 O , CO, C O 2 and O 2 . Calibration data of H 2 , C H 4 , CO, and C O 2 are shown in Appendix D. After the C H 4 decomposition step was completed, He was introduced to flush residual gases from the system. The temperature was again adjusted to the designated temperature for the subsequent partial oxidation step during this He flush period. During the partial oxidation step, O 2 was introduced to the T E O M and the mass change of the catalyst and gas concentration were both measured. In the present work, a single 33 CH4 decomposition step and a subsequent partial oxidation step is referred to as one cycle. At least 4 cycles were completed for all the experiments of the present study for cyclic performance evaluation. Once the partial oxidation step completed the 1st cycle, a He flush was required again to remove residual gases and stabilize the reactor temperature. Then, the 2nd C H 4 decomposition was started. The flow chart of the sequence of CH4 decomposition, partial oxidation and corresponding He flush for a single experiment is shown in Figure III-8. Once all cycles are completed In-Situ Reduction He Flush A C H 4 Decomposition He Flush Partial . Oxidation He Flushing + Cooling down system Perform at least 4 cycles for cyclic performance evaluation Figure III-8 Flow chart of the sequence of C H 4 decomposition, partial oxidation and corresponding He flushing The data collected from the T E O M and MS regarding the cyclic performance of the catalyst is presented in subsequent chapters. Pure errors associated with each measurement of interested were calculated from 5 sets of duplicate experiments. The details of the error analysis are shown in Appendix E. 34 Chapter IV Supported Co Catalyst 4.1 Introduction In this chapter, the results of CH4 decomposition followed by partial oxidation in a cyclic mode on supported Co catalyst are presented and discussed. The effect of MgO as a promoter is first discussed in Section 4.2. The interaction between the cobalt oxide catalyst precursor and the alumina support is relatively strong. As a result, the cobalt oxide precursor does not easily reduce to Co and, therefore, the degree of reduction is low. In general, the lower the degree of reduction, the lower the number of active sites of the catalyst. By applying MgO to modify the alumina support, the interaction between the metal oxide precursor and the modified alumina support is reduced and hence, the degree of reduction is increased. The cyclic performance of supported Co catalyst is discussed in Section 4.3. Calcined supported Co catalyst was loaded into the T E O M and reduced in situ. CH4 decomposition and subsequent partial oxidation were carried out in a cyclic mode. The experimental parameters such as CH4 concentration, O2 concentration, and temperature for both the C H 4 decomposition and partial oxidation steps were the same for all experiments whereas the duration for each step was varied. In order to investigate the stability of the catalyst, the same calcined supported Co catalyst sample was used for all experiments. Note, however, that for each set of conditions given in Table IV-1, the used catalyst was re-reduced after the final oxidation step of the pervious experiment. 35 With the data collected from the T E O M and MS, the mass change of the catalyst and the effluent gas analysis provide answers to the following critical questions regarding the objective of the present study: 1. Can H2 be produced during CH4 decomposition in a cyclic mode and does deactivation occur at the chosen experimental conditions? 2. Can the active sites be regenerated by removing the deposited carbon during the subsequent partial oxidation and is the deposited carbon easily removed under the current experimental conditions? 3. What is the product distribution of CO and CO2 during partial oxidation? 4.2 The Effect of MgO as a Promoter on the Degree of Reduction of the Co Catalyst Following impregnation of the alumina support with Mg (N03)2, the support was dried and calcined in air at 1073 K for 8 hours to decompose the Mg (N03)2 to MgO. The function of MgO is to modify the surface of the alumina support such that the interaction between the alumina support and the metal oxide precursor is reduced. Hence, the degree of reduction of the cobalt oxide precursor can be increased. As shown in Figure IV-1, the degree of reduction for the Co catalyst supported on MgO /A l203 was higher (72 %) compared to the catalyst supported directly on AI2O3 (45 %). 36 Figure IV-1 The effect of M g O used to enhance the degree of reduction of cobalt oxide. The catalyst sample was heated at a rate of 10 K/min to 1007 K in 60 mL/min of the 10 % H 2/90 % A r reducing gas stream C03O4 reduces to Co in two steps (Jacobs et al., 2004): StepT. C03O4 + H 2 3CoO + H 2 0 Step2: 3CoO + 3H 2 3Co + 3H 2 0 As shown in Figure IV-1, the TCD voltage curve for the Co-MgO/Al 2 03 TPR shows two peaks. The first step is fast and results in a sharp and low temperature peak. The second step involves reduction of the cobalt oxide (CoO). The cobalt oxide (CoO) tends to have a stronger interaction with the support which results in a broad, high temperature peak. 37 The theoretical H 2 consumption ratio for step 1 to step 2 is 0.33. The integrated area under the peaks represents the H_ consumption. The ratio of the peaks associated with the two reduction steps was 0.29, which is in reasonable agreement with the theoretical ratio. X R D identifies the different phases of metal and metal oxide present in the catalyst sample. Whether the cobalt oxide precursor was completely reduced or not can be revealed by comparing the X R D patterns of the calcined and reduced supported Co catalysts. As shown in Figure IV-2, the reduced supported Co catalyst had a weaker C03O4 peak compared to the calcined supported Co catalyst and the Co phase was also detected in the reduced catalyst. Clearly, some C03O4 phase was reduced to Co, but not all of the C03O4 was reduced by the TPR treatment. Calcined Co-MgO/AI203 Reduced Co-MgO/AI203 1 ' 1 ' 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 • 1 ' 1 ' 1 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 20 (degree) Figure IV-2 X R D pattern comparison of the 12 wt. % C o - M g O / A l 2 0 3 catalyst precursor after calcination at 723 K in air and reduction at 1007 K in 10 % H 2/90 % Ar reducing gas 38 4.3 Cyclic Performance of the Supported Co Catalyst For experiment Co-1, calcined Co catalyst was loaded into the T E O M and reduced in 40 vol. % H 2 at 823 K for 2 hours in situ followed by C H 4 decomposition and subsequent partial oxidation carried out in a cyclic mode with 4 cycles. After completing the 4th partial oxidation, experiment Co-1 was considered complete and the catalyst in the TEOM was cooled to room temperature under He flow. Without unloading or replacing the used catalyst, after the treatment of 4th partial oxidation of Co-1, experiment Co-2 was initiated by first completing an in situ reduction as for Co-1, followed by 4 cycles of CH4 decomposition and subsequent partial oxidation in 0 2 . Similarly, the catalyst after the 4th partial oxidation of Co-2 was re-reduced prior to beginning experiment Co-3. The experimental parameters of each experiment are listed in Table IV-1. A l l the experiments were carried out at atmospheric pressure. The total flow rate passing through sample bed was 200 seem and the purge gas was He at 400 seem. The reason for using the same catalyst was to investigate the stability of the catalyst. If the performance during the 1st C H 4 decomposition of the supported Co catalyst among Co-1, Co-2 and Co-3 was similar, the catalyst would be considered stable since the catalyst was re-usable after the in situ reduction. On the other hand, i f any deactivation occurred for the 1 st CH4 decomposition among the 3 experiments, it would indicate that the supported Co catalyst can't reach the same degree of reduction after the in situ TPR and can not be re-used. 39 Table IV-1 Experimental parameters for supported Co catalyst Experiment a , b CH4 Decomposition Partial Oxidation C H 4 Cone. (vol.%) Reactor Temperature (K) Duration (min) 0 2 Cone. (vol.%) Reactor Temperature (K) Duration (min) Co-1 5 773 45 5 . 773 45 Co-2 30 4 Co-3 45 4 a: Catalyst was re-reduced in situ in 40 vol. % H 2 at 823 K for 2 hours before the start of experiment Co-2 and Co-3 b: Same catalyst was used in each experiment without replacing catalyst Since the difference among the experiments was only the duration of each step, the legend used to describe each experiment in the following figures refers to the duration of the CH4 decomposition and partial oxidation steps. For example, the legend used to describe the experiment Co-2 was Co2-30min-4min. The first time (30 min) indicates the duration of the C H 4 decomposition step and the second time (4 min) indicates the duration of the partial oxidation step. By assuming the ratio of H 2 production rate to C H 4 decomposition rate was 2:1, the measured initial rate of C H 4 decomposition was calculated from the initial change in H 2 molar flow in the effluent gas. The initial oxidation rate of deposited carbon was measured from the initial change in CO and C O 2 molar flow rate in the effluent gas, as well. 40 4.3.1 Hydrogen Production and Catalyst Deactivation Figure IV-3 shows an example of the combined data collected by the T E O M and the MS. Effluent gas analysis indicated that C H 4 was decomposed and H 2 was produced. The mass of the catalyst increased as the carbon accumulated during C H 4 decomposition. Figure IV-3 also shows that the H 2 molar flow rate declined as time proceeded, indicating some catalyst deactivation during CH4 decomposition. 0x10° 0x10"3H 0.04 co co 05 -2.oxio H -4 -6.0x10° H -8. -1 1.0x10 .0x10"2 <o o E aT -1—• 05 CC $ o LL o 1.0x10 • 8.0x10"6 • 6.0x10"6 • 4.0x10"6-2.0x10'6-0.0-5.0x10"7-i 4.0x10"7-3.0x10"7-2.0x10'7- | 1.0x10"7-| !'-. 0.0-500 1000 1500 2000 T i m e ( s e c ) 500 2500 3000 C H 4 ( m o l / s ) H 2 (mo l / s ) — i — 1 — 1 — • — 1 — 1000 1500 2000 T ime(sec ) — i — 1 — r -2500 3000 C O ( m o l / s ) C0 2 (mol /s ) 500 1000 1500 2000 2500 3000 T i m e ( s e c ) Figure IV-3 Measurement of H 2 , C H 4 , C O , C 0 2 and the mass of the supported Co catalyst during C H 4 decomposition (5 vol. % C H 4 , 773 K) in the 1st cycle of experiment Co-1 41 Trace amounts of CO were produced during C H 4 decomposition as shown in Figure IV-3. Since C H 4 and He were the only gases introduced into the T E O M during C H 4 decomposition, the catalyst must be the source of oxygen to form CO. Presumably, oxygen from un-reduced cobalt oxide catalyst precursor or the metal oxide support reacted with C H 4 to produce CO. Since the detailed mechanism of CO formation during C H 4 decomposition was not the focus of the present study, the mechanism was not investigated further. Figure IV-4 shows the initial C H 4 decomposition rate as a function of cycle number of each experiment. Deactivation also occurred as the cycle number increased. The catalyst was completely deactivated after the 2nd cycle under the conditions of experiment Co-2. The initial rates of C H 4 decomposition for all experimental conditions in the 1st cycle were similar, but these rates were significantly different from the 2nd cycle onwards. These data indicated that the conditions of the partial oxidation step was an important contributor to the catalyst deactivation. Total H 2 production was also compared for each cycle. Figure IV-5 shows that the total amount of H 2 produced during C H 4 decomposition decreased with increased cycle number due to the deactivation of the catalyst. Note that even though the initial C H 4 decomposition rates were very similar for the 1 st cycle among all experiments (Figure IV-4), the amounts of H 2 produced for the 1st cycle decreased as the experiments proceeded from Co-1 to Co-2 and Co-3 (Figure IV-5). 42 Comparing experiment Co-1 and Co-3, the amount of H 2 produced in the 1st cycle for experiment Co-3 was only 27 % of that of experiment Co-1, even though both experiments had a similar initial CH4 decomposition rate and the same duration for CH4 decomposition. As shown in Figure IV-6, the molar flow rate of H 2 declined faster for experiment Co-3 compared to Co-1 during C H 4 decomposition in the 1st cycle. This indicated a faster deactivation occurred for experiment Co-3. -5T 1 . 0 x 1 0 • I 9 .0X10" 6 E. 8 . 0 x 1 0 " 6 7 .0x10" 6 -g 5.0x10"6 Q. § 4.0X10"6 <§ 3.0x10"' X* 2.0x10"' O ro 1.0x10"°H \ \ -•— Co1-45min-45min Co2-30min-4min Co3-45min-4min 2 3 Cycle Number Figure IV-4 Initial C H 4 decomposition rate vs cycle numbers for supported Co catalyst for experiment Co-1, Co-2 and Co -3. All 3 experiments were performed at 773 K with 5 vol. % C H 4 in the C H 4 decomposition step and 5 vol. % 02 in the partial oxidation step. The only difference among the 3 experiments was the duration of each step The data of Figure IV-6 suggest that the same supported Co catalyst can not be re-used for all the experiments even though in situ reduction was applied between each set of experiments. Supported Co catalysts were not able to achieve the same degree of 43 reduction under the in situ reduction conditions. As shown in Figure IV-7, the amount of water produced during in situ TPR decreased as the experiment proceeded. The direct measurement of water intensity was used here for integration due to the difficulty of water calibration. ° 3.0x10 -I 2.5x10"3-o Q . § 2.0x10"3-T3 1.5x10"3-I 1.0x10"3-T3 T3 % 5.0x10"4-x> o i _ Q _ CN 0.0-- •— Co1-45min-45min • Co2-30min-4min - A Co3-45min-4min 2 3 Cycle Number Figure IV-5 H 2 production during C H 4 decomposition for supported Co catalyst for experiment Co-1, Co-2 and Co-3. All 3 experiments were performed at 773 K with 5 vol. % C H 4 for the C H 4 decomposition step and 5 vol. % 0 2 for the partial oxidation step. The only difference among these 3 experiments was the duration of each step 44 Figure IV-6 H 2 production during C H 4 decomposition in the 1st cycle for experiment Co-1 and Co-3. Same experimental parameters was applied for both experiments during C H 4 decomposition The amount of water produced during in situ TPR was from the reduction of cobalt oxide. In other words, the less water produced, the less cobalt oxide reduced. The results indicate that an unused calcined catalyst was required for each experiment i f a valid comparison of initial rates and H_ production among different experimental conditions were to be made. 45 I • Integrated value of water peak during in-situ reduction ro Q . 6-o ro > •o ro CD * cn 34 Co-1 Co-2 Experiment Co-3 Figure IV-7 Water peak integrated area during in situ reduction for experiment Co-1, Co-2 and Co -3 4.3.2 The Effects of Subsequent Partial Oxidation The carbon deposit from C H 4 decomposition reacted rapidly with O 2 during subsequent partial oxidation. Figure IV-8 shows that the carbon deposit reacted all the O 2 initially to produce CO and C O 2 . The mass of the catalyst decreased corresponding to the amount of carbon reacted. The complete consumption of 0 2 initially indicated that the production of CO and C O 2 was limited by the concentration of O 2 introduced. 46 3 a* CO •^ 2 o E, of -*-» CO cr o ro o 4.0x10"3 2.0x10 H o.o A -2.0x10"J -4.0x10 H -6.0x10"J 4500 I ' 1 ' 1 • 1 1 1 1 1 1 1 5000 5500 6000 6500 7000 7500 8000 Time(sec) 7.0x10-1 6.0x10"6-5.0x10"6-4.0x10' 3.0x10" 2.0x10" 1.0x10° H 0.0. A CO(mol/s) • C0 2(mol/s) O2(mol/s) 4500 5000 I — I — ' — I — ' — I — I — I — I — I 5500 6000 6500 7000 7500 8000 Time(sec) Figure IV-8 Measurements of C O , C 0 2 , 0 2 and the mass of the catalyst during partial oxidation (5 vol. % 0 2 , 773 K) in the 1st cycle for experiment Co-1 Figure IV-9 also shows that the initial oxidation rate of the carbon deposit was much faster than the initial C H 4 decomposition rate (Figure IV-4). Fast oxidation rate indicates that the carbon deposit was very active and readily removed. The results suggest that a long reaction duration for partial oxidation is not required. 47 ID X CO I f 7 . 0 x 1 c r % 6.0x10 • o o. Q 5.0x10"5 c | 4.0x10"5-I o I 3.0x10"5H -«— CD * 2.0x10"5-g 1 1.0x10"5-x O p| _ J 0.0 —•— Co1-45min-45min • Co2-30min-4min A Co3-45min-4min Cycle Number Figure IV-9 Initial oxidation rate of carbon deposit vs cycle numbers for supported Co catalyst of experiment Co-1, Co-2 and Co-3. All 3 experiments were performed at 773 K with 5 vol. % C H 4 in the C H 4 decomposition step and 5 vol. % 0 2 in the partial oxidation step. The only difference among the 3 experiments was the duration of each step The previous discussion concluded that the subsequent partial oxidation affected the following C H 4 decomposition. In other words, the C H 4 decomposition in the 2nd cycle after the 1st partial oxidation would be different from the C H 4 decomposition in the first cycle. As shown in Figure IV-10, 3 significant differences were observed during C H 4 decomposition in the 2nd cycle compared to the 1st cycle. 48 CH4 Decomposition in the 1st Cycle 0.004-, CH4 Decomposition in the 2nd Cycle 0.0 , Time(sec) Time(sec) 1.0x10 • 9.0x10"6 • 8.0x10"°-7.0x10"°-6.0x10"6-5.0x10"6-4.0x10"6 -3.0x10"6-2.0x10'"-1.0x10"°-0.0. • CH4(mol/s) - H,(mol/s) 1000 1500 2000 Time(sec) 2500 3000 1 LL 1.0x10" 9.0x10' 8.0x10'°-| 7.0x10"° 6.0x10"6 5.0x10"° 4.0x10"° 3.0x10"6 2.0x10"" 1.0x10"* 0.0 I 500 •CH4(mol/s) -H(mol/s) 1000 1500 2000 Time(sec) 2500 3000 5.0x10"'-4.5x10"'-4.0x10''-3.5x10"'-3.0x10"'-2.5x10"'-2.0x10"'-1.5x10"'-1.0x10"'-5.0x10""-0.0-500 • CO(mol/s) - CO,(mol/s)| 1000 1500 2000 Time(seo) 2500 3000 | o 5.0x10"' 4.5x10"' 4.0x10'' 3.5x10"' 3.0x10"' 2.5x10"'-2.0x10"'-1.5x10"' 1.0x10"' 5.0x10"' 0.0 • CO(mol/s) - CO,(mol/s)| 500 1000 1500 2000 Time(sec) 2500 3000 Figure IV-10 Comparison of H 2 , C H 4 , C O , C 0 2 and the mass of catalyst during C H 4 decomposition (5 vol. % C H 4 , 773 K) in the 1st and 2nd cycle of experiment Co-1 (Different Y scales were applied for H 2 - C H 4 and C O - C 0 2 ) 49 Firstly, the mass change of catalyst was different. Unlike the smooth increase in sample mass that occurred in the 1 st cycle, in the 2nd cycle the mass of the catalyst was unchanged initially, then decreased briefly and then increased. Secondly, an induction period occurred in the 2nd cycle before CFL, decomposition and H 2 , CO and C O 2 production were observed. The induction period corresponded to the period of a stable mass. Thirdly, more CO and C O 2 were produced during the 2nd cycle compared to the 1st cycle. During the partial oxidation cycle, the carbon deposit reacted with O 2 to form CO and C O 2 , and Co metal reacted with O 2 to form cobalt oxide (CoO or C 0 3 O 4 ) . During the subsequent CH4 decomposition, C H 4 did not decompose presumably because the active site, Co metal, was not available. An induction period was therefore required to affect a reduction of the cobalt oxide to Co by C H 4 . Once Co became available, rapid C H 4 decomposition occurred, producing H 2 , deposited carbon and possibly further reduction of the metal. The carbon deposit also reacted with the solid oxygen on the catalyst, producing CO and C O 2 . The observed initial decrease in the mass of the catalyst was caused by the reaction of solid oxygen to gas products. Expanding the time scale of Figure IV-8, Figure IV-11 shows that the mass of the catalyst increased initially as a result of O 2 uptake by Co metal. Figure IV-11 also shows that the time required to oxidize the deposited carbon to CO and C O 2 was less than 250 (4700-4950) seconds under the current experimental conditions. This also indicates that the carbon deposit from the CH4 decomposition was very active. Therefore, it was not necessary to apply an extended reaction time period to remove the deposited carbon. 50 4.0x10'3-i 3.0x10'3-2.0x10"3-1.0x10"3-0.0-V) Ma -1.0x10'3--2.0x10'3--3.0x10"3--4.0x10'3--5.0x10"3-o E af -»—i CD CH <: o u_ \ o i—'— i . 4500 4550 4600 4650 4700 4750 4800 4850 4900 4950 5000 Time(sec) - i f—i—.—i—i—|—i—-\—i—|—i—| 4500 4550 4600 4650 4700 4750 4800 4850 4900 4950 5000 Time(sec) Figure IV-11 Measurement of C O , C 0 2 , 0 2 and the mass of the catalyst during the partial oxidation (5 vol. % 0 2 , 773 K) in the 1st cycle for experiment Co-1 with the expanded time scale 4.3.3 C O , C 0 2 Production, Carbon Removal Percentage and Selectivity to C O during Oxidation CO and C 0 2 were produced during the partial oxidation step of each reaction cycle. As shown in Figure IV-12, the production of CO and C 0 2 decreased as the number of cycles increased. 51 0) o E . 4.0x10"4-| o 'ro 3.5x10"" H •g | 3.0x10"" 1 2.5x10"4^ ro -1 £ 2.0x10'4-| f 1.5x10"" -j | 1.0xir/H = 5.0x10"5 Q. O o 0.0- I - r~ — • — Co1-45min-45min —©— Co3-45min-4min Cycle Number E 1.0x10"3 -i | 8.0x10"4 .2 6.0x10 •c ro Q. g> 4.0x10""H ? 2.0x10"" 3 •a o O O 0.0. 0-Cycle Number Figure IV-12 C O and C 0 2 production during partial oxidation on supported Co catalyst vs cycle number for experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % C H 4 for C H 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of partial oxidation Since the overall amount of O 2 introduced during the oxidation step was more than the amount of carbon deposited during the CH4 decomposition step under the current experimental conditions, the amount of CO and C O 2 produced depended on the amount of carbon deposited from the preceding CH4 decomposition. The amount of carbon deposit was directly related to the amount of H 2 produced due to C H 4 decomposition. Figure IV-13 shows a strong positive correlation between the CO and C O 2 produced 52 during partial oxidation and the H 2 produced during C H 4 decomposition at the same cycle for both experiments Co-1 and Co-3. 