Modelling future refineries on the path to net-zero CO2 emissions Master’s thesis in Innovative and Sustainable Chemical Engineering REYHANEH YAGHCHI SAGHAKHANEH DEPARTMENT OF SPACE, EARTH AND ENVIRONMENT DIVISION OF ENERGY TECHNOLOGY CHALMERS UNIVERSITY OF TECHNOLOGY Gothenburg, Sweden 2021 www.chalmers.se MASTER’S THESIS 2021 Modelling future refineries on the path to net- zero CO2 emissions REYHANEH YAGHCHI SAGHAKHANEH Department of Space, Earth and Environment Division of Energy Technology CHALMERS UNIVERSITY OF TECHNOLOGY Gothenburg, Sweden 2021 Modelling future refineries on the path to net-zero CO2 emissions REYHANEH YAGHCHI SAGHAKHANEH © REYHANEH YAGHCHI SAGHAKHANEH, 2021. Supervisors: Åsa Eliasson, Space, Earth and Environment, Chalmers Tharun Roshan Kumar, Space, Earth and Environment, Chalmers Examiner: Simon Harvey, Space, Earth and Environment, Chalmers Master’s Thesis 2021 Department of Space, Earth and Environment Division of Energy Technology Chalmers University of Technology SE-412 96 Gothenburg Telephone: + 46 (0)31-772 1000 Printed by Chalmers Digitaltryck Gothenburg, Sweden 2021 I Modelling future refineries on the path to net-zero CO2 emissions Reyhaneh Yaghchi Saghakhaneh Department of Space, Earth and Environment, Chalmers University of Technology Abstract Refineries are major emitters of carbon dioxide, thus major mitigation measures are required for this industrial sector. In this research, a model was developed to evaluate mitigation options for refineries. The model is able to capture the interplay between multiple mitigation options and their total combined effect when applied to a refinery. The mitigation options investigated include carbon capture and storage and bio-based feedstock co-processing. Furthermore, the model is capable of quantifying the effect of recovering excess heat available in the refinery to cover the heat demand of the carbon capture unit. The performance indicators calculated by the model include the CO2 mitigation potential and changes in the refinery’s energy demand resulting from application of selected mitigation options. The model was tested through a case study of the Preemraff Lysekil refinery, with a focus on mitigation options for the hydrogen production unit which accounts for 20.6% of the refinery’s total on-site emissions. The results indicate that implementing carbon capture and storage could potentially mitigate 53.5% of the total emissions of this unit. Furthermore, recovery and use of excess heat could potentially cover the full energy demand of the carbon capture unit, thereby increasing the CO2 mitigation potential by 55.7%. The bio-based feedstock co-processing option considered was hydrotreating of lipid- based feedstocks. The method is able to quantify the amount of on-site biogenic CO2 emissions generated within the upgrading process of the bio-based feedstock. The analysis was conducted for hydrotreating of a mixture of light gas oil and 17 wt% rapeseed oil. Compared to the effects of carbon capture and storage applied to the hydrogen production unit, the mitigation potential of the co-processing was around 2.5 times higher whereas the energy demand increase was shown to be only 9.6%. The interplay between the two mitigation options was analysed based on a number of test points. The best trade-off was identified as a low share of applying carbon capture (62.2%) coupled with co-processing (17 wt% of rapeseed oil). Additional analysis was conducted to evaluate the effects of capturing the on-site biogenic emissions. It was revealed that the rate of increase of the energy demand is notably higher than that of the CO2 emissions mitigation potential. This could be moderated if excess heat covers the energy demand of the carbon capture unit. Keywords: Modelling, MATLAB, CO2 emissions, Mitigation potential, Carbon capture, Bio-based feedstock co-processing, Petroleum refineries II III Acknowledgements This thesis was achieved by the help and support of people who accompanied me from the outset of this path. I would like to express my deep gratitude towards Simon Harvey, my examiner, for his scholarly guidance throughout the project, and meticulous scrutiny of the report to make it smoother and enriched. I gratefully thank Henrik Thunman for his supportive conduct. Also, it is my genuine pleasure to highly appreciate Karin Lundqvist from Preem for her kind support and provision of the useful data required for the project. Moreover, my sincere thanks go to Åsa Eliasson and Tharun Roshan Kumar, my supervisors, for their great assistance in all the steps taken in this work. Reyhaneh Yaghchi Saghakhaneh, June 2021 IV V TABLE OF CONTENTS NOMENCLATURE VIII LIST OF FIGURES XIV LIST OF TABLES XV 1. INTRODUCTION 1 1.1. Background 1 1.2. Aim and scope 2 1.3. Limitations 2 1.4. Research questions to be addressed 3 2. THEORY 5 2.1. Preemraff Lysekil refinery 5 2.2. Hydrogen Production Unit 6 2.3. Carbon Capture and Storage 7 2.4. Co-processing of bio-based feedstocks in existing refineries 8 2.4.1. Application 8 2.4.2. Classification of bio-based feedstocks 8 2.5. Potential insertion points of bio-based feedstock in oil refineries 9 2.5.1. Possible insertion points 9 2.5.2. Potential process units 10 3. METHODOLOGY 12 3.1. System boundaries 12 3.2. Modelling 13 3.3. Method for quantifying CO2 emissions 15 3.4. Method for quantifying the CO2 mitigation potential of implementing carbon capture and storage (CCS) technology 19 3.5. Method for quantifying the CO2 mitigation potential of co-processing bio-based feedstock 20 3.5.1. Composition 20 3.5.2. Reaction pathways 20 VI 3.6. Overview of analyses that can be conducted by the model 28 3.7. Input data to the model 30 4. RESULTS AND DISCUSSION 32 4.1. CO2 emissions of the hydrogen production unit (HPU) 32 4.2. CO2 mitigation potential and energy consumption of CC technology for the HPU 32 4.3. CO2 mitigation potential and energy consumption of CC technology for the refinery 33 4.4. CO2 mitigation potential of co-processing rapeseed oil in the hydrotreating unit 34 4.5. Interplay between CCS and co-processing of rapeseed oil 34 4.6. Analysis of the interplay between CCS and co-processing rapeseed oil including biogenic CO2 emissions 36 5. SUMMARY AND CONCLUSIONS 40 6. FUTURE WORK 43 7. REFERENCES 45 APPENDIX A: THE RAPESEED OIL HYDROTREATING PRODUCT COMPOSITION 50 VII VIII NOMENCLATURE Abbreviations ATR Autothermal reforming CC Carbon Capture CCS Carbon Capture and Storage CFCs Chlorofluorocarbons CHP Combined Heat and Power EU European Union FCC Fluid Catalytic Cracking FFA Free Fatty Acids FT Fischer-Tropsch liquids GHGs Greenhouse Gases HDC Catalytic hydrocracking HDT Catalytic hydrotreating HHV Higher Heating Value HPU Hydrogen Production Unit HTL Hydrothermal Liquefaction HTLO Hydrothermal Liquefaction Oil IPCC Intergovernmental Panel on Climate Change LGO Light Gas Oil LHV Lower Heating Value LLGHGs Long-Lived Greenhouse Gases LPG Liquefied Petroleum Gas MEA Monoethanolamine MHC Mild Hydro Cracker PSA Pressure-Swing Adsorption SMR Steam Methane Reforming Synsat Synergetic saturation IX Symbols 𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝒎𝒎𝑯𝑯𝑯𝑯𝑯𝑯 Additional on-site emissions to the HPU due to hydrotreating of the bio-based feedstock (106 t/year) 𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑯𝑯𝑯𝑯𝑯𝑯𝑯𝑯 Hydrogen consumption to be supplied by steam reforming of fossil-based methane (106 mole/year) 𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝑩𝒎𝒎𝑯𝑯𝑯𝑯𝑯𝑯 Biogenic CO2 emissions formed within the processes involved in upgrading the bio-based feedstock in the refinery (106 t/year) 𝑪𝑪𝑭𝑭𝑭𝑭 Number of carbon atoms in fatty acid FA 𝑪𝑪𝑪𝑪𝑪𝑪 CO2 mitigation potential of CC technology (106 t/year) 𝑪𝑪𝑪𝑪𝑪𝑪𝒃𝒃𝒃𝒃𝒃𝒃−𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇 CO2 mitigation potential related to the bio-feedstock co- processing (106 t/year) 𝑪𝑪𝑪𝑪 Captured CO2 emissions using CC technology (106 t/year) 𝑪𝑪𝑪𝑪 Captured rate (%) 𝑫𝑫𝑩𝑩𝑭𝑭𝑭𝑭 Number of double bounds available in 1 mole of the fatty acid FA 𝑬𝑬𝑬𝑬𝒖𝒖 Electricity consumption of each process unit u (GWh/year) 𝑬𝑬𝑬𝑬𝒃𝒃𝒃𝒃 Emission factor related to by-product bp (tCO2/GJ) 𝑬𝑬𝑭𝑭𝒇𝒇 Emission factor related to fossil fuel (f) (tCO2/GJ) 𝑬𝑬𝑭𝑭𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴−𝒖𝒖𝒖𝒖 𝒇𝒇 Emission factor of make-up fossil fuel (f) (tCO2/GJ) 𝑬𝑬𝒎𝒎𝑹𝑹𝑬𝑬𝒃𝒃𝒃𝒃𝒃𝒃−𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇 CO2 emissions from the resource extraction phase of bio- feedstock (106 t/year) 𝑬𝑬𝒎𝒎𝑹𝑹𝑬𝑬𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴−𝒖𝒖𝒖𝒖 𝒇𝒇 CO2 emissions from the resource extraction phase of make-up fossil fuel (f) (106 t/year) 𝑬𝑬𝒎𝒎𝑹𝑹𝑬𝑬𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹 𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇 𝒔𝒔𝒔𝒔𝒔𝒔𝒔𝒔𝒔𝒔 CO2 emissions from the resource extraction phase of the replaced fossil-based share (106 t/year) 𝑬𝑬𝒎𝒎𝒖𝒖 CO2 emissions associated with each process unit u (106 t/year) 𝑬𝑬𝒎𝒎𝒖𝒖,𝒐𝒐𝒐𝒐−𝒔𝒔𝒔𝒔𝒔𝒔𝒔𝒔 On-site CO2 emissions from each process unit u (106 t/year) 𝑬𝑬𝒎𝒎𝑼𝑼𝑷𝑷𝒃𝒃𝒃𝒃𝒃𝒃−𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇 CO2 emissions from the user phase of bio-feedstock (106 t/year) 𝑬𝑬𝑬𝑬𝑪𝑪𝑪𝑪𝑪𝑪 Addition to the energy demand of the refinery of focus due to the energy consumption of CCS (GJ/year) 𝑬𝑬𝒏𝒏𝑯𝑯𝑯𝑯𝑼𝑼𝒃𝒃𝒃𝒃𝒃𝒃−𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇 Energy demand of HPU related to the bio-based feedstock (106 GJ/year) 𝑭𝑭𝑭𝑭𝒖𝒖,𝒇𝒇 Consumption of fuel f in process unit u (106 GJ/year) X 𝑭𝑭𝑭𝑭 Fuel penalty that is the energy consumption of the applied CC technology (GJ/t CO2) 𝑯𝑯𝑭𝑭𝑭𝑭 Number of hydrogen atoms in fatty acid FA 𝒊𝒊 − 𝟏𝟏 Number of carbon atoms for Ci-1 n-alkane in the liquid product 𝒊𝒊 Number of carbon atoms for Ci n-alkane in the liquid product 𝑳𝑳𝑳𝑳𝑽𝑽𝒃𝒃𝒃𝒃 Lower Heating Value of by-product bp (GJ/Sm3) 𝑳𝑳𝑳𝑳𝑽𝑽𝑴𝑴𝑴𝑴 Lower Heating Value of fossil-based methane (GJ/mole) 𝒎𝒎𝒃𝒃𝒃𝒃𝒃𝒃−𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇 Mass flow of bio-based feedstock (106 t/year) 𝑴𝑴𝒃𝒃𝒃𝒃𝒃𝒃−𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇 Molar mass of lipid-based feedstock (t/mole) 𝑴𝑴𝒄𝒄 Molar mass of carbon (t/mole) 𝑴𝑴𝑪𝑪𝒊𝒊 Molar mass for Ci n-alkane in the liquid product (t/mole) 𝑴𝑴𝑪𝑪𝒊𝒊−𝟏𝟏 Molar mass of Ci-1 n-alkane in the liquid product (t/mole) 𝑴𝑴𝑪𝑪𝑪𝑪 Molar mass of CO (carbon monoxide) (t/mole) 𝑴𝑴𝑪𝑪𝑪𝑪𝟐𝟐 Molar mass of CO2 (carbon dioxide) (t/mole) 𝒎𝒎𝑮𝑮𝑮𝑮 Mass flow of gaseous product of hydrotreating the bio-based feedstock (106 t/year) 𝑴𝑴𝑯𝑯 Molar mass of hydrogen (t/mole) 𝒎𝒎𝑯𝑯𝑯𝑯𝑯𝑯𝑯𝑯 Total hydrogen consumption due to hydrotreating of the lipid- based feedstock (106 t/year) 𝒎𝒎𝑳𝑳𝑳𝑳 Mass flow of liquid product of hydrotreating the bio-based feedstock (106 t/year) 𝑴𝑴𝑴𝑴𝑴𝑴 Molar mass of methane (t/mole) 𝒎𝒎𝑴𝑴𝑴𝑴 𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴 Mass flow for methane formed by methanation (106 t/year) 𝑴𝑴𝑶𝑶 Molar mass of oxygen (t/mole) 𝒎𝒎𝒕𝒕𝒕𝒕𝒕𝒕 Mass flow of total liquid input to the process unit (106 t/year) 𝒎𝒎𝒘𝒘𝒘𝒘𝒘𝒘𝒘𝒘𝒘𝒘 Mass flow of water produced through hydrotreating the bio- feedstock (106 t/year) 𝒎𝒎𝒎𝒎𝒎𝒎𝑯𝑯𝑫𝑫𝑫𝑫 Hydrogen consumption for hydrogenation of double bounds in the lipid-based feedstock (106 mole/year) 𝒎𝒎𝒎𝒎𝒎𝒎𝑯𝑯𝑫𝑫𝑫𝑫 Hydrogen consumption for decomposition of triglycerides in the lipid-based feedstock into the corresponding fatty acids (106 mole/year) XI 𝒎𝒎𝒎𝒎𝒎𝒎𝑯𝑯𝑫𝑫𝑫𝑫𝑫𝑫 Hydrogen consumption of decarbonylation (106 mole/year) 𝒎𝒎𝒎𝒎𝒎𝒎𝑯𝑯𝑯𝑯𝑯𝑯𝑯𝑯 Hydrogen consumption resulted from hydrodeoxygenation (106 mole/year) 𝒎𝒎𝒎𝒎𝒎𝒎𝑯𝑯𝑯𝑯𝑯𝑯𝑯𝑯 Total hydrogen consumption due to hydrotreating of the lipid- based feedstock (106 mole/year) 𝒎𝒎𝒎𝒎𝒍𝒍𝑯𝑯𝑴𝑴𝑴𝑴 Hydrogen consumption to be supplied by steam reforming of methane (106 mole/year) 𝒎𝒎𝒎𝒎𝒎𝒎𝑯𝑯𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴 Hydrogen consumption for methanation (106 mole/year) 𝒎𝒎𝒎𝒎𝒍𝒍𝑯𝑯𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷 Hydrogen consumption to be supplied by steam reforming of propane (106 mole/year) 𝒏𝒏𝑪𝑪𝑪𝑪 Molar flow of CO produced by decarbonylation of the lipid- based feedstock (106 mole/year) 𝒏𝒏𝑪𝑪𝑪𝑪 𝑮𝑮𝑮𝑮 Molar flow of CO in the gaseous