Power-to-Gas concepts integrated with syngas production through gasification of forest residues Process modelling Master’s thesis in Sustainable Energy System Programme Andrea Gambardella & Yahya Yasin Yahya Department of Energy and Environment CHALMERS UNIVERSITY OF TECHNOLOGY Göteborg, Sweden 2017 Master’s thesis 2017 Power-to-Gas concepts integrated with syngas production through gasification of forest residues Process modeling Andrea Gambardella & Yahya Yasin Yahya Department of Energy and Environment Division of Energy Technology Chalmers University of Technology Göteborg, Sweden 2017 Power-to-Gas concepts integrated with syngas production through gasification of forest residues Process modelling Master’s Thesis within the Sustainable Energy System Andrea Gambardella & Yahya Yasin Yahya © Andrea Gambardella & Yahya Yasin Yahya, 2017. Supervisor: Johan Ahlström, Department of Energy and Environment Examiner: Stavros Papadokonstantakis, Department of Energy and Environment Master’s Thesis 2017 Department of Energy and Environment Division of Energy Technology Chalmers University of Technology SE-412 96 Göteborg Telephone +46 31 772 1000 Cover: Gasification of biomass with integrated power-to-gas concept. Chalmers Reproservice Göteborg, Sweden 2017 iv Power-to-Gas concepts integrated with syngas production through gasification of forest residues Process modelling Master’s Thesis within the Sustainable Energy System Andrea Gambardella & Yahya Yasin Yahya Department of Energy and Environment Division of Energy Technology Chalmers University of Technology Abstract Climate change and global warming caused by the emission of CO2 from fossil fuels utilization are one of the most challenging environmental threat mankind is fac- ing nowadays. As motives to reduce CO2 emissions, the attentions to replace the fossil fuels with renewable energy sources such as biomass, sun and wind is high on agenda in the recent decades. However, these renewable resource suffer from some drawbacks: the intermittent nature of the electricity from the sun and the wind destabilize the electric grid and energy from biomass is not easily accessible to be used in the transport sector for example, which is the major CO2 emitter. By a thermo-chemical process called gasification, biomass can be reacted to produce syngas (mixture of CO2 CO, H2, CH4) which can be used to synthesize secondary bio-fuels such as biogas, methanol and Fischer-Tropsch. Through power-to-gas tech- nology the intermittent electricity can be used in the gasification system. Gasifi- cation and Power-to-gas technology together can work in synergy enhancing the production of secondary biofuels, while increasing the integration of intermittent energy sources in the energy system and even stabilizing the electric grid. This project is concerned with the modeling of biomass gasification in a pressurized, oxygen-blown, fluidized bed gasifier and integration of power-to-gas technology in the gasification system. The model is based on experimental data available in a literature and developed by flowsheeting in ASPEN PLUS. The model includes pro- cesses such as biomass drying, biomass gasification, methanation of the syngas and the Sabatier process. Four different layouts of the Sabatier process are developed to investigate and compare the thermodynamic performance (energy and exergy ef- ficiency), economic performance (operational profits) and operational flexibility of layouts when integrating power-to-gas concept in the gasification system. The lay- out upstream of the Sabatier reactor is identical for every scenario and it has a CH4 yield of 0, 24kgCH4/kgdrybiomass. An important aspects of the different layouts is in the CO2 removal unit position and utilization: the CO2 can be fed to the Sabatier reactor either mixed in the gas coming from the methanation unit, or pure. In the former case all the CO2 is injected but only the unreacted has to be separated at the end of the process, whereas in the latter almost all the CO2 is removed, but then only the desired amount is injected into the reactor. Energy and exergy effi- ciencyof the system is in the range of 0,55 - 0,8 and 0,35 - 0,4 respectively, while the operational revenues can peak 0, 22USD/kWhdrybiomass. Concerning the operational v performances (economic, thermodynamic and flexibility) it was noticed that feeding the Sabatier reactor with the entire mix of gases coming from the methanation unit, and separating the unreacted CO2 afterwards was the most advantageous scenario, but it may lead to higher investment cost. Keywords: gasification, biomass, gasifier, product gas, biomethane, power-to-gas, Sabatier reactor, biogas, ASPEN PLUS, methanation, vi Acknowledgements Our sincere gratitude goes to Stavros Papadokonstantakis and Johan Ahlström who patiently guided us throughout the progression of this work. Andrea Gambardella & Yahya Yasin Yahya, Gothenburg, May 2017 vii viii Contents List of Figures xi List of Tables xv 1 Introduction 1 1.1 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.2 Scope of the project . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 Theory 5 2.1 Gasification of biomass . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.1.1 Drying of biomass . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.1.2 Pyrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.1.3 Char gasification and combustion . . . . . . . . . . . . . . . . 8 2.1.4 Oxygen for direct gasification . . . . . . . . . . . . . . . . . . 8 2.2 Gas treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.2.1 Gas purification . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.2.2 Particulate matter . . . . . . . . . . . . . . . . . . . . . . . . 9 2.2.3 Tar destruction . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.2.4 Sulfur contaminants . . . . . . . . . . . . . . . . . . . . . . . 10 2.3 Water-gas-shift for the syngas . . . . . . . . . . . . . . . . . . . . . . 12 2.4 Methanation of the syngas . . . . . . . . . . . . . . . . . . . . . . . . 12 2.5 Gas conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.5.1 Sabatier process . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.5.2 Sabatier reactor . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.5.3 Carbon dioxide removal . . . . . . . . . . . . . . . . . . . . . 15 2.5.4 Hydrogen separation . . . . . . . . . . . . . . . . . . . . . . . 17 2.6 Power-to-gas technology . . . . . . . . . . . . . . . . . . . . . . . . . 18 2.6.1 Electrolyser . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.7 Intermittent electricity in Sweden . . . . . . . . . . . . . . . . . . . . 20 3 Methods 23 3.1 Modeling in ASPEN PLUS . . . . . . . . . . . . . . . . . . . . . . . . 23 3.1.1 Gasification modeling . . . . . . . . . . . . . . . . . . . . . . . 23 3.2 Model development . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 3.2.1 Biomass drying . . . . . . . . . . . . . . . . . . . . . . . . . . 25 3.2.2 Gasification process . . . . . . . . . . . . . . . . . . . . . . . . 26 3.2.3 Syngas cleaning and pre-methanation . . . . . . . . . . . . . . 28 ix Contents 3.2.4 Methanation (Base case) . . . . . . . . . . . . . . . . . . . . . 29 3.2.5 Sabatier process . . . . . . . . . . . . . . . . . . . . . . . . . . 30 3.2.5.1 Scenario 1 . . . . . . . . . . . . . . . . . . . . . . . . 32 3.2.5.2 Scenario 2 . . . . . . . . . . . . . . . . . . . . . . . . 32 3.2.5.3 Scenario 3 . . . . . . . . . . . . . . . . . . . . . . . . 33 3.2.5.4 Scenario 4 . . . . . . . . . . . . . . . . . . . . . . . . 34 3.3 Performance indicators . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.3.1 Thermodynamic performances . . . . . . . . . . . . . . . . . . 35 3.3.2 Economic performance . . . . . . . . . . . . . . . . . . . . . . 36 4 Results and Discussions 37 4.1 Performance of the methanation (base case) . . . . . . . . . . . . . . 37 4.2 Thermodynamic performance of scenarios . . . . . . . . . . . . . . . . 38 4.2.1 Efficiencies with alkaline Electrolyzer . . . . . . . . . . . . . . 38 4.2.2 Efficiencies with PEM electrolyzer . . . . . . . . . . . . . . . . 41 4.3 Economic performance of scenarios . . . . . . . . . . . . . . . . . . . 43 4.3.1 Profit with alkaline electrolyzer . . . . . . . . . . . . . . . . . 45 4.3.2 Profit with PEM electrolyzer . . . . . . . . . . . . . . . . . . . 47 4.4 Sensitivity to performance indicators . . . . . . . . . . . . . . . . . . 48 5 Conclusion 51 5.1 Future Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Bibliography 52 A LHHW model, Standard Exergy and Biomass analysis III B Description of acronyms (for dryer, methanator and gasfier) V C Description of acronyms and their values in the Sabatier ProcessXIII D Calculation of economic performance and description of the acronym used XVII D.1 Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XVII D.2 Revenues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XVIII E Model sensitivity and biogas compositions XXIII x List of Figures 1.1 Direct and indirect gasification reactor [5]. . . . . . . . . . . . . . . . 2 2.1 Path of a gasification process [2]. . . . . . . . . . . . . . . . . . . . . 6 2.2 The gasification sequence and temperature level in a typical moving bed-reactor gasifier [2]. . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.3 The belt dryer as proposed by Alamia [10]. . . . . . . . . . . . . . . . 7 2.4 Configuration of methanation reactor. Acronyms are: Ri is reactors; Cli is cooler; SEP is the flash separator. . . . . . . . . . . . . . . . . . 13 2.5 The absorption process of CO2 with chemical solvent, adapted from [39]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.6 The increase in electricity supply from wind power between 2003 and 2015 [55]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2.7 The installed cumulative solar power from 1992 to 2015 in Sweden [57]. 21 3.1 The ASPEN PLUS model of the belt dryer. The acronym of compo- nents and streams with their descriptions are given in Table B.1 in Appendix B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 3.2 Comparison between experimental and the model-produced gas com- positions for the main gas components of the product gas. The two dotted lines enclose the region of 90% confidence around the linear line of slope 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.3 The ASPEN PLUS model of the biomass (forest residues) gasification. The process is operated at the pressure of 2,5 bar. The acronym of components and streams with its description are given in Table B.5 in Appendix B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.4 The BaseCase ASPEN PLUS model for the methanation of the prod- uct gas. The acronym of components and streams with its description are given in Table B.6 in Appendix B. . . . . . . . . . . . . . . . . . 30 3.5 The temperature profile in the RPLUG reactor of the Sabatier process. 31 3.6 The ASPEN PLUS model for the simulation of the Sabatier process, Scenario 1. The acronyms of components and streams with the re- spective descriptions are given in Table C.1 in Appendix C. . . . . . . 32 3.7 The ASPEN PLUS model for the simulation of Sabatier process, Sce- nario 2. The acronyms of components and streams with the respective descriptions are given in Table C.1 in Appendix C. . . . . . . . . . . 33 xi List of Figures 3.8 The simulation of the Sabatier process in ASPEN PLUS according to scenario 3. The explanation to acronyms of components and streams are given in Table C.3 Appendix C. . . . . . . . . . . . . . . . . . . . 33 3.9 The ASPENPLUS model for the simulation of scenario 4. Descrip- tions to the acronym of streams and components are given in Table C.3 in Appendix C . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.1 The unpinched grand composite curve of the BaseCase. . . . . . . . . 38 4.2 The sytem energy comparison for Scenario 1, 3 and 4 when H2 is supplied by alkaline electrolyzer. . . . . . . . . . . . . . . . . . . . . . 39 4.3 The exergy efficiency comparison for Scenario 1, 3 and 4 when H2 is supplied by alkaline electrolyzer. . . . . . . . . . . . . . . . . . . . . . 39 4.4 The cold gas efficiency comparison for Scenario 1, 3 and 4 when H2 is supplied by alkaline electrolyzer. . . . . . . . . . . . . . . . . . . . 40 4.5 The thermodynamic performance of Scenario 2. . . . . . . . . . . . . 