1.4x10 Co-1 •c 1.2x10 o ro § 1.0x10'3 ••e 8.0x10 6.0x10 8 2.0x10'" + O o Linear Regression Y = A + B * X Parameter 5.0x10" 1.0X10"3 1.5x10'3 2.0X10"3 2.5x10"3 3.0x10'3 H 2 produced during CH 4 Decomposition(mole) Value Error A 1.51488E-5 3.32174E-5 B 0,47802 0.01988 R SD N P 0.99828 3.47765E-5 4 0.00172 2.8x10-1 2.6x10'" 2.4x1 0'" 2.2X10"4 2.0x10'" 1.8x10'" 1.6x10"" 1.4x10'" o Co-3 Linear Regression Y = A + B * X Parameter Value -i ' 1 1 1 1 1 1 1 4.50x10'" 5.25x10" 6.00x10J 6.75x10'" 7.50x10'" H 2 produced during CH 4 Decomposition(mole) Error A 1.11419E-5 1.66861E-5 B 0.32679 0.02754 R SD N P 0.99297 6.06992E-6 4 0.00703 Figure IV-13 The correlation between the C O and C 0 2 produced during partial oxidation and H 2 produced during the preceding C H 4 decomposition for experiment Co-1 and Co-3 53 It is obvious that the more H 2 produced, the more CH4 consumed and the more carbon deposited. As a result, more CO and C 0 2 were produced during the subsequent partial oxidation. The carbon removal percentage is introduced here to indicate the ability to remove the deposited carbon during the oxidation step. The carbon removal percentage is defined as follows: „ , , moles(CO + CO 1) produced in POX Carbon removal percentage = x 100% moles of carbon deposit in CH4 decomposition The amount' of carbon deposited during C H 4 decomposition was calculated from the H 2 production by assuming a ratio of H 2 production to deposited carbon of 2:1. Figure IV-14 shows almost a 100 % carbon removal at the experimental conditions of experiment Co-1. Hence complete removal of the deposited carbon from C H 4 decomposition was achieved by reaction with O 2 at the conditions indicated. 54 —•— Co1-45min-45min —o— Co3-45min-4min '°1 ° 0 % ! ^ — r r - r 1 2 3 4 C y c l e N u m b e r Figure IV-14 Carbon removal fraction vs cycle number of experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % C H 4 for C H 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of partial oxidation The selectivity to CO during the partial oxidation step was also calculated as follows: Selectivity to CO = A m o u n t ° f C O ( m o l e ) Amount of CO(mole) + Amount of COi(mole) A low selectivity to CO is shown in Figure IV-15 for both experiment Co-1 and Co-3, suggesting that at the current experimental conditions, the formation of C 0 2 during partial oxidation was favoured. 55 Co1-45min-45min o CO •g x o t co Q. CD C ZJ T3 O O o -4—' >* ;> +-» o _Q) CU CO 30% 25% 20% 15% 10%-5%-0% 30% -, 25% -20% -15%-10%-5%-0%--® - r -2 Cycle Number • Co3-45min-4min 2 3 Cycle Number Figure IV-15 Selectivity to C O during partial oxidation of experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % C H 4 for C H 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of partial oxidation 4.3.4 C O to H i Ratio and Induction Period during C H 4 Decomposition Another important value to examine is the CO to H 2 ratio ( C O / H 2 ) reported herein as a % during C H 4 decomposition. This value was used as an indicator of the purity of the H 2 stream. Figure IV-16 shows an increasing trend in C O / H 2 ratio as the cycle number increased in experiment Co-1 and Co-3. Data indicate that the C O / H 2 reached 56 approximately 10 % starting from the 2nd cycle. This relatively high CO impurity associated with the Co catalyst makes it less desirable for application. Figure IV-16 C O / H 2 during C H 4 decomposition of experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % C H 4 for C H 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of partial oxidation Even though the detailed mechanism of CO formation during CH4 decomposition was not determined, results of Section 4.3.1 suggest that the source of oxygen to form CO was the catalyst. In order to distinguish the oxygen on the catalyst from the gas stream oxygen, the oxygen associated with the catalyst is referred to as solid oxygen. Section 4.3.1 also suggests that the solid oxygen was possibly in the form of CoO or C03O4. Two assumptions are made in the following discussion. First, the solid oxygen was from cobalt 57 oxide. Second, the carbon deposit from CH4 decomposition reacted with solid oxygen to form CO mainly. Figure IV-17 shows the CO produced during the C H 4 decomposition for each cycle of experiment Co-1 and Co-3. The CO production was relatively low in the 1st cycle since the catalyst had been reduced in situ. With a 72 % degree of reduction for the calcined supported Co catalyst, solid oxygen from cobalt oxide was still present. As a result, a small quantity of CO was produced during the C H 4 decomposition step of cycle number 1. o •§ 1.8x1 (T-i o : | 1.6x10""-I §• 1.4x1V4-\ § 1.2X10"4 I* 1.0x10"4-o g> 8.0x10"5 •§ 6.0x10"5-3 4.0x10"5-2 2.0x10"5 Q. O o —•— Co1-45min-45min • Co3-45min-4min 2 3 Cycle Number Figure IV-17 Amount of CO produced during C H 4 decomposition vs cycle number for experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % C H 4 for C H 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of each step The longer duration for partial oxidation of experiment Co-1 provided more opportunity for the oxidation of the supported Co catalyst, compared to Co-3. Hence, it was possible 58 to form more cobalt oxide. Due to increased solid oxygen, more CO was produced during the subsequent CH4 decomposition of experiment Co-1 compared to Co-3 and this then applied for cycles 2 to 4. The induction period was the time delay that occurred between the introduction of C H 4 and the production of H2 during CH4 decomposition. Since it is assumed that during the subsequent partial oxidation, Co metal was oxidized to cobalt oxide, the cobalt oxide must be reduced during the CH4 decomposition before H2 production could be initiated. The induction time was therefore an indicator of the difficulty to reduce cobalt oxide. Figure IV-18 shows that there was no induction period observed in the 1st cycle of CH4 decomposition for both experiments Co-1 and Co-3. An induction period was observed starting from the 2nd CH4 decomposition and longer induction periods for Co-1 were observed compared to experiment Co-3. The reason for the longer induction period can be attributed to the longer duration of partial oxidation. From the two figures (Figure IV-17 and Figure IV-18), it is clear that catalyst performance for the C H 4 decomposition during the 1st cycle was different from the subsequent cycles. There was no induction period required and low CO production during the 1st CH4 decomposition. The difference was due to the fact that the catalyst was reduced in H 2 prior to the 1st CH4 decomposition cycle of each set of experiments. The supported Co catalyst can be assumed to have a minimum solid oxygen during the 1st CH4 decomposition. On the other hand, partial oxidation occurred prior to the subsequent 59 C H 4 decomposition cycles. The quantity of solid oxygen increased according to the experimental conditions. Therefore, the supported Co catalyst contained relatively more solid oxygen before the 2nd to 4th C H 4 decomposition. Figure IV-18 Induction period during C H 4 decomposition vs cycle number for experiment Co-1 and Co-3. Both experiments were performed at 773 K with 5 vol. % C H 4 for C H 4 decomposition and 5 vol. % 0 2 for partial oxidation. The only difference was the duration of each step 4.4 Summary C H 4 decomposition and partial oxidation can be achieved in a cyclic mode on supported Co catalysts. C H 4 decomposed to produce H2 gas and carbon. The deposited carbon was removed by introducing O2 to oxidize the deposited carbon to CO and CO2. However, fast deactivation occurred for supported Co catalyst during C H 4 decomposition. 60 The observed deactivation of the catalyst from cycle to cycle was attributed to a loss in active cobalt metal sites by oxidation. After the CH4 decomposition, the subsequent partial oxidation not only oxidized carbon deposit to CO and C O 2 , but also oxidized cobalt metal to cobalt oxide. Cobalt oxide needed to be reduced before initiating CH4 decomposition. As a result, an induction period was observed. CO and C O 2 were also detected during C H 4 decomposition with high selectivity to CO. The detailed mechanism of CO and C O 2 formation was not determined in the present study, but evidence provided suggests the CO and C O 2 were produced by reaction between CH4 and solid oxygen associated with the catalyst. In the original concept of the present study, the function of partial oxidation is to remove the deposited carbon and regenerate active metal sites. Insufficient oxidation may result in low carbon removal fraction, but excessive oxidation may further oxidize metal to metal oxide. Supported Co catalyst showed a fast deactivation during CH4 decomposition and decreasing initial C H 4 decomposition rate and overall H 2 production, as the cycle number increased. Co catalyst showed no ability to be completely regenerated by in situ re-reduction with H 2 . As a result, Co catalyst can not be re-used. Furthermore, a high CO to H 2 ratio during C H 4 decomposition on Co catalyst was identified. Based on the above observations, supported Co catalysts are not suitable for the cyclic C H 4 decomposition and partial oxidation reactions examined. 61 In order to reduce deactivation during C H 4 decomposition, the metal catalyst should be easily reduced by C H 4 with a minimum induction period. Furthermore, the same catalyst without reloading will be applied for the cyclic performance in order to identify any deactivation between experiments and hence, to investigate the stability of the catalyst. 62 Chapter V Supported Ni Catalyst 5.1 Introduction In this chapter, the results of CH4 decomposition followed by partial oxidation in a cyclic mode on supported N i catalyst are presented and discussed. Section 5.2 presents the degree of reduction of N i catalyst by TPR. CeC>2 was confirmed to be reduced during TPR as well. By accounting for the H 2 consumed by CeC>2 reduction, the degree of reduction of NiO supported on Ce0 2/MgO/Al203 was determined. The cyclic performance of the same supported N i catalyst is presented in Section 5.3. By comparing the initial C H 4 decomposition rate and total H 2 production at each cycle among 4 different experiments, the results of Section 5.3.1 show no significant deactivation for N i catalyst and indicate that the N i catalyst can perform in a cyclic mode. Furthermore, N i catalyst shows the ability to be regenerated by in situ re-reduction in H 2 . Hence, N i catalyst is reusable for the cyclic treatment. Section 5.3.2 presents a possible explanation of the oxygen source for CO and C O 2 formation during CH4 decomposition. Temperature programmed oxidation (TPO) was used to verify that the oxygen uptake by reduced NiO was much greater than by reduced Ce02. Hence, the source of oxygen for CO and C O 2 production during C H 4 decomposition is probably the oxygen associated with NiO. 63 Instead of reacting the deposited carbon on Ni catalyst b y oxidation with O2, CO2 can b e introduced to react with carbon to form CO. Hence, carbon can b e removed and the active sites can b e regenerated. Section 5.4 shows a comparison of two experiments in which the carbon on the catalyst deposited at the same temperature and concentration, is reacted with O2 or CO2. Interesting values such as: CO to H2 ratio, carbon removal percentage, initial CH4 decomposition rate, and carbon removal rate are compared between the two experiments. 5.2 The Degree of Reduction of the Supported Ni Catalyst It was assumed that Ce02 would b e reduced during TPR of the supported N i catalyst and two catalysts were used to verify this assumption. The first was calcined 12 wt. % Ni supported on C e 0 2 / M g O / A l 2 0 3 and the second was Ce02/MgO/Al203 with no Ni present. If cerium oxide were reduced during TPR of the latter sample, the H2 consumption would be revealed b y a change in the TCD voltage. As shown in Figure V - l , the TCD voltage indicated H 2 consumption during TPR of the Ce02/Mg0/Al203. The area under the TCD voltage signal represented the amount of H2 consumed to reduce Ce02. The reduction peaks of C e 0 2 observed in the present study were similar to the TPR data for 12 wt. % C e 0 2 - A l 2 0 3 presented b y Damyanova et al. (2002). Due to a strong interaction between Ce02 and alumina, their data showed reduction peaks at 818-975 K for the first peak and 1147-1190 K for the second peak. However, in the present study the catalyst was modified by MgO and consequently the interaction between C e 0 2 and AI2O3 was reduced. Therefore, the temperature of reduction was reduced to 600 K for the first peak and 900 K for the second peak of the Ce0 2 /MgO/Al 2 03. 64 By subtracting the H2 consumption used for Ce02 reduction, the degree of reduction of the NiO of the calcined 12 wt. % Ni-Ce0 2/MgO/Al 20 3 catalyst was calculated as 85 %, which was higher than that reported for the supported Co catalyst (72 %). Figure V- l Comparison of TCD voltage signal during TPR for 12 wt. % Ni -Ce0 2 /MgO/AI 2 0 3 and C e 0 2 / M g O / A l 2 0 3 Diskin et al. (1998) indicated that the TPR peak temperature is strongly influenced by the NiO particle size and the support material. The first peak of the reduction profile for N i -Ce0 2/MgO/Al203 is assigned to the reduction of bulk NiO. The reduction of bulk NiO normally generates a sharp peak at low temperature. The second and third peaks were attributed to the stronger interaction between smaller NiO particles and the surface of the support, Ce02- Smaller NiO particles, in intimate contact with Ce02, would be most 65 difficult to reduce and, hence, two high temperature peaks were observed. The data also showed that the NiO is relatively easy to reduce compared to Ce0 2 , since the temperature associated with the reduction peak of NiO (500 K) was lower than that of C e 0 2 (600 K). Figure V-2 shows the X R D pattern for the calcined and reduced Ni-Ce0 2 /MgO/Al 2 03. The X R D pattern indicated that NiO was present in the calcined Ni -Ce0 2 /MgO/Al 2 03 and Ni was present in the reduced Ni-Ce0 2 /MgO/Al 2 03. These data confirmed that most of the NiO was reduced to N i metal by TPR. In addition, the peak intensity for Ce0 2 decreased for the reduced N i - C e 0 2 / M g O / A l 2 0 3 compared to the calcined N i -Ce0 2 /MgO/Al 2 03. This indicated that the Ce0 2 was reduced during TPR as well. • C e 0 2 V NiO T Ni -|—i—I—'—I—'—I—'—I—'—I—1—I—1—I—1—I—1 I 1 I 1 I 1 T" 20 25 30 35 40 45 50 55 60 65 70 75 80 20 (degree) Figure V-2 XRD pattern comparison for 12 wt. % N i - C e 0 2 / M g 0 / A l 2 0 3 after calcination and reduction 66 5.3 Cyclic Performance of the Supported Ni Catalyst The same procedures described in Section 4.3 were applied to perform cyclic reactions on the supported N i catalyst, except that the cycle number was increased from 4 to 6. The detailed experimental parameters are listed in Table V - l . Table V-l Experimental parameters for supported Ni catalyst Experiment a ' b ' c Experimental Parameters CH4 Decomposition Step Partial Oxidation Step CH 4 Cone. (vol.%) Reactor Temperature (K) Duration (min) 0 2 Cone. (vol.%) Reactor Temperature (K) Duration (min) Ni-4 5 773 25 10 773 5 Ni-5 5 5 Ni-6 2.5 10 Ni-7 20 5 . The same catalyst sample was used for all experiments without replacing catalyst b: The catalyst was re-reduced in situ in 40 % H 2 at 823 K for 2 hours before the start of experiment Ni-5, Ni-6 and Ni-7 c: The total flow rate passing through sample bed was 200 seem and the purge gas was He at 400 seem. Since the differences among the experiments were only in the 0 2 concentration and the duration of the partial oxidation step, the legend to describe each experiment used in the following figures was abbreviated to reflect these conditions. For example, the legend to describe the experiment Ni-4 was Ni4-10%O2-5min. Also note that because of a computer failure during experiment Ni-5, some data from the 4th cycle of experiment N i -5 were not available. 67 5.3.1 Hydrogen Production and Catalyst Deactivation As shown in Figure V-3, there was no significant deactivation of the supported Ni catalyst during the 6 cycles of the experiments Ni-4, Ni-5, and Ni-6 since the initial CH4 decomposition rate remained approximately constant. A small deactivation was observed in the 2nd cycle of experiment Ni-7. However, the initial C H 4 decomposition rate reached a stable value after the 2nd cycle for experiment Ni-7 and no further deactivation occurred after the 2nd cycle of experiment Ni-7. These data show that the supported N i catalyst can perform in a cyclic mode without significant deactivation under certain experimental conditions. Figure V-3 also shows that the initial CH4 decomposition rates measured in the 1st cycle were similar among the 4 experiments under the same experimental conditions for CH4 decomposition. No deactivation occurred from experiment Ni-4 to experiment Ni-7. The stable initial CH4 decomposition rate achieved during the 1st cycle among 4 different experiments indicated that the N i catalyst can be regenerated by in situ re-reduction by H 2 . A small increasing trend in the initial CH4 decomposition rate was observed from the 2nd to 6th cycle of Ni-4. The profiles of molar flow rate of H 2 produced in the first cycle of CH4 decomposition of experiment Ni-4, Ni-5, Ni-6 and Ni-7 are shown in Figure V-4. The C H 4 decomposition in the first cycle was performed immediately after in situ reduction of the catalyst in H 2 following the last oxidation of the previous experiment. 68 £• 1.4x10 H o 1.2x10s o: 1.0X10 c 1 8.0x10"-I in |- 6.0x10° o 8 4.0x10" Q I* 2.0x10" O ro 0.0 J ? 16x10s-I 1.4x10s-o •p- 1.2x10s-15 . a: 10x10 -c i 8.0x10" o 6.0x10" o g 4.0x10" a I* 2.0x10"-] o ro 0.0-1 Ni4-10%O,-5min 8 2 3 4 Cycle Number 2 3 4 5 Cycle Number _ 1.6x10s oT S 1.4x10s O £ 1.2x10 5 1.0x10" c s 8.0x10" o | 6.0x10" 2.0x10" 0.0. § 1.4x10* o I 1-2x10" a: 1.0x10 H c i 8.0x10" W O g- 6.0x10" * Ni6-2.5% O-10min R 5 4.0x10" H Q I* 2.0x10" O ro 0.0 • Ni5-5% O-5min 2 3 4 5 6 Cycle Number • T T T T Ni7-20% 0 -5min 2 3 4 5 Cycle Number Figure V -3 Initial C H 4 decomposition rate vs cycle number for the supported Ni catalyst of experiment Ni-4, Ni-5, Ni-6, and Ni-7 The decline in H2 molar flow rate with time indicated a slight deactivation of the supported N i catalyst during C H 4 decomposition. However, the rates of decline appear similar even when experimental conditions were changed during the partial oxidation step of the experiments. It is concluded that under the chosen experimental conditions, the supported N i catalyst can be regenerated by in situ reduction, without significant deactivation. It is noted that the profile of H2 molar flow rate is unique for experiment Ni-4. The H 2 molar flow rate reached a maximum initially and dropped suddenly and then increased 69 slowly followed by a slow decrease due to deactivation. Zhang and Smith (2005) proposed a deactivation model of supported Co catalyst (Co/SiC^) during C H 4 decomposition (23 % CH4/I2 % H2/65 % Ar) which predicted similar characteristics. However, the reason for the absence of a similar profile in experiments Ni-5, Ni-6, and Ni-7 is unclear. The only noticeable difference between Ni-4 and Ni-5, Ni-6, Ni-7 was that the supported N i catalyst was freshly loaded and had not been exposed to CH4 in experiment Ni-4. More detailed discussion of this observation will be presented in Section 6.3. -Ni4-10%O,-5min - Ni5-5% 0,-5min 3.5x10" 3.0x10' 2.5x10" 2.0x10" 1.5x10' 1.0x10"_H 5.0x10'7 0.0 3.5x10 3.0x10"" 2.5x10* 2.0x10s 1.5x10" 1.0x10" 5.0x10"' 0.0 4=1 200 400 600 800 1000 1200 1400 1600 Time(sec) Ni6-2.5%O-10min 0 200 400 600 800 1000 1200 1400 1600 Time(sec) 200 400 600 800 1000 1200 1400 1600 Time(sec) 3.5x10* 3.0x10* 2.5x10* 2.0x10*-1.5x10*-1.0x10' 5.0x10"'-o.o4=i - Ni7-20% O -5min 200 400 600 800 1000 1200 1400 1600 Time(sec) Figure V-4 Comparison of H 2 molar flow rate during the 1st cycle of C H 4 decomposition of experiment Ni-4, Ni-5, Ni-6 and Ni-7 70 The total H 2 production as a function of cycle number among the 4 experiments is shown in Figure V-5. Experiment Ni-4 showed an increasing trend in H 2 production as the cycle number increased. The total H 2 production reached approximately 4.0 mmol at the 6th cycle. The H 2 production during the 1st cycle of C H 4 decomposition of experiment Ni-5, Ni-6, and Ni-7 was 4.0 mmol as well. Since the catalyst was assumed to have the highest activity toward C H 4 decomposition during the 1st cycle (except the Ni-4) because the catalyst was just re-reduced in situ, 4.0 mmol of H 2 produced in the 1st cycle can be considered as the maximum amount obtained at the current C H 4 decomposition conditions. Therefore, the H 2 production shown in Figure V-5 is either stable as the cycle number increases (Ni-5) or decreases to different extents as indicated for Ni-6 and Ni-7. 