product (106 mole/year) 𝒏𝒏𝑪𝑪𝑪𝑪𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴 Molar flow of methanized CO (106 mole/year) 𝒏𝒏𝑪𝑪𝑶𝑶𝟐𝟐 Molar flow of CO2 produced by deocarboxylation of the lipid- based feedstock (106 mole/year) 𝒏𝒏𝑪𝑪𝑶𝑶𝟐𝟐 𝑮𝑮𝑮𝑮 Molar flow of CO2 in the gaseous product (106 mole/year) 𝑵𝑵𝑫𝑫𝑫𝑫 Number of double bounds available in 1 mole of the lipid-based feedstock 𝒏𝒏𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴𝑴 Molar flow of methane in the gaseous product (106 mole/year) 𝒏𝒏𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷𝑷 Molar flow for propane in the gaseous product (106 mole/year) 𝑶𝑶𝑬𝑬𝑬𝑬 Overall efficiency inclusive of generation, transmission and distribution of electricity supplied from each source of power, E (%) 𝑶𝑶𝑬𝑬𝒈𝒈𝐫𝐫𝐫𝐫𝐫𝐫 Grid overall efficiency inclusive of generation, transmission and distribution of grid electricity (%) 𝑺𝑺𝑺𝑺𝑫𝑫𝑫𝑫𝑫𝑫 Share of Ci-1 n-alkanes that are received through decarbonylation (%) 𝑺𝑺𝑺𝑺𝑬𝑬 Share of electricity supplied from different sources of power E (%) 𝑺𝑺𝑺𝑺𝒈𝒈𝒈𝒈𝒈𝒈𝒈𝒈 Share of electricity supplied from the grid (%) 𝑻𝑻𝑻𝑻 Total energy consumption (106 GJ/year) 𝑻𝑻𝑻𝑻𝑻𝑻 CO2 emissions associated with all the process units (106 t/year) XII 𝑻𝑻𝑻𝑻𝒎𝒎𝒃𝒃𝒃𝒃𝒃𝒃−𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇 CO2 emissions from the value chain regarding bio-feedstock (106 t/year) 𝑻𝑻𝑻𝑻𝒎𝒎𝒐𝒐𝒐𝒐−𝒔𝒔𝒔𝒔𝒔𝒔𝒔𝒔 Total on-site CO2 emissions from all process units (106 t/year) 𝑽𝑽𝒖𝒖,𝒃𝒃𝒃𝒃 Volume of by-product of process unit u (106 GJ/year) 𝑽𝑽𝒖𝒖,𝒊𝒊 Input(s) i to unit process u (106 Sm3/year) 𝑽𝑽𝒖𝒖,𝒑𝒑 Outputs/product(s) p of process unit u (106 Sm3/year) 𝒘𝒘𝒃𝒃𝒃𝒃𝒃𝒃−𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇𝒇 Weight fraction the bio-based feedstock in the liquid input to the process unit (%) 𝒘𝒘𝑪𝑪𝒊𝒊−𝟏𝟏 𝑳𝑳𝑳𝑳 Weight fraction of n-alkane with Ci-1 carbons in the liquid product (%) 𝒘𝒘𝑪𝑪𝒊𝒊 𝑳𝑳𝑳𝑳 Weight fraction for n-alkane with Ci carbons in the liquid product (%) 𝒘𝒘𝑪𝑪𝑪𝑪 Weight fraction of CO in the gaseous product (%) 𝒘𝒘𝑪𝑪𝑶𝑶𝟐𝟐 Weight fraction of CO2 in the gaseous product (%) 𝒘𝒘𝑴𝑴𝑴𝑴 Weight fraction related to methane in the gaseous product (%) 𝒙𝒙𝑭𝑭𝑭𝑭 Molar fraction of fatty acid FA (%) 𝒀𝒀𝑳𝑳𝑳𝑳 Yield related to the liquid product of hydrotreating the bio- based feedstock (%) 𝒀𝒀𝒖𝒖,𝒊𝒊,𝒑𝒑 Volume yields corresponding to the production of product p in the process unit u with the input i (%) 𝒀𝒀𝒘𝒘𝒘𝒘𝒘𝒘𝒘𝒘𝒘𝒘 Yield related to water produced through hydrotreating the bio- feedstock (%) 𝜟𝜟𝑯𝑯𝑷𝑷𝑷𝑷 Enthalpy of reaction for propane reforming (GJ/mole) ∆HR Enthalpy of reaction 𝜟𝜟𝑯𝑯𝑺𝑺𝑺𝑺𝑺𝑺 Enthalpy of reaction for SMR (GJ/mole) Subscripts bp By-product f Fossil fuel XIII XIV LIST OF FIGURES Figure 2-1: Overview of the main process flows at the Preemraff Lysekil refinery ................. 5 Figure 2-2: Block flow diagram of an HPU based on steam reforming .................................... 6 Figure 2-3: Potential insertion points of bio-based feedstock for co-processing in existing refineries .................................................................................................................................... 9 Figure 3-1: System boundary defined in this study ................................................................. 12 Figure 3-2: Architecture of the modelling. .............................................................................. 14 Figure 3-3: Overview of information flow in the model. ........................................................ 14 Figure 3-4: Bottom-up model construction based on material, energy and CO2 balances. ..... 18 Figure 3-5: Simplified schematic block flow diagram of a hydrotreating unit ....................... 22 Figure 4-1: The interplay between the mitigation options compared with the reference case 36 Figure 4-2: The interplay between the mitigation options compared with the reference case including the biogenic CO2 emissions ..................................................................................... 37 XV LIST OF TABLES Table 1: Definition of the terms in Equation 3.1 ..................................................................... 15 Table 2: Definition of the terms in Equation 3.2 ..................................................................... 16 Table 3: Definition of terms in Equation 3.3, Equation 3.4 and Equation 3.5 ........................ 17 Table 4: Definition of terms in Equation 3.6 and Equation 3.7 ............................................... 18 Table 5: Definition of terms in the Equation 3.8, Equation 3.9, and Equation 3.10 ............... 19 Table 6: Composition of rapeseed oil (wt%) in terms of fatty acids ....................................... 20 Table 7: The main fatty acids and corresponding triglycerides constituting rapeseed oil ....... 21 Table 8: The reaction pathways of the main model compound constituting rapeseed oil ....... 21 Table 9: Definition of terms in Equation 3.17 to Equation Equation 3.22 .............................. 23 Table 10: Definition of terms in Equation 3.23 to Equation 3.35 .......................................... 25 Table 11: Definition of terms in Equation 3.38 to Equation 3.41 ........................................... 26 Table 12: Definition of terms in Equation 3.42 to Equation 3.46 ........................................... 27 Table 13: Input parameter values to the model ....................................................................... 30 Table 14: Comparison between CO2 emissions from HPU estimated by the model and the real site data .................................................................................................................................... 32 Table 15: CO2 mitigation potential and energy requirement of CC technology, applied to the HPU, with and without the effect of using excess heat ........................................................... 33 Table 16: CO2 mitigation potential and energy requirement of CC technology, applied to the refinery, estimated by the model and the effect of using excess heat on them ........................ 34 Table 17: Model outcomes for co-processing of rapeseed oil and LGO in the hydrotreating unit ........................................................................................................................................... 34 Table 18: Test points for analyzing the interplay between the mitigation options .................. 35 Table 19: Factorial design of the test points for analyzing the interplay between the mitigation options ..................................................................................................................................... 35 Table 20: The results considering interplay between CCS and rapeseed oil co-processing .... 35 Table 21: The results of the interplay between CCS and the rapeseed oil co-processing including the biogenic CO2 emissions ..................................................................................... 37 Table 22: Comparison between the interplay results for the two analyses of with and without biogenic emissions .................................................................................................................. 38 Table 23: Results of the data collection for the HPU of the Preemraff Lysekil refinery . Error! Bookmark not defined. Table 24: Excess heat sources available at Preemraff Lysekil that can be used to supply CC .................................................................................................. Error! Bookmark not defined. Table 25: The product composition of the hydrotreating of rapeseed oil ................................ 50 XVI 1 1. INTRODUCTION 1.1. BACKGROUND Since the trends in the climate change pose an urgent and potentially irreversible threat to man-kind and the planet, the Intergovernmental Panel on Climate Change (IPCC) strongly recommends that the global temperature rise should be limited to 1.5°C above pre-industrial levels to mitigate the potential impacts and associated risks of global warming as a result of human activities [1]. Accordingly, the European Union (EU) has adopted ambitious targets for transition to a sustainable and climate-neutral economy by 2050 [2]. Furthermore, the Swedish climate policy framework has adopted a more stringent approach in this regard, including a number of new climate goals one of which is that Sweden is to have zero net emissions of greenhouse gases (GHGs) to the atmosphere by 2045 followed by negative emissions thereafter. These targets entail that Sweden will contribute to reducing the concentration of GHGs in the atmosphere after 2045 [3]. Among greenhouse gases, carbon dioxide (CO2) contributes the most to global warming [4]. This is due to the fact that it is one of the long-lived greenhouse gases (LLGHGs), with a relatively high and steadily rising concentration in the atmosphere, although other LLGHGs such as methane and chlorofluorocarbons (CFCs) have much higher global warming potential [4],[5]. Regarding anthropogenic CO2 emissions, industry and petroleum refineries contribute the most [6]. In Sweden, the largest liquid fuel producer is Preem that accounts for 80% of the Swedish refinery capacity. Preem has adopted the strategy to have net zero carbon dioxide emissions from their refineries by 2040 and the same to be applied to its entire value chain by 2045 [7]. In this regard, Preem’s plans and investments are designed to address the CO2 emissions throughout the entire value chain (Well-To-Wheel). Scope 1 includes the direct CO2 emissions from the refinery sites. Scope 2 is related to the indirect CO2 emissions from purchased electricity and heating and cooling. Scope 3 encompasses indirect CO2 emissions from resource extraction, transport, filling stations, business travel, as well as emissions from the end- use of refinery products [7]. Preemraff Lysekil is one of Preem’s refineries located in Lysekil municipality [8]. In this thesis, the mitigation of CO2 emissions is evaluated and modelled with the focus on the Preemraff Lysekil refinery as a case study. Among the candidate measures to significantly reduce emissions, Carbon Capture and Storage (CCS) is a promising technology [9]. There are three industrial scale options for CO2 capture: post- combustion, pre-combustion and oxyfuel combustion [10]. Among those, post- combustion is a particularly appropriate option for retrofitting since there is no requirement for reconstructing the available facilities [5], [11]. Moreover, transition from fossil-based to renewable feedstock and use of fossil-free hydrogen for production of biofuels [7] as well as energy system integration and process intensification [9] are other possible technical measures for achieving large emissions reduction. In the near future crude oil will still be dominantly used for energy purposes especially in the 2 transport sector, which in turn accounts for 49% of the world’s oil demand [12]. In this regard, co-processing of bio-based feedstocks will enable continued use of existing refinery infrastructures and avoiding the need of constructing new plants while reducing emissions [13]. 1.2. AIM AND SCOPE This thesis aims to develop a methodology to assess proposed measures for mitigation of CO2 emissions along the refinery value chain, including Scope 1, 2, and part of Scope 3 emissions. The method focuses on analyzing the changes in CO2 emissions and energy consumption associated with the deployment of Carbon Capture (CC) technology and/or the introduction of bio-based feedstock in the refinery. The objective is to develop the methodology and model such that it can be used to evaluate CO2 mitigation options in complex refineries. The work was conducted using Preemraff Lysekil refinery as a case study. The results constitute a basis for decision makers to evaluate potential measures that can be implemented to reach the goal of zero fossil carbon emissions by 2045. The model is generic and easily adaptable for future changes regarding factors such as capacities, improved rate of CC, and reduced energy usage due to increased energy efficiencies and/or process integration. The model provides a platform for the user to decide on the choice of mitigation option(s) to be applied. The model is constructed so that it can capture the interplay of mitigation options instead of investigating just one option in isolation, thereby failing to capture the effect of the combination of different options. The model is able to estimate the effect of different mitigation options, in terms of CO2 mitigation potential and changes in the energy requirement of the focused refinery. The overall purpose of the model is to reduce time, effort, and costs when evaluating the effects of mitigation options and deciding on the potential measures to be taken. 