Model for a maritime logistic supply chain for captured CO2 in Sweden. Performance of a quantitative analysis of CO2 ship transport systems in a Swedish context by establishing a suitable model tested in collaboration with Swedish stakeholders Master’s thesis in Mobility Engineering Gemma Bruguera Matute Antonio Juan Castro Molinero DEPARTMENT OF MECHANICS AND MARITIME SCIENCE CHALMERS UNIVERSITY OF TECHNOLOGY Gothenburg, Sweden 2023 www.chalmers.se www.chalmers.se Master’s thesis 2023 Model for a maritime logistic supply chain for captured CO2 in Sweden. Performance of a quantitative analysis of CO2 ship transport systems in a Swedish context by establishing a suitable model tested in collaboration with Swedish stakeholders Gemma Bruguera Matute Antonio Juan Castro Molinero Department of Mechanics and Maritime Sciences Division of Maritime Engineering Chalmers University of Technology Gothenburg, Sweden 2023 Research of a logistic supply chain for captured CO2 within Sweden. Performance of a qualitative and quantitative analysis of ship CO2 transport systems in a Swedish context by establishing a suitable model tested in collaboration with Swedish stakeholders. GEMMA BRUGUERA MATUTE, ANTONIO JUAN CASTRO MOLINERO © GEMMA BRUGUERA MATUTE, 2023. © ANTONIO JUAN CASTRO MOLINERO, 2023. Supervisor/Examiner: Per Mottram Hogström, Department of Mechanics and Maritime Sciences, Chalmers Supervisor: Hannes von Knorring, DNV Co-supervisor: Stein Ove Erikstad, Department of Marine Technology, NTNU Master’s Thesis 2023 Department of Mechanics and Maritime Sciences Division of Maritime Engineering Chalmers University of Technology SE-412 96 Gothenburg Telephone +46 31 772 1000 Cover: Liquefied CO2 carrier Ulle Clover rendering visualisation. Typeset in LATEX, template by Kyriaki Antoniadou-Plytaria Printed by Chalmers Reproservice Gothenburg, Sweden 2023 iv Model for a maritime logistic supply chain for captured CO2 in Sweden. Performance of and quantitative analysis of ship CO2 transport systems in a Swedish context by establishing a suitable model tested in collaboration with Swedish stakeholders. GEMMA BRUGUERA MATUTE, ANTONIO JUAN CASTRO MOLINERO Department of Mechanics and Maritime Sciences Chalmers University of Technology Abstract This master’s thesis report presents the results of the research carried out for the design of an optimal marine logistic chain for CO2 transportation in Sweden. The effectiveness of CO2 transport systems is related to the output of the Carbon Capture (CC) facilities, inland feeder traffic, intermediate storage hubs and type and cost of liquefaction plants among others. The solution is reached by means of a mathematical model built up so that the optimal fleet logistic can be found for a specific defined scenario, i.e., number of vessels that compounds the fleet, sizes of the ships in terms of its cargo capacity and sailing speed. Additionally, a cost breakdown of the mentioned logistic chain is presented in order to identify the origin and percentage of each variable contributing to the total cost. Results from this study indicate that the largest contributor to the CCS chain considered is the liquefaction process, followed by the fleet cost. In addition, medium pressure transport is always the most expensive option in comparison with low pressure transport and using the largest ship capacity is almost always more economical. This is because the increase in the cost of buffer storage is not as great as the increase in the cost of the fleet due to the increase in the number of ships per fleet. Keywords: CCS, CCU, maritime logistics, CO2, maritime decarbonization, Sweden, sup- ply chain. v Acknowledgements I am grateful to Chalmers University of Technology for the opportunity to conduct my research and for all the academic and personal knowledge I have gained during my two- year Master’s degree. I would also like to express my sincere gratitude to my supervisor at Chalmers, Per Mottram Hogström and my supervisor at DNV, Hannes von Knorring, for their guidance and support throughout the master’s thesis. To my family for all the trust they have placed in me over the years and for the uncon- ditional support they have given me, as well as my friends. Finally, I express my gratitude to Xavier Fernández for the optimism that he has trans- mitted to me during this process and in my life. Gemma Bruguera Matute, Gothenburg, June 2023 I would like to express my gratitude to both universities, NTNU (Trondheim, Norway) and Chalmers University (Gothenburg, Sweden) for giving me the opportunity to live such an insightful experience, personally, academically and professionally, over the last two last years. As a former Nordic Master student, I will always be grateful. To my supervisors Per Mottram Hogström (Chalmers University), Hannes von Knorring (DNV) and Stein Ove Erikstad (NTNU) for the confidence placed in me, their help, guidance and their valuable feedback. To my parents and my brother for being my role models and for their effort in supporting me in many different ways. For their encouragement since I started this long journey which ends today, leading by example every single day, how far perseverance can take you. Last but not least, to my partner, Luisa, for inspiring me and for being my daily support over these two years living abroad. Antonio Juan Castro Molinero, Gothenburg, June 2023 List of Acronyms Below is the list of acronyms that have been used throughout this thesis listed in alpha- betical order: CAPEX Capital Expenses CC Carbon Capture COP Convention Of the Parties CCS Carbon Capture and Storage CCS-RI Carbon Capture and Storage Readiness Index CCU Carbon Capture and Utilisation CCUS Carbon