4.2x10"'i o 3.8x10 E § 3.6x10° •o O 3.4x10°-| O) '§ 3.2x10° 8 3.0x1 o° 1 o- 2.8x10" £ 4.2x10° o S 4.0x10'J-| S 3.8x10° Q. E | 3.6x10°-| "O O 3.4x10° o> c 'g 3.2x10°-I TJ 1 •o 8 3.0x10"' T3 O o- 2.8x10° Ni4-10% 0;-5niin | 3 4 5 Cycle Number ~ Ni6-2.5%O,-10min I 5" 4.2x10"'-tion(n 4.0x10"'-o 3.8x10'-E 8 3.6x10°-T3 o 3.4x10°-during 3.2x10° s 3.0x10°-prodL 2.8x10° X ^ 4.2x103 o 4.0X10'3 o o 3.8x10"3 E £ 3.6x10"3 TJ O 3.4x10"3 c '5 3.2x10"3 TJ T3 S 3.0x10'3 TJ O 2.8x10"3 Ni5-5% 0,-5min 2 3 4 5 6 Cycie Number t Ni7-20% O-5min T • T Cycle Number 2 3 4 5 Cycle Number Figure V-5 H 2 production during C H 4 decomposition vs cycle number for the supported Ni catalyst of experiment Ni-4, Ni-5, Ni-6, and Ni-7 71 The reason for the low H2 production during the 1 st cycle and the subsequent increasing trend in H2 production observed in experiment Ni-4 is unclear. Apparently the activity of the supported N i catalyst was enhanced by the cyclic treatment of CH4 decomposition and partial oxidation. More experiments with unused catalyst will be presented in Section 6.3 to allow better comparison with experiment Ni-4. The H2 production was stable from cycle to cycle for experiment Ni-5. The experimental conditions of Ni-5 and Ni-6 show that equal amounts of O2 were introduced to the reactor during partial oxidation. However, the combination of a long oxidation period with a low concentration of O2 for the Ni-6 experiment showed a decrease in H2 production with cycle number. One possible reason for the loss in H2 production may be that not all the carbon was removed in the oxidation step of the cycle because of the low O2 concentration, and hence deposited carbon blocks active N i sites. Comparing experiments Ni-5 and Ni-7 show that the H2 production decreased dramatically from the 1st to the 2nd cycle of experiment Ni-7. The only difference in experimental conditions between Ni-5 and Ni-7 was the O2 concentration during the partial oxidation. Hence, deactivation is attributed to the oxidation of N i to NiO by the 20 % O2 used in the experiment Ni-7. Experiments of the present work have shown that under some conditions, a steady H2 production was obtained during CH4 decomposition for a series of cycles. The above 72 discussion also suggests that the cyclic performance varied according to the different experimental conditions. 5.3.2 I n d u c t i o n P e r i o d Figure V-6 shows the mass and effluent gas analysis from the first 3 cycles of experiment Ni-4. Due to the difficulty calibrating the quantity of water, the direct water intensity measurement is presented instead of molar flow rate. During the 1st cycle, CH4 had decomposed, H2 was produced and the mass of the supported N i catalyst increased as the carbon accumulated during CH4 decomposition. There was no CO, CO2 nor water observed during the 1st cycle. Consequently, pure H2 was the only, product during CH4 decomposition. During the 2nd cycle, a mass change was observed similar to the cyclic performance of the supported Co catalyst. An initial brief decrease in mass was observed and at the same time significant quantities of CO and CO2 were produced. Figure V-6 also shows the increase in water intensity during the 2nd and 3rd CH4 decomposition. The water was ascribed to be the product of reduction of NiO by C H 4 . Unlike the cyclic performance of the supported Co catalyst, however, there was no induction period required for the supported N i catalyst. 73 CH 4 Decomposition in the 1st Cycle CH 4 Decomposition in the 2nd Cycle CH 4 Decomposition in the 3rd Cycle 1.6x10"'i 1.4x10'-] „ 1.2x10' O) "Jo 1.0x10' 0) ^ 8.0x10° 6.0x10"' 4.0x10"' 2.0X10"1 0.0 -2.0x10° 7.0x10* .~. 6.0x10* £ 5.0x10* "Hf ro 4.0x10* rr 3 3.0x10" o U_ i- 2.0x10* _ro o 400 800 1200 1 600 2000 2400 Time(sec) 400 800 1200 1600 Time(sec) 0 400 800 1200 1600 2000 2400 Time(sec) - H2(mol/s) - CH/mol/s)! 7.0x10* 6.0x10* £ 5.0x10* aT ro 4.0x10* rr 3 3.0x10* o u_ i- 2.0x10* J? o •g 1.0x10* - 0.0 - l-ymol/s) 7.0x10" 6.0x10" 5.0x10* 4.0x10* 3.0x10* 2.0x10* 1.0x10* 0.0 - H,(mol/s) CH4(mol/s)| 400 800 1200 1600 2000 2400 71me{sec) 400 800 1200 1600 Time(sec) 400 800 1200 1600 2000 2400 Time(sec) O £ ro or 8 1.0x10* 9.0x10"' 8.0x10' -I 7.0x10"' 6.0x10*-| 5.0x10"'-I 4.0x10' 3.0x10"' 2.0x10"' 1.0x10"' 0.0 - CO(mol/s) - CO Onol/sJ E. 0 15 rr 3 o 800 1200 1 600 2000 2400 Time(sec) 400 800 1200 1600 Time(sec) - CO(mol/s) CO (mol/s)| 400 800 1200 1600 2000 2400 Time(sec) „ 7.0x10' • |2 6.0x10™ 5.0x10" £ — 4.0x10" S 0 400 800 1200 1600 2000 2400 Time(sec) 400 800 1200 1600 Time(sec) 0 400 800 1200 1600 2000 2400 Time(sec) ure V-6 Comparison of H 2 , C H 4 , C O , C 0 2 , water and mass during C H 4 decomposition for the first 3 cycles of experiment Ni-4 74 Figure V-7 shows the H 2 and C H 4 molar flow rate during C H 4 decomposition for experiment Ni-7. Even though significant deactivation occurred in the 2nd C H 4 decomposition cycle of experiment Ni-7, there was no induction period. As soon as C H 4 was introduced, H 2 was produced immediately. 7.0x10* 6.5x10"° 6.0x10"° 5.5x10"° "o 5.0x10"° E, 4.5x10* Is 4.0x10* CC s 3.5x10* o LL 3.0X10"6 i_ CO 2.5X10"6 o 2.0x10"6 1.5x10"° 1.0x10* 5.0x10"7 0.0 N N H 2 i N 5000 10000 15000 20000 Time(sec) 25000 30000 Figure V-7 Molar flow rate of H 2 and C H 4 during C H 4 decomposition of experiment Ni-7 The absence of an induction period indicated that the supported N i catalyst was easier to reduce after oxidation compared to the supported Co catalyst which required an induction period as shown in Figure V-8. 75 1.0x10" 9.0x10"e ^ 8.0x10"6-o 7.0x10 s-E i f ro i— o 6.0x10"°H 5.0x10"* 4.0x10"* ro o f I Induction : period Induction period i r 1500 3000 9000 10500 12000 18000 19500 21000 27000 28500 30000 2 CH Time(sec) Figure V-8 Molar flow rate of H 2 and C H 4 during C H 4 decomposition of experiment Co-1. Induction period is required during C H 4 decomposition from the 2nd cycle for supported Co catalyst The X R D patterns of the supported N i catalyst after partial oxidation and C H 4 decomposition are shown in Figure V-9. The presence of the NiO peaks was indicative of oxidation of Ni after partial oxidation. On the other hand, the disappearance of NiO and the appearance of N i indicate that NiO was reduced to N i after CH4 decomposition. Furthermore, the appearance of a graphite peak indicates that deposited carbon was present after CH4 decomposition. 76 I 1 I 1 I 1 I 1 I 1 I 1 I 1 I 1 I ' I 1 I 1 I 1 I 20 25 30 35 40 45 50 55 60 65 70 75 80 26 (degree) Figure V-9 XRD pattern comparison for 12 wt. % Ni-Ce02/MgO/AI203 after partial oxidation (10 % 0 2 , 773 K) and C H 4 decomposition (5 % C H 4 , 773 K) 5.3.3 Possible Source of Solid Oxygen and C O , C 0 2 Production during C H 4 Decomposition Although the detailed mechanism of CO formation during CH4 decomposition is unclear, a reaction between the deposited carbon and solid oxygen present in the catalyst to form CO is one possible explanation. The solid oxygen may come from the support (AI2O3), the promoters (MgO and Ce0 2 ), and NiO formed by the oxidation of N i during the partial oxidation step. 77 Unfortunately, the techniques used in the present study are unable to identify the source of the solid oxygen. However, it is likely that the solid oxygen from NiO dominates the amount of CO and CO2 produced during CH4 decomposition. Figure V-6 shows that there is no CO or CO2 produced during the 1st CH4 decomposition cycle. Before CH4 was introduced to initiate the 1st CH4 decomposition, the supported N i catalyst had been reduced in H2 and the extent of reduction was 85 %. The un-reduced NiO can be treated as inert due to a strong interaction between NiO and the support. Hence, the absence of CO and CO2 in the product during the 1st CH4 decomposition can be attributed to the lack of oxygen available from the reduced N i catalyst. On the other hand, CO and CO2 were obtained with high selectivity to CO during the subsequent CH4 decomposition cycles. Unlike the CH4 decomposition of the 1st cycle that occurred on the reduced N i catalyst, C H 4 was introduced onto the oxidized catalyst during the 2nd and subsequent CH4 decomposition cycles. The oxygen from NiO on the catalyst may react with CH4 or the deposited carbon to produce CO and CO2. In addition, the oxygen from Ce02 can not be completely excluded as a source of solid oxygen. As shown in Figure V - l , TPR indicated that Ce02 is reduced in H2 at about 650 K. Figure V-9 shows that the peaks associated with Ce02 were slightly reduced after CH4 decomposition compared to after partial oxidation. However, unless the amount of oxygen present in CO and CO2 exceeded the amount of oxygen from NiO, Ce02 as a source of oxygen is considered less likely. 78 In order to confirm that the oxygen uptake by N i is much greater than that of reduced Ce02, a temperature programmed oxidation (TPO) was carried out on the following 3 N i catalysts: 12 wt. % N i - C e 0 2 / M g O / A l 2 0 3 (TPOl), 12 wt. % N i - M g O / A l 2 0 3 (TP02) and 5 wt. % N i - C e 0 2 / M g O / A l 2 0 3 (TP03). The catalyst for TPO was reduced in situ in 40 % H 2 at 823 K for 2 hours followed by cooling to 323 K under He flow and holding at this temperature for 15 minutes. The reactant gas was then switched from He to 1 % 0 2 / 99 % He (Praxair, UHP, 99.999 %) and the reduced N i catalyst was held at 323 K for 8 minutes followed by heating from 323 K to 823 K at a ramp rate of 5 K/min, holding at 823 K for 15 minutes. The mass profile obtained by T E O M during the TPO represents the mass change due to the oxygen uptake of the catalyst. A blank run was also performed in order to obtain the mass change due to the temperature rise. The adjusted mass change of the catalyst during TPO was obtained by subtracting the data of the blank run. Figure V-10 shows the adjusted mass of the N i catalyst with and without Ce0 2 . The difference between the two mass profiles (12 wt. % Ni-Ce02/MgO/Al 2 0 3 minus 12 wt. % Ni-MgO/Al 2 0 3 ) , which reveals the effect of reduced Ce02 under oxidation, is also presented. As shown in Figure V-10, a peak in the difference mass profile indicated by the arrow, is present at 450 K. The peak can be ascribed to the oxygen uptake by the reduced Ce02. In addition, calculation shows that the mass increase due to the reduced Ce02 is only 15 % of the overall mass increase of the reduced Ni-Ce02/MgO/Al20 3 . The 79 data suggest that the oxygen uptake by the reduced Ce02 is low compared to the oxygen uptake by the N i . 5.0x10 4.5x1 O H 4.0x10 H — 1 I I I I I — . — I — I — I — I — I I— - ° — TP01, 12wt.% Ni-CeO./MgO/ALO, 1000 2000 3000 4000 5000 6000 7000 8000 9000 Time(sec) in in ro £ CU 5.0x10"v 4.5x10"3 4.0x10 1 3.5x10"3H 3.0x10"^ 2.5x10"; £ 2.0x10" to -TP02, 12wt.% Ni-MgO/AI 20 3 - Difference of mass, TP01 minus T P 0 2 . 900 850 800 750 700 -650 -600 -550 500 -450 -400 350 300 1000 2000 3000 4000 5000 6000 7000 8000 9000 Time(sec) Figure V-10 The comparison of adjusted mass change of reduced 12 wt. % Ni -Ce0 2 /MgO/Al20 3 and 12 wt. % Ni -MgO/Al 2 0 3 during the T P O Furthermore, Figure V - l l shows the correlation between the overall increase in mass measured by T E O M of each catalyst during TPO and the corresponding increase in mass 80 due to the oxidation of N i calculated according to the N i content of the catalyst. Including the origin in the linear fit, a high correlation coefficient (R=0.99) was obtained indicating that the overall increase in mass during TPO is strongly related to the N i content of the catalyst. Thus, we conclude that the oxygen uptake of the reduced catalyst is dominated by the content of N i and the oxidation of N i to NiO. o CL 3 T 3 CO CO CO E d) > o T3 CD i_ CO CO d> E 0) h-4.0x10 3.5x10 3.0x10 2.5x10 2.0x10' 1.5x10 1.0x10"H 5.0x10"" 0.0 T P 0 1 T P 0 2 T P 0 3 00 1.0x10'3 2.0x10'3 3.0x10'3 4.0x10'3 5.0x10'3 Theoretical increase in mass by Ni oxidation (g) Linear Regression Y = A + B * X Parameter Value Error A 2.82E-5 2.15E-4 B 0.84 0.08 R S D N P 0.99 2.59E-4 4 0.01 Figure V - l l The linear fit to show the correlation between the overall increase in mass during T P O and the corresponding increase in mass due to the oxidation of Ni according to the Ni content of the catalyst 81 As soon as the reactant gas was switched from He to 1 % 0^99 %He at 323 K during the TPO experiment, a sharp increase in mass (such as AW1 and AW2 for TPOl and TP02, respectively, as shown in Figure V-10) was observed for all the TPO experiments. This increase is not related to Ce02 because one catalyst was free of Ce02. The initial sharp increase in mass can be attributed to the oxygen uptake by reduced N i . A linear fit is established to show the correlation between the initial mass increase and the N i content of the catalyst for all the TPO experiments. As shown in Figure V - l 2 , the regression coefficient is 0.98033, which indicates that the initial mass increase is strongly related to the N i content of the catalyst. Previously, the oxygen uptake by reduced Ce02 was shown to occur at 450 K. Since the initial mass increase at 323 K is confirmed to be the oxygen uptake by reduced N i , it is reasonable to assume that the reduced N i reacts with oxygen faster than reduced Ce02-Since the oxygen uptake during oxidation is dominated by the oxidation of N i , it is reasonable to assume that the majority of oxygen atoms for CO and CO2 production during C H 4 decomposition are provided by NiO. Figure V - l 3 shows a comparison of the amount of oxygen atoms obtained from CO and CO2 during CH4 decomposition and the oxygen atoms available from the NiO content of the oxidized catalyst. The dashed line indicates the amount of oxygen atoms available from NiO if N i was completely oxidized to NiO according to the degree of reduction. The solid symbols represent the measured amount of oxygen from CO and CO2 as a function of cycle number for experiments Ni-4, Ni-5, Ni-6 and Ni-7. 82 2.0x10"3-O 1.8x10"3-CL 1.6x10'3-t-D) C 1.4x10"3-1.2x10"3-se T P 0 3 CO 1.0x10 -0 i— • • o 8.0x10"4-ass 6.0x10"4-ass • / E 4.0x10"1- / ' "ro 2.0x10"" -0.0- • , , , r 0.0 TP01 T P 0 2 3.0x10"3 6.0x10"3 9.0x10"3 1.2x10"2 1.5x10"2 1.8x10'2 Ni content of the catalyst (g) A l ( X I ) A2(X2) A3 (Y2) Point Mass of Ni of C a t a l y s t ( g ) I n i t i a l mass i n c r e a s e (g) 1 O r i g i n 0 _0 2 TPOl 0.01654 .Lvl.lL-l.LLG.wi.L .l,:..6.1.ErjJJ.GW2 ) 3 TP02 0.01531 4 TP03 6.0051 8 . 73E-4 Linear Regression: Y = A + B * X Parameter Value Error A 1.51E-4 1.61E-4 B 0.10 0.01 R S D N P 0.98 1.94E-4 4 0.02 Figure V-12 The linear fit to show the correlation between the Ni content of the catalyst and the initial mass increasing due to oxygen uptake 83 2.2x10'" 2.0x10'" 1.8x10'" 1.6x10"" 1.4x10"" 1.2x10"" 1.0x10"" 8.0x10"5 6.0x10'5 * Oxygen atoms from CO and CO ; for NJ4-10% Q7-5min — Amount of Oxygen atoms from NiO I • Oxygen atoms from CO and CO for Ni5-5% O-5min 2.2x10 -iL 2.0x10'" 1.8x10'" 1.6x10"" 1.4x10'" 1.2x10'" 1.0x10'" 8.0x10'5-| 6.0x10': 3 4 Cycle Number Oxygen atoms from CO and CO for Ni6-2.5% O -10min - Amount of Oxygen atoms from NiO 3 4 5 Cycle Number 2.2x10'"n 2.0x10" 1.8x10"" 1.6x10'" 1.4x10"" 1.2x10'" 1.0x10'"-] 8.0x10' 6.0x10 2.2x10"' 2.0x10" 1.8x10"' 1.6x10"" 1.4x10'' 1.2x10"" 1.0x10'" 8.0x10'5 6.0x10s - Amount of Oxygen atoms from NiO Cycle Number Oxygen atoms from CO and CO for Ni7-20% O -5min I Amount of Oxygen atoms from NiO 3 4 5 Cycle Number Figure V-13 Comparison of oxygen (moles) obtained between NiO and C O , C 0 2 during C H 4 decomposition for each cycle of experiment Ni-4, Ni-5, Ni-6 and Ni-7 There was no CO and CO2 detected during the CH4 decomposition in the 1st cycle for all 4 experiments. Taking account of the errors associated with the (CO + CO2) measurement, all the measured data were below the calculated maximum. Hence the oxygen atoms needed to produce the measured amount of CO and CO2 in the product was always less than the amount of oxygen available from NiO. 84 Another observation from Figure V-13 is that the measured oxygen atoms from CO and CO2 decrease as the cycle number increases for all 4 experiments. If we assume that the oxygen for CO and CO2 formation came from NiO, we can conclude that the amount of NiO decreased as the cycle number increased. Calculation also shows that the amount of oxygen introduced during partial oxidation among experiments Ni-4, Ni-5, Ni-6 and Ni-7 was at least 10 times more than the oxygen needed to completely oxidize the Ni . Therefore, insufficient oxygen supplied during the partial oxidation step was not the likely cause of decreasing NiO as the cycle number increased. Consequently, a reduction in the amount of N i available for oxidation to NiO as the cycle number increased must be the reason for the decline in CO and CO2 with cycle number. One possible explanation for the loss in available N i is that the carbon deposited on N i during the CH4 decomposition cycles was not completely removed by the subsequent partial oxidation. Hence, N i associated with carbon, presumably as Ni3C, was unable to react with oxygen to form NiO. Based on this assumption, an increase in un-removed carbon would result in a decrease in NiO formation during partial oxidation; therefore, the oxygen available for formation of CO and CO2 during the CH4 decomposition cycles would decrease. Figure V-14 shows the deposited carbon removal percentage of each cycle of experiment Ni-4, Ni-5, Ni-6 and Ni-7. The theoretical amount of un-removed deposited carbon of each cycle can be calculated for all 4 experiments. The cumulative amount of un-85 removed carbon on the catalyst at the designated cycle number can be further calculated by adding the un-removed carbon from cycle number one to the designated cycle. 130%-120%-§> 110%-| 100%-g. 90%-1 80%-| 70%-| 60%-8 50%-40%-Ni4-10% 0 -5min 2 3 4 5 Cycle Number 130% 120% D> 110% | 100% g. 90% 1 80% | 70% | 60%-) O 50% 40% Ni5-5% 0,-5min 2 3 4 5 Cycle Number 130%-120%-§> 110%-§ 100%-I. 90%-to g 80%-| 70%-| 60%-<3 50%-Ni6-2.5%0-10min 2 3 4 5 Cycle Number 130% 120% oi 110% § 100% JL 90% > 80% | 70%. | 60% O 50% 40% • T 2 3 4 5 Cycle Number Figure V-14 Carbon removal percentage vs cycle number of experiment Ni-4, Ni-5, Ni-6 and Ni-7 Figure V-15 shows the correlation between the amount of oxygen obtained from CO and CO2 during C H 4 decomposition and the accumulated carbon for cycle 2 to 6 of experiment Ni-4, Ni-6 and Ni-7. A consistent strong negative correlation is observed for all 3 experiments. The analysis confirms the assumption that the amount of oxygen available for formation of CO and CO2 during CH4 decomposition decreases as the accumulated carbon on the catalyst increases. 86 The un-removed carbon associated with N i could be filamentous or encapsulating carbon. It is well known that the initial CH4 decomposition rate is proportional to active N i sites during C H 4 decomposition. Therefore, once active N i sites are blocked due to encapsulating carbon, the initial C H 4 decomposition rate decreases. In contrast, filamentous carbon formed on N i does not cause a decrease in the initial C H 4 decomposition rate since active N i sites are not blocked. Figure V-16 shows the correlation between the initial C H 4 decomposition rate and accumulated carbon for cycles 2 to 6 of experiment Ni-4, Ni-6 and Ni-7. No significant correlation or trend was observed corresponding to experiment Ni-4, 6 and 7. The inconsistent correlations suggest that the initial C H 4 decomposition rate does not relate to the accumulated carbon deposit. Hence, it is concluded that the un-removed deposited carbon must be filamentous carbon. 87 O ° 2.1x10'-E c £ :1 2.0x10"* • I 2.0x10"' o cu X3 l"l .9x10"'' O O) 'E 1.8x10"' Ni4-10%O-5min o £ 1.5x10 1.4x10' 1.3x10' 1.2x10' 1.1x10' 1.0x10* 9.0x10* 8.0x10' 7.0x10 s 6.0x10 s 5.0x10"* 4.0x10"' 1.0x10"' 2.0x10"' 3.0x10* 4.0x10* 5.0x10* Accumulated deposited carbon (mole) Ni6-2.5% 0,-1 Omin Linear Regress ion : Y = A + B * X Parameter Va lue Error A 2 .15427E-4 2 .40327E-6 B -0 .06369 0.00857 R S D N P -0.9739 2 .49049E-6 5 0.00504 Linear Regression: Y = A + B * X Parameter Value Error A 1.69552E-4 1.67476E-5 B -0.01937 0.00383 R S D N P -0.94613 1.09577E-5 5 0.01489 2.0x10* 3.0x10"' 4.0x10"' 5.0x10" 6.0x10' 7.0x10"' Accumulated deposited carbon (mole) o E 1.35x10 1.33x10' 1.30x10' 1.28x10* 1.25x10* 1.23x10* 1.20x10* 1.18x10* 1.15x10' 1.13x10* 1.10x10* 1.08x10* 1 T Ni7-20% 0 2-5min .0x10* 2.0x10* 3.0x10* 4.0x10* 5.0x10* Accumulated deposited carbon (mole) Linear Regression: Y = A + B * X Parameter Value Error A 1.41907E-4 2.6146E-6 B -0.05145 0.00666 R S D N P -0.97578 2.20509E-6 5 0.00451 Figure V-15 Correlations between the amount of oxygen obtained from C O and C 0 2 during C H 4 decomposition and the accumulated deposited carbon of Ni-4, Ni-6 and Ni-7 88 o 1.20x10°-I Ni4-10% 0-5min o 1.29x10 H |S 1.26x10"' -| 1.14x10*-J 1.0x10'" 2.0x10" 3.0x10" 4.0x10'" 5.0x10"' Accumulated deposited carbon (mole) Linear Regression: Y = A + B * X Parameter Value Error A 1.13055E-5 1.8248E-7 B 0.004 6.50707E-4 R SD N P 0.96256 1.89103E-7 5 0.00865 2 1.40x10 H £ 1.35x10"N I O Ni6-2.5% 0-10min 2.0x10" 3.0x103 4.0x10'3 5.0x103 6.0x10* Accumulated deposited carbon (mole) Linear Regression: Y = A + B * X Parameter Value Error A B 1.563E-5 3.52996E-7 -4.56507E-4 8.06755E-5 SD -0.95621 2.30959E-7 0.01093 Ni7-20% 0 2 -5min S 1.0X10"5 I 1 1.0x10* | 9.9x10*-c g :S 9.8x10*-o Q. g 9.7x10* 0) Q j-- 9.6x10* O S 9.5x10* 1.0x10'" 2.0x10" 3.0x10'" 4.0x10" S.OxlO'* 6.0x10" Accumulated deposited carbon (mole) Linear Regression: Y = A + B * X Parameter Value Error A 9.55485E-6 1.72532E-7 B 9.98767E-4 4.39465E-4 R SD N P 0.79535 1.45509E-7 5 0.10766 Figure V-16 Correlations between the initial C H 4 decomposition rate and the accumulated deposited carbon of Ni-4, Ni-6 and Ni-7 89 5.3.4 Carbon Removal Percentage and Selectivity to C O during Oxidation The carbon removal percentage varied according to the different operating conditions during partial oxidation as shown in Figure V - l 4 . 100 % carbon removal was achieved for experiment Ni-4 with 10 % 0 2 and 5 minutes reaction duration during partial oxidation, proving that the deposited carbon can be completely removed by oxidation with O2 under the chosen experimental conditions for supported N i catalyst. Equal amounts of O2 were introduced during partial oxidation for experiment Ni-5 and Ni-6. However, different carbon removal percentages for Ni-5 (approximately 60 %) and Ni-6 (approximately 50 %) were observed. Furthermore, Figure V - l 7 shows a different selectivity to CO during partial oxidation of Ni-5 (22-27 %) and Ni-6 (15-20 %). It is possible that the carbon removal percentage and selectivity to CO may be adjusted by changing the parameters such as concentration of O2 introduced or reaction time of the partial oxidation. 