1.3. LIMITATIONS This study only inestigated one type of CC technology i.e. post-combustion. There are various techniques for post-combustion separation of CO2 such as absorption by different solvents or by membranes. However, in this research only absorption by Monoethanolamine (MEA) solvent is considered. Moreover, it is assumed that the emissions from the energy supply to satisfy the energy requirements of the carbon capture process are not captured. Another area of limitation is that capital and operating costs are not considered, which could be significantly influential on the choice of technologies and CO2 mitigation options or combinations thereof. A further limitation of the study is related to the quality of data used in the model. For a more precise investigation, data inventory for emissions specific to the case study plant is needed. Data collected from the literature such as CO2 emission factors related to grid electricity or fuels, or energy requirements of CC facilities may be not sufficiently relevant for the studied system. In addition, the data related to Preemraff 3 Lysekil refinery (e.g. energy demands of unit operations) is associated with uncertainties since some data could be erroneous due to lack of instrumentation, accuracy in instrumentation (flow, temperature), incorrect physical data in terms of unknown composition, phase change, etc. [14]. 1.4. RESEARCH QUESTIONS TO BE ADDRESSED This project aims to address the following research questions:  To what extent could CO2 emissions be reduced along the entire value chain by applying carbon capture technology at the Preemraff Lysekil refinery?  Is there a reasonable trade-off between the CO2 mitigation potential and the increased energy consumption resulting from the implementation of the CO2 capture process?  How can the introduction of bio-based feedstock contribute to reducing CO2 emissions along the value chain?  How is the refinery energy demand affected by co-processing of bio-based feedstock?  What is the CO2 mitigation potential of the combination of carbon capture and bio-feedstock co-processing?  Does the combination of the two mitigation options lead to an increase or decrease of the refinery energy demand? Is there a satisfactory balance between the amount of CO2 emissions reduced and the energy demand? 4 5 2. THEORY In this research, the focus is on evaluating the effect of different CO2 mitigation options to be applied to refineries in order to move towards zero emissions. This chapter provides an overview of the refinery process units and the concepts of the selected mitigation options. 2.1. PREEMRAFF LYSEKIL REFINERY Preemraff Lysekil is a complex refinery with a crude oil capacity of 11.4 Mt/year [15] and CO2 emissions of 1.625 Mt/year [7]. A simplified process flow diagram of the refinery, showing the main process units, is shown in Figure 2-1. Crude oil is distilled in the Crude Distillation unit (atmospheric distillation column) to produce gas, naphtha, kerosene, light gas oil, heavy gas oil, and the residue [8], [9]. The gas undergoes further separation and purification resulting in fuel gas that can be used within the refinery, and Liquefied Petroleum Gas (LPG), which is sold or used as gasoline components. Figure 2-1: Overview of the main process flows at the Preemraff Lysekil refinery ( adapted from [8]) Naphtha is hydrotreated in the Naphtha Desulfurization unit, where it undergoes catalytic desulfurization and thereafter fractionation by the addition of hydrogen. The lighter fraction is upgraded in the Isomerization unit in which linear molecules are transformed into branched molecules with desired octane number, which constitute the 6 isomerate [8], [9], [16]. The heavier fraction is sent to the Platformer which involves catalytic reforming to produce reformate with a higher octane number. Both isomerate and reformate are applied as components to produce gasoline. The kerosene fraction is sent to the Synsat unit (Synergetic saturation unit) where it is desulfurized in order to be blended into diesel. Light and heavy gas oils undergo catalytic desulfurization and dearomatization using hydrogen in the Synsat unit and the Mild Hydro Cracker (MHC) unit, respectively. The outcome of the MHC unit is used in diesel, and the product from the Synsat unit is an important component in Environmental Class 1 diesel fuel (MK1 Diesel). The “Residue” (residual oil - bottom products) of the “Crude Distillation” column is sent to the “Vacuum Distillation” unit to be further distilled and separated into vacuum gas oil and vacuum residue. The vacuum gas oil is led to the Hydro Cracker where desulfurization and cracking take place using hydrogen, after which around 50% of it is turned into products that are lighter and more valuable including diesel and naphtha. The remainder of the vacuum gas oil, after desulfurization, is led to the Catalytic Crackerunit to be mainly broken down into gasoline components and to some extent to propene. The vacuum residue is sent to the Visbreaker unit where it is upgraded through thermal cracking to lighter products with lower viscosity and higher values, which are separated into heavy gasoil and heavy fuel oil [8], [9], [16]. 2.2. HYDROGEN PRODUCTION UNIT In order to fulfil the hydrogen demand in refineries, hydrogen is usually produced in a Hydrogen Production Unit (HPU) [17],[18]. In this research, the HPU is investigated in more detail since it is a large emission source of CO2, which makes it suitable for applying CCS. The introduction of bio-based feedstock is expected to affect the hydrogen balance in the refinery. Furthermore, previous studies of emission mitigation measures have been conducted recently for the HPU at the Preemraff Lysekil refinery, thus, significant amounts of data were available for this refinery unit. A schematic block flow diagram of an HPU is illustrated in Figure 2-2. Figure 2-2: Block flow diagram of an HPU based on steam reforming H2 , CO Reformer Feed: Natural gas and/or off-gases and/or butane Fuel gas Electricity Flue gases CO2 Water- gas-shift PSA unit operation H2 , CO2 H2 Off-gases (also called tail gases) Steam 7 In the HPU, hydrogen is produced through steam reforming of light hydrocarbons as per Equation 2.1, which is highly endothermic. It could also be produced through partial oxidation based on Equation 2.2, which is an exothermic reaction and can be utilized for heavier hydrocarbons. Hydrogen production can also be achieved by the combination of the two reactions, which is called autothermal reforming (ATR). In ATR, the energy released from the exothermic partial oxidation supplies the energy required by the endothermic steam reforming reaction. 𝐶𝐶𝑛𝑛𝐻𝐻𝑚𝑚 + 𝑛𝑛𝐻𝐻2𝑂𝑂 ⟶ �𝑛𝑛 + 𝑚𝑚 2 �𝐻𝐻2 + 𝑛𝑛𝑛𝑛𝑛𝑛 2.1 𝐶𝐶𝑛𝑛𝐻𝐻𝑚𝑚 + 𝑛𝑛 2 𝑂𝑂2 ⟶ 𝑚𝑚 2 𝐻𝐻2 + 𝑛𝑛𝑛𝑛𝑛𝑛 2.2 As can be seen in Figure 2-2, fuel gas and tail-gas are typically combusted to supply energy to the reforming process. Reforming is followed by the water-gas-shift reaction based on Equation 2.3, which leads to increased production of hydrogen [17],[18]. Finally, purification of hydrogen is done by separating hydrogen from other gases through pressure-swing adsorption (PSA). 𝐶𝐶𝐶𝐶 + 𝐻𝐻2𝑂𝑂 ⟷ 𝐻𝐻2 + 𝐶𝐶𝑂𝑂2 2.3 2.3. CARBON CAPTURE AND STORAGE Carbon Capture and Storage (CCS) involves capturing CO2 from exhaust gases, compressing it to high pressure and transporting it to a certain place for storage [11], [10]. There are three main carbon capture technologies [5], [11], [10]: Post-combustion: After fuel is combusted, CO2 is separated from the combustion flue gases. Pre-combustion: Before fuel is combusted, it is converted to hydrogen and CO2. Thereafter, CO2 is separated and the obtained hydrogen can be combusted as fuel with no CO2 emissions. Oxyfuel combustion: Fuel combustion is done with pure oxygen instead of air. Therefore, the generated flue gas is rich in CO2, which is then purified. CO2 separation can be done through various techniques, which can be categorized under absorption, adsorption, membrane and cryogenics [5], [19]. The proper technique must be selected considering the characteristics of the target flue gas on which carbon capture is supposed to be implemented. For low to medium CO2 concentrations, as is the case for most refinery flue gases [20], chemical absorption is more suitable. In order to retrofit the existing plants and equip them with CC technology, post-combustion is the prevailing option since no change in the upstream infrastructure is needed. Absorption by MEA is the most mature post-combustion capture technology. MEA absorption has been applied in natural gas refining processes for over half a century. It also has commercialized application for CO2 removal from combustion flue gases. Therefore, post-combustion based on MEA absorption was selected for carbon apture in this thesis. 8 2.4. CO-PROCESSING OF BIO-BASED FEEDSTOCKS IN EXISTING REFINERIES 2.4.1. Application Today, the fulfillment of increasing energy demand relies mostly on fossil-based resources [21]. This results in growing greenhouse gas emissions, depleting fossil sources, and thus, the rise in the price of raw materials [21]. Although there are a number of technologies to decrease CO2 emissions associated with the oil refining process, user phase is the main contributor to the total emissions within the life cycle of liquid fuels [12]. Use of biomass as an alternative feedstock to crude oil is, therefore, a potential solution in that the released carbon in the use phase of produced biofuels has been consumed in the growth phase of biomass through photosynthesis [12]. On the other hand, renewable nature and extensive availability of biomass makes for huge capabilities worldwide to produce and supply it in a sustainable manner [21]. Petroleum refineries can implement co-processing of bio-based feedstock together with fossil-based feedstock to produce fuels of hybrid origin [21], [22]. This gives the opportunity of utilizing the capacity of well-developed infrastructure of petroleum refineries and avoids the need for high capital investment, at least within the near-term future. In addition, the infrastructure of petroleum refineries is potentially able to process various types of bio-based feedstocks, which makes them flexible with respect to the availability of bio-feedstocks. Moreover, co-processing could result in the production of a variety of biofuels in different ranges such as LPG, gasoline, kerosene, diesel, or fuel oil. 2.4.2. Classification of bio-based feedstocks Biomass-derived feedstock can be categorized into oleaginous feedstock and carbohydrates [21]. Oleaginous feedstock, also known as lipid-based feedstock [22], mainly consists of triglycerides [21]. Furthermore, hydrolysis of triglycerides releases free fatty acids (FFA), which normally constitute processed and low-grade oleaginous feedstock [21]. Lipids are classified in four groups [21], [22]: 1) Edible oils among which the most common are palm oil, rapeseed oil, sunflower oil, and soybean oil 2) Non-edible oils such as Jatropha oil, and tall oil 3) Residual oils such as waste vegetable oils, waste cooking oils, and animal fats (lard, tallow). 