41 4.6 The system energy efficiency for Scenario 1, 3 and 4, when H2 is supplied by PEM electrolyzer. . . . . . . . . . . . . . . . . . . . . . . 41 4.7 The exergy efficiency for Scenario 1, 3 and 4, when H2 is supplied by PEM electrolyzer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 4.8 The cold gas efficiency for Scenario 1, 3 and 4, when H2 is supplied by PEM electrolyzer. . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 4.9 Thermodynamic performance for Scenario 2 when H2 is provided by a PEM electrolyzer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 4.10 Change in size with gas flow for the Sabatier reactor, when gas stream of CO2+CH4 is, without separation of CO2, sent to the Sabatier re- actor (Scenario 1 and 2). The blue and orange graphs show the pro- portional variation of reactor size and biogas production with regards to a given reference, respectively. . . . . . . . . . . . . . . . . . . . . 44 4.11 Change in size with flow for the Sabatier reactor, when CO2 is sep- arated from CO2+CH4 stream and injected to the Sabatier reactor (Scenario 3 and 4). The blue and orange graphs show the propor- tional variation of reactor size and biogas production with regards to a given reference, respectively. . . . . . . . . . . . . . . . . . . . . . . 44 4.12 Operational profits for Scenario 1, 3 and 4 when H2 is supplied by alkaline electrolyzer. . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 4.13 Biogas production for Scenario 1, 3 and 4 when H2 is supplied by alkaline electrolyzer. . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 4.14 Economic performance of Scenario 2 when H2 is supplied by an alka- line electrolyzer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 4.15 Operational profits from Scenario 1, 3 and 4 when H2 is supplied by a PEM elctrolyzer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 4.16 Biogas production from Scenario 1, 3 and 4 when H2 is supplied by a PEM electrolyzer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 4.17 Economic performance of Scenario 2 with PEM electrolyzer. . . . . . 48 xii List of Figures B.1 Comparison between experimental and the model-produced gas com- positions for the main gas components of the raw product gas. The two dotted lines enclose the region of 90% confidence around the lin- ear line of slope 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII E.1 The variation of biogas composition as H2 and CO2 varies in Scenario 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XXIV E.2 The ASPENPLUS model for the simulation of scenario 4. Descrip- tions to the acronym of streams and components are given in Table C.3 in Appendix C . . . . . . . . . . . . . . . . . . . . . . . . . . . . XXIV E.3 The variation of biogas composition as H2 and CO2 varies in Scenario 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XXV E.4 The variation of biogas composition as H2 and CO2 varies in Scenario 4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XXV xiii List of Figures xiv List of Tables 1.1 Properties of Biogas type A and B according to Swedish standard SS 15 54 38 [8]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.1 Comparison of different air separation technologies [16]. . . . . . . . . 9 2.2 Particle separators with their performance and operation temperature [1]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.3 Alkaline and PEM electrolysers’ specification. The economy of the electrolysers applies for 2016 [49][50][51][52]. . . . . . . . . . . . . . . 20 3.1 Some of the components, with their model and descriptions, that can be used in ASPEN PLUS for a gasification system [60]. . . . . . . . . 24 3.2 Feauters of different scenarios. . . . . . . . . . . . . . . . . . . . . . . 34 A.1 Equilibrium and adsorption constants for LHHW model [24]. . . . . . III A.2 Activation energy and kinetic factor values for LHHW model [24]. . . III A.3 Standard chemical exergy and molar mass of substances. (at T= 298.15K, p = 101.325 kPa) [64]. . . . . . . . . . . . . . . . . . . . . . IV A.4 The proximal analysis of the forest residues used used in the experi- ment [9]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV A.5 The ultimate analysis of the forest residues [9]. . . . . . . . . . . . . . IV B.1 Description of acronyms for the components and streams in for the drying flowsheet. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI B.2 Variables with their corresponding values as they are used for the simulation of biomass drying. . . . . . . . . . . . . . . . . . . . . . . VII B.3 Variables with their correponding values as it is used for the simula- tion of biomass gasification. . . . . . . . . . . . . . . . . . . . . . . . VII B.4 The experimental data of the wet gas composition (N2 free) as it is given in Hannula et al. [9] and the modified data that the model will produce, the yield of C3 − C5 in the experiment has been represented by C3H8, C6H6 and C10H8 in the modified data. . . . . . . . . . . . . VIII B.5 Description of acronyms for components and streams of the gasifica- tion model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX B.6 Descriptions for streams and components of the methanation reactor. X B.7 Variables and values for the operation parameters in the methanation process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XI xv List of Tables B.8 Properties of Biogas type A and B according to Swedish standard SS 15 54 38 [8]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XI C.1 The description of stream and component acronym in the simulation of Scenario 1 and 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . XIII C.2 Continuation of Table C.1. . . . . . . . . . . . . . . . . . . . . . . . . XIV C.3 The description of acronyms for streams and components in the sim- ulation of Scenario 3 and 4. . . . . . . . . . . . . . . . . . . . . . . . XV C.4 Variables with their respective values for the simulation of Sabatier process in all the Scenarios. . . . . . . . . . . . . . . . . . . . . . . . XVI D.1 Independent variables and their values to calculate operational costs. "Retrieved" corresponds to the values which are taken from the sim- ulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XIX D.2 Continuation of Table D.1 . . . . . . . . . . . . . . . . . . . . . . . . XX D.3 Descriptions of the dependent variables to calculate the operational costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XXI D.4 Independent variables and their values used in the calculation of ther- modynamic performance indicators. "Retrieved" means the value is gotten from the simulations. . . . . . . . . . . . . . . . . . . . . . . . XXI xvi 1 Introduction One of the most challenging problems that humanity is facing today is climate change, which is manifesting itself in the form of global warming and unusual weather patterns such as floods and storms. The main cause of the climate change is the emission of CO2 (greenhouse gas) from the fossil fuels which is the main energy source of the world. In order to slow down the climate change, it is important to consider the reduction of fossil fuel use and its eventual replacement by renewable and environmentally friendly energy sources. Biomass is one of the energy sources considered to be a viable and alternative to the fossil fuels. By a thermochemical process called gasification, biomass can be reacted into useful product gas. The product gas (the syngas) is a mixture of CO, H2, CH4, CO2, light hydrocarbons (ethane and propane) and heavier hydrocarbons such as tars [1]. The product gas from biomass can be used in the production of useful chemicals and second-generation biofuels such as biomethane, methanol, Fischer-Tropsch liquids and hydrogen. These biofuels can be used both in the transport sector (for example for natural-gas driven vehicles), and in heat and power production plants thereby paving the way for the implementation of renewable energy in the energy system [2][3][4]. The motive to increase the share of renewable energy in the energy system is lead- ing to the rise in production of electricity from energy sources such as wind and sun. However, the electric power generated by these sources destabilize the grid due to their intermittent nature. Therefore, it would be advantageous to develop a mechanism which helps utilize the full potential of intermittent energy sources while maintaining the stability of the electric grid at the same time. Power-to-gas (P2G) technology together with the gasification might be the promising system for that purpose. Indeed, P2G (like water electrolysis) is already considered as an option to stabilize the electric grid by smoothing out load peaks due to unpredicted power generation, furthermore this technology can work in synergy with gasification to enhance biogas production (as it will be shown later on in the report) for an even better energy recovery. There are two types of gasification technologies: autothermal (direct) and allother- mal (indirect) gasification. In the former technology the heat required by the process is only internally generated by the partial combustion of the feedstock, whereas in the latter technology energy is delivered to the process also through the gasifying agent (steam). Furthermore, in the direct gasification, all the reactions occur in the 1 1. Introduction same device, while in the indirect, combustion reactions occur in a separate cham- ber, which communicates with each other both with mass streams (bed material, char, ashes and feedstock to be combusted) and energy streams (heat carried by the thermal inertia of the bed material itself). Figure 1.1 shows the direct and indirect gasification system [5]. Figure 1.1: Direct and indirect gasification reactor [5]. In the indirect gasifier, the flue gases of the combustion (mainly H2O, CO2 and N2) do not mix with the product gas because the oxygen carrier to the gasifier is the oxidized recirculating bed material. However, in the direct gasifier, the flue gases are mixed into the products and combustion occurs only partially (lambda < 1) [3][6]. Be authothermal or allothermal, the most known type of reactors used for the gasification process are moving-bed, fluidized bed(reactor temperature of 800- 1000◦) and entrained flow reactor (reactor temperature over 1000◦) [2]. In a previous study [7], an indirect gasification plant of 100MW thermal power was simulated in ASPEN PLUS. In the simulation, biomass (forest residues) was gasified in order to produce biomethane. It is, therefore, of interest to simulate a direct gasification plant of biomass and analyze the integration of power-to-gas (P2G) technology to the gasification system, and thereby compare the performance of the two gasification technologies. Table 1.1 shows the two types of biogas with the corresponding specifications according to Swedish standard. 2 1. Introduction Table 1.1: Properties of Biogas type A and B according to Swedish standard SS 15 54 38 [8]. Properties Type A Type B Wobbeindex (MJ/m3) 44,7 – 46,4 43,9 – 47,3 CH4 content (vol-%) 97 ± 1 97 ± 2 Water content (mg/Nm3) 32 32 O2 content (max, vol-%) 1,0 1,0 CO2 + N2 + O2 (vol-%) 4,0 5,0 Sulfur contaminants (mg/Nm3) 23 23 NH3 (mg/Nm3) 20 20 1.1 Objectives In this project, the assessment of power-to-gas technology (P2G) integrated with direct gasification plant is carried out through modeling in ASPEN PLUS. The main product of the plant is biomethane, but other valuable products such as steam from process heat, surplus oxygen and electricity generation have been considered for the plant economic and energy performance also. Pure oxygen has been considered as gasification agent, while hydrogen is injected to the Sabatier reactor to enhance the CH4 yield of the process while simultaneously decreasing the CO2 emissions. The main goals of the projects are: • to develop a plant layout, through ASPEN PLUS, which resembles reality both at system and process level 1 • to develop different layouts for power to gas integration. • to assess the economic (operational profits) and energy performances of the plant and its power to gas integration scenarios. The above mentioned goals will be fulfilled by: • creating a framework in Excel where result data from ASPEN PLUS can easily be retrieved. • generating Matlab code which can use the Excel framework to perform heat integration and assess the performance of the model. 