90 % 15% Ni4-10% O-5min 40% c o I 35% o 1 30% ra o_ | 25% TJ R 20% 3 4 Cycle Number a Ni6-2.5%Cy10min | 40%-, o I 35%-j o I 30% ra c 25% *c u TJ 8 20% o f 15% 3 4 Cycle Number Ni5-5% Q.,-5min | 2 3 4 5 Cycle Number Ni7-20% 0;-5min | 2 3 4 5 Cycle Number Figure V-17 Selectivity to C O during partial oxidation of experiment Ni-4, Ni-5, Ni-6 and Ni-7 Figure V - l 8 shows the comparison of molar flow rate of CO, CO2 and O2 during the 1st, 3rd, 5th cycle of partial oxidation for the Ni-5 and Ni-6 experiments. Data indicate that the maximum formation of CO2 is determined by the concentration of O2 introduced. In addition, the introduced oxygen was completely consumed initially and reacted with deposited carbon to produce CO and CO2. The initial complete consumption was also observed as shown in Figure V-20 of experiment Ni-7, with 20 % O2 introduced. Hence, the initial oxidation rate of deposited carbon as shown in Figure V - l 9 is an apparent oxidation rate since the rate is limited by the O2 concentration. The initial oxidation rate of deposited carbon was measured from the initial change in CO and CO2 molar flow rate in the effluent gas analysis. 91 1.0x10 -9.0x10"6-j 8.0x10*: 7.0x10"6; 6.0x10'6-5.0x10'6-4.0x10"°: 3.0x10*-2.0x10"°: 1.0x10'6-0.0-Ni5-5% 0 -5min, cycle 1 •CO CO„ 200 400 600 Time(sec) o £ CD i— o l _ _CD O 5 800 1000 1.0x10 9.0x10"°-] 8.0x10* 7.0x10* 6.0x10* 5.0x10* 4.0x10* 3.0x10* 2.0x10* 1.0x10* -I 0.0 0 Ni6-2.5% O2-10min, cycle 1 200 400 600 Time(sec) CO CO, 800 1000 "a? ~o 1.0x10% 9.0x10"6-E 8.0x10*-/ rate 7.0x10*-/ rate 6.0x10*-ar flov 5.0x10"6-4.0x10'6-o 3.0x10*-2.0x10'6-1.0x10'6-0.0-Ni5-5% O -5min, cycle 3 200 400 600 Time(sec) o E, £ ro * 5 o c= *_ ro o 800 1000 1.0x10 n 9.0x10*-8.0x10"6-7.0x10*: 6.0x10*-5.0x10 4.0x10 3.0x10 2.0x10 1.0x10 0.0 Ni6-2.5% O -10min, cycle 3 400 600 Time(sec) 1000 1.0x10 9.0x10* 8.0x10* 7.0x10* 6.0x10"6 5.0x10"6 4.0x10* 3.0x10'6 2.0x10'6-1.0x10*-0.0-Ni5-5% O -5min, cycle 5 1.0x10" i ^ 9.0x10"° ro o _ro o 0 200 400 600 800 1000 Time(sec) 7.0x10 H 6.0x10* 5.0x10* 4.0x10* 3.0x10° Ni6-2.5% O -10min, cycle 5 1000 Time(sec) Figure V-18 Comparison of molar flow rate of C O , C 0 2 and 02 during partial oxidation of cycle 1, 3, and 5 of experiment Ni-5 and Ni-6 92 In general, the initial oxidation rate of deposited carbon is proportional to the O2 concentration and remains stable as the cycle number increases as shown in Figure V-19. However, a decrease in the initial oxidation rate at the 2nd cycle of experiment Ni-7 is observed, and this decrease is attributed to the decrease in the amount of deposited carbon for Ni-7, as shown in Figure V-5. The H2 production decreased in the 2nd cycle of CH4 decomposition and therefore, the amount of carbon deposit decreased. The deactivation in Ni-7 is attributed to the oxidation of N i to NiO in the 20 % 0 2 of Ni-7. Figure V-20 shows the molar flow rate of CO, CO2 and O2 during the partial oxidation at cycle 1 to 3 of experiment Ni-7. Unlike the stable profile for CO2, a decrease in CO formation from cycle 1 to 2 was observed. It is obvious that the decrease in initial oxidation rate of deposited carbon was dominated by the decrease in initial change of CO molar flow rate. Thus, the initial oxidation rate of deposited carbon during partial oxidation of Ni-7 was limited by the deposited carbon from cycle 2 to 6 instead of by O2 concentration at cycle 1. The formation of CO and CO2 is not only determined by experimental parameters of the partial oxidation step such as O2 concentration and the reaction duration, but also the amount and type of deposited carbon formed during the preceding CH4 decomposition. As shown in Figure V-20, the amount of CO decreases from cycle 1 to cycle 2 but CO2 does not. This suggests that more than one type of deposited carbon formed during CH4 decomposition and one of types of carbon favour the formation of CO. Djaidja et al. (2006) also reported that 4 different types of carbon including carbon containing 93 hydrogen (CH X species), and/or surface carbon, nickel carbide and carbon nanotubes formed during dry reforming of C H 4 on supported N i catalyst. 05 B) 5.0x10"4-I 4.5x10"4-~g 4.0x10'4-\ "I 3.5x10"4-o ]jj 3.0x10"4-1 2.5x10"4-j S 2.0x10"4-S 1.5x10"4-ro 1.0x10"' 5.0x10'5H 1 0.0-E ® ® -A. A-- • — Ni4-10% 02-5min - •— Ni5-5% 02-5min A Ni6-2.5% O2-10min Ni7-20% 02-5min ~~i 1 1 1 1 2 3 4 Cycle Number 5 Figure V-19 Comparison of initial oxidation rate of deposited carbon during partial oxidation of each cycle of experiment Ni-4, Ni-5, Ni-6 and Ni-7 The extent of carbon removal and oxidation of N i during the partial oxidation step can result in different deactivation and H2 production in the subsequent C H 4 decomposition step. The effect of interaction caused by experimental parameters between cycles can be examined as the cycle number increases. Therefore, an investigation of the effects of interactions between cycles at different experimental parameters is needed and is discussed in Chapter 6. 94 3.5x1(f 3.0x10"5 1* 2.5x10"6 o E, £ 2.0x10"5 CO 3 o 05 o 1.5x10 2 LOxlO'5-5.0x10'6-0.0 Cycle 1 Ni7-20% 02-5min Cycle 2 •CO CO, Cycle 3 T * 1 ' 1 ' 0 100 200 300 0 100 200 300 0 100 200 300 Time(sec) Time(sec) Time(sec) Figure V-20 Comparison of molar flow rate of C O , C 0 2 and 0 2 during partial oxidation of cycle 1, 2, and 3 of experiment Ni-7 5.4 Using CO2 Instead of 02 to Remove the Carbon Deposit In Section 5.3.2, the conversion of N i to NiO during the partial oxidation step was proven. The presence of NiO not only reduced the catalyst activity for CH4 decomposition, but also provided the source of solid oxygen which resulted in the formation of CO during the subsequent C H 4 decomposition step. Figure V-21 shows that the CO to H2 ratio ranged from 1 to 5 % during CH4 decomposition among the 4 experiments. There was no CO produced in the 1 st cycle of the CH4 decomposition since the catalyst was reduced in situ which minimized the presence of NiO. 95 c o X O £Z 5.0% 4.5% w o D. E o 8 4.0% -I 3.5% 3.0% -| 2.5% 3 T3 OJ eg 2.0% | 1.5% d) CL £Z o i_ CM X o O O 1.0% 0.5% H 0.0% A - r -4 • Ni4-10% 0 2 -5min • Ni5-5% 0 2 -5min A Ni6-2.5% O 2 -10min T Ni7-20% 0 2 -5min 5 - r -6 Cycle Number Figure V-21 Comparison of C O to H 2 ratio (in percentage) during C H 4 decomposition from the 2nd to 6th cycle of experiments Ni-4, Ni-5, Ni-6 and Ni-7 To reduce the CO in H2 during the CH4 decomposition step, the formation of NiO during the partial oxidation step must be minimized. A n alternative approach to oxidation of the carbon is reacting it with CO2 instead of O2. In other words, instead of the partial oxidation reaction as shown in Equation 5.1, the reaction shown in Equation 5.2 was carried out to remove the carbon deposit and produce CO. (n+m)C + QAn + m)02nCO + mC02 (5.1) C + C 0 2 - ^ 2 C O (5.2) Two experiments were conducted for comparison to investigate the feasibility of deposited carbon removal by CO2. The detailed experimental parameters are listed in Table V-2. 96 Table V-2 Experimental parameters of Ni catalyst for comparison between deposited carbon removal by 0 2 and C 0 2 Experiment Experimental Parameters CH4 Decomposition Step Carbon Removal Step C H 4 Cone. (vol.%) Reactor Temperature (K) Decompose Duration (min) Gas used Gas Cone. (vol.%) Reactor Temperature (K) Duration (min) Ni-8-0 2 5 748 25 0 2 10 773 5 Ni-9-C0 2 15 c o 2 40 100 Unused calcined supported N i catalyst was loaded for each experiment and reduced in situ before cyclic CH4 decomposition followed by carbon removal, reacting either 0 2 or C 0 2 for 5 cycles. High C 0 2 concentration and long reaction duration were applied for experiment Ni -9 -C0 2 to ensure that the reaction between deposited carbon and C 0 2 was complete. "Carbon removal by 0 2 " and "carbon removal by C 0 2 " are the legends used in the following figures to simply distinguish experiment Ni -8 -0 2 and Ni -9 -C0 2 . The main objective of these two experiments was to provide answers to the following questions: 1. Can the formation of NiO on the N i catalyst be reduced and result in the reduction of CO in H 2 during the subsequent CH4 decomposition i f deposited carbon is removed by reacting with C 0 2 instead of 0 2 ? 2. Can the active N i sites still be regenerated and is there any deactivation during CH4 decomposition? 97 3. How fast is deposited carbon removed by reacting with CO2 compared to O2? Is the carbon removal rate by CO2 reasonable compared to the initial C H 4 decomposition rate, a requirement for practical application with dual reactors? Figure V-22 shows a X R D comparison of the two catalysts after the final cycle of carbon removal by reacting with CO2 or O2. NiO is identified on the catalyst for which the deposited carbon was removed by partial oxidation with O2 and no N i peak was detected. On the other hand, the X R D of the catalyst reacted with C 0 2 shows the presence of N i and no NiO. The data indicate that by reacting deposited carbon with CO2, the formation of bulk NiO is prevented. 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 29 (degree) Figure V-22 XRD pattern comparison of the Ni catalyst after carbon removal by 0 2 and C 0 2 98 Without bulk NiO on the catalyst after carbon removal by CO2 (N1-9-CO2), the formation of CO during the subsequent CH4 decomposition should be reduced since the source of solid oxygen is reduced. Figure V-23 shows that CO to H2 ratio during CH4 decomposition of experiment Ni-9-C02 is lower than that with O2. 4.5%-| 4.0%-3.5% c o (75 o o. E o o CD T3 O 3.0% D) C *i_ T3 CD D) TO C CD O t_ CD Q. C O ro I _o O O 2.5% 2.0% H 1.5% 1.0% 0.5% 0.0% 2 i 3 carbon removal by 0 2 carbon removal by C O 4 Cycle Number Figure V-23 Comparison of C O to H 2 ratio (in percentage) during C H 4 decomposition from the 2nd to 5th cycle of experiment Ni-8-0 2 and Ni -9 -C0 2 As shown in Figure V-24, similar and high carbon removal percentage is achieved with CO2 and O2. Takenaka et al. (2004) reported that the carbon nanofibers produced during CH4 decomposition on supported N i catalyst were very active and the carbon nanofibers were converted into CO with a conversion higher than 95 % by gasification with CO2, similar to the carbon removal percentage reported herein. 99 With the similar carbon removal percentage achieved by 0 2 and CO2, the regeneration of active N i sites should be the same and consequently a similar initial C H 4 decomposition rate is observed for both experiments as shown in Figure V-25. 100% 90% -\ S, 80% H 70% H c CD O \— <D CL ro > o E CD C o 1_ ro O 6 0 % -50% -4 0 % -30% 20% 10% 0% A carbon removal by 0 2 • carbon removal by C 0 2 2 3 4 Cycle Number 5 Figure V-24 Comparison of carbon removal percentage of each cycle of experiment Ni-8-0 2 and Ni -9 -C0 2 However, the carbon removal rate by CO2 is much slower (by a factor of approximately 20) compared to that obtained with O2, as shown in Figure V-26. The slow carbon removal rate by CO2 will slow down the continuous process in which similar reaction rates of CH4 decomposition and carbon removal are desired. 100 x 1.4x10"5 3 | 1.2x10"5 & 5 « 1.0x10 • | 8.0x10"6 co | 6.0x10"6 o g 4.0x10"6 O 2.0x10"6 co ->—* 0.0 • A • A • * carbon removal by 0 2 • carbon removal by C O - r -2 i 3 4 - r -5 Cycle Number Figure V-25 Comparison of initial C H 4 decomposition rate of experiment Ni-8-0 2 and Ni -9 -C0 2 CD O "5 £ ro \ > o E CD C o i ro O 1.50x10"4i 1.25x10"4 1.00x10"4 9.0x10"6 6.0x10"6 3.0x10"6 0.0 carbon removal by 0 2 carbon removal by C 0 2 2 —T— 4 -r— 5 Cycle Number Figure V-26 Comparison of carbon removal rate of experiment Ni-8-0 2 and Ni -9 -C0 2 101 Although the amount of CO in the H2 is reduced by reacting deposited carbon with CO2 instead of O2, thereby minimizing the formation of NiO, the carbon removal rate by CO2 is too slow for any practical application. Furthermore, unlike the exothermic O2 reaction (Equation 5.1), which can provide heat for the CH4 decomposition reaction, the CO2 reaction is endothermic (Equation 5.2). 5.5 Summary The degree of reduction of the supported N i catalyst is higher than the supported Co catalyst. Ce02 is confirmed to be reduced during TPR. The N i catalyst is shown to operate in a cyclic mode without significant deactivation. Steady H2 production was achieved during CH4 decomposition for a series of cycles. The supported N i catalyst is re-usable since the catalyst can be regenerated by in situ re-reduction by H2. X R D identified the oxidation of N i to NiO during the partial oxidation step. However, NiO can be rapidly reduced by reacting with C H 4 and, hence, no induction period was observed for the N i catalyst at the current experimental conditions. TPO confirmed that the oxygen uptake by reduced NiO is much greater than by reduced Ce02. Hence, the source of oxygen for CO and CO2 production during CH4 decomposition is most likely the oxygen associated with N i . The solid oxygen decreased as the accumulated carbon on the catalyst increased. By introducing C02to remove the deposited carbon, the formation of NiO was minimized. Hence, the CO to H2 ratio during CH4 decomposition decreased when carbon removal 102 was regenerated by C O 2 compared to 0 2 . Furthermore, similar regeneration ability of catalyst was achieved by introducing O 2 and C O 2 . However, the rate of carbon removal by C O 2 was too slow for any practical application of the cyclic reaction. A unique H 2 production profile is identified during the 1st cycle of C H 4 decomposition on a freshly loaded catalyst. This feature is qualitatively in agreement with the deactivation model proposed by Zhang and Smith (2005) on supported Co catalyst during CH4 decomposition. Based on the conclusion from the results presented in this chapter, the next series of experiments should include the following: 1. The supported N i catalyst should be used. 2. For each experiment, unused calcined catalyst, with in situ reduction, is required. 3. An experimental design which allows one to investigate the effects of interactions between different experimental parameters, with a minimum number of experiments, is required. 103 Chapter VI Fractional Factorial Analysis on the Supported Ni Catalyst 6.1 Introduction The results from Chapter 5 demonstrate that N i catalyst can operate in a cyclic mode between CH4 decomposition and partial oxidation without significant deactivation as the number of cycles increases. To investigate the effects of interaction between experimental parameters, Section 6.2 describes a half-fractional of a full factorial design with five factors at two levels. The factors are the experimental variables. The statistical analysis results are presented in Section 6.5. A l l the experiments presented in this chapter used fresh, calcined, supported N i catalyst. Results from the first C H 4 decomposition step can be isolated to verify the unique features of the H2 production profile identified in the 1 st cycle of CH4 decomposition in the previous chapter. Section 6.3 shows that the unique feature was observed consistently in the first CH4 decomposition step with fresh N i catalysts. Section 6.4 analyzes the N i particle size change that occurred at different steps of the cyclic operation. X R D and T E M were both used to reveal the evolution of N i particle size. The size of N i may be related to the activity of CH4 decomposition. The BET surface area measurements of some selected N i catalysts with different carbon removal percentage are presented in Section 6.5.3. 104 6.2 Method A half-fractional of a full factorial design with five factors at two levels was applied in order to investigate the main effect and the interactions among these factors. The factors examined were the experimental variables: gas composition during oxidation, reaction temperatures and the duration of the CH4 decomposition and partial oxidation steps. The factors with low and high levels are listed in Table VI-1. Due to the temperature limit of the TEOM, 823 K was the maximum reaction temperature investigated. Table VI-1 The factors with low and high levels used in the factorial design Factors: Low Level High Level CH4 decomposition temperature (K) 723 823 CH4 decomposition duration (min) 3 50 POX 0 2 concentration (%) 2.5 20 POX temperature (K) 723 823 POX duration (min) 3 15 Table VI-2 shows the list of the 16 (25"1) experiments in a randomized run order and the experimental variables employed from the factors listed in Table VI-1 for the half-fractional factorial design. The concentration of CH4 introduced during C H 4 decomposition was 5 % for all experiments. Unused calcined supported N i catalyst was loaded for each experiment and reduced in situ in 40 % H 2 at 823 K for 2 hours before each experiment. Each experiment consisted of 6 cycles of C H 4 decomposition followed by partial oxidation. Once all 16 experiments were completed, the following values were calculated and treated as responses for statistical analysis: 105 1. Initial CH4 decomposition rate (mol/(g-s)) 2. Average H 2 production rate (mol/s) 3. CO to H 2 ratio during the C H 4 decomposition step 4. Selectivity to CO during the partial oxidation step 5. Carbon removal percentage during oxidation In addition, the CH4 decomposition in the 1st cycle was extracted to compare with the deactivation model on supported Co catalyst proposed by Zhang and Smith (2005), since the N i catalysts were all un-used and reduced in situ prior to the C H 4 decomposition. Excluding partial oxidation, the variables among the 16 experiments for the first C H 4 decomposition are temperature and duration. Table VI-2 A 25"1 fractional factorial design at two levels of the supported Ni catalyst CH 4 decomposition CH 4 decomposition POX 0 2 POX POX Experiment temperature duration Cone. temperature duration (K) (min) (%) (K) (min) FD-1 823 3 20 723 15 FD-2 823 3 2.5 823 15 FD-3 823 50 20 823 15 FD-4 723 50 2.5 823 15 FD-5 723 50 20 723 15 FD-6 823 3 2.5 723 3 FD-7 723 3 2.5 723 15 FD-8 723 3 20 723 3 FD-9 823 50 2.5 823 3 FD-10 . 823 50 20 723 3 FD-11 723 50' ' 2.5 723 3 FD-12 823 50 2.5 723 15 FD-13 ... 723 50 20 823 3 FD-14 723 3 " 2.5 823 3 FD-15 723 3 20 823 15 FD-16 823 3 20 823 3 106 6.3 Methane Decomposition and Deactivation of Catalyst In Section 5.3.1, Figure V-4 showed a unique feature of the H 2 production profile during CH4 decomposition of experiment Ni-4, in which the H 2 molar flow rate reached a maximum initially, decreased rapidly and then increased slowly before decreasing again. However, this feature was only observed during the first CH4 decomposition step of experiment Ni-4. No such feature was observed after the 1st cycle of experiment Ni-4 and any other cycle of experiment Ni-5, Ni-6 and Ni-7, all of which were done with the same sample of catalyst. Figure VI-1 shows the H 2 production profile during the first CH4 decomposition on the N i catalyst at 723 K and 823 K. The feature was observed for each of the experiments with a long CH4 decomposition duration. A l l the activity profiles show that the H 2 molar flow rate reached a maximum initially because the N i catalyst, having been reduced had a maximum number of active N i sites available for C H 4 decomposition. The H 2 molar flow rate decreased rapidly after reaching a maximum due to a decrease in active N i sites because of site blocking by carbon. According to Zhang and Smith (2005), the deposited carbon can either diffuse through the metal particle and subsequently nucleate and grow filaments at the tailing face of the particle, or it can form encapsulating carbon on the leading face. If the carbon diffuses through N i , the surface active sites become available and an activity increase is observed. The H 2 production profile observed is the net rate among the various rate processes. These processes include stepwise CH4 dehydrogenation as described by Zhang and Smith 107 (2005), the rate of encapsulating carbon formation, the rate of carbon diffusion through the metal particle, the rate of carbon nucleation at the tailing face and the rate of filament growth. Figure VI-2 shows a T E M image of the N i catalyst after CH4 decomposition. The formation of filamentous carbon with N i located at the tip of the filamentous carbon is clearly visible. The metal particle is pear-like in shape, and similar observations have been made by others (Otsuka et al., 2003; Takenaka et al, 2004; L i et al. 2006). Figure VI-3 shows the filamentous carbon at a higher magnification. The angle between the graphene layer and the longitudinal axis of the carbon fiber is glancing, and a similar observation was presented by Toebes et al. (2002) and L i et al. (2006). Zhang and Smith (2005) also reported that the activity profiles showed a period of either stable activity or decreasing activity on supported Co catalysts, depending on the reaction conditions after the period of increasing activity. As shown in Figure VI-1, stable activity at 723 K and deactivation at 823 K were also observed on the supported N i catalyst. Moreover, similar to experiment Ni-4, this feature was only observed in the 1 st cycle of C H 4 decomposition. H2 production profiles for all 6 cycles of FD-4 (723 K) and FD-3 (823 K) indicating the typical difference between the 1st and other cycles are shown in Figure VI-4 and Figure VI-5. The major difference is that the H2 molar flow rate did not show the sharp maximum activity initially, followed by a rapid decline, during the 2nd to 6th cycles of CH4 decomposition. 