4) Algae. Non-edible oils have the advantage of not competing with food sources. This also applies to algae, which also have the advantage of high lipid contetn and rate of growth compared to crops used to obtain vegetable oils [23]. While large-scale production of animal fats and vegetable oils are readily achievable, large-scale production of algae is in its infancy [21], [22]. Carbohydrates, as the other category of bio-based feedstock, consist of molecules containing carbon, hydrogen, and oxygen, which have so far found 9 to be the dominant constituent of biomass [21]. They constitute sources such as sugars, starch mainly derived from crops, and cellulose, hemicellulose and lignin as the main components of lignocellulose. A number of bio-based intermediates can be received from carbohydrates such as pyrolysis bio-oils, hydrothermal liquefaction oils (HTLO), and Fischer-Tropsch (FT) liquids, which are obtained by thermochemical conversion of lignocellulosic feedstocks through pyrolysis, hydrothermal liquefaction (HTL), and gasification respectively [24], [25]. 2.5. POTENTIAL INSERTION POINTS OF BIO-BASED FEEDSTOCK IN OIL REFINERIES 2.5.1. Possible insertion points Three insertion points are of particular relevance for co-processing bio-based feedstock, as shown in Figure 2-3. Figure 2-3: Potential insertion points of bio-based feedstock for co-processing in existing refineries (adapted from [26]) These are associated with different levels of risk to the refinery operations [16]. Insertion Point 1 involves feeding bio-derived feedstock to the atmospheric and vacuum distillation units. This is applicable when the characteristics of the bio-feedstock are similar to that of crude oil [26]. Considering that in these units separation is the main operation, and not chemical transformation, bio-feedstock should be almost oxygen- free and with minimal content of reactive species such as olefins, alcohols, carbonyls, and aldehydes [26], [12]. Moreover, this insertion point could lead to spread of contaminants within the whole refinery. Thus, bio-feedstock should not be contaminated. On the other hand, large amounts of non-volatile compounds e.g. sugars and oligomeric phenols are present in many biomass-derived feedstocks, which are problematic for distillation operation. Since bio-feedstocks are not thermally stable, at elevated temperatures polymerization increases resulting in a high level of viscosity and solid residuals. Therefore, blending biomass-derived feedstock with crude oil at Insertion Point 1 poses the highest risk. Insertion point 2 Insertion point 1 Insertion point 3 10 Blending biofuels with fossil-based fuels at the Insertion Point 3 is associated with much lower risk since it only affects operations downstream of the main unit operations [16],[26]. Nevertheless, technical challenges as well as high investment costs impede the commercial applicability of the Insertion Point 3 [12]. Insertion Point 2 is applied when bio-based feedstocks are blended with intermediate streams of the refinery in existing unit processes [12]. This could potentially result in lower capital costs and promote upgrading refinery flows to desirable product qualities. This involves medium risk in terms of distribution of oxygenates, impurities and process performance [16]. 2.5.2. Potential process units When considering Insertion point 2, the main possible process units include the fluid catalytic cracking (FCC) and catalytic hydroprocessing which in turn is divided to two categories of catalytic hydrotreating (HDT) and catalytic hydrocraking (HDC) [22]. FCC is usually applied to crack heavy fractions of crude oil i.e. it is typically fed by heavy gas oil, vacuum gas oil or residues. The main products of this process are gasoline and propylene [22], [26]. The HDT process is employed to remove undesirable heteroatoms as well as hydrogenation (saturation) of olefins and limited hydrogenation of aromatic compounds [26], [22]. It includes removal of oxygen, sulfur (hydrodesulfurization, HDS), nitrogen (hydrodenitrogenation, HDN), metals (hydrometalation, HDM), and halide [21], [26]. The feed to HDT includes intermediate flows within the refinery prior to being converted e.g. feed of FCC unit [22]. The catalytic hydrotreating process is exothermic involving high temperatures and pressures using hydrogen [26]. HDC process is utilized to convert heavy fractions to lighter ones with decreased boiling points [16], [22]. Compared with HDT, the feeds to HDC are typically refinery heavier intermediate flows e.g. heavy gas oil and vacuum gas oil. Reactions similar to those in hydrotreater are performed using hydrogen, but under more severe conditions [26]. Normally when there is the need to further decrease the size of bio-feedstock to be upgraded, HDC is applied as a secondary stage. Hydroprocessing has the advantage of being highly flexible in terms of various bio-feedstocks such as lipids and pyrolysis oils [22]. 11 12 3. METHODOLOGY The purpose of the methodology is to estimate, analyse, and model the changes in CO2 emissions and energy consumption associated with the two CO2 mitigation options including CCS and bio-based feedstock. The related procedures are elaborated in this chapter. 3.1. SYSTEM BOUNDARIES Figure 3-1 shows the system boundary applied in this thesis. The system boundary includes all the refinery operations necessary to convert the feedstock (crude oil or bio- based input) to refined products including the major unit processes as well as CC facilities within the refinery. Since the Preemraff Lysekil refinery is used as a case study in this research, process units are considered as shown in Figure 2-1 as well as the hydrogen production unit (HPU) and utility unit. The boundary is also inclusive of resource extraction for refinery inputs (crude oil, make-up fuel, and bio-feedstock), electricity production and supply to the refinery, CO2 capture, compression, transport and storage, and the user phase. The analysis accounts for on-site CO2 emissions and emissions associated with electricity purchased from the grid, the potential of reduction in CO2 emissions as well as changes in the energy demand in terms of fuel and electricity consumption due to deployment of CCS, bio-based feedstock, or both options as compared to the plant without CCS and using crude oil feedstock. The study focuses on the effects of CO2 mitigation options in terms of changes in the energy demand of production as well as emissions reduction potential along the value chain, which are aligned with Preem’s Scope 1, 2, and part of Scope 3 as denoted in Figure 3-1. It should be noted that the emissions associated with the processes/activities of CO2 capture, compression, transport, and storage relate to the CO2 emissions associated with providing the heat and electricity required by these operations. This choice of system boundary is justified by the fact that the mitigation options are applied in the production stage and that most of the emissions take place in the user phase; approximately 85% of the CO2 emissions from Preem’s value chain comes from the combustion of fossil fuels in the user phase [7]. Figure 3-1: System boundary defined in this study 13 3.2. MODELLING The model to quanitfy and evaluate different mitigation options was developed in MATLAB. In order to conduct the modelling, first the data required as the input to the model must be collected. Data collection must be adapted to the purpose of its usage [27]. If the precision of required details are not of great importance in the simplified model or they are hard to collect from the case study refinery (e.g. when a mitigation option is not yet applied in practice to have the respective real data), nominal values from literature are sufficient. On the other hand, for specific details of the target refinery real site data needs to be collected. Therefore, since this thesis aims at both a simplified refinery as well as conducting a case study on the Preemraff Lysekil refinery, data was collected from both industry and literature. A bottom-up approach was adopted to build the model so that mass and energy balances were calculated for each process unit of the refinery and used as a basis to build a model of the complete refinery. Within the procedure of modelling, block flow diagrams were applied to depict the structure of aimed units in terms of involved unit operations through representing them by blocks in a simplified input-output diagram; also, to determine the extent of details needed for building the model [27]. The modelling was conducted at several layers (adapted from [27] and [28]). The micro layer deals with thermodynamics of reactions and energy balances at a molecular level. The meso layer involves mass and energy balances at the level of unit processes. The macro layer provides the user interface, which directs the procedure based on the decisions and input from the user side. In this regard, the modelling involves creating several modules (adapted from [28]). The module-based approach provides the possibility to assemble and add different computing procedures and mitigation options to the model. This enables constructing the model in a flexible way so that it can be tuned to the selected mitigation options and any additional mitigation option can be combined with the model in the future. Therefore, the modelling was performed through the following steps: • Required data was obtained. • The module for quantifying CO2 emissions was established on the basis of the method described in Section 3.3. • The module for CCS was defined following the method under Section 3.4. • The module for bio-based feedstock mitigation option was created based on the method described in Section 3.5. Figure 3-2 illustrates the modelling architecture. The algorithm through which the analyses mentioned in Section 3.6 were conducted is depicted in Figure 3-3. Validation of the model was done through comparison of results from modelling with data received from plant and literature. 14 Method for quantifying CO2 emissions Figure 3-3: Overview of information flow in the model. Figure 3-2: Architecture of the modelling. 15 3.3. METHOD FOR QUANTIFYING CO2 EMISSIONS In refineries, CO2 is emitted from a number of different sources [9]. The emission sources have various flowrates and CO2 concentrations [9]. The emission points include fired heaters and furnaces available in each process unit, flares, the process units which involve direct emissions of CO2 (e.g. catalytic cracking unit), and utility systems (e.g. production of steam) [6], [9]. It is assumed that post-combustion CO2 capture is applied to specific individual CO2 emission sources, and that the capture facilities are sized to handle the full flue gas streams from these sources. However, the energy supply for CC facilities is associated with generation of emissions, which are accounted for when estimating the mitigation potential of the applied CC technology. To estimate the amount of CO2 emitted from different process units within the refinery, a bottom-up model was constructed. In the first step, an Input-Output model was developed for each process unit u in terms of material, energy, and CO2 balances. Thereafter, regarding the connected process units within the focused refinery shown in Figure 2-1 as well as the hydrogen production unit (HPU) and utility unit, the model for the whole refinery was built. Finally, the CO2 reduction potential using CCS was investigated. The details and procedures are elaborated as follows and illustrated in Figure 3-4. In order to quantify CO2 emissions in this project and validate the data, among all units within the refinery as described under Section 3.3, hydrogen production unit (HPU) is focused more in detail. This is due to the importance of the unit and the availability of data. Material balances The material balance for each process unit is modelled based on Equation 3.1 with terms defined in Table 1 [9]: 𝑉𝑉𝑢𝑢,𝑝𝑝 = ∑ 𝑌𝑌𝑢𝑢,𝑖𝑖,𝑝𝑝 × 𝑉𝑉𝑢𝑢,𝑖𝑖𝑖𝑖 3.1 Table 1: Definition of the terms in Equation 3.1 1 Standard cubic meter per year Regarding Equation 3.1, the material balances require the data related to yield of each product in terms of each input. Since such data is usually not accessible in practice, the data collection was conducted on energy consumption level, which is described hereafter, knowing that the fuel consumptions of units are directly connected to the material balance and any change in the throughput of a unit process would directly affect its fuel consumption. Therefore, the model’s ability to handle variable material flows is maintained. Term Unit Denotation Description 𝑉𝑉𝑢𝑢,𝑝𝑝 106 Sm3/year 1 Outputs Product(s) p of process unit u 𝑉𝑉𝑢𝑢,𝑖𝑖 106 Sm3/year Inputs Input(s) i to process unit u 𝑌𝑌𝑢𝑢,𝑖𝑖,𝑝𝑝 % Volume yields Corresponding to the production of product p in the unit process u with the input i 16 Energy balances The energy consumption of each process unit depicted in Figure 2-1 as well as the utility system is estimated on the basis of the material balance and fuel consumption factors, which are thereafter combined to yield the total energy consumption (𝑇𝑇𝑇𝑇: 106 GJ/year) according to Equation 3.2, which is defined in Table 2 [9]. Since electricity consumption is normally reported in GWh/year, to facilitate data collection and input to the model, the same unit is used for the input, but the factor 3.6/103 is embedded in the equation to convert it to 106 GJ/year. 