1.2 Scope of the project What fall outside the project objectives is the economic evaluation of the plant size, investment costs, logistic etc. Nevertheless, concepts concerning the size variation of the plant units for different scenarios will be provided as a base for future studies. The environmental impact of the plant is also not assessed. The plant performances are assessed under steady state conditions, both energy and material wise, while no fluid or process dynamics have been investigated. Sensitivity analysis have been conducted by varying hydrogen and carbon dioxide feed into the Sabatier reactor simultaneously, all the parameters affected by this streams (such as reactor size or 1The model is based on the experimental data produced by Hannula et al [9] by gasifying forest residues in an oxygen-blown fluidized bed gasifier. 3 1. Introduction catalyst load) are varied accordingly, whereas all the non affected parameters are kept constant. The experimental data acquired from Hannula et al. [9] is produced from the direct gasification forest residues (lignocellulosic), and the produced biogas quality is evaluated on its chemical composition and categorized as type A or B according to the Swedish standard [8] for the biogas. Regarding the environmental condition that could affect the performance, the plant is supposed to be located close to the city of Göteborg in Sweden. 4 2 Theory 2.1 Gasification of biomass Generally, the gasification process of biomass undergoes the following steps: drying, pyrolysis (thermal decomposition), partial combustion, and gasification of volatile products. Figure 2.1 depicts a typical biomass gasification process [2]. The sequence of gasification reactions depend up on the type of reactor used such as moving-bed reactor, fluidized-bed reactor and entrained-flow reactor, see Figure 2.2 for the se- quence of gasification process in a moving-bed reactor [2]. The gasifying agent is a medium used to gasify the feedstock and it can either be air, pure O2, steam or a mixture of them. However, using air as a gasifying agent in au- tothermal gasification results in a product gas rich in N2 (and therefore results in low heating value). Studies have shown that a pressurized oxygen-blown autothermal gasifier outperforms an indirect gasifier economically, thermodynamically and exer- getically [4]. Furthermore, an authothermal system is simpler and easier to operate. Neverthless, the allothermal gasifier can produce a N2-free product gas (because air is not used as gasifying agent) and the system reaches a complete carbon conversion without production of problematic waste products [6]. The quality of a product gas produced is determined by properties such as H2/CO ratio, amount of inerts in the product gas, amount of methane and amount of poisonous agent such as sulfur- and chlorine component. The content of sulfur and chlorine impurities depends on the type of feedstock used in the gasification [6]. The H2 content in the product gas for indirect gasification is higher than that of direct gasification. However, the content of CO is to the contrary [1]. The H2/CO of the product gas is therefore has to be adjusted in order to get the required synthesis ratio, which normally is between 1,5 and 3 for the synthesis of biomethane. The ratio adjustment can be done by water-gas-shift reaction in a separate catalytic reactor [6]. 5 2. Theory Figure 2.1: Path of a gasification process [2]. Figure 2.2: The gasification sequence and temperature level in a typical moving bed-reactor gasifier [2]. 2.1.1 Drying of biomass One major difference between a conventional fuel such as coal and biomass is that the later has significantly higher content of moisture, with 50-60% wet basis (w.b) of moisture in the woody biomass. Fuel drying helps to sustain the gasification and decreases the need of auxiliary fuel. Additionally, drying biomass prior to gasifica- tion is beneficial because it reduces the size of the gasifier and the feeding system required. Usually, the biomass moisture prior to feeding the biomass into the gasifier 6 2. Theory is reduced below 15% w.b [10]. Definition for the heat required in the drying process varies. Usually it is defined to be the heat required to evaporate the moisture of the biomass, while the rising of the moisture’s temperature to the saturation is also accounted for in some other cases [11]. In these definitions, the energy required to extract the water molecules bonded to the cellular structure of the biomass is not considered. The consideration to this energy requirement necessitate the development and usage of an empirical formula or the usage of advanced computer simulation such as Computational Fluid Dynamics (CFD) which helps determine the needed energy. For a typical biomass with Lower Heating Value (LHV) of 19 MJ/kg and moisture content of 50% w.b, the energy required to completely evaporate the moisture and heat the biomass up to the gasifier temperature of 900◦ is around 22% of the LHV. However, if the biomass, for example, is dried to 10% w.b prior to gasification, the energy demand can be reduced to 2,5% [10]. Therefore, it is advantageous from the energy point of view to integrate the drying system with the gasification system so that the waste heat at lower temperature is used. The heat required in the drying process can be internally recovered and is supplied with the drying agents (examples of drying agents are: steam, hot air, or hot flue gases)[10][11]. Heyne et al. [11] investigated the integration of drying system where steam, flue gas and air is used as drying agent with an indirect gasification plant of 100 MWth. It was shown that steam drying performs better, covering around 50% of the energy demand to dry biomass from 50% w.b to 10% w.b [11]. In the drying process, precaution has to be taken as there could be a risk of explosion and fire, specially in the case of air drying, due to the collection of volatile organic compounds [10]. The most common types of dryer used for biomass drying are fluidized bed dryers, rotary dryers, super-heated steam dryers and belt dryers. Alamia et al. [10] proposes the belt dryer for gasification process because of low fire risk, high exploitation of low temperature heat (below 130◦), and thereby offers chances of heat recovery. In a belt dryer, biomass with a bed thickness of 2-30 cm is conveyed over a porous belt whereby the drying medium is blown from the bottom by fans through the belt. Figure 2.3 illustrates a two staged belt dryer where air at 100◦ (or below) is used to dry the biomass from 50% w.b to 10-20% w.b in the first stage, and steam with temperature of 120-150◦ is used in the second stage [10]. Figure 2.3: The belt dryer as proposed by Alamia [10]. 7 2. Theory 2.1.2 Pyrolysis Pyrolysis is a process which occurs also in gasification. In the pyrolysis step the feedstock (biomass) is thermally degraded in absence, or low presence of oxygen to produce various gas components such as H2, CO, CO2, H2O, C2H2, C2H6, C6H6, tar and solid char [2]. During pyrolysis, 70-80% of the biomass weight is devolatalized at the temperature of around 500◦ [12]. Tar is a complex organic product that consists predominantly of different types of aromatic hydrocarbons. The formation and composition of tar during gasification process depends on conditions such as operation temperature and the feedstock [13]. In the temperature range of 400-700◦, biomass gives rise to the so called primary tar which constitutes oxygenated organic molecules such as acetic acid, hydroxypropanone and methanol, and aromatic compounds such as toluene and benzene. For steam gasification at higher temperature (900-1000◦), tar consists aromatic hydrocarbons such as benzene, toluene and naphthalene [14]. 2.1.3 Char gasification and combustion The pyrolysis step is followed by the step where chemical reactions between the volatile matters such as steam, hydrocarbons, carbon dioxide, hydrogen and char gasification take place. The char of biomass contains primarily carbon (85%) and a little inorganic ashes with oxygen and hydrogen. The biomass char is more reactive than the coke (char of coal), owing to alkali metals which work as catalysts. The possible reaction that arises during the gasification of char are described as follows [2] Char + O2 −−→ CO2/CO (2.1) Char + CO2 −−→ CO (2.2) Char + H2O −−→ CH4 + CO (2.3) Char + H2 −−→ CH4 (2.4) Both exothermic and endothermic reactions occur during gasification. Therefore, it is important for the process in order to be autothermal (heat self-sustained) that the energy released by the exothermic reaction is sufficient to satisfy the energy needs of the endothermic ones. Reaction 2.1 is the most exothermic reaction of the gasification process which bestows 394 kJ/mol and 111 kJ/mol of heat energy depending on if CO2 and CO produced respectively [2]. 2.1.4 Oxygen for direct gasification Of the gasifying agents (oxygen, steam and air) oxygen is the best medium in gener- ating a product gas with high heating value (12-28 MJ/Nm3). Moreover, the use of pure oxygen as gasifying medium shrinks the size of reactor and auxiliary systems recommended [2]. Nevertheless, pure oxygen has to be generated, bringing energet- ic/economic penalty on the the system. Commercially, oxygen is produced through 8 2. Theory separating air into its components; mainly nitrogen and oxygen. The most common technologies used for air separation are cryogenic air separation (ASU), membrane and pressure swing adsorption (PSA). In ASU technology, air is cooled down until it liquefies, and the components of the air are then selectively separated based on their boiling points. Even though ASU is an energy intensive process, it is the most widely used technology for the voluminous production of oxygen with high purity [15]. Table 2.1 shows the comparison of energy performance for different types of air separation units [16]. Table 2.1: Comparison of different air separation technologies [16]. Technology O2 purity Capacity (ton/day) Energy demand kWh/ton O2 Driving force Cryogenic 99+ up to 4000 200 Electricity PSA 95+ up to 300 245 [4] Electricity Heat (70-90◦C) Membrane (polymer) 99+ up to 20 — Electricity 2.2 Gas treatment The product gas leaving the gasifier is a mixture of CO, CO2, H2, CH4, H2O, inorganic impurities ( H2S, HCl etc), and organic compounds such as tars, aliphatic hydrocarbons, toluene and benzene (if the gasification process is operated below 1000◦C). The produced gas has to be, therefore, purified of impurities and treated in the downstream process in order to meet the specification of the end-product, biomethane in this case. Additionally, the cleaning and upgrading systems employed for these purposes have to be efficient and economically feasible. [1]. 2.2.1 Gas purification The product gas has to be purified from catalyst poisoning impurities in order to decrease downstream process maintenance costs. The most known impurities in the product gas from direct gasification of biomass are particulate matter, volatile metal oxides (NaCl and KCl), tars and sulfur compounds (H2S and COS). The removal of these impurities in the downstream process takes place according to the presented order [1]. 