108 CH decompose at 723K 3.0x10 2.5X10"6 2.0x1 0* 1.5x10* 1.0X10"6 5.0x10'7 0.0 -FD-4| 1000 1200 1400 1600 Time(sec) 1800 2000 E 3.0x10 2.5x10* 2.0x10* 1.5X10"6 1.0x10* 5.0X10'7 0.0 -FD-5| 1000 1200 1400 1600 Time(sec) 1800 2000 3.0x10 2.5x10"6 2.0x10* 1.5x10* 1.0x10* 5.0x10"7 0.0 1000 1200 1400 1600 Time(sec) 3.0x10 2.5x10* 2.0x10* 1.5x10* 1.0x10* 5.0x10'7 0.0 1800 2000 -FD-13 1000 1200 1400 1600 1800 2000 Time(sec) CH decompose at 823K 5.0x10 4.0x10* 3.0x10* 2.0x10* 1.0x10* 0.0-1000 1200 1400 1600 1800 2000 Time(sec) 5.0x10 4.0x10* 3.0x10* 2.0x10* 1.0x10* 0.0 - FD-9| 1000 1200 1400 1600 1800 2000 Time(sec) 5.0x10' 4.0x10" 3.0x10*-2.0x10*-1.0x10*-0.0 -FD-10 1000 1200 1400 1600 1800 2000 Time(sec) 5.0x10* 4.0x10* 3.0x10* 2.0x10* 1.0x10* 0.0 -FD-12 1000 1200 1400 1600 1800 2000 Time(sec) Figure VI-1 H 2 molar flow rate indicating an unique feature during the 1st C H 4 decomposition (5 % C H 4 , 723 K and 823 K) 109 Figure VI-2 Filamentous carbon formation on supported Ni catalyst after C H 4 decomposition (5 % C H 4 , 773K) o E, c g w o o. E o o CD T3 i * O •o "CD 1 3 o 5= i_ ro o FD-4 —L7J— During —Q— During —ru—• During —a— During — 0 — During — E — During the 1 st cycle the 2nd cycle the 3rd cycle the 4th cycle the 5th cycle the 6th cycle 400 800 1200 1600 2000 2400 2800 3200 3600 Time(sec) Figure VI-4 H 2 molar flow rate during C H 4 decomposition to show the activity of experiment FD-4 (5 % C H 4 , 723 K, 50 minutes) FD-3 03 o 4. E c o '55 o a. E o o cu •o I* O cn CL T3 0) TO 3 o 5x10 4.0x10 3.5x10 3.0x10 2.5x10"°H 2.0x10 £ 1.5x10" 1.0x10 % 5.0x107 During the 1st cycle During the 2nd cycle During the 3rd cycle During the 4th cycle During the 5th cycle During the 6th cycle 600 800 1000 1200 1400 1600 Time(sec) Figure VI-5 H 2 molar flow rate during C H 4 decomposition to show the activity of experiment FD-3 (5 % C H 4 , 823 K, 50 minutes) This difference could be attributed to the presence of NiO formed during the partial oxidation step. As shown in Figure VI-6, unlike CH4 decomposition in the 1 st cycle for which the N i catalyst had been reduced to yield the maximum number of available N i active sites, CH4 decomposition in the subsequent cycles followed partial oxidation and NiO being formed during this oxidation step. Due to the presence of NiO, the available Ni active sites have decreased. As a result, the H 2 production activity did not reach the same maximum in the 2nd to 6th cycles of CH4 decomposition compared to the 1st cycle. NiO was easily reduced during the CH4 decomposition step and N i active sites would be regenerated as CH4 started to decompose. However, the maximum number of N i active sites would not be available initially, and therefore the maximum achieved in the 1st cycle was not observed in the subsequent cycles. Since the carbon produced from CH4 decomposition reacted with the solid oxygen associated with the NiO, forming CO and CO2 initially, the rate of carbon formation to encapsulate N i active sites was reduced. Hence, no rapid decline in H2 production due to encapsulating carbon formation was observed from the 2nd to 6th cycle. 112 In-Situ Reduction; CH,. Decomposition Partial Oxidation Cycle 1 CH< Decomposition J Partial i Oxidation Cycle 2 CH 4 r , Partial i Decomposition i Oxidation j i | Cycle 3 Cycle 4 CHt Decomposition Partial Oxidation Cycle 5 CH, Decomposition Partial Oxidation Cycle 6 CH. Decomposition Partial Oxidation Figure VI-6 Flow chart of the sequence of C H 4 decomposition and partial oxidation (He flush steps are not included) for 6 cycles Deactivation of the catalyst due to carbon deposition is also relevant here. Therefore, it is important to review experiments with a minimum degree of oxidation, especially the experiment which combined a long reaction time (50 minutes) of C H 4 decomposition and a short duration of oxidation (3 minutes) with dilute O2 (2.5 %) such as experiments FD-11 and FD-9. Under these conditions, a significant quantity of carbon would be expected to accumulate on the catalyst. Figure VI-7(a) shows the carbon recovery percentage during partial oxidation of experiment FD-11. The quantity of 0 2 introduced was unable to remove all the deposited 113 carbon produced from CH4 decomposition. Hence, the carbon recovery percentage is low and un-removed carbon accumulated on the catalyst. The quantity of accumulated carbon increased as the cycle number increased. As shown in Figure VI-7(b), the initial CH4 decomposition rate decreased significantly from the 1st to 2nd cycle but did not decrease further from the 2nd to 6th cycle even though accumulated carbon increased as the cycle number increased. This indicates that the initial CH4 decomposition rate, determined by the N i active sites, was not a function of the accumulated carbon and the similar results was shown in Figure V-16 which suggested that the initial CH4 decomposition rate does not relate to the accumulated carbon deposit. Rather oxygen associated with N i as a result of the oxidation step decreases the number of available N i active sites. The data also show that N i can oxidize to NiO with very dilute O2 in a short period, even though the deposited carbon is present. The deactivation as the cycle number increased, caused by the accumulated carbon, can be observed from the H 2 activity profiles shown in Figure VI-7(c). One or more rates, including the rate of encapsulating carbon formation, the rate of carbon diffusion through the metal particle, the rate of carbon nucleation at the tailing face and the rate of filament growth, could decrease because of the accumulated carbon. Hence, the H 2 activity profiles, affected by the net rate of CH4 decomposition, showed a decline as the cycle number increased and the overall H 2 production during C H 4 decomposition decreased. 114 (a) (b) 100%-, A FD-11 1.1x10 FD-11 80% - "5 1.0x10 60%- 9.0x10 co 40%-O 20%-0%-2 3 4 5 Cycle Number a s. E I O 0x10* , 7.0x10 6.0x10* 3 4 Cycle Number (c) FD-11 G— During the 1st cycle During the 2nd cycle Q During the 3rd cycle During the 4th cycle B During the 5th cycle During the 6th cycle 1800 2400 3000 3600 Time (sec) (d) FD-11 During During a During During Q During During the 1st cycle the 2nd cycle the 3rd cycle the 4th cycle the 5th cycle the 6th cycle 120 180 240 Time(sec) 300 360 Figure VI-7 (a) Carbon recovery percentage during oxidation, (b) Initial C H 4 decomposition rate, (c) H 2 activity profiles during C H 4 decomposition for all 6 cycles of experiment FD-11 (5 % C H 4 , 723 K for 50 minutes, 2.5 % 0 2 , 723 K for 3 minutes), (d) H 2 activity profiles during C H 4 decomposition for all 6 cycles during the first 6 minutes Figure VI-7(d) shows the H 2 activity profiles at the initial period (first 6 minutes) during CH4 decomposition of FD-11. It is obvious that H 2 activity of the 1st CH4 decomposition increased slowly after the rapid decline due to fast formation of encapsulating carbon on 115 Ni active sites. The H 2 molar flow rate profiles from cycle 2 to cycle 6 all reached the similar maximum activity but the extent of declining rate increased as the cycle number increased. Similar observations were also made for experiment FD-9. Figure VI-8(a) shows a low percentage of carbon removal during partial oxidation; hence, accumulated carbon was present on the catalyst. However, the initial CH4 decomposition rate did not decrease significantly from the 2nd to 6th cycle as shown in Figure VT-8(b). The H 2 activity profiles as shown in Figure VI-8(c) and Figure VI-8(d) are the same as discussed for experiment FD-11. Therefore, deactivation due to accumulated carbon can be observed from the experiments with a relative low carbon recovery percentage. Note that for experiment FD-15, a combination of the minimum quantity of deposited carbon and the maximum degree of oxidation, showed no cyclic ability. No C H 4 was able to decompose starting from the 2nd cycle because the experimental parameters applied during CH4 decomposition (5 % CH4, 723 K for 3 minutes) were unable to reduce the NiO formed during the previous partial oxidation (20 % 0 2 , 823 K for 15 minutes). Without the presence of active N i sites, CH4 decomposition can not be initiated. 116 (a) (b) 100%-c g TO 1 80%-ra a 60% -40% • 20% 0% FD-9 3 4 Cycle Number 1.8x10"5-co (mol/ rate 1.7x10'5-tion w »mpc 1.6x10"5-0 -o i * o Initial 1.5x10'5-FD-9 3 4 Cycle Number (c) (d) FD-9 —Q— During the 1st cycle —Q— During the 2nd cycle — ( 3 — During the 3rd cycle —O— During the 4th cycle —Q— During the 5th cycle During the 6th cycle FD-9 400 600 Time(sec) 1200 • During the 1st cycle,-- During the 2nd cycle,-• During the 3rd cycle-During the 4th cycle During the 5th cycle During the 6th cycle 120 180 240 Time(sec) 360 Figure VI-8 (a) Carbon recovery percentage during oxidation, (b) Initial C H 4 decomposition rate, (c) H 2 activity profiles during C H 4 decomposition for all 6 cycles of experiment FD-9 (5 % C H 4 , 823 K for 50 minutes, 2.5 % 0 2 , 823 K for 3 minutes), (d) H 2 activity profiles during C H 4 decomposition for all 6 cycles during the first 6 minutes 117 6.4 The Change of Ni Particle Size and the Activity The Ni particle size at different steps calculated from the data obtained by X R D , is shown in Table VI-3. The data show that the N i particle became bigger during CH4 decomposition and then decreased in size after oxidation. Note that only the NiO peak was identified for the sample after oxidation. From crystallographic data, the equivalent Ni particle size can be calculated from the particle size of NiO, based on the molar volume of NiO and N i . Table VI-3 Comparison of Ni particle size at different steps Treatment of the N i catalyst Peak identified from X R D N i Particle Size (A) N i (26) NiO (29) After calcination (calcined in air, 723 K) 43.191 * 120 After reduction ( 1 0 % H 2 /90% Ar, 1007 K) 51.272 159 After CH4 decomposition (5 % C H 4 , 773 K) 51.839 261 After oxidation (carbon removed by 0 2 ) (10%O 2 , 773 K) 43.280 154* After carbon removed by C 0 2 (40 % C 0 2 , 773 K) 51.765 182 *: Equivalent Ni particle size is calculated from particle size of NiO based on the assumption that the number of unit cell per particle for NiO is equal to that of Ni Figure VI-9 shows the T E M images which also indicate the same trend of N i particle size change among the samples. 118 After reduction: Estimated Ni particle is 159 A After C H 4 decomposition: Estimated Ni particle is 261 A After oxidation: Estimated Ni particle is 154 A Figure VI-9 T E M images show the change of Ni particle size after reduction, after C H 4 decomposition and after oxidation 119 A similar change in N i particle size during CH4 decomposition and activity associated with the N i particles size has been reported in the literature. Avdeeva et al. (1996) reported a N i particle size of 15 nm after heating in H2 at 1073 K for 2 hours. The particle size increased during the CH4 decomposition reaction to 16.5, 22.5 and 40 nm after 20, 60 and 180 minutes of reaction, respectively. Takenaka et al. (2002) reported that the mean particle size of N i supported on Si02 increased from 24 to 27 nm and became almost constant during CH4 decomposition. Takenaka et al. (2004) proposed that the size of N i metal particles of supported N i catalysts determines the catalytic activity for CH4 decomposition. They claimed that N i metal particles with diameters from 60 to 100 nm were the most effective for C H 4 decomposition. Increasing N i crystallite size up to 26 nm during CH4 decomposition was also reported by L i et al. (2006). They proposed that metallic N i with crystallite size of 10-20 nm is suitable for CH4 decomposition on unsupported N i catalysts and speculated that N i crystals larger than 26 nm are not active for CH4 decomposition. Therefore, we can conclude that N i particle size increases during CH4 decomposition and there is a certain range of N i particle size which is most effective for CH4 decomposition. A number of explanations for increasing N i particle size during CH4 decomposition have been presented in the literature. Avdeeva et al. (1996) proposed that the dissolution of the deposited carbon into the N i metal causes the N i particle size to increase. Takenaka et al. (2002) suggested that the aggregation of N i metal particles on the catalyst occurs by 120 contact with CH4. L i et al. (2006) proposed that the deposited carbon would induce aggregation of N i crystallites into larger particles. However, based on the NiO peaks identified from X R D after the oxidation of deposited carbon from the present study, the estimated N i particle size was calculated to be 15 nm, the same as that measured after reduction in H 2 and prior to contacting with CH4. Hence, it is more likely that the dissolution of the deposited carbon into the N i metal particle resulted in the increase in N i particle size after C H 4 decomposition and the N i particle size decreases after oxidation. In Section 5.3.1, Figure V-5 showed an increasing trend of H 2 production during C H 4 decomposition of experiment Ni-4. A similar trend was observed for FD-4 and FD-10 as shown in Figure VI-10 and Otsuka et al. (2004) also reported similar observations. It is highly possible that the evolution of the N i particle size results in an increase in H 2 production as the cycle number increases. The N i particle size may increase toward the optimal range for C H 4 decomposition by repeated cyclic CH4 decomposition and oxidation reaction. 121 J D 4.2x10 • o 3.9x10"3 I 3.6x10"3 o E 3.3x10"3 o 8 3.0x10"3 2.7x10"3 c? 2.4x10"3H -3 "° 2.1x10 T3 y i.8xio" 3 "CS I. 1.5x10"3 I FD-4 FD-10 ^ • I ' I ' I < T ~ 2 3 4 5 6 Cycle Number Figure VI-10 H 2 production as a function of cycle number of experiment FD-4 (5 % C H 4 at 723 K for 50 minutes and 2.5 % 0 2 at 823 K for 15 minutes) and FD-10 (5 % C H 4 at 823 K for 50 minutes and 20 % 0 2 at 723 K for 3 minutes) However, more investigation and catalyst characterizations (XRD, STEM) as a function of C H 4 decomposition and oxidation cycle are needed to verify the N i particle size and confirm this speculation. 6.5 Statistical Analysis Fractional factorial design analysis was carried out by MINITAB statistical software. The interested values were calculated and treated as responses which were inputted into the software. The factors that had a significant effect on the response were first identified from the normal probability plot. Once the factor and effect was identified, main effects and interaction effects were interpreted from the corresponding main effects plot and interaction plot. Only two-factor interactions are considered and three-factor and higher interactions are ignored. 122 6.5.1 Hydrogen Production Statistical analysis indicated that the C H 4 decomposition temperature was the only significant factor to affect the initial CH4 decomposition rate. Figure VI-11 shows the initial C H 4 decomposition rate as a function of cycle number in 4 groups. Group (a) and (c) are both at 823 K and group (b) and (d) are at 723 K. It is obvious that the higher decomposition temperature applied resulted in a higher initial CH4 decomposition rate. From Figure VI-11, few experiments also show a decrease in initial CH4 decomposition rate as the cycle number increased. However, since the cycle number was not a factor of the factorial analysis, the effect of the cycle number can not be revealed statistically among these 16 experiments. * 2.4x10' =f 2.2x10s-H 2.0x10*-co 1.8x10 s-g 1.6x10s-1 1.4x10s-1 1.2x10*-8 1.0x10s-•S 8.0x10*-1" 6.0x10* 2 4.0x10* I 2.0x10* 0.0. (a) *, 2.4x10* f 2.2x10s § 2.0x10s CD . c; 1.8x10* g 1.6x10* ; l 1.4x10s I 1.2x10* fi 1.0x10s g 6.0x10 • 2 4.0x10*-| 2.0x10*-0.0. —a- -o--A--. . . 0 -A... V — — v —0— FD-1 0-- FD-2 ...a- FD-6 — FD-16 (c) 3 4 Cycle Number - O — FD-3 • « - - FD-9 ->~ FD-10 -o—FD-12 3 4 Cycle Number Figure VI-11 Initial C H 4 decomposition rate as a function of cycle number, (a) Decomposition of C H 4 at 823 K for 3 minutes, (b) Decomposition of C H 4 at 723 K for 3 minutes, (c) Decomposition of C H 4 at 823 K for 50 minutes, (d) Decomposition of C H 4 at 723 K for 50 minutes 123 The overall quantity of H 2 produced during CH4 decomposition is affected by the decomposition duration only. The quantity of H 2 production increased with increased decomposition duration. Note that the overall H 2 production and the rate of H 2 production are important since catalyst deactivation may be involved. For example, Figure VI-12 shows a comparison of H 2 production profiles with the same duration but at two temperatures. The H 2 production profile of FD-10 shows a high activity with a much faster deactivation, compared to FD-4 showing a relative steady but low H 2 production activity. In order to evaluate the H 2 production ability including the deactivation, the H 2 production rate (mol/s) is then introduced as an alternative response instead of the overall quantity of H 2 produced. cn "o E, Q) ro o ro o E 4 . 5 x 1 0 *1 4.0x10" 6H 3.5x10"6 3.0x10"6 2.5x10" 6 2.0x10" 6-1.5X10" 6-1.0x10"6-5.0x10" 7-0.0 FD-10, 823 K FD-4, 723 K -1 I • 1 • 1 • 1 • 1 • 1 1 1 , 1 • , 0 400 800 1200 1600 2000 2400 2800 3200 3600 Time(sec) Figure VI-12 H 2 production profile during the first C H 4 decomposition for FD-10 (decomposition of C H 4 at 823 K) and FD-4 (decomposition of C H 4 at 723 K). The duration for both experiment are 50 minutes 124 The average H 2 production rate was calculated by dividing the overall quantity of H 2 produced by the decomposition duration and can be treated as an average H 2 molar flow rate. Figure VI-13(b) shows that the higher H 2 production rate can be obtained by conducting the CH4 decomposition at a higher temperature with a shorter duration. (a) CH, decomposed at 723 K for 50 minutes (b) C H 4 decomposed at 823 K for 3 minutes 2.2x1 0' 2.0x10'M 1.8x10'" 1.6x10"'-] 1.4X10"1 1.2x10'-| 1.0x10"' 8.0x10"5H 6.0x10": 4.0x10"! 2.0x10 s-| 0.0 y — a — FD-4 0 FD-5 * FD-11 — Y — FD-13 9 -A-3 4 Cycle Number I 2.2x10"*, I" 2.0x10"H | 1.8x10" I" 1.6x10" § 1.4X10"4-! ™* 1.2x10"' " 1.0x10"' 1 8.0x10"5 ~Z 6.0x10"5-i 2 4.0x10"5 I 2.0x10 s •a 0.0 o-A --FD-1 FD-2 FD-6 -FD-16 3 4 Cycle Number -O - A Figure VI-13 H 2 production rate as a function of cycle number, (a) Decomposition of C H 4 at 723 K for 50 minutes, (b) Decomposition of C H 4 at 823 K for 3 minutes 6.5.2 C O Impurity during the C H 4 Decomposition Step As shown in Figure VI-14, none of the main or two-factor interaction effects significantly affected the CO to H 2 ratio (CO/H 2) and the overall quantity of CO produced during the C H 4 decomposition. Hence the experimental parameters used in the analysis do not affect the CO to H 2 ratio and amount of CO during C H 4 decomposition. From previous discussions, we concluded that the source of oxygen for CO formation during CH4 decomposition was mainly the solid oxygen associated with NiO. Therefore, 125 it is reasonable that neither the CO to H2 ratio nor the overall CO production will be affected significantly for the range of experimental conditions investigated since the same Ni catalysts were used for all 16 experiments. Normal Probability Plot of the Effects (response is CO /H2 , Alpha = .10) H A: CH4Temp B: C H 4 D u r a C: POXconc D: POXTemp E: POXDura CD o o CO o -H -0.04 -0.03 -0.02 -0.01 0.00 0.01 0.02 0.03 0.04 Effect Figure VI-14 Normal probability plot with C O / H 2 as the response. 5 main effects and 10 two-factor interaction effects are all insignificant 6.5.3 CO Selectivity during Partial Oxidation Statistical analysis indicated that there was no significant main or two-factor interaction effect for the selectivity to CO during the partial oxidation step. It is highly possible that the effect was not significant enough due to the narrow range of oxidation temperatures (100 K) used here. 126 6.5.4 Carbon Removal Percentage Three main effects and one interaction effect were identified from the normal probability plot shown in Figure VI-15(a) as factors affecting the carbon removal percentage. Figure VI-15(b) shows the main effects are 0 2 concentration and duration of the partial oxidation step and both had a positive effect on the carbon removal percentage. However, C H 4 decomposition duration had a negative effect. Consequently a higher carbon removal f percentage can be achieved by introducing a higher concentration of O 2 with longer duration during oxidation, with shorter duration for C H 4 decomposition, since less carbon is deposited on the catalyst. Also, the interaction plot as shown in Figure VI-15(c), indicates that with a higher concentration of O 2 (20 %) introduced for the oxidation of deposited carbon, the effect of duration was insignificant. On the other hand, longer duration of oxidation is required to achieve higher carbon removal percentage if low concentration of O 2 is introduced. The reduced N i catalyst, the N i catalyst after CH4 decomposition and catalysts with different carbon removal percentages were selected for BET surface area measurements and the results are listed in Table VI-4. The results show that BET surface area of the catalysts, with a range of 10 % to 100 % carbon removal percentage after the partial oxidation step, does not change significantly but the surface area of the catalyst after CH4 decomposition decreased significantly. 