𝑇𝑇𝑇𝑇 = ∑ ∑ 𝐹𝐹𝐹𝐹𝑢𝑢,𝑓𝑓 +𝑓𝑓𝑢𝑢 ∑ ∑ 𝐸𝐸𝐸𝐸𝑢𝑢 × 𝑆𝑆ℎ𝐸𝐸/𝑂𝑂𝑂𝑂𝐸𝐸𝐸𝐸 × 3.6/103𝑢𝑢 3.2 Table 2: Definition of the terms in Equation 3.2 1 Fuel f includes refinery by-products that are potentially used as fuels for different process units (e.g. methane, ethane) and fossil fuels. 2 This can be electricity purchased from grid or electricity generated on-site. CO2 balances CO2 emissions from different process units and the utility system are estimated using the emission factors for the different fuels (fossil fuels and refinery by-products) consumed within the units and the energy consumption obtained from the energy balance(s). Thus, the amount of CO2 emissions associated with each process unit u within the scope of this project (𝐸𝐸𝑚𝑚𝑢𝑢: 106 t/year) is calculated based on Equation 3.3 with terms described in Table 3 (adapted from [9]). 𝐸𝐸𝑚𝑚𝑢𝑢 = ∑ 𝐸𝐸𝐸𝐸𝑏𝑏𝑏𝑏 × 𝐿𝐿𝐿𝐿𝑉𝑉𝑏𝑏𝑏𝑏 × 𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏 + [∑ �𝐹𝐹𝐹𝐹𝑢𝑢,𝑓𝑓 × 𝐸𝐸𝐹𝐹𝑓𝑓� − ∑ �𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏 × 𝐿𝐿𝐿𝐿𝑉𝑉𝑏𝑏𝑏𝑏 × 𝐸𝐸𝐹𝐹𝑏𝑏𝑏𝑏�𝑏𝑏𝑏𝑏 +𝑓𝑓𝑏𝑏𝑏𝑏 ∑ �𝐸𝐸𝐶𝐶𝑢𝑢 × 𝑆𝑆ℎ𝐸𝐸 𝑂𝑂𝑂𝑂𝐸𝐸 × 3.6 103 × 𝐸𝐸𝐹𝐹𝑓𝑓�𝑓𝑓 ] 3.3 The first term in Equation 3.3 is related to the CO2 emissions from refinery by-products used as fuel, which are part of the on-site emissions framed in Scope 1. The combination of the remaining terms in this equation is related to the CO2 emissions due to combusting purchased fossil fuels for satisfying the heat demand as well as fossil fuel usage for generation of required power for each unit. The second and third terms account for the CO2 emissions from purchased fossil fuels that are combusted in the refinery to fulfil the energy demands, which are part of the on-site emissions reflecting Term Unit Denotation Description 𝐹𝐹𝐹𝐹𝑢𝑢,𝑓𝑓 106 GJ/year Fuel consumption Consumption of fuel f 1 in process unit u ��𝐹𝐹𝐹𝐹𝑢𝑢,𝑓𝑓 𝑓𝑓𝑢𝑢 106 GJ/year Fuel consumption Consumption of fuel f in all the process units shown in Figure 2-1, the HPU and utility unit 𝐸𝐸𝐸𝐸𝑢𝑢 GWh/year Electricity consumption Electricity consumption of each process unit u 𝑆𝑆ℎ𝐸𝐸 % Share of electricity Supplied from different sources of power, E 2 𝑂𝑂𝑂𝑂𝐸𝐸 % Overall efficiency Inclusive of generation, transmission and distribution of electricity supplied from each source of power, E 𝑇𝑇𝑇𝑇 106 GJ/year Total energy consumption Related to all the process units shown in Figure 2-1, the HPU and utility unit 17 Scope 1. The fourth term is related to the CO2 emissions from purchased fossil fuels combusted for electricity generation. These emissions are inclusive of the on-site emissions (Scope 1) when the power generation is taking place in the refinery as well as the offsite emissions (Scope 2) for the power generated outside of the refinery and purchased from the grid. Regarding the scope of this project to account for CO2 emissions along the value chain of refinery products, CO2 emissions from resource extraction (Scope 3) and user phase (Scope 3) must be considered as well. The CO2 emissions from resource extraction (𝐸𝐸𝑚𝑚𝑅𝑅𝑅𝑅: 106 t/year) can be calculated from the data available in the literature. The CO2 emissions from combustion of refinery products (fuels) in the user phase (𝐸𝐸𝑚𝑚𝑈𝑈𝑈𝑈: 106 t/year) can be estimated based on Equation 3.4 defined in Table 3. 𝐸𝐸𝑚𝑚𝑈𝑈𝑈𝑈 = ∑ 𝐸𝐸𝐸𝐸𝑝𝑝 × 𝑉𝑉𝑝𝑝𝑝𝑝 3.4 Therefore the total emissions of CO2 (𝑇𝑇𝑇𝑇𝑇𝑇: 106 t/year) within the system boundary of this study is obtained by Equation 3.5, and the terms are also defined in Table 3. 𝑇𝑇𝑇𝑇𝑇𝑇 = ∑ (𝐸𝐸𝑚𝑚𝑢𝑢)𝑢𝑢 + 𝐸𝐸𝑚𝑚𝑅𝑅𝑅𝑅 + 𝐸𝐸𝑚𝑚𝑈𝑈𝑈𝑈 3.5 Table 3: Definition of terms in Equation 3.3, Equation 3.4 and Equation 3.5 1 By-product bp, which is used as a fuel, e.g. methane and ethane The on-site CO2 emissions (Scope 1) do not include the emissions associated with electricity purchased from the grid (Scope 2) and other emissions within the value chain of refinery products (Scope 2 and 3). Therefore, on-site CO2 emissions from each process unit u (𝐸𝐸𝑚𝑚𝑢𝑢,𝑜𝑜𝑜𝑜−𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠: 106 t/year) and in total (𝑇𝑇𝑇𝑇𝑚𝑚𝑜𝑜𝑜𝑜−𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠: 106 t/year) are calculated based on Equation 3.6 and Equation 3.7 according to the definitions presented in Table 4. 𝐸𝐸𝑚𝑚𝑢𝑢,𝑜𝑜𝑜𝑜−𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 = ∑ 𝐸𝐸𝐸𝐸𝑏𝑏𝑏𝑏 × 𝐿𝐿𝐿𝐿𝑉𝑉𝑏𝑏𝑏𝑏 × 𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏 + [∑ �𝐹𝐹𝐹𝐹𝑢𝑢,𝑓𝑓 × 𝐸𝐸𝐹𝐹𝑓𝑓� − ∑ �𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏 × 𝐿𝐿𝐿𝐿𝑉𝑉𝑏𝑏𝑏𝑏 × 𝐸𝐸𝐹𝐹𝑏𝑏𝑏𝑏�𝑏𝑏𝑏𝑏 +𝑓𝑓𝑏𝑏𝑏𝑏 ∑ �𝐸𝐸𝐶𝐶𝑢𝑢 × 𝑆𝑆ℎ𝐸𝐸 𝑂𝑂𝑂𝑂𝐸𝐸 × 𝐸𝐸𝐹𝐹𝑓𝑓 − 𝐸𝐸𝐶𝐶𝑢𝑢 × 𝑆𝑆ℎ𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔 𝑂𝑂𝑂𝑂𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔 × 𝐸𝐸𝐹𝐹𝑓𝑓�𝑓𝑓 ] 3.6 𝑇𝑇𝑇𝑇𝑚𝑚𝑜𝑜𝑜𝑜−𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 = ∑ (𝑢𝑢 𝐸𝐸𝑚𝑚𝑢𝑢,𝑜𝑜𝑜𝑜−𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠) 3.7 Term Unit Denotation Description 𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏 106 Sm3/year Volume of byproduct1 Related to refinery by-product bp of process unit u 𝐿𝐿𝐿𝐿𝑉𝑉𝑏𝑏𝑏𝑏 GJ/Sm3 Lower Heating Value (LHV) Related to by-product bp 𝐸𝐸𝐸𝐸𝑏𝑏𝑏𝑏 tCO2/GJ Emission factor Related to by-product bp 𝐸𝐸𝐹𝐹𝑓𝑓 tCO2/GJ Emission factor Related to fossil fuel (f) 𝐸𝐸𝑚𝑚𝑢𝑢 106 t/year CO2 emissions Associated with each process unit u 𝐸𝐸𝐸𝐸𝑝𝑝 tCO2/Sm3 Emission factor Related to product p 𝑉𝑉𝑝𝑝 106 Sm3/year Volume of product Related to refinery product p 𝐸𝐸𝑚𝑚𝑈𝑈𝑈𝑈 106 t/year CO2 emissions From user phase 𝐸𝐸𝑚𝑚𝑅𝑅𝑅𝑅 106 t/year CO2 emissions From resource extraction phase 𝑇𝑇𝑇𝑇𝑇𝑇 106 t/year CO2 emissions From value chain inclusive of emissions associated with all the unit processes shown in Figure 2-1, the HPU and utility unit 18 Table 4: Definition of terms in Equation 3.6 and Equation 3.7 Term Unit Denotation Description 𝑆𝑆ℎ𝑔𝑔rid % Share of electricity Supplied from the grid 𝑂𝑂𝑂𝑂𝑔𝑔rid % Grid overall efficiency Inclusive of generation, transmission and distribution of grid electricity 𝐸𝐸𝑚𝑚𝑢𝑢,𝑜𝑜𝑜𝑜−𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 106 t/year On-site CO2 emissions From each unit process u 𝑇𝑇𝑇𝑇𝑚𝑚𝑜𝑜𝑜𝑜−𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 106 t/year Total on-site CO2 emissions From all the unit processes shown in Figure 2-1, HPU and utility unit Figure 3-4: Bottom-up model construction based on material, energy and CO2 balances. Refinery 𝐸𝐸𝑚𝑚𝑢𝑢,𝑜𝑜𝑜𝑜−𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 On-site CO2 emissions from process unit u Output(s) 𝑉𝑉𝑢𝑢,𝑝𝑝 Input(s) 𝑉𝑉𝑢𝑢,𝑖𝑖 Process unit u �𝐹𝐹𝐹𝐹𝑢𝑢,𝑓𝑓 𝑓𝑓 Total fuel consumption of process unit u Total electricity consumption of process unit u supplied from different sources of power, g �𝐸𝐸𝐸𝐸𝑢𝑢 × 𝑆𝑆ℎ𝐸𝐸/𝑂𝑂𝐸𝐸𝐸𝐸 𝐸𝐸 Process unit u …. …. …. 𝑇𝑇𝑇𝑇𝑚𝑚𝑜𝑜𝑜𝑜−𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 Total on-site CO2 emissions from all process units ��𝐹𝐹𝐹𝐹𝑢𝑢,𝑓𝑓 𝑓𝑓𝑢𝑢 Total fuel consumption of all process units Total electricity consumption of all process units supplied from different sources of power, E ��𝐸𝐸𝐸𝐸𝑢𝑢 × 𝑆𝑆ℎ𝐸𝐸/𝑂𝑂𝐸𝐸𝐸𝐸 𝐸𝐸𝑢𝑢 19 3.4. METHOD FOR QUANTIFYING THE CO2 MITIGATION POTENTIAL OF IMPLEMENTING CARBON CAPTURE AND STORAGE (CCS) TECHNOLOGY In order to quantify the CO2 mitigation potential using CCS technology, it is first necessary to quantify the amount of on-site CO2 emissions through the method described in Section 3.3. Thereafter, it can be decided which process units can be equipped with CCS technology. Then the amount of captured CO2 can be calculated according to Equation 3.8 with terms defined in Table 5. 𝐶𝐶𝐶𝐶 = ∑ 𝑆𝑆𝑆𝑆𝑆𝑆.𝐸𝐸𝑚𝑚𝑢𝑢,𝑜𝑜𝑜𝑜−𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 × 𝐶𝐶𝑅𝑅𝑢𝑢𝑢𝑢 3.8 The energy consumption of the CC technology itself leads to CO2 emissions, which are not captured in this study as mentioned previously. Thus, it is shown as a fuel penalty, which can be estimated based on literature data. In addition, there are CO2 emissions due to transport and storage of the captured CO2, which can be obtained from the literature as well. Therefore, the CO2 mitigation potential (𝐶𝐶𝐶𝐶𝐶𝐶: 106 t/year) is estimated by Equation 3.9 described in Table 5. 𝐶𝐶𝐶𝐶𝐶𝐶 = 𝐶𝐶𝐶𝐶 × �1 − 𝐹𝐹𝐹𝐹 × 𝐸𝐸𝐹𝐹𝑓𝑓 − 𝐸𝐸𝑚𝑚𝐶𝐶𝐶𝐶� − 𝐸𝐸𝑚𝑚𝑅𝑅𝐸𝐸𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀−𝑢𝑢𝑢𝑢 𝑓𝑓 = 𝐶𝐶𝐶𝐶 × (1 − 𝐹𝐹𝐹𝐹 × 𝐸𝐸𝐹𝐹𝑓𝑓 − 𝐸𝐸𝑚𝑚𝐶𝐶𝐶𝐶) − (𝐶𝐶𝐶𝐶 × 𝐹𝐹𝐹𝐹 × 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑚𝑚𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀−𝑢𝑢𝑢𝑢 𝑓𝑓) 3.9 The addition to the energy demand of the refinery in focus is due to the energy consumption of CCS i.e. 𝐸𝐸𝐸𝐸𝐶𝐶𝐶𝐶𝐶𝐶 (GJ/year), which can be derived from Equation 3.10, also defined in Table 5. 𝐸𝐸𝐸𝐸𝐶𝐶𝐶𝐶𝐶𝐶 = 𝐶𝐶𝐶𝐶 × 𝐹𝐹𝐹𝐹 3.10 Table 5: Definition of terms in the Equation 3.8, Equation 3.9, and Equation 3.10 Term Unit Denotation Description 𝐶𝐶𝐶𝐶 106 t/year Captured CO2 emissions Using CC technology Sel. Emu,on−site 106 t/year On-site CO2 emissions Related to the selected process unit u for applying CC technology 𝐶𝐶𝑅𝑅𝑢𝑢 % Capture rate Related to the applied CC technology to each unit process u, which can be obtained by consulting literature 𝐶𝐶𝐶𝐶𝐶𝐶 106 t/year CO2 mitigation potential Related to the applied CC technology 𝐹𝐹𝐹𝐹 GJ/t CO2 Fuel penalty The energy consumption of the applied CC technology 𝐸𝐸𝑚𝑚𝐶𝐶𝐶𝐶 t/t CO2 CO2 emissions Related to transport and storage of the captured CO2 𝐸𝐸𝐸𝐸𝐶𝐶𝐶𝐶𝐶𝐶 106 GJ/year Addition to the energy demand of the refinery of focus Due to the energy consumption of CCS 𝐸𝐸𝑚𝑚𝑅𝑅𝐸𝐸𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀−𝑢𝑢𝑢𝑢 𝑓𝑓 106 t/year CO2 emissions From the resource extraction phase of make-up fossil fuel (f) SpecEmMake−up f tCO2/GJ Specific CO2 emissions Related to the resource extraction phase of make-up fossil fuel (f) 20 3.5. METHOD FOR QUANTIFYING THE CO2 MITIGATION POTENTIAL OF CO-PROCESSING BIO-BASED FEEDSTOCK The type of bio-feedstock that was considered in this research is lipid-based, as described under Section 2.4. The input data for the calculations relate primarily to rapeseed oil. This is because it is used as one of the common lipids for co-processing [22] as well as being used in the Preemraff Lysekil refinery for this purpose. The hydrotreating unit is a suitable insertion point for vegetable oils [12] such as rapeseed oil. Among all possible insertion points, hydrotreating has been commercialized [12], which is also applied for upgrading rapeseed oil in the Preemraff Lysekil refinery. Therefore, in this study the insertion point of focus is the hydrotreating unit. 3.5.1. Composition The category of oleaginous/lipid-based feedstock mainly consists of triglycerides [29]. Triglyceride in turn is an ester formed by the combination of glycerol and fatty acids [18]. Thus, the composition of lipids can be defined by fatty acids as model compounds and their representative triglycerides [23]. The typical composition of rapeseed oil is shown in terms of fatty acids in Table 6. Table 6: Composition of rapeseed oil (wt%) in terms of fatty acids [29], [30] Fatty acid Structure1 Refined rapeseed oil composition (wt%) Myristic acid C14:0 0.06 Myristoleic acid C14:1 0.00 Palmitic acid C16:0 4.64 Palmitoleic acid C16:1 0.24 Stearic acid C18:0 1.96 Oleic acid C18:1 63.47 Linoleic acid C18:2 20.01 Linolenic acid C18:3 6.97 Arachydic acid C20:0 0.60 Arachidonic acid C20:1 1.18 Behenic acid C22:0 0.15 Erucic acid C22:1 0.07 Lignoceric acid C24:0 0.13 Nervonic acid C24:1 0.14 1 Cx:y is a fatty acid with x carbon atoms and y double bonds Vegetable oils and animal fats can contain other compounds such as metals, phospholipids, polyphenols, and sterols to a minor extent and thus, need to be pretreated to avoid negative effects on the activity of catalysts [29], [23]. This leads to refined oils rich in triglycerides by more than 99%. 