2.2.2 Particulate matter Particulate matter in the product gas consists of ash particles, unreacted carbon, soot (in the case of direct gasification) and particles of bed materials when fluidized bed gasifier is used for gasification. Technologies used to remove particulate matter from the product gas are cyclones, filters (electrostatic and barrier) and scrubbers. The volatile metal oxides in the product gas is also removed when removing partic- ulate matter with the application of filters and scrubbers. Table 2.2 shows different 9 2. Theory technologies for removing particulate matter with their efficiency and operation tem- perature [1]. Table 2.2: Particle separators with their performance and operation temperature [1]. Technology Temperature(degC) Particle reduction (%) Cyclone 20-900 45-70 Sandbed filter 20-900 80-95 Bag filter 150-750 90-99 Scrubber 20-200 40-65 Electrostatic precipitator 40-50 95-99 2.2.3 Tar destruction The existence of tar in the gas product is undesirable as tar may condense on the surface of heat exchangers and block gas cleaning filter in the downstream process, leading to operational problems. Therefore it is necessary to remove or destroy the tar. Tar of the product gas at 400-900◦C is catalytically disintegrated into CO and H2 with the help of nickel-based catalyst. The main disadvantage of this method is that inorganic impurities in the product gas may impair the performance of the catalyst. Therefore, these impurities have to be cleaned off by particulate separators prior to catalytical destruction of the tar. In a direct gasification with an entrained flow gasifier, tar can be thermally cracked at high temperature (over 1000◦C) [1] [14]. The most promising method of scrubbing tar from the product gas is the so called physical scrubbing, where organic liquids such as biodiesel is used to wash away the tar from the product gas [1]. This method is employed at GoBiGas where rape methyl ester (RME) of 0,03-0,035 MWRME/MWbiomass amount is used to scrub the tar from the product gas. The RME together with the captured tar is recirculated to the combuster to be burned [7]. 2.2.4 Sulfur contaminants Generally, the percentage amount of sulfur in biomass feedstock is low: ranging from 0,3-0,4% (w/w) in herbaceous biomass to <0,1% in woody biomass. Sulfur in the biomass feedstock is converted during gasification to H2S, COS or sulfur oxides(SOx), depending on the type of gasification used. The low concentration of sulfur in the biomass does not entail the removal of sulfur contaminants when the biomass product is used as fuel for example. However, when the product gas of the biomass is used for synthesis, the removal of sulfur contaminants is mandatory because these contaminants jeopardize the performance of catalysts in the down- stream. H2S is the major sulfur contaminants which is harmful and corrosive and has to be removed from the product gas of gasification. Not all the H2S can be removed (due to the efficiency of the removal unit), nevertheless the acceptable H2S 10 2. Theory content in the produced biomethane has to be kept below 1 ppm [17]. There are a number of methods used for the removal of H2S from the product gas. The method is chosen based upon the specification of the intended product, the characteristics of the gas composition and amount of gas to be treated. The method used to remove H2S can be biological or chemical-physical, where chemical and physical solvents are jointly used to remove H2S [1]. Absorption with chemical solvents is favourable when the partial pressure of the acidic gases (H2S, CO2) is low, while the usage of physical solvents perform better at higher partial pressure [1]. The removal of acidic gases from a gas stream is conventionally done at lower temperature (20-25◦C) by gas-liquid absorption process where aqueous solutions of amines such as MEA (monoethanolamine), DEA (diethanolamine) and MDEA (methyldiethanolamine) are used, (this process is similar to the CO2 capture, see Figure 2.5). The advantage of using alkanolamines for the removal of acidic gases is that it can be used in such a way so that the absorption of the desired gas is favoured [18][1]. For exemple, non-aqueous MDEA reacts only with H2S, however, as the water content of MDEA solution increases, the co-absorption of CO2 also increases, leading to the decrease in the selectivity of H2S as it is shown in the experiment done by Hong et al. [19]. According to the experiment done by Zhi et al. [20], the optimal MDEA concentration for the best selectivity and efficiency for removing H2S is 20% (w/w). They have experimentally demonstrated that one can, when using 15% (w/w) aqueous MDEA solution at operation temperature of 40◦C, achieve a 98.7% of H2S removal from a gas stream. In such process, 9,50% of CO2 in the gas stream is co-absorbed [20]. The regeneration energy required when using MDEA as solvent is less than that of MEA, which is 2,4 kg of saturated steam (at temperature of 100-150◦C [21]) per kg of H2S eliminated. The absorption capacity of MDEA is around 0,7-0,8 moleH2S/moleMDEA [22]. MDEA reacts with H2S according to reaction 2.5 [1]. R3−N + H2S←−→ R3NH+ + HS− (2.5) When using MDEA to remove H2S, the absorber and desober function at high pressure (7-70 bar) and atmospheric pressure respectively, and the operation tem- perature of lower than 60◦C is recommended for an adequate absorption. The main benefit of using amines for the scrubbing of H2S, in comparison to CO2 scrubbing, is that the rate of reaction between amines and H2S is faster, the heat transfer coeffi- cient is also higher. This enables the usage of a relatively smaller absorption unit [1]. Generally, using amines in the gas-liquid absorption process is considered energy intensive, and amines are caustic and degradable. Studies are ongoing in order to develop a superior, energy efficient and economically feasible solid sorbent which can remove H2S from a gas stream at higher temperature (>300◦C). Various metal oxides such as ZnO, CaO, Al2O3 and MgO, and compounds such as zeolites are being tested and studied [18]. 11 2. Theory 2.3 Water-gas-shift for the syngas The most important components of syngas are H2 and CO. The composition ra- tio (H2/CO) of these two gases has to be adjusted depending on the types of the chemicals to be synthesized in the downstream process, and for the production of biomethane this ratio is between 1,5 and 3. The H2/ CO ratio of the product gas can be adjusted by the water-gas-shift: CO + H2O←−→ CO2 + H2 ∆H =−41,2 kJ/mol (2.6) The equilibrium of reaction 2.6 in the mixture of the product gas depends on the temperature, not on the pressure (if the pressure is below 70 bar). Without a catalyst the equilibrium of the reaction is instantaneously reached at a temperature above 950◦C. With the presence of catalyst temperature can be lowered, and two temperature ranges are used for the water-gas-shift reaction at industrial level: 300- 510◦C (copper catalysts) and 180-270◦C (aluminium oxide-copper-zinc catalysts) [1]. 2.4 Methanation of the syngas The need for the production of synthetic natural gas (SNG) from biomass is high because of the surging price of natural gas and the will to reduce the reliance on natural gas. At the core of producing SNG is the methanation process where the dry and purified syngas is converted to methane according to reaction 2.7 [23][24]. CO + 3 H2 −−→ CH4 + H2O ∆ H ◦ =−206,2 kJ/mol (2.7) CO2, which is present in the syngas, can also react with H2 according to the Sabatier reaction (reaction 2.8) to form methane. However, in a gas mixture that contain CO, the methanation of CO, reaction 2.7, is more favoured. The (exothermic) heat re- leased during the reaction favours the reverse water-gas-shift (reaction 2.6), thereby increasing CO concentration, and suppresses CO2 methanation [25]. Therefore the methanation of syngas, unlike that of CO2, can be operated at higher temperature. Two types of reactors are used for the methanation process: fixed and fluidized bed [23][24]. Companies such as Kernforschungszentrum Jülich (Germany), Haldor Top- søe (Denmark) [26] and Sasolburg (South Africa) have, since the 1970s, extensively worked on the development of methanation reactors at both demonstration and pilot scales. They have conducted methanation in a seriously connected adiabatic fixed bed reactors. The operation temperature and pressure in reactors are 250-700◦C and up to 30 bar, respectively. The temperature of the gas stream in such system is controlled by inter-cooling or recycling of the product gas, see Figure 2.4 [23]. The research and studies done by these companies culminated in the development of the first and only commercial methanation plant which was commissioned in 1984 in Dakota USA. The plant is operated by Dakota Gasification Company and can produce 4,8 million m3 of SNG per day [27]. 12 2. Theory An other methanation reactor built, as a demonsntartion plant, is that of the Gobi- Gas (Gothenburg Biomass Gasification Project) plant. The constraction of GoBiGas was started in 2011 and completed in november 2013. This plant is the first of its kind where biomass is gasified and converted into biomethane. Currently, the plant has the capacity of 32MWbiomass, and there is a plan to develop the plant into a commercial level with a capacity of 100MWbiomass. Figure 2.4: Configuration of methanation reactor. Acronyms are: Ri is reactors; Cli is cooler; SEP is the flash separator. The usage of fluidized bed reactors for the methanation process is advantageous for large scale operation when the involved reactions of the methanation has to be heterogeneously catalyzed. The turbulence condition in the fluidized bed offer, in comparison to the fixed bed, higher heat transfer coefficient, thereby facilitating a better temperature control, an isothermal operation and requirement of fewer reac- tors [28]. Additionally, the continuous removal and addition of catalysts are possible in the case of fluidized bed. However, care has to be taken when using fluidized bed as catalysts may entrained. In the fixed bed reactor, the problem related with cat- alysts is sintering due to high temperature in the reactor. Thermodynamically, the methanation process is favoured at low temperature and high pressure [23]. Different types of catalysts are used for the methanation of the syngas in fixed bed and fluid bed. The catalysts for the methanation in fixed bed are commercially avail- able, however, there is no catalyst ready to be used for the commercial methanation in a fluidzed bed [29]. The most widely used catalyst for the commercial methana- tion of syngas is Ni based catalyst, modified with addition compounds such as MgO, Al2O3 and SiO2. The modification of Ni catalyst with these oxides has shown to maintain the performance of the catalyst which would otherwise be deteriorated by the formation of carbon due to the dissociation of CO during the methanation [30]. It is reported that the life time of a catalyst could be between 3 and 10 years de- pending on the operating condition and the presence of impurities in the gas stream [31]. 2.5 Gas conditioning After the methanation process of the syngas, the gas stream contains mainly CO2 (formed mainly due to water-gas-shift) and CH4 (biomethane). The produced CO2 can either be injected with H2 to the Sabatier process for further enhancement of biomethane production or separated and ejected out to the atmosphere. With the choice of the former alternative, pure CO2 (after separation) or with other gas can 13 2. Theory be injected to the Sabatier process. The gas stream after the Sabatier reactor may contain unreacted CO2 and H2. These gases have to be removed from the gas stream in order to produce a standardized biomethane which will be ready to be injected to a gas grid to be utalized. 2.5.1 Sabatier process The Sabatier reaction, first found by Paul Sabatier in 1910s, is an exothermic re- action where CO2 reacts with H2 to form CH4 and H2O, according to reaction 2.8. This reaction occurs in the presence of catalysts [32]. CO2 + 4 H2 −−→ CH4 + 2 H2O ∆H =−165 kJ/mol (2.8) The formation of carbon according to reaction 2.9, 2.10 and 2.11 also occurs during the Sabatier process. Carbon formation is dependent on the process temperature and pressure. At atmospheric pressure carbon forms at 365◦C, and with increasing pressure its formation temperature also rises quickly. The formation of carbon dur- ing the process is undesirable because it gets deposited on catalyst and consumes the reactants, thereby decreasing the CH4 yield. Therefore it is recommended the process is operated at higher pressure [33]. CH4 + H2 ←−→ C (s) + 3 H2 ∆ H ◦ = 74,9 kJ/mol (2.9) 2 CO←−→ C (s) + CO2 ∆ H ◦ =−172,4 kJ/mol (2.10) CO2 + 2 H2 ←−→ C (s) + 2 H2O ∆ H ◦ =−90 kJ/mol (2.11) The Sabatier process is thermodynamically bounded, and lower operating temper- ature in the range of 250-400◦C is required for higher conversion of CO2. Various studies have been done to investigate and compare the performance and selectivity of different catalysts, such as Rh, Ru and Ni, for the reaction [32]. Ni and Rh have been proven to be the most suitable and economically competitive catalyst used in the Sabatier process at industrial level [33]. 