127 These results indicated that the surface area generated by deposited carbon was not significant. Rather, the blocking of pores of the alumina support by the deposited carbon is expected to cause a decrease in surface area. A CH4Temp B: CH4Dura C: POXconc D: POXTemp E: POXDura Main Effects Plot (data means) for Carbon Remov 1.28m . . . . . (c) Interaction Plot (data means) for Carbon Remov 1 \*i CH4Dura 1.3 • 30 0.9 • 2 0.5 POXconc - 1.3 •20 0.9 •2.5 05 POXDura Figure VI-15 Plots generated from statistical analysis when carbon removal percentage was chosen to be the response, (a) normal probability plot, (b) main effects plot and (c) interaction effects plot 128 Table VI-4 Results of B E T surface area measurements Sample Description Carbon removal percentage Surface area (m2/g) reduced n/a 111 Ni catalyst Ni catalyst after CH 4 decomposition n/a 76 100% 89 10-30% 109 Ni catalyst after partial oxidation 100% 100 20-60% 109 Completely deactivated 103 6.6 Summary The H 2 production rate during C H 4 decomposition reached a maximum initially, decreased rapidly and then increased slowly before decreasing again for the 1 st cycle of C H 4 decomposition for all the experiments loaded with unused reduced catalyst. The observed profiles are in qualitative agreement with that predicted by the deactivation model proposed by Zhang and Smith (2005). However, due to the formation of NiO during partial oxidation, this profile changed from the 2nd cycle of CH4 decomposition. Namely, the H 2 molar flow rate did not show a sharp maximum activity initially, followed by a rapid decline, during the 2nd to 6th 129 cycles of C H 4 decomposition. The presence of NiO reduced the available N i active sites. As a result, the activity for H 2 production did not reach the maximum initially from the 2nd to 6th cycle of C H 4 decomposition. Since NiO was easily reduced during the GH4 decomposition step and N i active sites could be regenerated as CH4 started to decompose, the carbon produced from C H 4 decomposition reacted with the solid oxygen associated with the NiO, forming CO and C O 2 initially, the rate of carbon formation to encapsulate N i active sites was reduced. Hence, no rapid decline in H 2 production due to encapsulating carbon formation was observed from the 2nd to 6th cycle. The evolution of N i particle size was revealed by X R D and T E M and shows that the N i particle size increased during CH4 decomposition but decreased after oxidation. Since the oxidation of carbon can cause the reduction of N i particle size, it is more likely that the dissolution of the deposited carbon into the N i metal particle caused the increase in N i particle size. It is highly possible that the N i particle size evolved toward the optimal range for CH4 decomposition as repeated cyclic CH4 decomposition and oxidation occurred. The results obtained from statistical analysis are summarized below: 1. The initial CH4 decomposition rate is dominated by the decomposition temperature in each cycle. None of the experimental parameters during partial oxidation affect the initial C H 4 decomposition rate significantly. 130 2. Higher H 2 production rate can be obtained by conducting the CH4 decomposition at a higher temperature with a shorter duration. 3. There is no significant main or two-factor interaction effects on the CO to H 2 ratio, overall quantity of CO produced during the CH4 decomposition and the CO selectivity during partial oxidation, within the chosen range of the applied experimental parameters. 4. Higher carbon removal percentage can be achieved by introducing higher concentration of 0 2 with longer duration on the catalyst that has been exposed to a shorter duration CH4 decomposition. 131 C h a p t e r V I I C o n c l u s i o n s a n d R e c o m m e n d a t i o n s f o r F u t u r e W o r k 7.1 Conclusions CH4 decomposition and partial oxidation in a cyclic mode over supported Co and N i catalysts has been investigated. CH4 decomposed to produce H2 gas and carbon. The deposited carbon was removed by introducing O2 to oxidize the deposited carbon to CO and CO2. However, the subsequent partial oxidation not only oxidized the carbon deposit, but also oxidized the catalyst metal (Co, Ni) to metal oxide. The presence of metal oxide impacts the subsequent CH4 decomposition step. The metal oxide formed during the partial oxidation step has to be reduced in order to initialize the subsequent CH4 decomposition. Data show that both metal oxides can be reduced by CH4. However, an induction period was observed for Co catalyst, indicating that the Ni catalyst was easier to reduce after oxidation compared to the Co catalyst. The deposited carbon can react with the oxygen associated with metal oxide during CH4 decomposition. As a result, CO and CO2 were detected in the H2 production stream during C H 4 decomposition, with high selectivity to CO. Note that there was no CO or CO2 produced during the first CH4 decomposition on N i catalysts since the oxygen associated with NiO was minimized after the in situ reduction in H2. Co catalyst showed a fast deactivation during CH4 decomposition and decreasing initial CH4 decomposition rate and total H 2 production as the cycle number increased. Co 132 catalyst showed no ability to be completely regenerated by in situ re-reduction in H 2 . Therefore, Co catalyst is not suitable for the cyclic C H 4 decomposition and partial oxidation reactions examined. On the other hand, steady H 2 production was achieved during C H 4 decomposition for a series of cycles over N i catalyst. Moreover, N i catalyst was re-usable since the catalyst can be regenerated by in situ re-reduction in H 2 . As a result, N i catalyst was proven to be better than Co catalyst in terms of better stability and activity for the cyclic reaction. The addition of MgO as a promoter reduced the interaction between the alumina support and the metal oxide precursor. Hence, the degree of reduction of supported metal catalyst can be enhanced. However, the addition of C e 0 2 did not prevent the oxidation of metal during partial oxidation. The TPO results of N i catalysts indicated that the oxygen uptake was mainly dominated by reduced N i . In order to reduce the formation of NiO during oxidation of the deposited carbon, C O 2 was introduced instead of O 2 to remove the carbon. Therefore, the CO to H 2 ratio during C H 4 decomposition decreased when carbon removal was by C O 2 compared to O 2 . Furthermore, compared to O 2 , similar regeneration ability of catalyst was achieved by introducing C O 2 . However, the rate of carbon removal by C O 2 was too slow for any practical application of the cyclic reaction. A unique H 2 production profile was identified during the first cycle of C H 4 decomposition on a freshly loaded N i catalyst. The H 2 production rate reached a 133 maximum initially, decreased rapidly and then increased slowly before decreasing again. The observed profiles are in qualitative agreement with that predicted by the deactivation model proposed by Zhang and Smith (2005). However, due to the formation of NiO during partial oxidation, this feature was not observed in the subsequent C H 4 decomposition. The statistical analysis results of the fractional factorial design indicated that higher H 2 production rate can be obtained by conducting the CH4 decomposition at a higher temperature with a shorter duration. However, within the chosen range of the applied experimental parameters in the present study, there is no significant effect on the CO to H 2 ratio, overall quantity of CO produced during the CH4 decomposition and the CO selectivity during partial oxidation. 7.2 Recommendations for Future Work The kinetic data of the present study have been collected at a very high sample rate (4 seconds for gas analysis and 10 seconds for mass change measurement). The activity data, especially obtained during the first CH4 decomposition with freshly loaded N i catalyst, should be used to fit the deactivation model proposed by Zhang and Smith (2005). Moreover, the data collected during the C H 4 decomposition which did not show some of the features obtained in the first cycle (such as the CH4 decomposition activity during the 2nd to 6th cycles of N i catalyst) can also be used to verify the explanation proposed in the present study for the change in the activity profile. Due to the presence of NiO, the carbon reacted with oxygen associated with N i to produce CO and C 0 2 . Hence, the rate 134 of encapsulating carbon formation and the rate of carbon bulk diffusion through the metal can be assumed to be zero with the presence of NiO. Therefore, the deactivation model describing the CH4 decomposition on N i catalyst with the presence of NiO could be established. The T E O M used in the present study showed a remarkable ability to perform activity measurement and record the real time mass change simultaneously. The mass change measurements were in a good qualitative agreement with the gas analysis results. However, the mass change can't be isolated when simultaneous reactions occur. Therefore, the mass change measurement is not so critical for activity measurement since the mass change involved simultaneous carbon deposition and carbon removal reactions during CH4 decomposition. Rather, direct gases analysis measured by MS is more reliable. However, the highest operating temperature allowed for the T E O M is 823 K. Consequently, a fixed bed reactor made with stainless steel is recommended for future studies. An external heater equipped to heat the reactor heat up to 1300 K and a thermocouple located inside the catalyst bed should be used to control and measure the reaction temperature precisely. A future factorial design experiment could be employed with a wider range of reaction temperatures in order to reveal the effects on CO selectivity or CO production. In the present study, data showed that NiO was formed during the partial oxidation step. Without applying H 2 reduction to reduce NiO to N i prior to the subsequent CH4 decomposition, NiO was reduced by reacting with CH4 during the C H 4 decomposition 135 step. This suggests that the calcined N i catalyst could be used without in situ reduction. An activity comparison should made by decomposing CH4 over reduced and calcined N i catalyst under same experimental conditions. Furthermore, the catalysts with different N i loading should be used in order to investigate the relationship between the N i loading and the CO production during the CH4 decomposition step. In order to investigate the mechanism of CO and C O 2 formation during the C H 4 1 o decomposition step and the partial oxidation step, labeled oxygen ( O 2 ) can be used as the gas introduced during the partial oxidation step for cyclic reactions. By analyzing the product gases using a mass spectrometer, the oxygen source for production of CO and C O 2 can be easily distinguished. For example, i f the gas analysis during the partial oxidation shows CO and C 0 2 peaks at m/z = 28 (C 1 6 0) and 48 (C 1 8 0 2 ) , then we can conclude that the oxygen to form CO is mainly from the solid oxygen associated with the catalyst but the oxygen to form C O 2 is from gaseous 0 2 introduced. However, this is just a simplified situation used here to provide an idea for the proposed approach. Once the mechanism of CO and C O 2 formation during each step is discovered, then controlling the selectivity to CO will be more understandable. 136 References Aiello, R., Fiscus, J. E., zur Loye, H. , & Amiridis, M . D. (2000). Hydrogen production via the direct cracking of methane over Ni/SiC^: Catalyst deactivation and regeneration. Applied Catalysis A: General, 192(2), 227-234. Armor, J. N . (2005). Catalysis and the Hydrogen Economy. Catalysis Letters, 101 (3), 131-135. Avdeeva, L. B., Goncharova, O. V. , Kochubey, D. I., Zaikovskii, V . I., Plyasova, L. M . , & Novgorodov, B. N . , et al. (1996). Coprecipitated Ni-alumina and Ni—Cu-alumina catalysts of methane decomposition and carbon deposition. II. evolution of the catalysts in reaction. Applied Catalysis A: General, 141(1-2), 117-129. Boskovic, G., & Smith, K. J. (1997). Methane homologation and reactivity of carbon species on supported co catalysts. Catalysis Today, 37(1), 25-32. Bueno - Lopez, A. , Krishna, K. , Makkee, M . , & Moulijn, J. A . (2005). Active Oxygen from Ce02 and its Role in Catalysed Soot Oxidation Chen, J., L i , Y . , L i , Z., & Zhang, X . (2004). Production of COx-free hydrogen and nanocarbon by direct decomposition of undiluted methane on ni-cu-alumina catalysts. Applied Catalysis A: General, 269(1-2), 179-186. Chen, Y. , & Zhang, L. (1992). Surface interaction model of g-alumina-supported metal oxides. Catalysis Letters, 72(1-3), 51-62. Chernavskii, P. A. , Pankina, G. V. , & Lunin, V . V . E. -. (2000). The influence of oxide-oxide interaction on the catalytic properties of C0/AI2O3 in CO hydrogenation. Catalysis Letters, 66(3), 121-124. Choudhary, T. V . , & Goodman, D. W. E. -. (1999). Stepwise methane steam reforming: A route to CO-free hydrogen. Catalysis Letters, 59(2 - 4), 93-94. 137 Choudhary, T. V. , Sivadinarayana, C , Chusuei, C. C , Klinghoffer, A. , & Goodman, D. W. (2001). Hydrogen production via catalytic decomposition of methane. Journal of Catalysis, 199(1), 9-18. Damyanova, S., Perez, C. A. , Schmal, M . , & Bueno, J. M . C. (2002). Characterization of ceria-coated alumina carrier. Applied Catalysis A: General, 234(1-2), 271-282. Diskin, A . M , Cunningham, R. H. , & Ormerod, R. M . (1998). The oxidative chemistry of methane over supported nickel catalysts. Catalysis Today, 46(2-3), 147-154. Djaidja, A. , Libs, S., Kiennemann, A. , & Barama, A . (2006). Characterization and activity in dry reforming of methane on NiMg/Al and Ni/MgO catalysts. Catalysis Today, 113(3-4), 194-200. Dong, W., Jun, K. , Roh, H. , Liu, Z., & Park, S. (2002). Comparative Study on Partial Oxidation of Methane Over Ni /Zr0 2 , N i /Ce0 2 and N i / Ce0 2 -Zr02 Catalysts. Catalysis letters, 78(1), 215-222. Fathi, M . , Bjorgum, E., Vi ig , T., & Rokstad, O. A . (2000). Partial oxidation of methane to synthesis gas: Elimination of gas phase oxygen. Catalysis Today, 63(2-4), 489-497. Ferreira-Aparicio, P., Rodriguez-Ramos, I., & Guerrero-Ruiz, A . (1997). Methane interaction with silica and alumina supported metal catalysts. Applied Catalysis A: General, 148(2), 343-356. Jacobs, G., Chaney, J. A. , Patterson, P. M . , Das, T. K. , & Davis, B. H. (2004). Fischer-tropsch synthesis: Study of the promotion of re on the reduction property of Co/A1203 catalysts by in situ E X A F S / X A N E S of co K and re LIII edges and XPS. Applied Catalysis A: General, 264(2), 203-212. Jiang, P., Shang, Y . , Gheng, T., B i , Y . , Shi, K. , & Wei, S., et al. (2003). Methane decomposition over Ni /a -Al 2 03 promoted by La 2 03 and Ce0 2 . Journal of Natural Gas Chemistry, 12(3), 183-188. 138 L i , Y. , Zhang, B., Xie, X . , Liu, J., Xu , Y. , & Shen, W. (2006). Novel N i catalysts for methane decomposition to hydrogen and carbon nanofibers. Journal of Catalysis, 238(2), 412-424. Liao, M . , & Zhang, Q. (1998). Dissociation of methane on different transition metals. Journal of Molecular Catalysis A: Chemical, 136(2), 185-194. Loong, C. -K. , & Ozawa, M . (2000). The role of rare earth dopants in nanophase zirconia catalysts for automotive emission control. Journal of Alloys and Compounds, 303-304, 60-65. Murata, K., Inaba, M . , Mik i , M . , & Yamaguchi, T. (2005). Formation of Filamentous Carbon and Hydrogen by Methane Decomposition Over Al 203-Supported N i Catalysts. Reaction Kinetics and Catalysis Letters, 55(1), 21-28 Otsuka, K., Ogihara, H. , & Takenaka, S. (2003). Decomposition of methane over N i catalysts supported on carbon fibers formed from different hydrocarbons. Carbon, 41(2), 223-233. Otsuka, K., Takenaka, S., & Ohtsuki, H . (2004). Production of pure hydrogen by cyclic decomposition of methane and oxidative elimination of carbon nanofibers on supported-ni-based catalysts. Applied Catalysis, A: General, 273(1-2), 113-124. Otsuka, K., Wang, Y . , Sunada, E., & Yamanaka, I. (1998). Direct partial oxidation of methane to synthesis gas by cerium oxide. Journal of Catalysis, 175(2), 152-160. Pantu, P., Kim, K. , & Gavalas, G. R. (2000). Methane partial oxidation on Pt/Ce0 2 -Zr0 2 in the absence of gaseous oxygen. Applied Catalysis A: General, 795(1-2), 203-214. Piao, L., L i , Y . , Chen, J., Chang, L. , & Lin, J. Y . S. (2002). Methane decomposition to carbon nanotubes and hydrogen on an alumina supported nickel aerogel catalyst. Catalysis Today, 74(1-2), 145-155. 139 Poirier, M . G., & Sapundzhiev, C. (1997). Catalytic decomposition of natural gas to hydrogen for fuel cell applications. International Journal of Hydrogen Energy, 22(4), 429-433. Roh, H., Jun, K. , Baek, S., & Park, S. (2001). Extraordinary catalytic behavior of CeC»2 in partial oxidation of methane. Chemistry Letters, (10), 1048-1049. Solymosi F., Erdohelyi A . , Cserenyi J., & Felvegi A. , (1994). Decomposition of CH4 over supported Pd catalysts. Journal of Catalysis, 147(1), 272-278. Takenaka, S., Ishida, M . , Serizawa, M . , Tanabe, E., & Otsuka, K. (2004). Formation of carbon nanofibers and carbon nanotubes through methane decomposition over supported cobalt catalysts. Journal of Physical Chemistry B, 108(31), 11464-11472. Takenaka, S., Kato, E., Tomikubo, Y. , & Otsuka, K. (2003). Structural change of N i species during the methane decomposition and the subsequent gasification of deposited carbon with CO2 over supported N i catalysts. Journal of Catalysis, 219(1), 176-185. Takenaka, S., Ogihara, H. , & Otsuka, K. (2002). Structural change of N i species in N i / S i 0 2 catalyst during decomposition of methane. Journal of Catalysis, 208(1), 54-63. Takenaka, S., Tomikubo, Y. , Kato, E., & Otsuka, K. (2004). Sequential production of H 2 and CO over supported N i catalysts. Fuel, 83(1), 47-57. Toebes, M . L., Bitter, J. H. , van Dillen, A . J., & de Jong, K. P. (2002). Impact of the structure and reactivity of nickel particles on the catalytic growth of carbon nanofibers. Catalysis Today, 76(1), 33-42. Wang, H. Y. , & Ruckenstein, E. (2002). Formation of filamentous carbon during methane decomposition over co-MgO catalysts. Carbon, 40(11), 1911-1917. 140 Xu, G., Shi, K. , Gao, Y. , Xu , H. , & Wei, Y . (1999). Studies of reforming natural gas with carbon dioxide to produce synthesis gas: X . the role of CeO"2 and MgO promoters. Journal of Molecular Catalysis A: Chemical, 147(1-2), 47-54. Zadeh, J. S. M . , & Smith, K. J. (1999). Two-step homologation of methane on supported Co catalysts. Journal of Catalysis, 183(2), 232-239. Zadeh, J. S. M . , 8c Smith, K. J. (1998). Kinetics of CH4 decomposition on supported cobalt catalysts. Journal of Catalysis, 176(1), 115-124. Zhang, T., & Amiridis, M . D. (1998). Hydrogen production via the direct cracking of methane over silica-supported nickel catalysts. Applied Catalysis A: General, 167(2), 161-172. Zhang, Y. , & Smith, K. J. (2005). A kinetic model of CH4 decomposition and filamentous carbon formation on supported co catalysts. Journal of Catalysis, 231(2), 354-364. Zhang, Y. , & Smith, K. J. (2002). CH4 decomposition on co catalysts: Effect of temperature, dispersion, and the presence of H 2 or CO in the feed. Catalysis Today, 77(3), 257-268. Zhu, T., & Flytzani-Stephanopoulos, M . (2001). Catalytic partial oxidation of methane to synthesis gas over Ni-Ce02. Applied Catalysis A: General, 208(1-2), 403-417. 