3.5.2. Reaction pathways The major heteroatom in bio-based feedstock is oxygen [26]. Thus, upgrading mainly involves removing oxygen, which lowers the energy intensity. This is done through 21 decarboxylation, decarbonylation, and hydrodeoxygenation. The aforementioned reactions for fatty acids in the lipid-based feedstock are shown respectively as follows: 𝑅𝑅 − 𝐶𝐶𝐻𝐻2 − 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 ⟶ 𝑅𝑅 − 𝐶𝐶𝐻𝐻3 + 𝐶𝐶𝑂𝑂2 3.11 𝑅𝑅 − 𝐶𝐶𝐻𝐻2 − 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 + 𝐻𝐻2 ⟶ 𝑅𝑅 − 𝐶𝐶𝐻𝐻3 + 𝐶𝐶𝐶𝐶 + 𝐻𝐻2𝑂𝑂 3.12 𝑅𝑅 − 𝐶𝐶𝐻𝐻2 − 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 + 3𝐻𝐻2 ⟶ 𝑅𝑅 − 𝐶𝐶𝐻𝐻3 + 2𝐻𝐻2𝑂𝑂 3.13 In addition to the aforementioned deoxygenation reactions, methanation is an important side reaction that takes place according to Equation 3.14 [29],[31]. 𝐶𝐶𝐶𝐶 + 3𝐻𝐻2 ⇆ 𝐶𝐶𝐻𝐻4 + 𝐻𝐻2𝑂𝑂 3.14 Another side reaction taking place is as per Equation 3.15 [29],[31]. 𝐶𝐶𝑂𝑂2 + 𝐻𝐻2 ⇆ 𝐶𝐶𝐶𝐶 + 𝐻𝐻2𝑂𝑂 3.15 During hydrotreating, hydrogenation and deoxygenation take place [18]. Also, using hydrogen, the model triglycerides are decomposed to fatty acids [23]. Regarding the main fatty acids in rapeseed oil, the corresponding model triglycerides are described in Table 7. Table 7: The main fatty acids and corresponding triglycerides constituting rapeseed oil Fatty acid Structure Formula Corresponding triglycerides Formula Oleic acid C18:1 C18H34O2 Triolein C57H104O6 Linoleic acid C18:2 C18H32O2 Trilinolein C57H98O6 Linolenic acid C18:3 C18H30O2 Trilinolein C57H92O6 Palmitic acid C16:0 C16H32O2 Tripalmitin C51H98O6 The decomposition of model triglycerides to the associated fatty acids is assumed to follow the general reaction as below [23]: 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 + 3𝐻𝐻2 ⟶ 3𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 + 𝐶𝐶3𝐻𝐻8 3.16 The reaction pathways regarding the main triglyceride constituting rapeseed oil, triolein, are shown in Table 8. Table 8: The reaction pathways of the main model compound constituting rapeseed oil (adapted from [23],[18]) Reaction Description 𝐶𝐶57𝐻𝐻104𝑂𝑂6 + 3𝐻𝐻2 → 𝐶𝐶57𝐻𝐻110𝑂𝑂6 Hydrogenation 𝐶𝐶57𝐻𝐻110𝑂𝑂6 + 3𝐻𝐻2 → 3𝐶𝐶18𝐻𝐻36𝑂𝑂2 + 𝐶𝐶3𝐻𝐻8 Decomposition of triglycerides to fatty acids 𝐶𝐶18𝐻𝐻36𝑂𝑂2 ⟶ 𝑛𝑛 − 𝐶𝐶17 + 𝐶𝐶𝑂𝑂2 Decarboxylation 𝐶𝐶18𝐻𝐻36𝑂𝑂2 + 𝐻𝐻2 ⟶ 𝑛𝑛 − 𝐶𝐶17 + 𝐶𝐶𝐶𝐶 + 𝐻𝐻2𝑂𝑂 Decarbonylation 𝐶𝐶18𝐻𝐻36𝑂𝑂2 + 3𝐻𝐻2 ⟶ 𝑛𝑛 − 𝐶𝐶18 + 2𝐻𝐻2𝑂𝑂 Hydrodeoxygenation 22 Material balances A simplified schematic block flow diagram of a hydrotreating unit is illustrated in Figure 3-5. The method accounts for the changes made by the bio-feedstock when added to the fossil-based feedstock for co-processing. Thus, the calculations focus on the bio-share of the system. Figure 3-5: Simplified schematic block flow diagram of a hydrotreating unit Regarding the ratio of bio-based feedstock to fossil-based feedstock (wbio−feedstock), the mass flow of the former is obtained by Equation 3.17. The description of the terms in Equation 3.17 to Equation 3.22 is provided in Table 9. 𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 = 𝑤𝑤𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓.𝑚𝑚𝑡𝑡𝑡𝑡𝑡𝑡 3.17 The lipid-based feedstock to the hydrotreating unit is converted to liquid and gaseous mixtures of compounds [29]. The mass flow of liquid product is obtained based on Equation 3.18. In addition, water generated due to hydrotreating is separated and its mass flow can be estimated based on Equation 3.19. The yields of liquid product and water generation can be obtained by consulting the experimental results in the literature for similar process conditions. Since the mass flow of hydrogen required for hydrotreating of lipid-based feedstock in these equations is unknown and is supposed to be calculated by the proposed method, the method is based on an iterative calculation in which first an estimate is considered for the aforementioned hydrogen consumption. The calculated hydrogen consumption for hydrotreating of lipid-based feedstock is then compared with the estimate and if they are not in agreement, a revised estimate is used until a common value for the hydrogen consumption is received. 𝑚𝑚𝐿𝐿𝐿𝐿 = (𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 + 𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻).𝑌𝑌𝐿𝐿𝐿𝐿 3.18 𝑚𝑚𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤 = (𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 + 𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻).𝑌𝑌𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤 3.19 Using the mass balance regarding the input lipid-based feedstock and hydrogen as well as products as shown in Equation 3.20, the mass flow of gaseous product can be calculated. 𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 + 𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 = 𝑚𝑚𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤 + 𝑚𝑚𝐿𝐿𝐿𝐿 + 𝑚𝑚𝐺𝐺𝐺𝐺 3.20 The amount of hydrogen required for hydrogenation of double bounds in the lipid-based feedstock is calculated by Equation 3.21 (adapted from [31]). 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐷𝐷𝐷𝐷 = 𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑀𝑀𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 .𝑁𝑁𝐷𝐷𝐷𝐷 3.21 23 Table 9: Definition of terms in Equation 3.17 to Equation Equation 3.22 The number of double bounds available in 1 mole of the lipid-based feedstock is derived by Equation 3.22. 𝑁𝑁𝐷𝐷𝐷𝐷 = ∑ 𝐷𝐷𝐵𝐵𝐹𝐹𝐹𝐹.𝑥𝑥𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 3.22 The molar mass of the lipid-based feedstock is estimated by Equation 3.23 (adapted from [31]). 𝑀𝑀𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 = ∑ (3𝑀𝑀𝑐𝑐 + 5𝑀𝑀𝐻𝐻). 𝑥𝑥𝐹𝐹𝐹𝐹 3𝐹𝐹𝐹𝐹 + ∑ 𝑥𝑥𝐹𝐹𝐹𝐹(𝐶𝐶𝐹𝐹𝐹𝐹.𝑀𝑀𝑐𝑐 + 𝐻𝐻𝐹𝐹𝐹𝐹.𝑀𝑀𝐻𝐻 + 2.𝑀𝑀𝑂𝑂)𝐹𝐹𝐹𝐹 3.23 In addition, the hydrogen consumption for decomposition of the triglycerides in the lipid-based feedstock into the corresponding fatty acids is derived by Equation 3.24. The equation is defined considering that based on Equation 3.16, the number of moles of hydrogen consumed for decomposition of a triglyceride equals the number of moles of fatty acids formed. The terms in Equation 3.23 to Equation 3.35 are defined in Table 10. 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐷𝐷𝐷𝐷 = ∑ 𝑥𝑥𝐹𝐹𝐹𝐹.𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑀𝑀𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝐹𝐹𝐹𝐹 3.24 Term Unit Denotation Description 𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 106 t/year Mass flow Related to the bio-based feedstock 𝑤𝑤𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 % Weight fraction Related to the bio-based feedstock in the liquid input to the process unit 𝑚𝑚𝑡𝑡𝑡𝑡𝑡𝑡 106 t/year Mass flow Total liquid input to the unit process 𝑚𝑚𝐿𝐿𝐿𝐿 106 t/year Mass flow Liquid product of hydrotreating the bio- based feedstock 𝑌𝑌𝐿𝐿𝐿𝐿 % Yield Related to the liquid product of hydrotreating the bio-based feedstock 𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 106 t/year Total hydrogen consumption Due to hydrotreating of the lipid-based feedstock 𝑚𝑚𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤 106 t/year Mass flow Water produced through hydrotreating the bio-feedstock 𝑌𝑌𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤 % Yield Related to water produced through hydrotreating the bio-feedstock 𝑚𝑚𝐺𝐺𝐺𝐺 106 t/year Mass flow Gaseous product of hydrotreating the bio- based feedstock 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐷𝐷𝐷𝐷 106 mole/year Hydrogen consumption For hydrogenation of double bounds in the lipid-based feedstock 𝑀𝑀𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 t/mole Molar mass For lipid-based feedstock 𝑁𝑁𝐷𝐷𝐷𝐷 - Number of double bounds Available in 1 mole of the lipid-based feedstock 𝐷𝐷𝐵𝐵𝐹𝐹𝐹𝐹 - Number of double bounds Available in 1 mole of the fatty acid FA 𝑥𝑥𝐹𝐹𝐹𝐹 % Molar fraction Related to fatty acid FA 𝑀𝑀𝑐𝑐 t/mole Molar mass Carbon 𝑀𝑀𝐻𝐻 t/mole Molar mass Hydrogen 𝑀𝑀𝑂𝑂 t/mole Molar mass Oxygen 𝐶𝐶𝐹𝐹𝐹𝐹 - Number of carbon atoms In fatty acid FA 𝐻𝐻𝐹𝐹𝐹𝐹 - Number of hydrogen atoms In fatty acid FA 24 Similarly, the number of moles of propane that are formed due to the decomposition of triglycerides can be accounted according to Equation 3.25. 𝑛𝑛𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 = ∑ 𝑥𝑥𝐹𝐹𝐹𝐹 3 .𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑀𝑀𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝐹𝐹𝐹𝐹 3.25 Based on Equation 3.13, each mole of fatty acid is deoxygenated through hydrodeoxygenation using 3 moles of hydrogen. Therefore, the hydrogen consumption by hydrodeoxygenation is estimated by Equation 3.26. 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 = 3.∑ 𝑤𝑤𝐶𝐶𝑖𝑖 𝐿𝐿𝐿𝐿.𝑚𝑚𝐿𝐿𝐿𝐿 𝑀𝑀𝐶𝐶𝑖𝑖 𝐶𝐶𝑖𝑖 3.26 The mass flow of methane resulted from methanation is obtained by Equation 3.27. 𝑚𝑚𝑀𝑀𝑀𝑀 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 = 𝑤𝑤𝑀𝑀𝑀𝑀 .𝑚𝑚𝐺𝐺𝐺𝐺 3.27 Considering Equation 3.14, the hydrogen consumption related to methanation reaction is estimated by Equation 3.28 (adapted from [31]). 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 = 3.𝑚𝑚𝑀𝑀𝑀𝑀 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑀𝑀𝑀𝑀𝑀𝑀 3.28 The share of Ci-1 n-alkanes that are received through decarbonylation is accounted as per Equation 3.29 (adapted from [31]). 𝑆𝑆ℎ𝐷𝐷𝐷𝐷𝐷𝐷 = 𝑛𝑛𝐶𝐶𝐶𝐶 𝑛𝑛𝐶𝐶𝑂𝑂2+𝑛𝑛𝐶𝐶𝐶𝐶 3.29 The CO that is formed through decarbonylation reaction is partly converted to methane by the methanation reaction. Thus, the molar flow of CO due to deoxygenation of the lipid-based feedstock is calculated as per Equation 3.30 [31]. 𝑛𝑛𝐶𝐶𝐶𝐶 = 𝑛𝑛𝐶𝐶𝐶𝐶 𝐺𝐺𝐺𝐺 + 𝑛𝑛𝐶𝐶𝐶𝐶𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 3.30 The molar flow of CO in the gaseous product is received by Equation 3.31. 𝑛𝑛𝐶𝐶𝐶𝐶 𝐺𝐺𝐺𝐺 = 𝑤𝑤𝐶𝐶𝐶𝐶 .𝑚𝑚𝐺𝐺𝐺𝐺 𝑀𝑀𝐶𝐶𝐶𝐶 3.31 The molar flow of the methanized CO is equal to the molar flow of methane that is received according to Equation 3.32 [31]. 𝑛𝑛𝐶𝐶𝐶𝐶𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 = 𝑛𝑛𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 = 𝑚𝑚𝑀𝑀𝑀𝑀 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑀𝑀𝑀𝑀𝑀𝑀 3.32 In addition, to decarbonylation, CO can be formed by the side reaction as per Equation 3.15. Based on Equation 3.11 each mole of fatty acid is deoxygenated through decarboxylation without using hydrogen. Thus, for each mole of CO2 formed by decarboxylation that is converted to CO according to Equation 3.15, 1 mole of hydrogen is consumed. Since generation of 1 mole of CO by decarbonylation also consumes 1 mole of hydrogen, in order to calculate the corresponding hydrogen consumption all the generated CO is attributed to decarbonylation. Therefore, total CO2 formed by decarboxylation based on Equation 3.11 is assumed to be the amount of CO2 available in the gaseous product. The molar flow of CO2 can be estimated by Equation 3.33. 25 𝑛𝑛𝐶𝐶𝑂𝑂2 = 𝑛𝑛𝐶𝐶𝐶𝐶2 𝐺𝐺𝐺𝐺 = 𝑤𝑤𝐶𝐶𝐶𝐶2 . 𝑚𝑚𝐺𝐺𝐺𝐺 𝑀𝑀𝐶𝐶𝐶𝐶2 3.33 Based on Equation 3.12, each mole of fatty acid is deoxygenated through decarbonylation using 1 mole of hydrogen. Therefore, the hydrogen consumption by decarbonylation is estimated by Equation 3.34. 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐷𝐷𝐷𝐷𝐷𝐷 = ∑ 𝑤𝑤𝐶𝐶𝑖𝑖−1 𝐿𝐿𝐿𝐿 . 𝑚𝑚𝐿𝐿𝐿𝐿 𝑀𝑀𝐶𝐶𝑖𝑖−1 . 𝑆𝑆ℎ𝐷𝐷𝐷𝐷𝐷𝐷𝐶𝐶𝑖𝑖−1 3.34 The total hydrogen consumption due to hydrotreating of the lipid-based feedstock is estimated by Equation 3.35. 