2.5.2 Sabatier reactor The two types of reactors used for the implementation of the Sabatier process are packed-bed (the traditional one) and the microchannel reactor. The microchannel reactor has an improved performance for the conversion of CO2 as a result of the better mass and heat transfer between reactants and channel walls [34]. One major problem in the Sabatier process is the rise of temperature because of the exothermicity of the reaction. The increase in temperature in the reactor unfavour the formation of the product (CH4 and H2O) due to the deactivation of the cata- lysts at the temperature over 450◦C [34]. Therefore the Sabatier reactor has to be designed so that the temperature in the reactor is controlled. Solar Fuel devised the Sabatier reactor in which at least two reactors are connected in series. In such reactor configuration the temperature of the reactor is controlled by inter-cooling 14 2. Theory and condensation of the water which will be removed partially before the gas stream is reheated and enters into the next reactor. The partial removal of water favours the formation of CH4. Turbo SE and MAN Diesel invented in 2011 an advanced single reactor for the Sabatier process. The reactor has two separate regions with shell-and-tube reactor which is filled by catalyst pellets. Each region of the reactor is cooled solely by cooling medium that flow externally on tubes [24]. The Sabatier process can be described by kinetic models such as power laws and Langmuir-Hinshelwood-Hougen-Watson LHHW. The process could be described by presenting a model for WGS and methanation of CO separately, or by a single model for the reaction 2.8 [24][33]. Schlereth [24] attempted to derive a kinetic model that reflects the Sabatier process (for reaction 2.8) at commercially acceptable conditions. He experimentally investigated the LHHW model for the Sabatier process when Ni- alumina is used as the catalyst of the process. Schlereth assumed the formation of formly functional group (HCO-X), as rate influencing factor, in the derivation of LHHW model. The descriptions of LHHW model, which can be formulated according to equation 2.12, is presented in Appendix A in detail. r = Kineticfactor. Drivingforce Adsorption (2.12) 2.5.3 Carbon dioxide removal There are different technologies to be used concerning the removal of CO2 from a gas stream. These technologies are adsorption (pressure and temperature swings), ab- sorption (using chemical solvents), membrane separation, cryogenic separation and biological fixation [35][36][37]. Of all these technologies chemical solvent absorption is the most reliable, widely studied and economically feasible method in the appli- cation of removing CO2 from a gas stream . The most acceptable and widely used chemical solvent in the CO2 scrubbing is the aqueous solution of monoethanolamine MEA[37][36][38]. In the MEA scrubbing process two fundamental reactions take place between amine and CO2[38]. 2 R−NH2 + CO2 ←−→ R−NH3+ + R−NH−COO− (2.13) R−NH2 + CO2 + H2O←−→ R−NH3+ + HCO3− (2.14) According to reaction 2.13 and 2.14, 2 moles and 1 mole of MEA, respectively, are stochiometrically required to absorb 1 mole of CO2. Which means that the stochiometric absorption capacity of MEA according to reaction 2.13 and 2.14 is 0,36 and 0,72 kgCO2/kgMEA. Studies have shown that the absorption capacity of MEA is dependent on operation temperature, presence of additional gases in the gas stream and the concentration of MEA in the aqueous solution. The presence of gases such as O2, H2S, HF, HCl, SO2 and NO2 in the gas stream degrade the absorbing capacity of MEA because these gases react with MEA to create unwanted 15 2. Theory byproducts and reduce the absorption process’ reaction rate [35][36][38]. Figure 2.5 illustrates the absorption process of CO2 with chemical solvent [39]. Figure 2.5: The absorption process of CO2 with chemical solvent, adapted from [39]. As it is seen on Figure 2.5, the gas stream to be purified is fed into the absorber. The chemical solvent, at relatively low temperature, is streamed from above in the absorber so that it absorbs CO2 from the counter flowing gas stream. The chemical solvent, rich in CO2, is then transported to the desorber through the heat exchanger where it absorbs heat from the regenerated solvent. In the desorber, which consti- tutes the main energy demand of the absorption unit (over 70%), CO2 is released from the solvent through stripping. In order to design an optimal CO2 scrubber experimental data for CO2 absorption process is required [38]. Yeh et al.[36] did an experiment to investigate how the absorption capacity of MEA and the efficiency of CO2 capturing are influenced by MEA concentration (7-35% w/w), CO2 level in the gas stream (8-16% v/v) and operation temperature (10-40◦C). It was shown that the MEA absorption capacity and efficiency varied between 0,36-0,38 kgCO2/kgMEA and 0,42-0,92, respectively, with the variation of MEA concentration while operation temperature and gas con- centration were kept constant at 25◦C and 16%(v/v) respectively. It was observed that the CO2 absorption capacity and efficiency were slightly improved, 0,35-0,40 kgCO2/kgMEA and 0,88-0,94 respectively, with the variation of operation temper- 16 2. Theory ature [36]. In an experiement done by Jose et al. [38] the absorption capacity of MEA is independent of the CO2 concentration in the inlet gas stream, provided that the inlet gas stream is free from the toxic gases (O2, H2S etc) which would have otherwise jeopardised the proper functionality of the MEA. Of the CO2 removal technologies used in the framework of producing SNG, amine- based absorption is superior in giving high CO2 removal and high CH4 recovery (99,96%) [40]. It is reported that the thermal energy requirement when using 30% (w/w) MEA for scrubbing of CO2 lies in the range of 3,2 - 5,2 MJ/kgCO2 [35]. Heyne et al. [41] calculated the energy demand for the regeneration of MEA when remov- ing CO2 from biogas (30-55 vol-% in concentration of CO2) to be 3,3MJ/kgCO2 [41]. This energy is often provided by steam with a temperature in the range of 120-150◦C. The loss of MEA during absorption process lies in the range of 0,3-0,8 kgMEA/ tonCO2 [42]. 2.5.4 Hydrogen separation After the Sabatier process, there could be unreacted H2 in the gas stream. In or- der to produce the biogas with the right Wobbe Index, the unreacted H2 has to be removed. Pressure swing adsorption PSA is the most widely studied and used technology for H2 purification at industrial level [43]. The principle of PSA is based on adsorption which arises when the gas molecules interacting with the neighboring solid (adsorbent) surface are physically bonded to the surface. The forces of at- traction between the adsorbent material and the gas molecules is a function of the nature of the adsorbing material, operating temperature and the partial pressure of the gas component. In the case of the physical adsorption the determinant force is the van der Waals forces between the adsorbent surface and the gas molecules. Volatile gases such as H2 and He are selectively not adsorbed in comparison with gases such as N2 and CO2 [43]. There are two pressure levels in the PSA system. The first is the high pressure level (usually between 10 and 40 bar) where the adsorption of undesired gases occur. The adsorption process continues until the equilibrium loading between the surface of ad- sorbent and contaminants are attained. When the adsorption process is completed, the desorption of impurities from the adsorbent material starts at lower pressure (slightly above atmospheric pressure), thus regenerating the adsorbent material. In such manner the PSA process cyclically swings between two pressure levels [44]. With PSA technology, 60-90% of hydrogen recovery can be achieved from a gas stream whose major constituent is hydrogen (around 50 vol-%) in the inlet stream. In this manner hydrogen purity of 99,99% can be achieved. If the level of the impurities in the gas stream is low, temperature swing adsorption (TSA) is the ad- vantageous way to purify hydrogen. In the TSA process, the system works between two different temperatures [44]. 17 2. Theory The most widely used adsorbents in hydrogen purification, depending on the op- eration temperature and pressure, is silica gels, aluminas, zeolites and activated carbons. The usage of these adsorbents is also a function of the type of impurities needed to be removed. Activated carbon, for example, is a very effective adsorbent in removing H2 from a mixture of CO2 and hydrocarbons, but they are less advan- tageous when it comes to separating H2 from a gas stream of CO and N2 [45]. From the data published by Mivechian et el. [46] in the study of hydrogen separation from off-gas streams (72 and 25 vol-% of hydrogen and methane respectively) using PSA technology at the Tehran refinery plant, the electric energy demand of the PSA system was calculated to be 0,53kWh/kgH2 [46]. 2.6 Power-to-gas technology In the concept of power-to-gas (P2G) technology, electric energy, through different applications, is converted into gas that could be stored and used as fuel. The most common application of P2G technology in the gasification process is to produce H2 and O2 through electrolysis according to reaction 2.15. The produced H2 can be made to react with CO2 of the product gas according to the Sabatier reaction to produce biomethane [47]. 2 H2O −−→ 2 H2 + O2 ∆ H ◦ = 286 kJ/mol (2.15) The other application of P2G technology is the direct heating of a gasifier by resis- tance heater which converts electric energy into heat energy. This would decrease the internal energy demand of the gasifier and leads to the decrease of char combus- tion and increases char gasification instead. The advantage of this application over the electrolysis is that it has a higher efficiency , because more amount of energy is stored in the product gas, and lower investment cost [7]. The electricity used in the P2G technology could come from the intermittent re- newable energy sources such as solar and wind energy whose installed capacity is continuing to increase. The integration of these renewable energy risks the stabiliza- tion of electric grid due to its alternating nature. Therefore a widespread research is being done as to how to fully utilize the potential of these energy sources while maintaining the stability of the grid. P2G technology is a suitable candidate that has got attention in this regard. In water electrolysis, electrical energy is used to turn water into its elemental components H2 and O2, which can be stored/used to produce biomethane that can be injected to the existing gas grid. This, furthermore, increases the attractiveness of P2G technology application as it makes use of the al- ready existing gas network [48]. Additionally, these intermittent energy sources are suitable candidates to work in synergy with direct gasification because the oxygen produced through electrolysis can be used as a gasifying agent. 