141 Appendix A: Calibration for TCD Voltage TCD Voltage and Temperature vs Time 0.008 0.006 0.004 0.002 0.000 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 Eclipse Time(sec) Figure A- l TCD voltage and temperature vs time for TPR of C u 2 0 Table A- l Summary results of TPR of C u 2 0 Experiment ID Amount of C u 2 0 (g) Theoretical H 2 (moles) required for 100% reduction31 Integreted area under TCD voltage Area of TCD voltage per mole of H 2 122231 0.260 0.002 20.38 11216.89 122234 0.250 0.002 18.47 10568.88 122236 0.250 0.002 19.67 11257.14 120479 0.11448 0.0008 12.24 15295.23 120493 0.11811 0.0008 10.54 12772.89 (Average of above values) 12222.21 a : Amount of H 2 required is calculated based on the following equation Cu20 + H2 > 2Cu + H20 142 Appendix B : Sample Calculations Degree of reduction For Co catalyst Sample Information Sample ID 125105 1 | Sample Description C o on modi f ied a lumina Weight of Sample(calcined) -l 0.32813 g loaded metal Co Weight percent of loaded metal 12.00% mole of metal oxide produced per mole of metal - , - \ 0.33 Weight percent of loaded metal oxide 16.046% MW of loaded metal(Co) 58.933 g/mole MW of loaded metal oxide(Co304) 247.797 g/mole Weight of loaded metal oxide 0.05265 g During TPR, the reaction is Co304 + 4H2 >3Co + 4H20 Calculation mole of loaded reactant (metal oxide) 0.000212 mole stoichiometr ic ratio for Hydrogen c o n s u m e d to reactant 4 mole of Hydrogen required(consumed) 0.000850 mole mole of water generated 0.000850 mole MW of water 18.0200 g/mole weight of water generated 0.0153 g Measured area under curve from integration based on TCD voltage 7.49983 Theoretical Area a : 10.38763 12222.21 Degree of reduction (%)b: 72.20% a : Theoretical area is calculated by multiplying H 2 required by 12222.21 (area/mole of H2) which is the the calibrated value shown in Appendix A b Degree of reduction is calculated by dividing measured area under TCD curve by theoretical area For N i catalyst, using the following equation to calculate instead NiO + H2 >Ni + H20 143 B E T surface area Sample ID 126278 | 125866 126115 126147 126263 126135 126133 FD3 FD11 FD4 FD9 FD14 Sample Description reduced Ni catalyst Ni catalyst after CH4 decomposition FD series Carbon removal percentage n/a n/a 100% 10-30% 100% 20-60% catalyst completely deactivated Weigh 1 Empty tube and holder (g) 13.0121 13.4191 13.0321 13.3316 13.4146 13.5839 13.0100 Weight 2 Un-dried sample, tube, and holder (g) 13.1528 13.4913 13.1345 13.4483 13.5310 13.7103 13.0975 Weight 3 Dried sample, tube, and holder (g) 13.1527 13.4851 13.1324 13.4478 13.5291 13.7089 13.0959 Surface area reading 1 The reading after appling warm H20 15.68 5.08 8.97 12.75 11.43 13.72 8.94 Surface area reading 2 The reading after appling warm H20 15.62 5.07 8.90 12.75 11.47 13.72 8.91 Dried Sample weight 0.1406 0.0660 0.1003 0.1162 0.1145 0.1250 0.0859 Average of area 15.65 5.08 8.94 12.75 11.45 13.72 8.93 Surface area 111.31 76.89 89.08 109.72 100.00 109.76 103.90 Dried sample weight (g) = weight 3 (g) - weight 1 (g) Surface area (m 2 / g) = average of area (m2) / dried sample weight (g) 144 TPO related calculation The following table is the summary of data collected from TPOl and TP02 Experiment Mass of Catalyst(g) Ni loading Ni(g) Ni(moles) degree of reduction Overall mass change(g) TP01 0.1378 12% 1.65E-02 2.82E-04 85% 3.48E-03 TP02 0.1276 12% 1.53E-02 2.61 E-04 85% 2.74E-03 Since there was no Ce02 in TP02, the overall mass increased in TP02 was done by N i oxidation only. Therefore, the theoretical mass increased by N i (mole) during oxidation is calculated below. overall mass increases (g) 2.74xlO" 3 , = A = 12.35 (g/mole) Ni(mole) 2.61 xlO" 4 x85% The mass increased by N i in TPOl according to its N i content is calculated below. A W (by N i in TPOl) = 12.35 (g/mole) x 2.82xl0"4 (mole) x 85% = 2.96xl0"3 (g) The mass increased by reduced Ce02 during oxidation is calculated below. A W (by Ce0 2 ) = 3.48xl0"3 (g) - 2.96xl0"3 (g) = 5.24xl0"4(g) (Overall mass increased in TPOl - mass increased by N i in TPOl) Therefore, the percentage of mass increased by Ce02 is calculated below. AW(byCeO) J - ™ * ^ m % = l 5 M % overall mass increased 3.48 x 10 145 Equivalent N i particle size obtained from NiO X R D Data • From crystallographic data on minerals in Handbook of Chemistry & Physics, the 87th edition, 2006-2007, the following information is obtained. Minerals Crystal system Unit cell length (A) Number of formula units per the unit cell NiO Cubic 4.177 4 N i Cubic 3.5238 4 • For the N i catalyst after oxidation, NiO peaks were identified by X R D . The main peak of NiO is at the 20 = 43.28 and the corresponding crystallite size is 183 A • By assuming that the equal mole of N i per crystallite of NiO and N i , the size was calculated as below 3.5238(A)x = 154(A) 146 Activity Measurement Example: experiment FD-3, CH4 decomposition during the 2nd cycle 1. Partial pressure of species i, P(i), is obtained directly from raw data of Prolab MS. 2. Mole fraction of species i, y(i), is calculated based on the following calibration results: ^ _ = m ( / ) x _ K 0 _ P(He) y(He) 3. Molar flow rate of species i, F(i), is then calculated based on the following equation: F(i) = F(He)x^77^x~x y ^ 22414 60 y(He) mole cm3 mole min = x — x sec min cm sec 4. The molar flow rate vs. time on stream is plotted for each species as Figure B - l . The overall quantity of species (i) produced is determined by the integrated area under the molar flow rate curve. The initial C H 4 decomposition rate is calculated by the following equation: initial CH 4 decompositon rate = initial max. H 2 molar flow rate x — x - ^ 2 mass of catalyst loaded 147 o E JS o Methane, mol/s 1.0x10"° 8.0x10"° H 6.0x10"° H | 4.0x10"° H • i 2.0x10"° H 0.0 n — 1 — i — < — i — < — i — > — i — • — i — 1 — ~ t — i — i — > — i 7000 7500 8000 8500 9000 9500 10000 10500 11000 11500 Time(sec) | f 1.0x10"5 £ 8.0x10"° H w 6.0x10"° H cc | 4.0x10"* • i 2.0x10* H ro o 0.0 Hydrogen, mol/s f — i — i — | — i — | — i — | — i — | — i — | — i — | 7000 7500 8000 8500 9000 9500 10000 10500 11000 11500 Time(sec) ^ 6.0x10" 7i | 5.0x10"7^ J D 4.0x10" 7 : S. 3.0x10"7 § 2.0x10"7 £ 1.0x10"7H JS 0.0 Carbon Monoxide, mol/s T ^ i — ' — r — ' — i — - - — i — > — i — • — i — ' 7000 7500 8000 8500 9000 9500 10000105001100011500 Time(sec) 6.0x10"'n 5.0x10"7-4.0x10"7--3.0x10"7--§ 2.0x1 CS7-^ 1.0x10"7-0.0-03 CC ro o Carbon Dioxide, mol/s i — | — i — | — , — | — i — | — , — i — i — | — , — i — , — i 7000 7500 8000 8500 9000 9500 10000 10500 11000 11500 Time(sec) Figure B-l Molar flow rate vs time of C H 4 , H 2 , C O and C 0 2 during the 2nd cycle of C H 4 decomposition of experiment FD-3 148 A p p e n d i x C : C a l i b r a t i o n f o r M a s s F l o w C o n t r o l l e r Mass Flow Controller 1 2 2 2 3 4 5 Gas applied He 5%CH4/Ar CH, 4.82%H 2 10.4%CH 4 2.01%C 2 H 4 3.75%C 2 H 6 /Ar H 2 o2 He Deliver Pressure (psi) 50 220 130 180 200 50 50 Setup flow rate (ml/min) 40 50 50 50 40 20 50 Time 1(sec) 27.88 15.37 24.94 19.19 27.93 57.25 20.81 Time 2(sec) 27.81 15.38 25.03 19.03 28.06 57.19 20.75 Time 3(sec) 27.87 15.43 25.03 19.06 28.09 57.22 20.72 Ave. Time 27.85 15.39 25.00 19.09 28.03 57.22 20.76 Measured flow rate(ml/min) 43.08 77.96 48.00 62.85 42.82 20.97 57.80 Setup flow rate (ml/min) 80 100 100 100 80 40 100 Time 1(sec) 14.00 7.91 13.56 9.78 13.91 29.00 10.50 Time 2(sec) 14.03 7.94 13.54 9.75 13.94 29.06 10.56 Time 3(sec) 14.00 7.94 13.50 9.78 13.97 29.06 10.50 Ave. Time 14.01 7.93 13.53 9.77 13.94 29.04 10.52 Measured flow rate(ml/min) 85.65 151.32 88.67 122.82 86.08 41.32 114.07 Setup flow rate (ml/min) 120 150 150 150 120 60 150 Time 1(sec) 9.35 5.40 9.22 6.54 9.44 19.47 7.00 Time 2(sec) 9.38 5.44 9.18 6.55 9.37 19.56 7.03 Time 3(sec) 9.37 5.38 9.18 6.53 9.31 19.53 7.06 Ave. Time 9.37 5.41 9.19 6.54 9.37 19.52 7.03 Measured flow rate(ml/min) 128.11 221.95 130.53 183.49 128.02 61.48 170.70 Setup flow rate (ml/min) 160 200 200 200 160 80 200 Time 1(sec) 7.13 4.09 6.94 4.97 7.03 14.67 5.25 Time 2(sec) 7.15 4.07 6.91 4.97 7.00 14.65 5.22 Time 3(sec) 7.12 4.09 6.97 5.00 7.00 14.72 5.25 Ave. Time 7.13 4.08 6.94 4.98 7.01 14.68 5.24 Measured flow rate(ml/min) 168.22 293 88 172.91 240.96 171.18 81.74 229.01 Setup flow rate (ml/min) 200 250 250 250 200 100 250 Time 1(sec) 5.62 3.22 5.65 3.93 5.59 11.75 4.28 Time 2(sec) 5.69 3.23 5.63 3.93 5.66 11.72 4.28 Time 3(sec) 5.72 3.22 5.63 3.97 5.50 11.75 4.28 Ave. Time 5.68 3.22 5.64 3.94 5.58 11.74 4.28 Measured flow rate(ml/min) , 211.39 372.29 212.89 304.31 214.93 102.21 280.37 149 Table C - l Summary results of measured and setup flow rates for calibration of 5 mass flow controllers in different gases Mass Flow Controller 1 2 2 2 3 4 5 He 5%CH4/Ar CH„ 4.82%H 2 10.4CH 4 2 .01%C 2 H 4 3.75%C 2 H 6 /Ar H 2 o2 He Setup Fow Rate(ml/min) in MFC 40 50 50 50 40 20 50 80 100 100 100 80 40 100 120 150 150 150 120 60 150 160 200 200 200 160 80 200 200 250 250 250 200 100 250 Measured Flow Rate(ml/min) 43.08 77.96 48.00 62.85 42.82 20.97 57.80 85.65 151.32 88.67 122.82 86.08 41.32 114.07 128.11 221.95 130.53 183.49 128.02 61.48 170.70 168.22 293.88 172.91 240.96 171.18 81.74 229.01 211.39 372.29 212.89 304.31 214.93 102.21 280.37 Convert measured flow rate to SCCM Pressure of the day(kPa) 101.4 Ambient temperature(K) 300.72 For MFC2I4 82'^H2 i Pressure of the day(kPa) Ambient temperature(K) 296.89 39.15 70.84 43.62 38.91 19.06 52.53 77.83 137.51 80.58 113 50 78.23 37.55 103.66 116.42 201.69 118.61 169.55 116.34 55.86 155.12 152.87 267.05 157.13 222 67 155.56 74.28 208.10 192.10 338.30 193.46 281 21 i 195.31 92.88 254.78 150 Setup flow rate(SCCM) vs Measured flow rate(SCCM) for MFC1(He) 250 200 E o o in £ 150 100 50 0.00 y = 1 . 0 3 9 6 X R2 = 0.9998 50.00 100.00 150.00 Measured flow rate(SCCM) 200.00 250.00 Setup flow rate(SCCM) vs Measured flow rate(SCCM) for MFC2(5%CH„/Ar) 3 0 0 • 2 5 0 • 2 0 0 • 1 5 0 • 1 0 0 -5 0 • 0 • 0.00 V = 0.741X = 0.9994 5 0 . 0 0 1 0 0 . 0 0 1 5 0 . 0 0 2 0 0 . 0 0 2 5 0 . 0 0 Measured flow rate(SCCM) 3 0 0 . 0 0 3 5 0 . 0 0 400.00 Setup flow rate(SCCM) vs Measured flow rate(SCCM) for MFC2(CH„) 3 0 0 2 5 0 2 0 0 1 5 0 3 o t 100 5 0 0.00 y = 1.2748X R 2 = fl 9979 50.00 100.00 150.00 Measured flow rate(SCCM) 200.00 250.00 Figure C - l Calibration curve for MFC1 (He), M F C 2 (5 % CH 4 /A r ) and M F C 2 (CH 4 ) 151 Setup flow rate(SCCM) vs Measured flow rate(SCCM) for MFC2(4.82%H2, 10.4%CH4, 2.01%C2H„, 3.75%C2H6/Ar) f 300 O 250 co "o 200 E 150 100 50 0 0.00 y = 0.8898X R 2 = 0.9997 . ^ - " 50.00 100.00 150.00 200.00 Measured flow rate(SCCM) 250.00 300.00 Setup flow rate(SCCM) vs Measured flow rate(SCCM) for MFC3(H2) 250 8 200 o co £ 150 o 100 •S 50 co 0.00 y = 1.0265X R 2 = 1 50.00 100.00 150.00 Measured flow rate(SCCM) 200.00 250.00 Setup flow rate(SCCM) vs Measured flow rate(SCCM) for MFC4(02) 120 5 100 o o <« 80 60 40 20 y = 1.0749x R 2 = 0. 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 Measured flow rate(SCCM) Figure C-2 Ca l i b ra t i on curve for M F C 2 (4.82 % H 2 , 1 0 . 4 % C H 4 , 2 . 0 1 % C 2 H 4 , 3 . 7 5 % C 2 H 6 / A r ) , M F C 3 ( H 2 ) and M F C 4 ( 0 2 ) 152 300 Setup flow rate(SCCM) vs Measured flow rate(SCCM) for MFC5(He) S 250 O o W 200 o re y = 0.9711X 150 R = 0.9995 100 50 0.00 50.00 100.00 150.00 200.00 Actural flow rate(cc/min) 250.00 300.00 Figure C-3 Calibration curve for M F C 5 (He) Table C-2 Summary results of calibration equation for mass flow controllers Mass Flow Controller 1 2 2 2 3 4 5 Gas applied He 5%CH„,Ar C H , 4.82%H 2 10.4%CH 4 2.01%C 2H 4 3.75%C 2H 6/Ar H 2 o2 He Calibration equation y=1.0396x y=0.741x y=1.2748x y=0.8898x y=1.0265x y=1.0749x y=0.9711x 153 Appendix D: Calibration for Prolab Mass Spectrometer Calibrated Gas Used Tank cone. Tank cone. CH„ cone. (%) 3.95% C 0 2 cone. (%) 0.495% 20 cone, (ppm 501 H 2 cone, (%) 1.02% Design(SCCM) Flow mole fraction(base on He) MFC1, He (seem) MFC2, calibrated gas (seem) MFC5, He (seem) Total Flow (seem) CH4 C02 CO H2 200 100 300 600 0.0079 0.0010 0.00010 0.0020 200 150 250 600 0.0132 0.0017 0.00017 0.0034 200 200 200 600 0.0198 0.0025 0.00025 0.0051 200 250 150 600 0.0282 0.0035 0.00036 0.0073 100 250 0 350 0.0988 0.0124 0.00125 0.0255 50 250 0 300 0.1975 0.0248 0.00251 0.0510 mole fraction (based on He) Partial pressure (based on He) CH 4 C 0 2 CO H 2 Methane(16) /He Intensity Carbon Dioxide(44) /He Intensity Carbon monoxide(28) /He Intensity Hydrogen(2) /He Intensity 0.0079 0.0010 0.00010 0.0020 8.03E-02 8.54E-03 1.49E-03 5.79E-03 0.0132 0.0017 0.00017 0.0034 1.18E-01 1.39E-02 2.14E-03 9.40E-03 0.0198 0.0025 0.00025 0.0051 1.70E-01 2.03E-02 2.95E-03 1.38E-02 0.0282 0.0035 0.00036 0.0073 2.34E-01 2.84E-02 4.00E-03 1.95E-02 0.0988 0.0124 0.00125 0.0255 7.10E-01 8.92E-02 1.19E-02 6.36E-02 0.1975 0.0248 0.00251 0.0510 1.33E+00 1.69E-01 2.25E-02 1.22E-01 154 Partial Pressure vs Mole Fraction 0.00 0.05 0.10 0.15 0.20 0.25 Mole Fraction(based on He) — Methane(16) /He Intensity - Linear (Methane(16) /He Intensity) Partial Pressure vs Mole Fraction -Carbon Dioxide(44) /He Intensity • Linear (Carbon Dioxide(44) /He Intensity) 0.00 0.01 0.02 0.03 Mole Fractionfbased on He) Partial Pressure vs Mole Fraction 0.0000 0.0010 0.0020 0.0030 Mole Fraction(based on He) -Carbon monoxide(28) /He Intensity • Linear (Carbon monoxide(28) /He Intensity) 1.40E-01 1.20E-01 1.00E-01 8.00E-02 6.00 E-02 4.00E-02 2.00E-02 O.OOE+00 Partial Pressure vs Mole Fraction y=2.416x R2 = 0.9987 0.00 0.01 0.02 0.03 0.04 0.05 0.06 Mole Fractionfbased on He) Hydrogen(2) /He Intensity 1 Linear (Hydrogen(2) /He Intensity) Figure D-l Calibration curves for C H 4 , C 0 2 , C O and H 2 of Prolab mass spectrometer 155 Appendix E: Error Analysis 5 sets of duplicate experiments (FD-4, FD-6, FD9, FD-12 and FD-14) were conducted in order to obtain the pure error of the measurements. The pure error was calculated as shown in the following equation. A Q-test was also used to identify statistical outliers. k 2 pure error = 1=1 u=l i=i The pure error is summarized in Table E-. The details of each error analysis are shown below. Table E - l Summary of pure error for the measurements Measurement Pure Error H2 produced during CH4 decomposition (moles) 9.91E-05 CO produced during CH4 decomposition (moles) 4.21E-06 CO2 produced during CH4 decomposition (moles) 3.95E-06 H2 production rate during CH4 decomposition (mol/s) 3.87E-06 Initial CH4 decomposition rate (mol/(g»s)) 8.47E-07 CO produced during partial oxidation (moles) 2.12E-05 CO2 produced during paritial oxidation (moles) 2.60E-05 Carbon removal percentage after oxidation (%) 5% 156 Error Analysis (95%) Measurement: H 2 produced (moles) during C H 4 decomposition Cycle Variance Variance(sorted) Range G a p Qcalc 1 2.88E-08 0.00E+00 3.92E-08 0.00E+00 0.00 2 3.92E-08 0.00E+00 0.00E+00 0.00 3 3.13E-08 0.00E+00 0.00E+00 0.00 4 2.42E-08 5.86E-11 5.86E-11 0.00 5 3.92E-08 1.55E-10 1.63E-11 0.00 6 3.92E-08 1.71E-10 1.63E-11 0.00 1 5.86E-11 2.00E-10 2.91E-11 0.00 2 4.56E-10 3.12E-10 1.12E-10 0.00 3 1.07E-08 4.40E-10 1.57E-11 0.00 4 0.00E+00 4.56E-10 1.57E-11 0.00 5 0.00E+00 6.57E-10 1.43E-10 0.00 6 0.00E+00 8.00E-10 3.39E-21 0.00 1 5.46E-09 8.00E-10 3.39E-21 0.00 2 4.40E-10 9.88E-10 1.88E-10 0.00 3 6.57E-10 1.25E-09 2.62E-10 0.01 4 1.55E-10 2.54E-09 6.59E-10 0.02 5 3.12E-10 3.20E-09 6.59E-10 0.02 6 9.88E-10 5.46E-09 5.93E-10 0.02 1 8.00E-10 6.05E-09 5.93E-10 0.02 2 1.62E-08 8.70E-09 2.04E-09 0.05 3 1.45E-08 1.07E-08 2.04E-09 0.05 4 8.00E-10 1.45E-08 1.75E-09 0.04 5 1.25E-09 1.62E-08 1.75E-09 0.04 6 2.00E-10 1.91 E-08 2.92E-09 0.07 1 6.05E-09 2.42E-08 4.60E-09 0.12 2 3.20E-09 2.88E-08 2.45E-09 0.06 3 1.91E-08 3.13E-08 2.45E-09 0.06 4 1.71E-10 3.92E-08 1.69E-21 0.00 5 8.70E-09 3.92E-08 0.00E+00 0.00 6 2.54E-09 3.92E-08 0.00E+00 0.00 Sum of varience 2.95E-07 Degrees of freedom 30 Er ror 9.91 E-05 157 Error Analysis (95%) Measurement: C O produced (moles) during C H 4 decomposition Cycle Variance Variance(sorted) Range G a p Qcalc 1 0.00E+00 0.00E+00 6.77E-10 0.00E+00 0.00 2 1.89E-11 0.00E+00 0.00E+00 0.00 3 5.26E-13 0.00E+00 0.00E+00 0.00 4 4.69E-13 0.00E+00 0.00E+00 0.00 5 6.72E-12 0.00E+00 0.00E+00 0.00 6 1.95E-13 0.00E+00 0.00E+00 0.00 1 0.00E+00 0.00E+00 0.00E+00 0.00 2 3.47E-11 1.87E-14 1.87E-14 0.00 3 6.77E-10 1.95E-13 1.76E-13 0.00 4 0.00E+00 3.93E-13 7.56E-14 0.00 5 0.00E+00 4.69E-13 5.78E-14 0.00 6 0.00E+00 5.26E-13 5.78E-14 0.00 1 0.00E+00 7.04E-13 1.61E-13 0.00 2 3.44E-10 8.65E-13 1.61E-13 0.00 3 4.91E-11 4.92E-12 1.80E-12 0.00 4 1.96E-11 6.72E-12 1.80E-12 0.00 5 4.17E-11 1.14E-11 9.18E-13 0.00 6 4.92E-12 1.23E-11 9.18E-13 0.00 1 8.65E-13 1.89E-11 6.80E-13 0.00 2 9.15E-11 1.96E-11 6.80E-13 0.00 3 3.29E-11 2.47E-11 5.11E-12 0.01 4 4.37E-11 3.29E-11 1.77E-12 0.00 5 1.00E-10 3.47E-11 1.77E-12 0.00 6 1.14E-11 4.17E-11 2.05E-12 0.00 1 0.00E+00 4.37E-11 2.05E-12 0.00 2 7.04E-13 4.91 E-11 5.38E-12 0.01 3 2.47E-11 9.15E-11 8.60E-12 0.01 4 3.93E-13 1.00E-10 8.60E-12 0.01 5 1.23E-11 3.44E-10 2.44E-10 0.36 6 1.87E-14 6.77E-10 3.33E-10 0.49 Sum of varience 4.95E-10 Degrees of freedom 28 Er ror 4.21 E-06 158 Error Analysis (95%) Measurement: C 0 2 produced (moles) during CH 4 decomposition Cycle Variance Variance(sorted) Range Gap Qcalc 1 0.00E+00 0.00E+00 4.38E-10 0.00E+00 0.00 2 6.27E-16 0.00E+00 0.00E+00 0.00 3 1.28E-13 0.00E+00 0.00E+00 0.00 4 2.56E-13 0.00E+00 0.00E+00 0.00 5 6.34E-14 0.00E+00 0.00E+00 0.00 6 8.21E-16 0.00E+00 0.00E+00 0.00 1 0.00E+00 0.00E+00 0.00E+00 0.00 2 5.51E-12 0.00E+00 0.00E+00 0.00 3 4.38E-10 6.27E-16 1.95E-16 0.00 4 2.45E-12 8.21E-16 1.95E-16 0.00 5 6.61E-12 6.34E-14 6.20E-14 0.00 6 3.71E-12 1.25E-13 2.26E-15 0.00 1 0.00E+00 1.28E-13 2.26E-15 0.00 2 5.57E-12 1.86E-13 5.03E-14 0.00 3 8.83E-13 2.37E-13 1.96E-14 0.00 4 1.08E-12 2.56E-13 1.96E-14 0.00 5 8.38E-13 8.38E-13 4.54E-14 0.00 6 1.25E-13 8.83E-13 4.54E-14 0.00 1 0.00E+00 1.08E-12 2.02E-13 0.00 2 7.20E-11 2.45E-12 1.26E-12 0.00 3 4.06E-11 3.71E-12 1.26E-12 0.00 4 9.60E-11 5.51E-12 5.49E-14 0.00 5 1.04E-10 5.57E-12 5.49E-14 0.00 6 1.12E-10 6.61E-12 1.05E-12 0.00 1 0.00E+00 4.06E-11 3.14E-11 0.07 2 1.86E-13 7.20E-11 2.41 E-11 0.05 3 2.37E-13 9.60E-11 7.66E-12 0.02 4 0.00E+00 1.04E-10 7.66E-12 0.02 5 0.00E+00 1.12E-10 8.56E-12 0.02 6 0.00E+00 4.38E-10 3.26E-10 0.74 Sum of varience 4.52E-10 Degrees of freedom 29 Error 3.95E-06 159 Error Analysis (95%) Measurement: C O produced (moles) during partial oxidation Cycle Variance Variance(sorted) Range G a p Qcalc 1 5.44E-10 2.57E-13 2.21 E-09 1.66E-13 0.00 2 7.52E-11 4.22E-13 2.62E-14 0.00 3 2.11E-10 4.49E-13 2.62E-14 0.00 4 2.81E-10 5.19E-13 7.06E-14 0.00 5 3.40E-10 1.80E-12 7.06E-13 0.00 6 5.19E-13 2.51E-12 2.53E-13 0.00 1 2.76E-12 2.76E-12 2.53E-13 0.00 2 2.51E-12 1.53E-11 3.88E-12 0.00 3 1.53E-11 1.92E-11 3.88E-12 0.00 4 2.57E-13 7.52E-11 7.64E-12 0.00 5 4.49E-13 8.29E-11 7.64E-12 0.00 6 4.22E-13 1.12E-10 1.31E-11 0.01 1 2.44E-10 1.25E-10 1.31E-11 0.01 2 8.29E-11 2.11E-10 3.29E-11 0.01 3 1.80E-12 2.44E-10 3.29E-11 0.01 4 1.25E-10 2.81E-10 2.72E-12 0.00 5 1.12E-10 2.83E-10 2.72E-12 0.00 6 1.27E-09 3.40E-10 5.71E-11 0.03 1 2.21E-09 4.92E-10 2.67E-11 0.01 2 1.83E-09 5.18E-10 2.61E-11 0.01 3 5.18E-10 5.44E-10 2.61E-11 0.01 4 6.55E-10 6.55E-10 1.07E-11 0.00 5 1.92E-11 6.66E-10 1.07E-11 0.00 6 8.93E-10 7.58E-10 9.26E-11 0.04 1 2.83E-10 8.93E-10 8.45E-14 0.00 2 8.93E-10 8.93E-10 8.45E-14 0.00 3 4.92E-10 8.97E-10 4.40E-12 0.00 4 7.58E-10 1.27E-09 3.69E-10 0.17 5 6.66E-10 1.83E-09 3.78E-10 0.17 6 8.97E-10 2.21 E-09 3.78E-10 0.17 Sum of varience 1.34E-08 Degrees of freedom 30 Er ror 2.12E-05 160 Error Analysis (95%) Measurement: C 0 2 produced (moles) during paritial oxidation Cycle Variance Variance(sorted) Range G a p Qca lc 1 2.28E-09 1.70E-13 2.45E-09 4.46E-14 0.00 2 4.38E-10 2.14E-13 4.46E-14 0.00 3 4.50E-10 6.28E-13 1.05E-13 0.00 4 1.80E-09 7.32E-13 1.05E-13 0.00 5 1.25E-09 2.97E-12 7.67E-13 0.00 6 1.80E-09 3.74E-12 7.67E-13 0.00 1 5.97E-11 8.78E-12 5.04E-12 0.00 2 3.35E-11 1.90E-11 1.02E-11 0.00 3 2.97E-12 3.35E-11 1.45E-11 0.01 4 2.14E-13 5.00E-11 9.76E-12 0.00 5 8.78E-12 5.97E-11 9.76E-12 0.00 6 6.28E-13 1.71E-10 2.88E-11 0.01 1 1.71E-10 2.00E-10 2.88E-11 0.01 2 5.04E-10 2.33E-10 3.25E-11 0.01 3 7.32E-13 4.38E-10 1.21E-11 0.00 4 2.33E-10 4.50E-10 4.23E-22 0.00 5 3.74E-12 4.50E-10 4.23E-22 0.00 6 1.60E-09 5.04E-10 4.16E-11 0.02 1 2.00E-10 5.45E-10 9.06E-12 0.00 2 1.80E-09 5.54E-10 9.06E-12 0.00 3 8.00E-10 8.00E-10 1.24E-10 0.05 4 1.80E-09 9.24E-10 1.24E-10 0.05 5 2.45E-09 1.25E-09 3.26E-10 0.13 6 4.50E-10 1.60E-09 1.99E-10 0.08 1 1.90E-11 1.80E-09 4.23E-22 0.00 2 5.45E-10 1.80E-09 0.00E+00 0.00 3 1.70E-13 1.80E-09 0.00E+00 0.00 4 9.24E-10 1.80E-09 0.00E+00 0.00 5 5.00E-11 2.28E-09 1.65E-10 0.07 6 5.54E-10 2.45E-09 1.65E-10 0.07 Sum of varience 2.02E-08 Degrees of freedom 30 Er ror 2.60E-05 161 Error Analysis (95%) Measurement: Carbon removal percentage (%) after oxidation Cycle Variance Variance(sorted) Range G a p Qca lc 1 0.00001 0.00000 4.37E-02 3.82E-07 0.00 2 0.00126 0.00000 3.82E-07 0.00 3 0.00087 0.00001 1.02E-05 0.00 4 0.00003 0.00003 1.48E-05 0.00 5 0.00059 0.00004 1.60E-05 0.00 6 0.00051 0.00011 6.80E-05 0.00 1 0.00955 0.00026 2.20E-05 0.00 2 0.00475 0.00028 2.05E-05 0.00 1 0.00030 0.00030 2.05E-05 0.00 2 0.00501 0.00036 4.64E-05 0.00 3 0.00065 0.00041 4.64E-05 0.00 4 0.00088 0.00051 8.40E-05 0.00 5 0.00028 0.00059 5.32E-05 0.00 6 0.04375 0.00065 3.13E-05 0.00 1 0.00000 0.00068 3.13E-05 0.00 2 0.00011 0.00087 4.36E-06 0.00 3 0.00004 0.00088 4.36E-06 0.00 4 0.00068 0.00126 2.51 E-04 0.01 5 0.00026 0.00151 2.51 E-04 0.01 6 0.00036 0.00302 1.51E-03 0.03 1 0.00041 0.00475 2.56E-04 0.01 2 0.00151 0.00501 2.56E-04 0.01 3 0.01147 0.00955 1.40E-03 0.03 4 0.00000 0.01095 5.25E-04 0.01 5 0.01095 0.01147 5.25E-04 0.01 6 0.00302 0.04375 3.23E-02 0.74 Sum of varience 5.35E-02 Degrees of freedom 25 Error 5% 162 Error Analysis (95%) Measurement: H 2 production rate (mol/s) during C H 4 decomposition Cycle Variance Variance(sorted) Range G a p Qcalc 1 1.25E-11 0.00E+00 1.05E-09 0.00E+00 0.00 2 1.70E-11 0.00E+00 0.00E+00 0.00 3 1.36E-11 0.00E+00 0.00E+00 0.00 4 1.05E-11 7.42E-14 1.26E-14 0.00 5 1.70E-11 8.68E-14 1.26E-14 0.00 6 1.70E-11 3.47E-13 1.65E-24 0.00 1 5.72E-12 3.47E-13 1.65E-24 0.00 2 4.45E-11 5.43E-13 1.95E-13 0.00 3 1.05E-09 1.10E-12 2.86E-13 0.00 4 0.00E+00 1.39E-12 2.86E-13 0.00 5 0.00E+00 2.63E-12 1.15E-12 0.00 6 0.00E+00 3.78E-12 1.15E-12 0.00 1 5.33E-10 5.72E-12 5.47E-13 0.00 2 4.30E-11 6.27E-12 5.47E-13 0.00 3 6.42E-11 7.03E-12 7.60E-13 0.00 4 1.51E-11 8.30E-12 1.27E-12 0.00 5 3.04E-11 1.05E-11 2.00E-12 0.00 6 9.65E-11 1.25E-11 1.06E-12 0.00 1 3.47E-13 1.36E-11 1.06E-12 0.00 2 7.03E-12 1.51E-11 1.53E-12 0.00 3 6.27E-12 1.70E-11 0.00E+00 0.00 4 3.47E-13 1.70E-11 0.00E+00 0.00 5 5.43E-13 1.70E-11 8.30E-25 0.00 6 8.68E-14 3.04E-11 1.26E-11 0.01 1 2.63E-12 4.30E-11 1.53E-12 0.00 2 1.39E-12 4.45E-11 1.53E-12 0.00 3 8.30E-12 6.42E-11 1.96E-11 0.02 4 7.42E-14 9.65E-11 3.24E-11 0.03 5 3.78E-12 5.33E-10 4.36E-10 0.42 6 1.10E-12 1.05E-09 5.16E-10 0.49 Sum of varience 4.19E-10 Degrees of freedom 28 Er ror 3.87E-06 163 Error Analysis (95%) Measurement: Initial C H 4 decomposition rate (mol/(g*s)) Cycle Variance Variance(sorted) Range G a p Qcalc 1 1.0789E-12 2.1733E-17 1.75E-11 5.12E-17 0.00 2 4.1165E-13 7.2937E-17 5.12E-17 0.00 3 3.8000E-13 1.3948E-16 6.65E-17 0.00 4 3.7656E-13 7.0117E-T5 5.09E-15 0.00 5 6.1056E-13 1.2099E-14 5.09E-15 0.00 6 6.0169E-13 2.4055E-14 6.68E-15 0.00 1 3.0734E-14 3.0734E-14 5.00E-15 0.00 2 1.2099E-14 3.5732E-14 5.00E-15 0.00 3 1.7528E-11 4.8078E-14 1.23E-14 0.00 4 1.3948E-16 9.6617E-14 4.85E-14 0.00 5 7.2937E-17 1.8401E-13 8.74E-14 0.00 6 2.1733E-17 3.7656E-13 3.43E-15 0.00 1 4.9938E-13 3.8000E-13 3.43E-15 0.00 2 9.7089E-13 4.1165E-13 3.17E-14 0.00 3 4.1378E-12 4.9938E-13 8.77E-14 0.01 4 1.1760E-12 6.0169E-13 8.87E-15 0.00 5 4.4153E-12 6.1056E-13 8.87E-15 0.00 6 9.5446E-13 6.6754E-13 1.16E-14 0.00 1 8.4367E-13 6.7916E-13 1.16E-14 0.00 2 6.6754E-13 7.0165E-13 2.25E-14 0.00 3 7.0165E-13 7.9339E-13 5.03E-14 0.00 4 6.7916E-13 8.4367E-13 5.03E-14 0.00 5 7.9339E-13 9.5446E-13 1.64E-14 0.00 6 1.0549E-12 9.7089E-13 1.64E-14 0.00 1 1.8401E-13 1.0549E-12 2.40E-14 0.00 2 3.5732E-14 1.0789E-12 2.40E-14 0.00 3 7.0117E-15 1.1760E-12 9.71E-14 0.01 4 9.6617E-14 4.1378E-12 2.77E-13 0.02 5 4.8078E-14 4.4153E-12 2.77E-13 0.02 6 2.4055E-14 1.7528E-11 1.31E-11 0.75 Sum of varience 2.08E-11 Degrees of freedom 29 Er ror 8.47E-07 164 A p p e n d i x F : S u m m a r y R e s u l t s o f F a c t o r i a l D e s i g n E x p e r i m e n t s RunOrder CHt decompose temperature CC) CH, decompose duration (min) POX Oxygen cone. (%) POX