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 = 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐷𝐷𝐷𝐷 + 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐷𝐷𝐷𝐷 + 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 + 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐷𝐷𝐷𝐷𝐷𝐷 + 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 3.35 Table 10: Definition of terms in Equation 3.23 to Equation 3.35 Term Unit Denotation Description 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐷𝐷𝐷𝐷 106 mole/year Hydrogen consumption For decomposition of triglycerides in the lipid-based feedstock into the corresponding fatty acids 𝑛𝑛𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 106 mole/year Molar flow For propane in the gaseous product 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 106 mole/year Hydrogen consumption Resulted from hydrodeoxygenation 𝑤𝑤𝐶𝐶𝑖𝑖 𝐿𝐿𝐿𝐿 % Weight fraction For n-alkane with Ci carbons in the liquid product 𝑀𝑀𝐶𝐶𝑖𝑖 t/mole Molar mass For Ci n-alkane in the liquid product 𝑚𝑚𝑀𝑀𝑀𝑀 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 106 t/year Mass flow For methane formed by methanation 𝑤𝑤𝑀𝑀𝑀𝑀 % Weight fraction Related to methane in the gaseous product 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 106 mole/year Hydrogen consumption For methanation 𝑀𝑀𝑀𝑀𝑀𝑀 t/mole Molar mass Methane 𝑆𝑆ℎ𝐷𝐷𝐷𝐷𝐷𝐷 % Share of decarbonylation Share of Ci-1 n-alkanes that are received through decarbonylation 𝑛𝑛𝐶𝐶𝐶𝐶 106 mole/year Molar flow For CO produced by decarbonylation of the lipid-based feedstock 𝑛𝑛𝐶𝐶𝑂𝑂2 106 mole/year Molar flow For CO2 produced by deocarboxylation of the lipid-based feedstock 𝑛𝑛𝐶𝐶𝑂𝑂2 𝐺𝐺𝐺𝐺 106 mole/year Molar flow For CO2 in the gaseous product 𝑤𝑤𝐶𝐶𝑂𝑂2 % Weight fraction For CO2 in the gaseous product 𝑀𝑀𝐶𝐶𝐶𝐶2 t/mole Molar mass CO2 (carbon dioxide) 𝑛𝑛𝐶𝐶𝐶𝐶 𝐺𝐺𝐺𝐺 106 mole/year Molar flow For CO in the gaseous product 𝑤𝑤𝐶𝐶𝐶𝐶 % Weight fraction For CO in the gaseous product 𝑀𝑀𝐶𝐶𝐶𝐶 t/mole Molar mass CO (carbon monoxide) 𝑛𝑛𝐶𝐶𝐶𝐶𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 106 mole/year Molar flow Related to the methanized CO 𝑛𝑛𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 106 mole/year Molar flow For methane in the gaseous product 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐷𝐷𝐷𝐷𝐷𝐷 106 mole/year Hydrogen consumption For decarbonylation 𝑤𝑤𝐶𝐶𝑖𝑖−1 𝐿𝐿𝐿𝐿 % Weight fraction For n-alkane with Ci-1 carbons in the liquid product 𝑀𝑀𝐶𝐶𝑖𝑖−1 t/mole Molar mass For Ci-1 n-alkane in the liquid product 𝑚𝑚𝑚𝑚𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 106 mole/year Total hydrogen consumption Due to hydrotreating of the lipid-based feedstock 26 Energy balances The changes in the energy demand of the refining process is mainly caused by the additional hydrogen consumption due to hydrotreating the bio-share of the feedstock, which must be supplied by the HPU. The hydrogen consumption for hydrotreating lipid-based feedstock is much higher than that of fossil-based feedstock [31], [32]. On the other hand, for the common ranges of co-processing, the major part of the liquid feed to a hydrotreating unit is fossil-based. Therefore, the major change in hydrogen consumption is due to hydrotreating of bio-share. Hence, the change in the hydrogen consumption of fossil-based feedstock is neglected. The gaseous product of lipid-based feedstock contains methane and propane that can be used for hydrogen production through steam reforming. The corresponding reactions are considered as per Equation 3.36 and Equation 3.37 𝐶𝐶𝐻𝐻4 + 𝐻𝐻2𝑂𝑂 ⟶ 3𝐻𝐻2 + 𝐶𝐶𝐶𝐶 3.36 𝐶𝐶3𝐻𝐻8 + 3𝐻𝐻2𝑂𝑂 ⟶ 7𝐻𝐻2 + 3𝐶𝐶𝐶𝐶 3.37 As was mentioned under Section 2.2, reforming is followed by water-gas-shift based on Equation 2.3. Thus, steam reforming and water-gas-shift will lead to 4 moles of hydrogen for each mole of methane, and 10 moles of hydrogen for each mole of propane. Therefore, the moles of hydrogen required for hydrotreating of bio-feedstock that can be obtained by reforming the methane and propane in the gaseous product of the hydrotreating are calculated according to Equation 3.38 and Equation 3.39 described in Table 11. 𝑚𝑚𝑚𝑚𝑙𝑙𝐻𝐻𝑀𝑀𝑀𝑀 = 𝑛𝑛𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 × 4 3.38 𝑚𝑚𝑚𝑚𝑙𝑙𝐻𝐻𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑛𝑛𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 × 10 3.39 The additional hydrogen that must be produced using fossil-based feedstock to the HPU is calculated according to Equation 3.40 (terms are defined in Table 11). Since natural gas is normally fed to the HPU, which mainly consists of methane [17], the additional hydrogen is considered to be produced by steam methane reforming (SMR) as per Equation 3.36. 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 = 𝑚𝑚𝑚𝑚𝑙𝑙𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 − 𝑚𝑚𝑚𝑚𝑙𝑙𝐻𝐻𝑀𝑀𝑀𝑀 − 𝑚𝑚𝑚𝑚𝑙𝑙𝐻𝐻𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 3.40 Table 11: Definition of terms in Equation 3.38 to Equation 3.41 Term Unit Denotation Description 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 106 mole/year Hydrogen consumption To be supplied by steam reforming of fossil-based methane 𝑚𝑚𝑚𝑚𝑙𝑙𝐻𝐻𝑀𝑀𝑀𝑀 106 mole/year Hydrogen consumption To be supplied by steam reforming of methane 𝑚𝑚𝑚𝑚𝑙𝑙𝐻𝐻𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 106 mole/year Hydrogen consumption To be supplied by steam reforming of propane 𝐸𝐸𝑛𝑛𝐻𝐻𝐻𝐻𝑈𝑈𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 106 GJ/year Energy demand of HPU Related to the bio-based feedstock Δ𝐻𝐻𝑆𝑆𝑆𝑆𝑆𝑆 GJ/mole Enthalpy of reaction For SMR 𝛥𝛥𝐻𝐻𝑃𝑃𝑃𝑃 GJ/mole Enthalpy of reaction For propane reforming 𝐿𝐿𝐿𝐿𝑉𝑉𝑀𝑀𝑀𝑀 GJ/mole Lower Heating Value Related to fossil-based methane 27 The required energy to supply the additional hydrogen due to co-processing bio-based feedstock is calculated according to Equation 3.41 with terms defined in Table 11. This is regarding the point that the reforming reactions are endothermic and must be supplied by energy. The third term is related to the energy content of fossil-based methane that is converted to hydrogen. 𝐸𝐸𝑛𝑛𝐻𝐻𝐻𝐻𝑈𝑈𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 = (𝛥𝛥𝐻𝐻𝑆𝑆𝑆𝑆𝑆𝑆). (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻+𝑚𝑚𝑚𝑚𝑙𝑙𝐻𝐻𝑀𝑀𝑀𝑀) 4 + (𝛥𝛥𝐻𝐻𝑃𝑃𝑃𝑃). 𝑚𝑚𝑚𝑚𝑙𝑙𝐻𝐻𝑃𝑃𝑟𝑟𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 10 + 𝐿𝐿𝐿𝐿𝑉𝑉𝑀𝑀𝑀𝑀 × 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 4 3.41 CO2 balances The additional on-site CO2 emissions due to hydrotreating of bio-based feedstock is obtained by Equation 3.42. It is related to the emissions that arise from the additional energy requirement of the HPU. The terms in Equation 3.42 to Equation 3.46 are described in Table 12. 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻 = 𝐸𝐸𝑛𝑛𝐻𝐻𝐻𝐻𝑈𝑈𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 .𝐸𝐸𝐹𝐹𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀−𝑢𝑢𝑢𝑢 𝑓𝑓 3.42 To account for the CO2 mitigation potential of the bio-based feedstock, the associated CO2 emissions throughout the value chain are considered. From a cradle-to-grave perspective, the biogenic CO2 emissions result from uptake of CO2 from the atmosphere through photosynthesis during the growth step of the biogenic feedstock, which makes this cycle carbon neutral. The biogenic CO2 emissions formed within the processes involved in upgrading the bio-based feedstock in the refinery are, therefore, considered as neutral. The corresponding biogenic CO2 emissions, which all end up in the HPU, are quantified according to Equation 3.43. 𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻 = (𝑛𝑛𝐶𝐶𝑂𝑂2 + 𝑛𝑛𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 + 3 × 𝑛𝑛𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃) × 𝑀𝑀𝐶𝐶𝑂𝑂2 3.43 Table 12: Definition of terms in Equation 3.42 to Equation 3.46 Term Unit Denotation Description 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻 106 t/year Additional on-site emissions Related to HPU due to hydrotreating of the bio-based feedstock 𝐸𝐸𝐹𝐹𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀−𝑢𝑢𝑢𝑢 𝑓𝑓 tCO2/GJ Emission factor Related to make-up fossil fuel (f) 𝐸𝐸𝑚𝑚𝑈𝑈𝑃𝑃𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 106 t/year CO2 emissions From the user phase of bio-feedstock 𝑖𝑖 - Number of carbon atoms For Ci n-alkane in the liquid product 𝑖𝑖 − 1 - Number of carbon atoms For Ci-1 n-alkane in the liquid product 𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻 106 t/year Biogenic CO2 emissions Formed within the processes involved in upgrading the bio-based feedstock in the refinery 𝐸𝐸𝑚𝑚𝑅𝑅𝐸𝐸𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 106 t/year CO2 emissions From the resource extraction phase of bio-feedstock 𝐸𝐸𝑚𝑚𝑅𝑅𝐸𝐸𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀−𝑢𝑢𝑢𝑢 𝑓𝑓 106 t/year CO2 emissions From the resource extraction phase of make-up fossil fuel (f) 𝐸𝐸𝑚𝑚𝑅𝑅𝐸𝐸𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑠𝑠ℎ𝑎𝑎𝑎𝑎𝑎𝑎 106 t/year CO2 emissions From the resource extraction phase of the replaced fossil-based share 𝑇𝑇𝑇𝑇𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 106 t/year CO2 emissions From the value chain regarding bio- feedstock 𝐶𝐶𝐶𝐶𝐶𝐶𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 106 t/year CO2 mitigation potential Related to the bio-feedstock co- processing 28 The amount of CO2 emissions arising from bio-share of the produced fuels in the user phase can be considered as CO2 reduction in this case. This is because the biogenic share of the final fuel produced through co-processing replaces the same amount of fossil-based fuel, which means the corresponding fossil-based CO2 emissions are replaced by neutral biogenic CO2 emissions. Considering each mole of Ci and Ci-1 n- alkanes are converted to CO2 when the bio-share of the produced fuel is combusted in the user phase, the amount of associated CO2 emissions related to the user phase (𝐸𝐸𝑚𝑚𝑈𝑈𝑃𝑃𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 : 106 t/year) is derived by Equation 3.44. 𝐸𝐸𝑚𝑚𝑈𝑈𝑃𝑃𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 = �∑ 𝑖𝑖.𝑤𝑤𝐶𝐶𝑖𝑖 𝐿𝐿𝐿𝐿.𝑚𝑚𝐿𝐿𝐿𝐿 𝑀𝑀𝐶𝐶𝑖𝑖 𝑖𝑖 + ∑ (𝑖𝑖 − 1).𝑤𝑤𝐶𝐶𝑖𝑖−1 𝐿𝐿𝐿𝐿 . 𝑚𝑚𝐿𝐿𝐿𝐿 𝑀𝑀𝐶𝐶𝑖𝑖−1 𝑖𝑖 � .𝑀𝑀𝐶𝐶𝑂𝑂2 3.44 Therefore, the total value chain CO2 emissions associated with the bio-feedstock being co-processed (𝑇𝑇𝑇𝑇𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓: 106 t/year) within the system boundary of this study is obtained by Equation 3.45. 𝑇𝑇𝑇𝑇𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 = 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝐻𝐻𝐻𝐻𝐻𝐻 + 𝐸𝐸𝑚𝑚𝑅𝑅𝐸𝐸𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 + 𝐸𝐸𝑚𝑚𝑅𝑅𝐸𝐸𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀−𝑢𝑢𝑢𝑢 𝑓𝑓 3.45 The CO2 mitigation potential due to applying bio-based co-processing as the mitigation option (𝐶𝐶𝐶𝐶𝐶𝐶𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓: 106 t/year) is estimated by Equation 3.46. 𝐶𝐶𝐶𝐶𝐶𝐶𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 = 𝐸𝐸𝑚𝑚𝑈𝑈𝑃𝑃𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 + 𝐸𝐸𝑚𝑚𝑅𝑅𝐸𝐸𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑠𝑠ℎ𝑎𝑎𝑎𝑎𝑎𝑎 − 𝑇𝑇𝑇𝑇𝑚𝑚𝑏𝑏𝑏𝑏𝑏𝑏−𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 3.46 3.6. OVERVIEW OF ANALYSES THAT CAN BE CONDUCTED BY THE MODEL The model enables the following analyses for CCS and bio-based feedstock mitigation options. Carbon Capture and Storage: Based on the procedure described in Section 3.3, the on-site CO2 emissions from each unit process and in total are calculated. Therefore, the emissions associated with unit processes can be monitored by the user to decide whether to deploy CCS in all process units or in selected units. Thereafter, based on the selected units to which CCS is to be applied, the total on-site CO2 emissions captured, the CO2 mitigation potential, and the energy consumption related to CCS, can be quantified according to the procedure described in Section 3.4. In addition, the model can account for the opportunity to use available excess heat at the refinery to supply the heat demand of the CC unit. It should be noted that the amount of excess heat available in a refinery could be affected by co-processing bio-feedstock. Therefore, the option to use available refinery excess heat, which is estimated in the absence of co-processing of bio-feedstock, should be considered only when CCS is the only mitigation option, and not in combination with the bio-feedstock mitigation option. 