18 2. Theory 2.6.1 Electrolyser The core of P2G technology is the electrolyser which through electrolysis converts electrical energy and water into H2 and O2. There are three types of electrolysers: alkaline, PEM ( polymer electrolyte membrane ) ans SOEC (solid oxide electrol- yser). The first two electrolyser technologies are currently being used both at pilot and commercial level for water electrolysis. They are called lower temperature elec- trolysers because they operate below 100◦C. On the other hand, SOEC operates at higher temperature and has energy efficiency of 90-95%. However, SOEC technol- ogy is not accessible for the commercial application yet as it is still at the research stage [47]. The efficiency of an electrolysis technology can be described according to equation 2.16 ηefficiency = LHVH2 Eelectricity (2.16) where ηefficiency is the electrolytic efficiency, LHVH2 is the lower heating value of hydrogen and and Eelectricity is the amount of energy used in the electrolysis. Among electrolysis technologies, alkaline is considered to be the most mature and cheap technology which is used at the industrial level. Aqueous potassium hydrox- ide, steel and nickel-plated steel are respectively used as electrolyte, cathode and anode in the akaline electrolyser. The main drawbacks related with the usage of al- kaline electrolyser is that it has limited operational pressure (below 30 bar), limited range of load and environmental problem as a result of the used caustic electrolyte. With having the limited range of load, alkaline electrolyser has been considered to be unfit for the application in the P2G technology due to the varying nature of power output from intermittent renewable resources. However, alkaline manufac- turers recently claimed to have designed the electrolyser that can adapt itself with ranging load of 5-100% of nominal capacity with increased hydrogen production and fast starting time in the range of seconds [47]. PEM electrolyser is an electrolysis technology used at both commercial and pilot level. This technology is developed to overcome high pressure operation (up to 100 bar) and partial load (as low as 5%) which alkaline electrolyser suffers from. In PEM electrolysis, a polymer membrane which is proton conductive material is used as an electrolyte. The polymer lso helps to separate the product gases. Conventionally, platinum alloy (with either ruthenium or iridium) and platinum is used as anode and cathode respectively. The usage of this material increases the investment cost and durability of the PEM electrolyser in comparison with the alkaline. Table 2.3 shows the specification and operation parameters of alkaline and PEM electrolysers [47]. 19 2. Theory Table 2.3: Alkaline and PEM electrolysers’ specification. The economy of the electrolysers applies for 2016 [49][50][51][52]. Specification Alkaline PEM Capacity range (Nm3/hr) 301 – 485 100 - 400+ Production capacity (% of nominal flow rate) 20 – 100% 0 – 100% H2 yield (kg/MWhel) 23,7 18,69 H2 purity 99.9% ± 0.1 99.99% Efficiency (LHV) 40-70% 48-72% Investment cost (e/MW) 1,07 2,55 Operating cost (% of inv.cost) 4 4 Life span (year) 25 20 2.7 Intermittent electricity in Sweden The annual production of electricity in Sweden is normally between 140 and 150 TWh, with hydroelectric power and nuclear power being the dominant electricity producers in the country [53]. However, driven by the energy policy (in the form of subventions, researching and electricity certificate system) the share of renewable energy resources in the total electricity supply of the country is increasing. Wind and solar power, which are intermittent energy sources, are the main sources of electricity production that are increasing significantly. Wind power is a renewable energy source and it exists over the whole country. In the year 2000, the installed wind power in Sweden was 241 MW, covering 0,2% of the total electricity of the country. However, the installed power raised to around 6000 MW in 2015, covering about 16 TWh, 10% of the total electricity supply [54]. The installed power is expected to rise as the governmen pushes toward the Swedish goal of 30 TWh (approximately 20% of the total electricity) energy from wind power by the year 2020. Figure 2.6 shows the exponential increase in the electricity supply from wind power between 2003 and 2015 [55]. Figure 2.6: The increase in electricity supply from wind power between 2003 and 2015 [55]. 20 2. Theory Solar power on the other hand is a renewable energy source which has got consider- able attention in the move to change the energy system of the country to be more renewable. Even though Sweden is located far from the equator, the region on Earth which receives highest solar energy, an installed solar power of 1 kW in Sweden give rises to an annual energy production of 1000 kWh, according to the right-hand rule [56]. This amount of solar power installation require an area of 7 m2. As for 2016 there was around 200 MW of solar power installed in Sweden, producing 0,2% (0,3 TWh) of the total electricity supply, and the political ambition from the Swedish Energy Authority was to increase the energy production from solar power to 7 - 14 TWh [56]. In order to create awareness and thereby increase the private usage of solar power in residential houses and villas, the Swedish government has introduced in 2015 the tax reduction for the private installation of solar power. The reduction meant that the micro-producers receive the payment of 60 cent/kWh for the excess solar energy they feed into the electric grid. Figure 2.7 illustrates the cumulative installed solar power in Sweden between the year 1992 and 2015. The total installed power of each year is divided according to the type of the systems to which the power is connected to [57]. Figure 2.7: The installed cumulative solar power from 1992 to 2015 in Sweden [57]. The amount of electricity produced at every moment has to be equal with the amount of electricity consumed at that moment. Otherwise the frequency of the system cannot be kept at 50 Hz. The increase in the electricity production from wind and sun, in a sense, contributes to a sustainable development. However, the intermittent nature of these sources destabilise the frequency of the electric grid. 21 2. Theory 22 3 Methods 3.1 Modeling in ASPEN PLUS ASPEN PLUS is a computer program that can be used to simulate a gasification process. It helps to model and simulate biological, physical and chemical processes that concern gaseous, liquid and solid streams under defined settings [12]. Using computer program for simulation and modeling of a process is more cost effective than carrying out experiments, even though not always as accurate. Furthermore, computer simulations can also help to answer different questions that cannot be answered by experiments. 3.1.1 Gasification modeling Modeling of a gasification process in fluidized bed is primarily implemented with two modeling ways: dynamic and equilibrium modeling. The dynamic modeling of a process is done by taking the kinetics of reactions and hydrodynamics (reactor’s geometry and design with the residence time of the gas) of the reactor into con- sideration. Fulfilling the detail of all parameters required by the dynamic model could be complex and tedious, but the model gives an acceptable reflection of a real process. In the equilibrium modeling, only thermodynamic concepts of the process is treated, and modeling with such approach is advantageous for exploratory design and process development [2][58]. The simulation and modeling of biomass gasification, using ASPEN PLLUS, in a flu- idized bed have been done by [9],[12],[13],[28],[58] and [59] among others. In these studies, a far-reaching kinetic models of gasification were developed and certified with the experimental data. When modeling the gasification system, it is common that different steps of gasification, drying, pyrolysis, volatile combustion and char gasification are separately modeled. It is common that part of the gasification pro- cess such as pyrolysis and combustion of volatiles are modelled with equilibrium model, assuming the produced species can attain the minimum Gibbs free energy. Different types of approaches are taken when it comes to tar modeling. The common assumptions made for tar modelling is that it is considered either to be inert or repre- sented by heavy cyclic hydrocarbons such as benzene, toluene and naphthalene [12]. For the simulation of gasification process in the ASPEN PLUS program, different types of components are used. Table 3.1 presents types and model of components with descriptions as to why/when they used [60]. 23 3. Methods Table 3.1: Some of the components, with their model and descriptions, that can be used in ASPEN PLUS for a gasification system [60]. Model Description Purpose Reactors RStoic Stochiometric reactor Helps to model stochiometric a reactor with known reactions, kinetics is not important here. RYield Yield reactor Helps model a reactor with known yield distribution, stochiometry and kinetics are not important. REquil Equilibrium reactor Helps to model a reactor with simultaneous chemical and phase equilibrium. RGibbs Equilibrium reactor with minimization of Gibbs free energy Helps to model a reactor with phase and chemical equilibrium with minimized free Gibbs energy. It Calculates phase equilibrium for vapor-liquid-solid systems and solid solutions. RCSTR Continuously stirred reactor Helps to model a reactor which is continuously stirred in one-, two- or three-phase, with known kinetics and stochiometry. RPlug Plug flow reactor Helps to model a reactor which is one-, two- or three-phase plug flow reactors, with known kinetics and stochiometry. Separators Sep Component separator Helps to separates inlet stream into different outlet streams according to specified fraction or flows. Flash2 Two-outlet flash Helps to separates inlet stream into two output streams by using rigorous vapor-liquid-liquid or vapor-liquid equilibrium. Heat Exchangers Heater Heater or cooler It is used as a heater, a cooler or a condenser. It determines phase and thermal conditions of output stream. HeatX Two-stream heat exchanger It helps to exchange heat between two streams. 24 3. Methods 3.2 Model development The process simulation is carried out by flowsheeting in ASPEN PLUS, beginning with the handling of wet biomass to the production of the end product, biomethane. The whole process is divided into different stages: drying, gasification of biomass, syngas cleaning, methanation of the product gas and the Sabatier process where the CO2 of the gas stream after the methanation reactor is made to react with H2 for four different scenarios. Thus, only the Sabatier reactor configuration differ between all the models. In the simulation, the PR − BM (Peng-Robinson equation of state with Boston- Mathias modification) property method was used to calculate the physical property of conventional components, while the DCOALIGT and HCOALGEN (property model for the non-conventional components) are used to calculate the density and enthalpy of non-conventional components (biomass, ash and char). The ultimate and proximal analysis of the biomass (forest residues) used in the model are given in Table A.5 and Table A.4 in Appendix A respectively. 3.2.1 Biomass drying Assumptions: Pressure drop in the dryer is neglected. The drying process is isothermal and in steady state. Figure 3.1 is the representation of the ASPEN PLUS model of the belt dryer as proposed by Alamia et al. [10]. The RStoic reactor, AIRDRYER, represents the air belt of Figure 2.3 where the hot air is blown over the biomass, drying the biomass to a certain level. The partially dried biomass is transported further to be dried by the two consecutive RStoic reactors, DRYER1 and DRYER2, which represent the steam belt dryer. More in detail, in the RStoic blocks, the biomass moisture properties are changed while water is generated accordingly and separated afterwards in the separator blocks. The drying agent (steam) is represented in the subsequent steam cycle where the energy duty is also calculated and where the biomass moisture is first injected and then purged out. The stream DRYB contains biomass with the moisture content of 10.4% which will be transported to the gasifier. Alamia et al. [10] calculated using CFD the required energy to dry 1 kg of biomass, from a given moisture content to the desired moisture content, in the belt dryer they proposed. Based on the presented calculation, the amount of energy required as a form of hot air and steam to dry the biomass from 40% to 10,4% moisture content was determined. 25 3. Methods Figure 3.1: The ASPEN PLUS model of the belt dryer. The acronym of components and streams with their descriptions are given in Table B.1 in Appendix B. 3.2.2 Gasification process Assumptions: Pressure drop in the gasifier is neglected. The gasification process is isothermal and attains steady state. Devolatalization takes place instantaneously. The formed tar is assumed to be disintegrated into CO and H2 by an ideal catalyst. Costs for cleaning the syngas from particulate matter is negligible. As illustrated in the theory section, gasification is a complex process, the modelling of which would require detailed reaction kinetics and well defined operational pa- rameters such as internal local temperature and pressure, feedstock particles size and shape etc. Our modelling approach was aimed to reproduce the exact exper- imental data of an existing direct gasifier running on similar biomass and using identical operational conditions. We set an atom balance and we adjusted the inlet parameters accordingly, being confident that the experimental results would lay in a safe 10% error margin: we imposed the results of our model to be identical to the experimental data but, at the same time, we changed the inlet parameters of air-to- fuel and steam-to-fuel ratio in order to match the atom balance, assuming that, in a real case, the small adjustments in the inlet parameters would not have led to a deviation in the results greater than 10%. Figure 3.2 shows how the simulation was imposed to mimic the experimental results. 