temperature CO POX duration (min) Exp ID 1 550 3 20 450 15 FD_1 2 550 3 2.5 550 15 FD_2 3 550 50 20 550 15 FD_3 4 450 50 2.5 550 15 FD_4 5 450 50 20 450 15 FD_5 6 550 3 2.5 450 3 FD_6 7 450 3 2.5 450 15 FD_7 8 450 3 20 450 3 FD_6 9 550 50 2.5 550 3 FD_9b 10 550 50 20 450 3 FD_10 11 450 50 2.5 450 3 FD_11 12 550 50 2.5 450 15 F D J 2 13 450 50 20 550 3 FD_13 14 450 3 2.5 550 3 FD_14 15 450 3 20 550 15 FD_15 16 550 3 20 550 3 FD.16 repeat 550 50 2.5 450 15 FD_12 12 550 50 2.5 450 15 FD_12 repeat 450 3 2.5 550 3 FD_14 14 450 3 2.5 550 3 FD_14 repeat 550 3 2.5 450 3 FD_6b 6 550 3 2.5 450 3 FD_6 repeat 450 50 2.5 550 15 FD_4b 4 450 50 2.5 550 15 FD_4 repeat 550 50 2.5 550 3 FD_9 9 550 50 2.5 550 3 FD_9b During CH 4 Decomposition [Hydrogen produced(mol) Cycle_1 Cycle_2 Cycle_3 Cycle_4 Cycle_5 Cycle_6 4.93E-04 5.01 E-04 5.17 E-04 5.14E-04 5.16E-04 5.00E-04 6.31 E-04 6.30E-04 6.28E-04 6.30E-04 6.46E-04 6.51E-04 2.82E-03 2.05E-03 2.03E-03 1.96E-03 2.12E-03 2.13E-03 3.10E-03 3.33E-03 3.63E-03 3.80E-03 3.93E-03 4.00E-03 2.98E-03 1.80E-03 2.30E-03 2.34E-03 2.39E-03 2.37E-03 5.83E-04 6.06E-04 6.07E-04 6.07E-04 6.14E-04 6.02E-04 1.88E-04 1.24E-04 1.12E-04 1.13E-04 1.12E-04 1.10E-04 1.99E-04 5.00E-05 7.30E-05 8.64E-05 9.20E-05 9.67E-05 1.91E-03 1 OOE-03 9.04E-04 9.02E-04 7.74E-04 7.67E-04 1.94E-03 2.09E-03 2.15E-03 2.26E-03 2.33E-03 2.39E-03 2.96E-03 2.76E-03 2.33E-03 1.63E-03 1.48E-03 1.19E-03 2.13E-03 2.51 E-03 2.47E-03 2.52E-03 2.52E-03 2.55E-03 3.19E-03 1.38E-03 2.31 E-03 2.08E-03 2.28E-03 2.14E-03 1.90E-04 1.80E-04 0.00E+00 O.OOE+00 0.00E+00 0.00E+00 1.92E-04 0.00E+00 0.00E+00 O.OOE+00 0.00E+00 0.00E+00 4.82E-04 4.98E-04 5.38E-04 5.09E-04 5.20E-04 5.26E-04 2.37E-03 2.79E-03 2.72E-03 2.74E-03 2.80E-03 2.83E-03 2.13E-03 2.51 E-03 2.47E-03 2.52E-03 2.52E-03 2.55E-03 2.01 E-04 2.11 E-04 1.47E-04 0.00E+00 0.00E+00 O.OOE+00 1.90E-04 1.80E-04 O.OOE+00 0.00E+00 0.00E+00 0.00E+00 4.78E-04 6.35E-04 6.44E-04 6.25E-04 6.39E-04 6.46E-04 5.83E-04 6.06E-04 6.07E-04 6.07E-04 6.14E-04 6.02E-04 3.14E-03 3.51 E-03 3.80E-03 3.84E-03 3.98E-03 4.02E-03 3.10E-03 3.33E-03 3.63E-03 3.80E-03 3.93E-03 4.00E-03 2.02E-03 1.08 E-03 1.10E-03 9.21 E-04 9.06E-04 8.39E-04 1.91E-03 1.00 E-03 9.04E-04 9.02E-04 7.74E-04 7.67E-04 Figure F-l Summary results of total H 2 production during C H 4 decomposition for F D series 165 CO produced(mol) RunOrder CH4 decompose temperature (°C) CH4 decompose duration {min) POX Oxygen cone. (%) POX temperature CO POX duration (min) Exp ID 1 550 3 20 450 15 FD_1 2 550 3 2.5 550 15 FD_2 3 550 50 20 550 15 FD_3 4 450 50 2.5 550 15 FD_4 5 450 50 20 450 15 FD_5 6 550 3 2.5 450 3 FD_6 7 450 3 2.5 450 15 FD_7 8 450 3 20 450 3 FD_8 9 550 50 2.5 550 3 FD_9b 10 550 50 20 450 3 FD_10 11 450 50 2.5 450 3 FD_11 12 550 50 2.5 450 15 FD_12 13 450 50 20 550 3 FD_13 14 450 3 2.5 550 3 FD_14 15 450 3 20 550 15 FD_15 16 550 3 20 550 3 F D J 6 repeat 550 50 2.5 450 15 FD_12 12 550 50 2.5 450 15 FD_12 repeat 450 3 2.5 550 3 FD_14 14 450 3 2.5 550 3 FD_14 repeat 550 3 2.5 450 3 FD_6b 6 550 3 2.5 450 3 FD_6 repeat 450 50 2.5 550 15 FD_4b 4 450 50 2.5 550 15 FD_4 repeat 550 50 2.5 550 3 FD_9 9 550 50 2.5 550 3 FD_9b During CH, Decomposition Cyclejt Cycle_2 Cycle_3 Cyclejt Cycle_5 Cye/e_6 2.87E-06 7.43E-05 5.15E-05 4.12E-05 3.09E-05 2.44E-05 0.00E+00 1.14E-04 1.05E-04 9.85E-05 9.28E-05 8.82E-05 2.82E-05 8.34E-05 7.05E-05 4.61 E-05 4.28E-05 3.08E-05 7.93E-06 1 40E-04 1.23E-04 1.30E-04 1.04E-04 1 03E-04 0.00E+00 2.16E-05 1.99E-05 1.51 E-05 1.15E-05 9.10E-06 0.00E+00 1.61 E-05 3.75E-05 2.84E-05 2.78E-05 2.02E-05 0.00E+00 1.82E-05 1.49E-05 1.37E-05 1.46E-05 1.13E-05 0.00E+00 7.81 E-07 2.12E-06 3.05E-06 2.46E-06 2.66E-06 0.00E+00 1.39E-05 5.79E-06 3.99E-06 3.90E-06 3.65E-06 O.00E+00 9.71 E-05 8.77E-05 6.99E-05 5.91 E-05 5.71 E-05 0.00E+00 2.72E-05 1.54E-05 2.16E-05 1.48E-05 2.04E-05 0.00E+00 3.31 E-05 1.55E-05 1.60E-05 1.18E-05 1.30E-05 1.05E-05 2.77E-05 3.67E-05 2.71 E-05 2.76E-05 1.92E-05 0.00E+00 6.27E-05 0.00E+00 0.OOE+00 0.00E+00 0.00E+00 O.OOE+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 5.77E-05 5.28E-05 4.92E-05 4.85E-05 4.07E-05 0.00E+00 3.93E-05 1.65E-05 1.69E-05 1.54E-05 1.24E-05 0.00E+00 3.31 E-05 1.55E-05 1.60E-05 1.18E-05 1.30E-05 O.00E+O0 7.11 E-05 3.68E-05 0.00E+00 O.00E+00 O.OOE+OO 0.00E+00 6.27E-05 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 4.23E-05 2.76E-05 3.46E-05 1.87E-05 2.34E-05 0.00E+00 1.61 E-05 3.75E-05 2.84E-05 2.78E-05 2.02E-05 6.62E-06 1.53E-04 1.31E-04 1.21 E-04 1.18E-04 1.08E-04 7.93E-06 1.40E-04 1.23E-04 1.30E-04 1.04E-04 1.03E-04 0.00E+00 1.27E-05 1.28E-05 4.87E-06 8.86E-06 3.85E-06 0.00E+00 1.39E-05 5.79E-06 3.99E-06 3.90E-06 3.65E-06 Figure F-2 Summary results of total C O production during C H 4 decomposition for F D series 166 RunOrder CH4 decompose temperature ro CH, decompose duration (min) POX Oxygen cone. (%) POX temperature CC) POX duration (min) Exp ID 1 550 3 20 450 15 FD_1 2 550 3 2.5 550 15 FD_2 3 550 50 20 550 15 F D _ 3 4 450 50 2.5 550 15 FD_4 5 450 50 20 450 15 FD_5 6 550 3 2.5 450 3 FD_6 7 450 3 2.5 450 15 F D _ 7 8 450 3 20 450 3 FD_8 9 550 50 2.5 550 3 FD_9b 10 550 50 20 450 3 FD_10 11 450 50 2.5 450 3 FD_11 12 550 50 2.5 450 15 FD_12 13 450 50 20 550 3 FD_13 14 450 3 2.5 550 3 FD_14 15 450 3 20 550 15 FD_15 16 550 3 20 550 3 FD_16 repeat 550 50 2.5 450 15 FD_12 12 550 50 2.5 450 15 FD_12 repeat 450 3 2.5 550 3 FD_14 14 450 3 2.5 550 3 FD_14 repeat 550 3 2.5 450 3 FD_6b 6 550 3 2.5 450 3 F D _ 6 repeat 450 50 2.5 550 15 FD_4b 4 450 50 2.5 550 15 FD_4 repeat 550 50 2.5 550 3 FD_9 9 550 • 50 2.5 . 550 3 FD_9b During CH 4 Decomposition CO j producod(mol) Cyctejt Cycle_2 Cycte_3 Cyc/e_4 Cycle_S Cycle_6 0.00E+00 2.08E-05 3.52E-05 4.36E-05 4.71 E-05 5.21 E-05 O.OOE+00 3.74E-05 4.57E-05 4.43E-05 4.71 E-05 4.89E-05 0.00E+00 2.31 E-05 2.21 E-05 3.04E-05 3.05E-05 2.78E-05 0.00E+00 2.90E-05 3.25E-05 3.61 E-05 2.90E-05 3.22E-05 0.00E+00 1.06E-05 1.99E-05 2.37E-05 2.05E-05 2.48E-05 0.00E+00 4.81 E-07 3.98E-06 2.03E-06 2.61 E-06 1.21 E-06 0.00E+00 4.72E-05 4.16E-05 4.07E-05 4.14E-05 4.33E-05 0.00E+00 1.66E-05 2.30E-05 2.53E-05 2.80E-05 2.93E-05 0.00E+00 0.00E+00 0.00E+00 O.OOE+00 0.00E+00 0.00E+00 0.00E+00 1.50E-05 1.71E-05 1.66E-05 8.25E-06 1.76E-05 0.00E+00 4.58E-06 3.66E-07 4.30E-06 3.43 E-07 3.88E-06 0.00E+00 3.06E-06 7.70E-07 6.41 E-07 9.53E-07 6.73E-07 0.00E+00 8.57E-06 1.46E-05 1.16E-05 1.03 E-05 1.11E-05 0.00E+00 2.57E-05 7.20E-06 7.83E-06 8.13E-06 8.59E-06 0.00E+00 5.39E-06 6.00E-06 1.98E-06 6.97E-06 9.41 E-06 0.00E+00 2.63E-05 3.32E-05 3.78E-05 3.08E-05 3.78E-05 0.00E+00 3.02E-06 1.28E-06 1.36E-06 5.97E-07 6.32E-07 O.00E+00 3.06E-06 7.70E-07 6.41 E-07 9.53E-07 6.73E-07 0.00E+00 2.24E-05 3.68E-05 1.00E-05 1.18E-05 1.13E-05 0.00E+00 2.57E-05 7.20E-06 7.83E-06 8.13 E-06 8.59E-06 0.00E+00 3.82E-06 2.65E-06 3.51 E-06 1.32E-06 1.71 E-06 0.00E+00 4.81 E-07 3.98E-06 2.03E-06 2.61 E-06 1.21 E-06 O.OOE+00 4.10E-05 4.15E-05 5.00E-05 4.34E-05 4.72E-05 0.00E+00 2.90E-05 3.25E-05 3.61 E-05 2.90E-05 3.22E-05 0.00E+00 6.10E-07 6.88E-07 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 O.OOE+00 0.00E+00 0.00E+00 Figure F-3 Summary results of total C 0 2 production during C H 4 decomposition for F D series 167 RunOrder CH, decompose temperature CO CH, decompose duration (min) POX Oxygen cone. (%) POX temperature CO POX duration (min) Exp ID 1 550 3 20 450 15 FD_1 2 550 3 2.5 550 15 FD_2 3 550 50 20 550 15 FD_3 4 450 50 2.5 550 15 FD_4 5 450 50 20 450 15 FD_5 6 550 3 2.5 450 3 FD_6 7 450 3 2.5 450 15 FD_7 8 450 3 20 450 3 FD_8 9 550 50 2.5 550 3 FD_9b 10 550 50 20 450 3 FDJO 11 450 50 2.5 450 3 FD_11 12 550 50 2.5 450 15 FD_12 13 450 50 20 550 3 F D J 3 14 450 3 2.5 550 3 FD_14 15 450 3 20 550 15 FD_15 16 550 3 20 550 3 FD_16 repeat 550 50 2.5 450 15 FD_12 12 550 50 2.5 450 15 FD_12 repeat 450 3 2.5 550 3 FD_14 14 450 3 2.5 550 3 FD_14 repeat 550 3 2.5 450 3 FD_6b 6 550 3 2.5 450 3 FD_6 repeat 450 50 2.5 550 15 FD_4b 4 450 50 2.5 550 15 FD_4 repeat 550 50 2.5 550 3 FD_9 9 550 50 2.5 550 3 FD_9b During POX CO produced(mol) Cycle J CycieJL Cyclojl Cyclejt Cycled Cycled 7.46E-05 1.17E-05 1.61 E-05 1.26E-05 1.47E-05 4.80E-06 2.39E-04 7.76E-05 8.41 E-05 8.58E-05 8.47E-05 8.99E-05 8.15E-04 3.94E-04 3.79E-04 4.04E-04 3.91 E-04 3.60E-04 6.27E-04 6.19E-04 6.80E-04 7.16E-04 7.17E-04 6.53E-04 7.54E-04 6.57E-05 3.18E-04 2.67E-04 2.49E-04 1.22 E-04 8.98E-05 4.58E-05 4.38E-05 3.79E-05 4.34E-05 2.18E-05 2.30E-05 3.36E-06 1.92E-06 2.54E-06 3.21 E-06 2.79E-06 2.88E-05 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 1.38E-04 1.79E-04 1.60E-04 1 67E-04 1.57E-04 1.66E-04 3.98E-04 3.15E-04 3.24E-04 3.31 E-04 3.23E-04 3.03E-04 4.21 E-05 4.63E-05 2.71 E-05 3.15E-05 1.46E-05 1.23E-05 7.40E-05 7.40E-05 6.82E-05 6.27E-05 5.89E-05 6.08E-06 9.92E-04 6.70E.05 4.67E-04 3.04E-04 3.73E-04 2.56E-04 5.38E-05 2.51 E-06 9.14E-07 7.16E-07 9.47E-07 9.19E-07 2.64E-05 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 9.81E-05 2.91 E-05 6.47E-05 5.64E-05 4.91 E-05 5.37E-05 1.07E-04 8.63E-05 8.87E-05 8.64 E-05 8.50E-05 7.10E-06 7.40E-05 7.40E-05 6.82E-05 6.27E-05 5.89E-05 6.08E-06 5.62E-05 4.75E-06 6.45E-06 0.00E+00 0.00E+00 0.00E+00 5.38E-05 2.51 E-06 9.14E-07 7.16E-07 9.47E-07 9.19E-07 6.77E-05 5.87E-05 4.57E-05 5.38E-05 5.84E-05 7.21 E-05 8.98E-05 4.58E-05 4.38E-05 3.79E-05 4.34E-05 2.18E-05 6.94E-04 6.79E-04 7.12E-04 7.52E-04 7.23E-04 6.95E-04 6.27E-04 6.19E-04 6.80E-04 7.16E-04 7.17E-04 6.53E-04 1.14E-04 1.36E-04 1.28E-04 1.28E-04 1.21 E-04 1.23E-04 1.38E-04 1.79E-04 1.60E-04 1.67 E-04 1.57 E-04 1 66E-04 Figure F-4 Summary results of total C O production during partial oxidation for F D series 168 RunOrder CH« decompose temperature CO CH4 decompose duration (min) POX Oxygen cone. <%> POX temperature ro POX duration (min) Exp ID 1 550 3 20 450 15 FD_1 2 550 3 2.5 550 15 FD_2 3 550 50 20 550 15 FD_3 4 450 50 2.5 550 15 FD_4 5 450 50 20 450 15 FD_5 6 550 3 2.5 450 3 FD_6 7 450 3 2.5 450 15 FD_7 8 450 3 20 450 3 FD_8 9 550 50 2.5 550 3 FD_9b 10 550 50 20 450 3 FD_10 11 450 50 2.5 450 3 FD_11 12 550 50 2.5 450 15 FD_12 13 450 50 20 550 3 FD_13 14 450 3 2.5 550 3 FD_14 15 450 3 20 550 15 FD_15 16 550 3 20 550 3 FD_16 repeat 550 50 2.5 450 15 FD_12 12 550 50 2.5 450 15 FD_12 repeat 450 3 2.5 550 3 FD_14 14 450 3 2.5 550 3 FD_14 repeat 550 3 2.5 450 3 FD_6b 6 550 3 2.5 450 3 FD_6 repeat 450 50 2.5 550 15 FD_4b 4 450 50 2.5 550 15 FD_4 repeat 550 50 2.5 550 3 FD_9 9 550 50 2.5 550 3 FD_9b - During'POX CO t produced(mol) Cye/eJ Cycle_2 Cyc/e_3 Cyc/e_4 Cycled Cycle_G 2.536-04 2.106-04 2.20E-04 2.52E-04 2.57E-04 2.20E-04 2.19E-04 1.52E-04 1.53E-04 1.60E-04 1.64E-04 1.74E-04 6.80E-04 6.88E-04 6.75E-04 6.92E-04 7.93E-04 8.70E-04 1.07E-03 1.03E-03 1.12E-03 1.17E-03 1.24E-03 1.33E-03 9.51 E-04 9.04E-04 9.94E-04 1.08 E-03 1.15E-03 1.26E-03 1.69E-04 2.18E-04 2.02E-04 2.16E-04 2.06E-04 1.32E-04 1.02 E-04 3.74E-05 3.37E-05 3.47E-05 3.82E-05 3.746-05 1.276-04 2.07E-05 2.43E-05 2.60E-05 3.00E-05 3.02E-05 5.64E-05 7.67E-05 8.34E-05 8.12E-05 8.05E-05 8.74E-05 6.59E-04 7.60E-04 8.27E-04 8.84E-04 8.98E-04 9.43E-04 1.58E-04 1.95E-04 1.89E-04 2.04 E-04 1.96E-04 1.89E-04 7.65E-04 9.32E-04 1.01 E-03 1.04E-03 1.09E-03 1.00E-03 7.00E-04 6.60E-04 8.00E-04 8.19E-04 8.54E-04 9.046-04 3.89E-05 2.33E-05 2.57E-05 2.34E-05 2.17E-05 1.756-05 1.05E-04 7.12E-06 6.37E-06 3.03E-06 4.44E-06 3.61E-06 1.83E-04 2.07E-04 2.24E-04 2.10E-04 2.126-04 2.18E-04 8.33E-04 9.61 E-04 1.04E-03 1.10E-03 1.146-03 1.06E-03 7.65E-04 9.32E-04 1.01E-03 1.04E-03 1.096-03 1.00E-03 2.79E-05 1.51 E-05 2.33E-05 2.40E-05 1.75E-05 1.86E-05 3.89E-05 2.33E-05 2.57E-05 2.34E-05 2.176-05 1.75E-05 1.51E-04 1.86E-04 2.03E-04 1.95E-04 2.09E-04 1.89E-04 1.69E-04 2.18E-04 2.02E-04 2.16E-04 2.066-04 1.32E-04 1.05E-03 1.09E-03 1.16E-03 1.23E-03 1.31 E-03 1.366-03 1.07E-03 1.03E-03 1.12E-03 1.17E-03 1.24E-03 1.336-03 6.2SE-05 1.10E-04 8.40E-05 1.24E-04 9.04E-05 1.216-04 5.64E-05 7.67E-05 8.34E-05 8.12E-05 8.05E-05 8.746-05 Figure F-5 Summary results of total C 0 2 production during partial oxidation for F D series 169 RunOrder CH, decompose temperature <°C) C H , decompose duration (min) POX Oxygen cone. (%) POX temperature CC) POX duration (min) Exp ID 1 550 3 20 450 15 FD_1 2 550 3 2.5 550 15 FD_2 3 550 50 20 550 15 FD_3 4 450 50 2.5 550 15 FD_4 5 450 50 20 450 15 FD_5 6 550 3 2.5 450 3 FD_6 7 450 3 2.5 450 15 FD_7 8 450 3 20 450 3 FD_8 9 550 50 2.5 550 3 FD_9b 10 550 50 20 450 3 FD_10 11 450 50 2.5 450 3 FD_11 12 550 50 2.5 450 15 FD_12 13 450 50 20 550 3 FD_13 14 450 3 2.5 550 3 FD_14 15 450 3 20 550 15 FD_15 16 550 3 20 550 3 FD_16 repeat 550 50 2.5 450 15 FD_12 12 550 50 2.5 450 15 FD_12 repeat 450 3 2.5 550 3 FD_14 14 450 3 2.5 550 3 FD_14 repeat 550 3 2.5 450 3 FD_6b 6 550 3 2.5 450 3 FD_6 repeat 450 50 2.5 550 15 FD_4b 4 450 50 2.5 550 15 FD_4 repeat 550 50 2.5 550 3 FD_9 9 550 50 2.5 550 3 FD_9b During CH 4 Decomposition Hydrogen produced rate(mol/sec) Cycle_1 Cycle_2 Cycle_3 Cycle_4 Cycle_5 Cycle_6 1.54E-04 1.57E-04 1.62E-04 1.61 E-04 1.61 E-04 1.56E-04 1.97E-04 1.97E-04 1.96E-04 1.97E-04 2.02E-04 2.03E-04 5.88E-05 4.27E-05 4.23E-05 4.08E-05 4.42E-05 4.44E-05 6.46E-05 6.94E-05 7.56E-05 7.92E-05 8.19E-05 8.33E-05 6.21 E-05 3.75E-05 4.79E-05 4.88E-05 4.98E-05 4.94E-05 1.82E-04 1.89E-04 1.90E-04 1.90E-04 1.92E-04 1.88E-04 5.86E-05 3.89E-05 3.51 E-05 3.54E-05 3.51 E-05 3.43E-05 6.22E-05 1.56E-05 2.28E-05 2.70E-05 2.87E-05 3.02E-05 3.98E-05 2.08E-05 1.88E-05 1.88E-05 1.61 E-05 1.60E-05 4.04E-05 4.35E-05 4.48E-05 4.71 E-05 4.85E-05 4.98E-05 6.17E-05 5.75E-05 4.85E-05 3.81 E-05 3.08E-05 2.48E-05 4.44E-05 5.23E-05 5.15E-05 5.25E-05 5.25E-05 5.31 E-05 6.65E-05 2.88E-05 4,81 E-05 4.33E-05 4.75E-05 4.46E-05 5.93E-05 5.64E-05 O.OOE+00 0.00E+00 0.00E+00 0.00E+00 6.01 E-05 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 1.50E-04 1.56E-04 1.68E-04 1.59E-04 1.62E-04 1.64E-04 4.94E-05 5.81 E-05 5.67E-05 5.71 E-05 5.83E-05 5.90E-05 4.44E-05 5.23E-05 5.15E-05 5.25E-05 5.25E-05 5.31 E-05 6.27E-05 6.58E-05 4.58E-05 O.OOE+00 0.00E+00 0.00E+00 5.93E-05 5.64E-05 0.00E+00 O.OOE+00 0.00E+00 0.00E+00 1.50E-04 1.99E-04 2.01 E-04 1.95E-04 2.00E-04 2.02E-04 1.82E-04 1.89E-04 1.90E-04 1.90E-04 1.92E-04 1.88E-04 6.54E-05 7.31 E-05 7.92E-05 8.00E-05 8.29E-05 8.38E-05 6.46E-05 6.94E-05 7.56E-05 7.92E-05 8.19E-05 8.33E-05 4.21 E-05 2.25E-05 2.29E-05 1.92E-05 1.89E-05 1.75E-05 3.98E-05 2.08E-05 1.88E-05 1.88E-05 1.61E-05 1.60E-05 Figure F-6 Summary results of average H 2 production rate during C H 4 decomposition for FD series 170 RunOrder CH4 decompose temperature ro C H , decompose duration (min) POX Oxygen cone. (%) POX temperature CC) POX duration (min) Exp ID 1 550 3 20 450 15 FD_1 2 550 3 2.5 550 15 FD_2 3 550 50 20 550 15 FD_3 4 450 50 2.5 550 15 FD.4 5 450 50 20 450 15 FD_5 6 550 3 2.5 450 3 FD_6 7 450 3 2.5 450 15 FD_7 8 450 3 20 450 3 FD_8 9 550 50 2.5 550 3 FD_9b 10 550 50 20 450 3 FD_10 11 450 50 2.5 • 450 3 FD.11 12 550 50 2.5 450 15 FD_12 13 450 50 20 550 3 FD_13 14 450 3 2.5 550 3 FD_14 15 450 3 20 550 15 FD_15 16 550 3 20 550 3 FD_16 repeat 550 50 2.5 450 15 FD_12 12 550 50 2.5 450 15 FD_12 repeat 450 3 2.5 550 3 FD_14 14 450 3 2.5 550 3 FD_14 repeat 550 3 2.5 450 3 FD_6b 6 550 3 2.5 450 3 FD_6 repeat 450 50 2.5 550 15 FD_4b 4 450 50 2.5 550 15 FD_4 repeat 550 50 2.5 550 3 FD_9 9 550 50 2.5 550 3 FD_9b During CH4 Decomposition Initial CH 4 decomposition Rate Cycteji Cycle_2 Cycle_3 Cycle_4 Cycle_5 Cycle_6 1.88E-05 1.76E-05 1.83E-05 1.78E-05 1.76E-05 1.74E-05 1.93E-05 2.09E-05 2.10E-05 2.09E-05 2.11E-05 2.12E-05 1.56E-05 1.31 E-05 1.39E-05 1.41 E-05 1.42E-05 1.38E-05 1.19E-05 5.75E-06 5.98E-06 6.23E-06 6.48E-06 6.58E-06 1.12E-05 3.37E-06 4.13E-06 4.12E-06 4.24E-06 4.13E-06 1.82E-05 1.86E-05 2.01 E-05 1.93E-05 1.96E-05 1.85E-05 1.06E-05 7.99E-06 7.30E-06 7.09E-06 7.05E-06 6.93E-06 1.19E-05 3.45E-06 4.82E-06 5.66E-06 6.09E-06 6.32E-06 1.78E-05 1.59E-05 1.56E-05 1.59E-05 1.56E-05 1.56E-05 1.86E-05 1.68E-05 1.72E-05 1.74E-05 1.76E-05 1.77E-05 1.06E-05 6.68E-06 6.96E-06 6.70E-06 6.64E-06 6.42E-06 1.75E-05 1.55E-05 1.61 E-05 1.66E-05 1.66E-05 1.71E-05 1.07E-05 2.35E-06 3.41 E-06 3.06E-06 3.27E-06 3.08E-06 1.13E-05 5.05E-06 1.78E-07 1.85E-07 1.69E-07 1.69E-07 1.09E-05 2.28E-07 1.44E-07 1.96E-07 2.13E-07 1.63E-07 1.66E-05 1.25E-05 1.39E-05 1.36E-05 1.39E-05 1.35E-05 1.90E-05 1.64E-05 1.70E-05 1.74E-05 1.79E-05 1.82E-05 1.75E-05 1.55E-05 1.61 E-05 1.66E-05 1.68E-05 1.71E-05 1.13E-05 5.05E-06 1.76E-07 1.85E-07 1.69E-07 1.69E-07 1.15E-05 4.90E-06 6.10E-06 2.02E-07 1.57E-07 1.62E-07 1.72E-05 1.72E-05 1.72E-05 1.78E-05 1.67E-05 1.71 E-05 1.82E-05 1.86E-05 2.01 E-05 1.93E-05 1.96E-05 1.85E-05 1.06E-05 4.60E-06 4.80E-06 5.06E-06 5.22E-06 5.13E-06 1.19E-05 5.75E-06 5.98E-06 6.23E-06 6.48E-06 6.58E-06 1.78E-05 1.59E-05 1.56E-05 1.59E-05 1.56E-05 1.56E-05 1.71 E-05 1.57E-05 1.57E-05 1.55E-05 1.53E-05 1.54E-05 Figure F-7 Summary results of initial C H 4 decomposition rate for FD series 171 A p p e n d i x G : D e s c r i p t i o n o f E x p e r i m e n t O p e r a t i o n a n d s u m m a r y o f e x p e r i m e n t p a r a m e t e r s Prolab MS 1. Turn on Prolab mass spectrometer 2. Start the emission by turning on the filament current 3. Turn on the inlet heater to maintain the capillary tube at 180-200 °C 4. Scan in MIM mode and record the data TEOM 5. Load the catalyst 6. Perform the TE check 7. Start the profile to run cyclic performance of CEL, decomposition and partial oxidation 8. This profile includes 3 programs a. Program 1: in situ EE: reduction b. Program 2: CEL; decomposition followed by partial oxidation. He flush is employed between each step. This program can be run repeatedly in order to reach the desired number of cycles c. Program 3: cool down the system to room temperature under the He flow 9. During the experiment, adjust the valves according to the following figure to introduce corresponding reactant gas 172 reactor a. For H2 reduction i . turn off V33 i i . turn on V55 i i i . turn V(Y) to "From H 2 / C H 4 " iv. turn V(G) to "To reactor" b. For He flush i . turn V(G) to vent c. For CH4 decomposition i . turn off V55 i i . turn on V33 ii i . turn V(Y) to "From H 2 / C H 4 " iv. turn V(G) to "To reactor" d. For Partial oxidation i . turn V(Y) to "From 0 2 " i i . turn V(G) to "To reactor" 10. Once the experiment is finished, turn off the Prolab MS and T E O M Refer to "Prolab Operations Manual" provided by ThermoONIX for the following topics: 1. Basic operating instructions of Prolab mass spectrometer 2. How to use the Gasworks process software • 3. How to construct a recipe to specify the mass for multiple ion monitoring (MIM) Refer to "Operating Manual of T E O M series 1500 Pulse Mass Analyzer" for the following topics: 1. Program and profile setup, data storage setting (Section 5: Setting up the software) 2. Sample loading (Section 4: Loading and unloading a sample). A video clip to show a demonstration of sample loading and unloading is available. Please ask it from Dr. Kevin Smith 3. TE checking (Section 3: Starting the PMA) 4. The operation of the T E O M Pulse Mass Analyzer (Section 6: Starting an experiment) 174 Table G-l Summary of experiment parameters Experimenta'b Mass of catalyst loaded (g) CH 4 decomposition temperature (K) CH 4 decomposition duration (min) POX o 2 Cone, (vol. %) POX temperature (K) POX duration (min) Co-1 0.14825 773 45 5 773 45 Co-2 773 30 5 773 4 Co-3 773 45 5 773 "4 Ni-4 0.1190 773 25 10 773 5 Ni-5 773 25 5 773 5 Ni-6 773 25 2.5 773 10 Ni-7 773 25 20 773 5 Ni-8-02 0.09433 748 25 10 773 5 Ni-9-C02 0.10187 748 15 40 (C02) 773 100 FD-1 0.11751 823 3 20 723 15 FD-2 0.12196 823 3 2.5 823 15 FD-3 0.13563 823 50 20 823 15 FD-4 0.11077 723 50 2.5 823 15 FD-5 0.11055 723 50 20 723 15 FD-6 0.11062 823 3 2.5 723 3 FD-7 0.12713 723 3 2.5 723 15 FD-8 0.11699 723 3 20 723 3 FD-9 0.11980 823 50 2.5 823 3 FD-10 0.10931 823 50 20 723 3 FD-11 0.12163 723 50 2.5 723 3 FD-12 0.12836 823 50 2.5 723 15 FD-13 0.13769 723 50 20 823 3 FD-14 0.11120 723 3 2.5 823 3 FD-15 0.12441 723 3 20 823 15 FD-16 0.14676 823 3 20 823 3 a: The catalyst was reduced in situ in 40 % H 2 at 823 K for 2 hours before the start of each experiment b: All the experiments were carried out at atmospheric pressure. The total flow rate passing through sample bed was 200 seem and the purge gas was He at 400 seem. 175 Appendix H : Activation Energy Measurement Activation Energy of C H 4 Decomposition on Co Catalyst Exp ID C H 4 decompose Temperature (°C) Decomposition Rate (mol/[gcat,s]) C H 4 decompose Temperature (K) 1/T (1/K) In(Rate) 120490 425 1.80E-06 698 0.00143 -13.23 120491 475 3.12E-06 748 0.00134 -12.68 120494 525 6.63E-06 798 0.00125 -11.92 Slope R(J/[mol KI) E/R E(KJ) -7232.98 8.314 7232.98 60.13 The activation energy of C H 4 decomposition for Co catalyst is 60.13 (KJ/mole) -11.80 -12.00 -12.20 ^ -12.40 $ -12.60 -12.80 -13.00 -13.20 -13.40 In(Rate) vs 1/T(K) for C H 4 Decomposition V = -7231.98x-2.91 R 2 = 0.98 0.00120 0.00125 0.00130 0.00135 1/T(1/K) 0.00140 0.00145 Figure H-l Activation energy measurement of C H 4 decomposition on Co catalyst 176 Activation Energy of C H 4 Decomposition on Ni Catalyst Exp ID C H 4 decompose Temperature CC) Decomposition Rate (mol/[gcat,s]) C H 4 decompose Temperature (K) 1/T (1/K) In(Rate) 125884 425 2.787E-06 698 0.001433 -12.79 125887 475 5.958E-06 748 0.001337 -12.03 125867 500 7.349E-06 773 0.001294 -11.82 125890 525 1.138E-05 798 0.001253 -11.38 Slope R(J/[mol KI) E/R E(KJ) -7595.35 8.314 7595.35 63.15 The activation energy of C H 4 decomposition for Ni catalyst is 63.15 (KJ/mole) -11.20 -11.40 -11.60 -11.80 .00 3 "1 2-03 CC -12.20 -12.40 -12.60 -12.80 -13.00 In(Rate) vs 1/T(K) for C H 4 Decomposition y =-7595.35x- 1.91 = 0.99 0.001200 0.001250 0.001300 0.001350 1/T(1/K) 0.001400 0.001450 Figure H-2 Activation energy measurement of C H 4 decomposition on Ni catalyst 177 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0058758/manifest

Comment

Related Items