29 Bio-based feedstock: The module is based on the procedure described in Section 3.5. The type of bio-based feedstock, the fraction of bio-based feedstock in the feed to the unit as well as the targeted product need to be defined as inputs to the model. The general idea of the analysis is that the bio-feedstock composition is supposed to be determined by the model considering the type of bio-feedstock specified by the user. Then the general idea is that based on the type of bio-feedstock and products, the model is supposed to select the insertion point of the bio-feedstock (as discussed previously, different types of feedstock are suitable for different insertion points) and the associated reaction sets taking place within the upgrading process of the bio-feedstock to biofuels. Due to limited time, the scope of the work was restricted to one product and one suitable insertion point. As described previously, the change in energy demand of the units in focus resulting from co-processing bio-fedstock is then calculated, followed by estimating the associated CO2 mitigation potential. Combinations of mitigation options: The model provides the opportunity to combine the two mitigations options to derive the total CO2 mitigation potential and changes in the energy demand. 30 3.7. INPUT DATA TO THE MODEL Important input parameter values are defined in Table 13. In this research, a case study is conducted on Preemraff Lysekil refinery, therefore the input variables are based on the corresponding data from this refinery, which are not revealed due to confidentiality. It should be noted that all the given input data listed in Table 13 can be revised by users based on the specific data of any certain refinery and other generic data according their sources of data collection. Each module is modelled based on the method described for the corresponding module. Table 13: Input parameter values to the model Input parameters to the model Unit Value Reference 𝐹𝐹𝐹𝐹𝑢𝑢,𝑓𝑓- Consumption of fuel f in process units a 106 GJ/year Confidential Preemraff Lysekil ∑ ∑ 𝐸𝐸𝐸𝐸𝑢𝑢 × 𝑆𝑆ℎ𝐸𝐸/𝑂𝑂𝑂𝑂𝐸𝐸𝐸𝐸𝑢𝑢 - Electricity consumption of all process units b GWh/year 522 [8] 𝑉𝑉𝑢𝑢,𝑏𝑏𝑏𝑏 - volume flow of refinery by-product bp, used as a fuel in each process unit u 106 Sm3/year Confidential Preemraff Lysekil 𝑉𝑉𝑝𝑝 - volume of each refinery product, p 106 Sm3/year Confidential Preemraff Lysekil 𝐸𝐸𝐹𝐹𝑓𝑓/ 𝐸𝐸𝐹𝐹𝑏𝑏𝑏𝑏/𝐸𝐸𝐹𝐹𝑝𝑝 - CO2 emission factor kg CO2/GJ Grid electricity: 13.1 Liquefied Natural Gas (LNG): 57 Cracker coke: 103 Gasoline: 72.6 MK1 diesel: 72 Diesel: 75.3 Heavy fuel oil: 79 LPG: 65.1 [33] [34], [35], [36] [34], [35], [36] [37] [38] [37] [39] [37] CRu- CO2 capture rate t/t reference emission (%) 85 [6], [40] Specific energy requirements and CO2 emissions (EmCS) for compression, transport and storage of captured CO2 -based on permanent storage beneath seabed off Norwegian coast tCO2/tCO2 captured MWh//tCO2 0.019-0.024 (avg.: 0.021) 0.4-0.5 (avg.: 0.45) Adapted from [41], [33] [41] FP- CO2 capture fuel penalty Based on: - CC heat demand - CC boiler efficiency GJ/t CO2 GJ/t CO2 (%) 4.5 3.3-4.4 (avg.: 3.85) 85 [6] [42] 31 Global volume-weighted average upstream (well-to- refinery) carbon intensity gCO2eq/MJ 10.3 [43] SpecEmMake−up f- Specific resource extraction emissions for LNG -based on Higher Heating Value (HHV) gCO2eq/MJ HHV 18.3 [44] Crude oil input to the refinery 106 GJ/year 416.2 Adapted from [8] Heat sources to supply CC unit: - Steam rates - Operation hours t/h h/year Confidential 8500 A study on Preemraff Lysekil Relative hydrogen consumption L/kg Rapeseed oil 294 [45] 𝑌𝑌𝐿𝐿𝐿𝐿- Estimated liquid product yield related to hydrotreating of rapeseed oil wt% 94 [29], [32], [46], [31] Y𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤- Estimated water yield related to hydrotreating of rapeseed oil wt% 5 [45], [46], [32] Estimated compositions of liquid and gas products wt% Appendix A Adapted from [47], [48] Specific emissions from extraction phase to production of refined rapeseed oil kgCO2eq/t refined rapeseed oil 262 [49] mtot- Mass flow of the liquid input to the hydrotreating unit process 106 t/year Confidential Preemraff Lysekil Composition of rapeseed oil in terms of fatty acids wt% Table 6 [29], [30] a Fuel f is inclusive of fossil fuels and the byproducts of the refinery that can be used as fuels for different unit processes (e.g. fuel gas) b Power source can be electricity purchased from grid or electricity generated on-site by combined heat and power (CHP) or steam turbines. 32 4. RESULTS AND DISCUSSION The model was constructed based on the methodology described in the previous sections. Thereafter the model was applied to quantify CO2 emissions and analyze the effect of mitigation options including CCS, bio-feedstock co-processing, and combinations thereof in terms of changes in the refinery energy demand and CO2 mitigation potential. The effect of recovering excess heat to satisfy the heat demand of the CC unit was also investigated. The results are presented in this chapter. 4.1. CO2 EMISSIONS OF THE HYDROGEN PRODUCTION UNIT (HPU) Based on the specified rate of input feed and corresponding fuel consumptions, on-site CO2 emissions in flue gases were quantified using the CO2 quantifying module. In order to validate the model, the estimated amount of CO2 emissions was compared with the corresponding site data for the Preemraff refinery [15] in Table 14. It can be concluded that the amount of estimated emissions is satisfactorily in agreement with the real data. Table 14: Comparison between CO2 emissions from HPU estimated by the model and the real site data 4.2. CO2 MITIGATION POTENTIAL AND ENERGY CONSUMPTION OF CC TECHNOLOGY FOR THE HPU The CO2 mitigation potential and required energy for CC technology were calculated using the CCS module. Input data was taken from a previous confidential study on the Preemraff Lysekil refinery regarding availability of excess heat which could be recovered to supply heat to the CC unit. The results regarding the CO2 mitigation potential and energy requirement of CC technology for the two cases of supplying CC technology with/without available excess heat are shown in Table 15. The share of the CO2 mitigation potential regarding the total on-site CO2 emissions from the Preemraff Lysekil refinery [7] are also shown. On-site CO2 emissions from HPU Estimated by the model Site data 106 t/year 0.624 0.6 33 Table 15: CO2 mitigation potential and energy requirement of CC technology, applied to the HPU, with and without the effect of using excess heat The CC unit’s energy demand adds around 62% to the total energy consumption of the HPU in the Preemraff Lysekil refinery, which is supplied by 2 MWof electricity [50], 29.4 MW of fuel gas [50], and offgas from PSA that is calculated based on the available site data. The use of excess heat reduces or eliminates the required primary energy supply for CC. As can be seen, in this case it was sufficient to fully cover the energy demand of CC. Therefore, the emissions related to the CC fuel penalty (defined in Equation 3.9) are eliminated, leading to a considerable increase in the CO2 mitigation potential. As can be seen in this case, the percentage of the HPU on-site emissions mitigated is slightly below 85%, which is the carbon capture rate applied. This is because there are some emissions associated with the compression, transport and storage, which are subtracted from the total carbon captured. By applying CC technology to the HPU, a significant share of the total on-site emissions of the refinery can be mitigated, and the mitigation potential can also be increased by using the available excess heat for supplying energy to the CC technology. 4.3. CO2 MITIGATION POTENTIAL AND ENERGY CONSUMPTION OF CC TECHNOLOGY FOR THE REFINERY The case study on the effect of excess heat on CO2 mitigation potential and the primary energy requirement of the CC unit was extended to investigate the effect of applying CC technology to the total on-site CO2 emissions from the Preemraff Lysekil refinery. Note that the input data to the model does not include data for all refinery process units, therefore, the value of the total on-site emissions was retrieved from Preem’s sustainability report [7]. The CO2 mitigation potential and the CC energy demand were estimated by the CCS module for the two cases of supplying CC technology with/without excess heat. The maximum available excess heat in the refinery that can be used to supply the CC unit was considered, which is around 2.3 × 106 GJ/year that approximately corresponds to 73 MW (a conservative value considering the CC operating conditions estimated based on [51]). The results are shown in Table 16. It can be seen that utilizing excess heat for this purpose can cover around 43.5% of the CC primary energy demand. This leads to a 13% increase in the CO2 mitigation potential of the CC technology. CCS mitigation option Total on-site emissions (106 tCO2/year) HPU on-site emissions (106 tCO2/year) CO2 mitigation potential (106 tCO2/year) Percentage of HPU on-site emissions mitigated (%) Percentage of total on-site emissions mitigated (%) CC primary energy consumption (106 GJ/year) Without excess heat 1.625 0.624 0.334 53.5 20.6 2.404 With excess heat 1.625 0.624 0.52 83.3 32 0 34 Table 16: CO2 mitigation potential and energy requirement of CC technology, applied to the refinery, estimated by the model and the effect of using excess heat on them 4.4. CO2 MITIGATION POTENTIAL OF CO-PROCESSING RAPESEED OIL IN THE HYDROTREATING UNIT Co-processing was evaluated for a feedstock mix consisting of rapeseed oil and light gas oil (LGO) with a ratio of 17:83 wt%. The hydrotreating was assumed to be conducted at 340 ˚C and 4 MPa over NiMo/Al2O3, which is a common commercial catalyst for hydrotreating. The results for CO2 mitigation potential, change in energy demand for the refining process as well as the amount of biogenic CO2 emissions generated within the process units involved in upgrading the bio-based feedstock are summarized in Table 17. Table 17: Model outcomes for co-processing of rapeseed oil and LGO in the hydrotreating unit Co-processing rapeseed oil and LGO CO2 mitigation potential (106 tCO2/year) Increased energy demand for HPU (106 GJ/year) Biogenic CO2 emissions (106 tCO2/year) 17:83 wt% 0.82 0.23 0.043 Compared to the results for CCS applied to the HPU described under Section 4.2, the co-processing mitigation option has a CO2 mitigation potential that is around 2.5 times higher whereas the increase of refining energy demand is around 9.6% of the corresponding value for the CCS option applied only to the HPU. This is to a large extent due to the fact that co-processing has reduction effect on the emissions in the user phase (Scope 3), which accounts for the major contribution to the decrease of value chain emissions. 4.5. INTERPLAY BETWEEN CCS AND CO-PROCESSING OF RAPESEED OIL In order to evaluate the combined effect of the CCS and co-processing mitigation options, a number of test points were considered, as defined in Table 18. The High value for CCS assumes that the technology is applied to the total on-site emissions from the HPU. The Low value assumes that the technology is applied to a fraction of the HPU flue gases for which the CC energy demand can be covered by excess heat from the refinery’s heat collection network (HCN). . This was considered as a basis to define CCS mitigation option Total on-site emissions (106 tCO2/year) CO2 mitigation potential (106 tCO2/year) CO2 mitigation potential (% of total on-site emissions) CC primary energy consumption (106 GJ/year) Percentage of CC primary energy consumption covered by excess heat (%) Without excess heat 1.625 0.868 53.4 6.256 0 With excess heat 1.625 1.079 66.4 3.533 43.5 35 a low value for CCS. However, in the analysis only the aforementioned fraction (62.2%) is applied as the Low value, and the energy for the CC technology was not satisfied by the HCN to be able to monitor the pure effect of the CC. The High value for the rapeseed oil co-processing corresponds to the indicative va