26 3. Methods Figure 3.2: Comparison between experimental and the model- produced gas compositions for the main gas components of the product gas. The two dotted lines enclose the region of 90% confidence around the linear line of slope 1. Figure 3.3 depicts the developed gasification process that reproduces the experi- mental data of the wet gas composition from gasification of forest residues as it is presented in Hannula et al. [9], see Table B.4 in Appendix. Figure 3.3: The ASPEN PLUS model of the biomass (forest residues) gasification. The process is operated at the pressure of 2,5 bar. The acronym of components and streams with its description are given in Table B.5 in Appendix B. As it can be seen on Figure 3.3 two RStoic reactors with acronym PREGASIF and GASIFIER are used. PREGASIF is used to decompose the dry biomass into its el- ementary constituents according to reaction 3.1. The coefficient of the products are 27 3. Methods calculated by dividing the molar amount of the component in the biomass with the amount of dry biomass feed. In GASIFIER, the product from PREGASIF together with other inputs (steam, oxygen, biomass moisture and char) reacts to produce the wet gas composition of the experiment as it is presented in Table B.4 (the modified column). The produced gas at the temperature of the gasifier is cooled down after ash and char have been removed by ideal separators. H2S is then separated from the produced gas. In the modeling, 97% of carbon conversion is assumed, and the char (which contains C, H2, and O2 in the w-% of 97, 1 and 2 respectively) is recycled. Table B.3 in Appendix B shows the variables and their value used in the model of Figure 3.3. It was possible to represent the gasification process with one RStoic reactor only; thus letting the whole process be represented by a single reaction where reactants are biomass, steam and oxygen, and the products are the wet gas composition from the experiment. However, the single reactor case simulation did not match the energy released during the decomposition of biomass with LHV of the dry biomass. The reason could be that ASPEN PLUS might not correctly determine the heat of reaction for the non-conventional component, biomass. Therefore, in order to solve this issue, the energy duty required in the reaction turning the biomass into its elemental components was calculated as follows: 1. the complete combustion of the biomass was modeled through 2 RStoic blocks, one to decompose the biomass into its elements and the other to combust these elements. The energy duty of the RStoic block in charge of the oxidation was reliable (since the reactor is operating with all ASPEN PLUS conventional materials), while the heat of the reaction for the biomass degradation into its elements was manually adjusted until the overall energy duty of the process corresponded to the actual LHV of the fed biomass. 2. the same heat of reaction calculated in the first RStoic was then used in the analogous RStoic (PREGASIF reactor) block in the gasification model. BIOMASS −−→ 0,0427108 C + 0,0123291 O2 + 1,5585 e-005 S + 0,000178396 N2 + 0,0302597 H2 + 0,025987 ASH (3.1) 3.2.3 Syngas cleaning and pre-methanation At the end of the gasification process the product gas is still not ready to be processed in the metanation section. Impurities (such as H2S) and ash have to be removed. Char has to be removed and recycled, and the gas has to be compressed to the methanation reactor’s operational pressure which in this model is between 14 and 10 bar. These steps occurs in series according to the following order: 1. ash removal; 2. char removal and recycle; 3. gas cooling (till the temperature required for the H2S scrubbing); 4. H2S scrubbing by using aqueous MDEA solution (15 wt%); 5. liquid phase separation (for a more efficient compression); 6. gas inter-refrigerated compression, liquid phase and condensate compression. 28 3. Methods The liquid phase and the condensate are mostly water and their re-injection inti the methanation process can be dosed and optimized according to the reaction needs (for example in the water-shift reaction). 3.2.4 Methanation (Base case) Assumptions: The process is isothermal and attains steady state with thermody- namic equilibrium while keeping the minimum Gibbs free energy. There is pressure drop in the process. Figure 3.4 depicts the flowsheet where the cleaned syngas (product gas) from the gasifier is dried and pressurized for methanation. As it is seen on the figure, four RGibbs reactors are seriously connected as it has been described in the literature [23][28]. The configuration of reactors, also, resemble that of which Alamia et al. [61] used for the simulation of methanation process of the GoBiGas plant. The first reactor of the methanation process is set so that the the water-gas-shift (WGS) takes place, consuming CO and boosting the H2 content of the gas stream. This is done by implying restricted chemical equilibrium with specified temperature approach of 5◦C while keeping certain species such as N2, S, H2S and C and all hy- drocarbons inert. The remaining RGibbs reactors are set with calculated phase and chemical equilibrium while specifying possible products (all components in the inlet gas are assumed to be present in the product). In these reactors the methanation of CO (reaction 2.7) occurs. RGibbs reactors, by keeping the minimum Gibbs free energy, resembles the methanation process in the fluidized bed which gives higher heat transfer and turbulence that favors the reactions. Since the reactions occurring in this process are exothermic, inter-stage cooling has been performed to restore the process temperature as the driving force of the reaction. The outlet gas tempera- ture after each reactor was monitored so that it lies in the recommended range of 250-700◦C. In the methanation process of Figure 3.4 the pressure drops from 14 bar at the inlet of HEATER2 to 10 bar at the outlet of COOLER6. The pressure drop only occur in the heat exchangers. Heat duty for all the reactors are set to be zero, and activation energy for the reactants at the inlet of of the first reactor was provided by a heater. Table B.7 in Appendix B presents the variables and values used in the simulation of methanation process as it is depicted in Figure 3.4. 29 3. Methods Figure 3.4: The BaseCase ASPEN PLUS model for the methanation of the product gas. The acronym of components and streams with its description are given in Table B.6 in Appendix B. 3.2.5 Sabatier process In the methanation reactor, both the WGS (reation 2.6) and the Sabatier reaction (reaction 2.8) occur. However, the high operation temperature of the methanation reactor (250-700◦C [23]) favor the rate of WGS much more than the Sabatier re- action, thus resulting in the accumulation of CO2 in the gas stream at the end of the reactor. The gas stream after the methanation reactor predominantly contains H2O, CO2 and CH4. In order to produce the standard biomethane according to Table 1.1, the gas product of the methanation reactor has to be dried. The CO2 has to be removed or the dried gas stream has to be further sent to a separate Sabatier reactor so that the CO2 will react with the supplied H2, enhancing the production of biomethane. In this case, the later option has been chosen, and the hydrogen sources for the Sabatier reactor are assumed to be two electrolyzer technologies: alkaline and PEM. Microchannel reactors give better yield for the Sabatier process [32]. Therefore the simulation of the process here is represented by multi-tubular Rplug reactor, which also has been used by Jürgensen et al. [33] for the simulation of the Sabatier process, with specified temperature according to Figure 3.5, see Table C.4 in Appendix C for the setting of the reactor. 30 3. Methods Figure 3.5: The temperature profile in the RPLUG reactor of the Sabatier process. The temperature range, on Figure 3.5, favors the carbon conversion of the Sabatier process [32], and the risk of carbon formation in the reactor with this operation temperature is negligible because the reactor is operating at the elevated pressure of 60 bar [33]. The Sabatier process is modeled kinetically based on LHHW model according to equation 2.12. The values for adsorption constants, equilibrium constants and ki- netic factor given in Table A.1 and A.2 in Appendix A are used to model the process. The reactor size is automatically updated with the flow through an internal code. Moreover, in order to achieve the same rate of reaction for different inlet flow condi- tions, the reactor has been always oversized; nevertheless the volume variation with the flow has been considered so that it is possible to asses the magnitude of the investment cost variation accordingly. Four different layouts, called scenarios, are developed for the simulation of the Sabatier process. Upstream, all the scenarios have the same gasifier and methana- tion reactor (base case). The operational parameters (such as pressure, temperature, catalyst type load and density) are the same for every scenario. In all the scenarios, heat exchangers HEATER3 and COOLER7 are placed respectively before and after the Sabatier reactor. The former ensures that the temperature of the stream enter- ing the RPlug reactor is high enough to trigger the reaction in a reasonable time while the latter cools down the product stream so that the condensate can be then easily removed by the separator H2OREM2. A turbine is placed immediately after the RPLUG reactor to lower the pressure of the outlet gas stream to 10 bar so that the CO2 removal columns that may come in the downstream process will perform at a plausible pressure. In all the scenarios, the produced biomethane is pressurized to 30 bar by COMP3 in order to fulfill the gas grid standards and be subsequently injected into it. 31 3. Methods The difference between the scenarios lies in the way the injection and ejection of CO2 and H2 are handled. The variables used, with their respective values, in the simulation of all the scenarios are presented in Table C.4 in Appendix C. The produced biogas compositions must fulfill the standards of the Wobbe Index in order to be sold and other specifications, see Table B.8 in Appendix B. 3.2.5.1 Scenario 1 In this scenario, a mixture of dried gas (contains mainly CO2 and CH4) from the methanation reactor and H2 from electrolysis are pressurized and then sent to the Sabatier reactor. The unconverted CO2 is removed using two separators (each at 90% efficiency of CO2 recovery), and some amount of CO2 can be recycled to the inlet gas stream increasing the plant operational flexibility, while the rest is ejected to the atmosphere. Biomethane is produced to be pressurized to 30 bar by COMP3 and injected into the gas grid. Figure 3.6: The ASPEN PLUS model for the simulation of the Sabatier process, Scenario 1. The acronyms of components and streams with the respective descriptions are given in Table C.1 in Appendix C. 3.2.5.2 Scenario 2 The dried gas stream of CO2 and CH4 from the methanation reactor together with the stream of H2 is mixed and pressurized to be sent to the reactor. In this scenario, the H2 feed is so that all the CO2 in the inlet gas stream is completely converted. In this way, the CO2 separator is no longer required and the CO2 emission are minimized, however, this will necessitate the removal of the unreacted H2 later in the downstream, since an excess amount of H2 is required to enhance the driving force of the reaction equilibrium towards the products for the complete conversion of CO2. 32 3. Methods Figure 3.7: The ASPEN PLUS model for the simulation of Sabatier process, Scenario 2. The acronyms of components and streams with the respective descriptions are given in Table C.1 in Appendix C. 3.2.5.3 Scenario 3 In this scenario, 99% of CO2 is recovered from the inlet gas stream, then only the decided amount is sent to the Sabatier process, providing operational flexibility to the layout and allowing a better size optimization of the reactor. The CO2 is removed both from the products of the methanation and the Sabatier processes in the same unit after that the two streams have been mixed. Removed the CO2, the gas is compressed according to the grid standards and then is ready to be sold. Furthermore, this layout consists also of a H2 separator located downstream the Sabatier reactor, in order to achieve the maximum operational flexibility. Figure 3.8: The simulation of the Sabatier process in ASPEN PLUS according to scenario 3. The explanation to acronyms of components and streams are given in Table C.3 Appendix C. 33 3. Methods 3.2.5.4 Scenario 4 In this scenario, the dried product stream of the Sabatier process is recycled and mixed with the gas stream from the methanation reactor whereafter CO2 is recovered and injected to the Sabatier process. Scenarios 3 and 4 are identical, but Scenario 4 has no H2 removal. It was interesting to find out how the increased flexibility in terms of being able to remove H2 will affect the economic performance between the scenarios. Figure 3.9: The ASPENPLUS model for the simulation of scenario 4. Descriptions to the acronym of streams and components are given in Table C.3 in Appendix C Table 3.2 summarize the main features of different scenarios and their operational flexibility. Table 3.2: Feauters of different scenarios. Scenario Sabatier reactor’s inlet stream CO2 removal H2 removal Flexibility NO CO2 emission after the methnation 1 CO2 and H2 and other gases YES NO LOW POSSIBLE 2 CO2 and H2 and other gases NO YES NONE YES 3 Only CO2 and H2 YES YES VERY HIGH POSSIBLE 4 Only CO2 and H2 YES NO HIGH POSSIBLE 34 3. Methods 3.3 Performance indicators All the scenarios, except for scenario 2 where CO2 was meant to be fully converted, were simulated while the amount of CO2 and H2 injection to the Sabatier process are independently varied. The H2 injection was varied in the range of 0-24 kg/hr, which is actually a range that exceeds the value needed to the complete conversion of CO2 for the given dry biomass feed of 100 kg/hr. 3.3.1 Thermodynamic performances For the methanation reactor (BaseCase), a thermodynamic performance called methane efficiency ηCH4 is defined according to equation 3.2. The methane efficiency indi- cates the energy in the dry biomass that can be recovered as a form of biogas after gasification. ηCH4 = (ṁ ∗ LHV )CH4 (ṁ ∗ LHV )biomass (3.2) where ṁ and LHV are mass flow and lower heating value respectively, with the in- dices CH4 and biomass standing for methane and biomass. In order to compare the thermodynamic performance of the four scenarios, three different types of efficiency namely cold gas efficiency ηcold, overall system efficiency ηsystem and exergy efficiency ηexergy whose mathematical descriptions are presented in equation 3.3, 3.4 and 3.5 respectively, were defined. ηcold = ∑(ṁp ∗ LHVp)∑(ṁf ∗ LHVf ) (3.3) where ṁ stands for the mass flow of a component and p respective f being the product (CH4 and H2) and the feed (dry biomass and the injected H2). ηsystem = ∑(ṁp ∗ LHVp) + Q̇− + Ė− ṁbiomass ∗ LHVbiomass + Q̇+ + Ė+ (3.4) where Q̇ and Ė are the thermal power (at 400K) and the electrical power, respec- tively. The plus (+) and minus (-) signs stand for the consumption and production respectively. The electrical power considered in the calculation was that of fans, compressors, PSA, electrolyzer and the turbine, while the thermal energy consumed was used for CO2 and H2S scrubbing. ηexergy = ∑ p ṅp ∗ ep + Q̇− exe + Ė−∑ f ṅf ∗ ef + Q̇+ exe + Ė+ [62] (3.5) where ṅ and ė is the molar flow (kmol/s) and the corresponding standard exergy content (MJ/kmol), while index f and p stand for the net feeds and net products respectively. Q̇exe is the exergy of the thermal power at 400K which can be calculated by multiplying Q̇ with τ (defined by equation 3.6). The standard exergy of the component in the feed and product are given in Table A.3 in Appendix A, and the 35 3. Methods molar flow of the of different components considered in the calculation are retrieved from simulations. τ = 1− Ta Ti (3.6) where Ta is the annual average temperature in Göteborg 281,15K [63] and Ti is the temperature at which the heat is transferred (in the case of the steam produced, for example, Ti is 400K). Throughout the gasification model developed, from the feeding of the wet biomass into the dryer to the production of biomethane at the end of the Sabatier process, there are hot and cold streams. Therefore, to know the net heat flow in the process, it was necessary to draw a Grand Composite Curves GCC which shows the net heat flow against the temperature (shifted). 3.3.2 Economic performance For the comparison of different scenarios from economic point of view, operational profits are considered. The operational profits are the subtraction of costs (money spent to buy biomass, electrical energy to operate the plant and to upgrade the biomethane in the methanation and Sabatier reactors) from the revenues (money received from selling biomethane, excess heat, by product oxygen and electricity). Operational parameters such as ash and char handling, bed-material regeneration, energy for biomass, oxygen and steam feeding, and tar scrubbing are not included here. Hydrogen is assumed to be supplied at the necessary pressure. Operational profits (as revenues-costs) are calculated using equations D.1-D.16 in Appendix D. 36 4 Results and Discussions In this chapter the simulation results of the BaseCase and all the scenarios are pre- sented with illustrating graphs. The cases are compared and contrasted according to performance indicators: thermodynamic performances with different types of ef- ficiency (system, cold and exergy), and economic performance (operational profits and biogas production), while varying the injection of H2 and CO2 into the Sabatier process. The presented graphs shows the performance of the scenarios while pro- ducing the biogas according to the Wobbe Index of Table B.8 in Appendix B, thus only economically valuable biogas is considered. 4.1 Performance of the methanation (base case) Using data from the simulation, the methane efficiency ηCH4 and system efficiency ηsystem calculated according to equation 3.2 and 3.4, respectively, are 0,65 and 0,80. This values are not affected by the choice of the downstream plant layout (the Sabatier unit). Moreover, we use these values to compare direct and indirect gasifi- cation: since no hydrogen from water electrolysis is utilized so far these performance indicators are mainly affected by the gasification technology utilized (the indirect gasification plant we used for the comparison was not integrated with a unit for water electrolysis). The methane efficiency in this case, is higher than that of the indirect gasifier simulated by Alamia [7], which is 0,57. The higher methane effi- ciency of the direct gasifier is due to the production of good syngas composition which can be attributed to the use of oxygen as gasifying agent. However, the system energy efficiency calculated, 0,80, is lower than that of Alamia’s which is around 0,9. It is clear that oxygen production penalized the over all system en- ergy of the direct gasifier. Nevertheless, in this project, no use has been found for the excess heat available at a temperature lower than 400K, otherwise, the system energy efficiency will increase if the heat at lower temperature is considered valuable. Given the exothermic nature of this process no external heat is required. Figure 4.1 shows the unpinched grand composite curve (GCC) of the BaseCase. 37 4. Results and Discussions Figure 4.1: The unpinched grand composite curve of the BaseCase. 4.2 Thermodynamic performance of scenarios 4.2.1 Efficiencies with alkaline Electrolyzer Since scenario 2 has no flexibility with respect to the hydrogen feed, it was compared separately. Figure 4.2 depicts the system energy efficiency for Scenario 1, 3 and 4, when H2 is supplied by an alkaline electrolyzer. The intervals reflect the gas pro- duction fulfilling the standards required by the Wobbe Index. First to notice is that, for each scenario, the efficiency changes with both H2 and CO2 injection variation are small (in the order of 1%). Furthermore, the efficiency always decreases with both CO2 and H2 injection, mainly for the following reasons: the Sabatier reaction and the electrolysis process have their own energy efficiency therefore, increasing the reactants flow rate gives more weight to these reaction efficiencies (reaction 2.15 and 2.8) in the overall energy efficiency of the process. Additionally, the injection of superfluous CO2 only causes an energy penalty due to its cyclic separation and circulation. As it can be seen on Figure 4.2, Scenario 1 has the highest system efficiency for all the H2 amount injected. This is mainly because Scenarios 3 and 4 suffer for an energy penalty related to the CO2 separation prior to the Sabatier reactor; in these scenarios all the CO2 is separated before it can be injected, while in Scenario 1 only the unreacted CO2 needs to be removed. Nevertheless, the energy advantages of Scenario 1 are counterbalanced by a heavier investment cost for the reactor which has to be sized according to the entire flow rather than by the only desired reactants flow. 38 4. Results and Discussions Figure 4.2: The sytem energy comparison for Scenario 1, 3 and 4 when H2 is supplied by alkaline electrolyzer. Figure 4.3 shows the exergy efficiency of Scenario 1, 3 and 4. The exergy efficiency resembles the system energy efficiency trends except for the trend related with the variation of H2 feed which is reversed instead. Giving a quality-weight to the dif- ferent forms of energy produced and consumed, more importance should be given to the biogas production, which increases with the H2 feed. A similar trend will be also notable in the revenues, which like the exergy, gives a different weight to the energy streams by different prices. Remember, the reference temperature for the exergy content of the heat flows was taken to be the annual average temperature of Göteborg, which is 8◦C. The slight difference in exergy for Scenario 3 and 4 is due to the hydrogen content in the biogas. Figure 4.3: The exergy efficiency comparison for Scenario 1, 3 and 4 when H2 is supplied by alkaline electrolyzer. Figure 4.4 shows the cold gas efficiency for Scenario 1, 3 and 4. Remember, in 39 4. Results and Discussions Scenario 2, CO2 in the gas stream from the methanation reactor is made to be com- pletely converted by injecting excess H2. As expected, the cold gas efficiency of all the scenarios increases with increasing H2 feed. The small difference (approximately 0,002) in the cold gas efficiency between different scenarios for the given amount of H2 is due to the H2 content remained in the biogas. Scenario 1 has the lowest amount of H2 in the biogas because of the high conversion due to high concentration of CO2 in the gas streams to the Sabatier reactor. The cold gas efficiency of the Scenario 3 increases gradually and then stabilizes, because of the removal of unconverted H2 from the system; hence the cold gas efficiency increases with increasing CO2 until H2 is totally consumed. Figure 4.4: The cold gas efficiency comparison for Scenario 1, 3 and 4 when H2 is supplied by alkaline electrolyzer. Figure 4.5 shows the thermodynamic performance of Scenario 2. In this scenario, the aim was to completely convert CO2, with the assumption that H2 is abundantly available. This reduces the flexibility of the plant (the size of the gasification plant is limited by the H2 availability) which can only operate within a short range of abundant hydrogen but, at the same time, it avoids the installation of a CO2 sepa- ration unit. All the graphs show an insignificant range of efficiency changes for the given range of H2 feed. Note that only the cases where the gas quality boundary condition is met are shown