Capture, Utilisation and Storage CII Carbon Intensity Indicator CO2 Carbon dioxide DC Direct Cost DWT Deadweight Tonnage EEDI Energy Efficiency Design Index EEXI Energy Efficiency for Existing Ships Index EU ETS European Union Emissions Trading System GHG Greenhouse Gas GWP Global Warming Potential HP High Pressure IMO International Maritime Organization LCO2 Liquefied CO2 LNG Liquefied Natural Gas logiCO2 Maritime logistic supply chain for captured CO2 LP Low Pressure LPG Liquefied Petroleum Gas MARPOL International Convention for the Prevention of Pollution from Ships MEPC Marine Environment Protection Committee MIP Mixed Integer Program MP Medium Pressure NLS Noxious Liquid Substances NPV Net Present Value OPEX Operating Expenses ROA Real Option Approach SEEMP Ship Energy Efficiency Management Plan TDC Total Direct Cost TPC Total Plant Cost TTW Tank-to-Well UN United Nations UNFCCC United Nations Framework Convention on Climate Change VLSFO Very Low Sulphur Fuel Oil WHO World Health Organisation WTT Well-to-Tank ix Nomenclature Below is the nomenclature of indices, sets, parameters, and variables that have been used throughout this thesis. Parameter Name Unit α Annual payment €m ρ Cargo density t/m3 P Pending loan €m j interest rate - m Loan return period y To Operational Time h Trt Round-trip Time h Tmoor Mooring Time h Tumoor Unmooring Time h Tpin Port Entry Time h Tpout Port Exit Time h Tl Loading Time h Tu Unloading Time h Tb Buffer Storage Time h Tc Cargo transportation temperature ◦C Pc Cargo transportation pressure bar d Sailing distance km v Sailing speed knots xi Number of ship type - cxi Ship capacity kt cb Buffer storage capacity kt a Total amount of CO2 to be transported within the planning horizon Mt axi amount of CO2 transported by ship type i during the planning horizon Mt Rb Flow rate of the buffer storage t/h CH Hiring Cost €m CS Sailing Cost €m CB Buffer Storage Cost €m CBO Buffer Storage Operational Cost €m xi CBCH Buffer Storage CAPEX in the planning horizon €m C CAPEX of the capacity under consid- eration €m Co CAPEX for the reference capacity €m S Capacity under consideration MtCO2/y So Reference capacity MtCO2/y n Scaling exponent - CL Liquefaction Cost €m CLC Liquefaction Investment Cost or Lique- faction CAPEX €m CLV O Liquefaction Variable Operating Cost €m CLF O Liquefaction Fixed Operating Cost €m CR Reconditioning Cost €m CRC Reconditioning Investment cost or Re- conditioning CAPEX €m CRF O Reconditioning Fixed Operating Cost €m CRV O Reconditioning Variable Operating Cost €m CT Total Cost €m xii Contents List of Acronyms ix Nomenclature xi List of Figures xv List of Tables xvii 1 Introduction 1 1.1 Background and problem description . . . . . . . . . . . . . . . . . . . . . 1 1.1.1 Current and upcoming regulations in shipping industry . . . . . . . 2 1.1.2 CCUS and shipping industry . . . . . . . . . . . . . . . . . . . . . . 3 1.2 Scope and objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.3 System boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.4 Outline of the report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.5 Limitations and assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2 Literature review 9 2.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.2 State of the art . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.3 Methodology for logistic analysis . . . . . . . . . . . . . . . . . . . . . . . 10 2.4 Techno-economic assessment . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.5 Legal aspect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.6 CCS potential in Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 3 Methodology 19 3.1 Model overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.2 Technical Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 3.2.1 Buffer storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.2.2 Pressure transport . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.2.3 Ship characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 3.2.4 Shipping logistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 3.3 Cost assessment methodology . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.3.1 CO2 ship cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.3.1.1 Hiring cost . . . . . . . . . . . . . . . . . . . . . . . . . . 29 3.3.1.2 Sailing cost . . . . . . . . . . . . . . . . . . . . . . . . . . 31 3.3.2 Buffer storage cost . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.3.3 Liquefaction and reconditioning process cost . . . . . . . . . . . . . 35 3.3.3.1 Liquefaction process cost . . . . . . . . . . . . . . . . . . . 35 3.3.3.2 Reconditioning process cost . . . . . . . . . . . . . . . . . 38 4 Results 41 4.1 Validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 4.2 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 4.3 Scenarios prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 4.4 Scenario 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 xiii Contents 4.5 Scenario 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 4.5.1 Stockholm - Kollsnes . . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.5.2 Västerås - Kollsnes . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 4.5.3 Södertälje - Kollsnes . . . . . . . . . . . . . . . . . . . . . . . . . . 54 4.6 Scenario 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 4.6.1 Nynäshamn - Kollsnes . . . . . . . . . . . . . . . . . . . . . . . . . 57 4.6.2 Stockholm - Nynäshamn - Kollsnes . . . . . . . . . . . . . . . . . . 58 4.6.3 Västerås - Nynäshamn - Kollsnes . . . . . . . . . . . . . . . . . . . 60 4.6.4 Södertälje - Nynäshamn - Kollsnes . . . . . . . . . . . . . . . . . . 62 4.7 Comparison between Scenario 2 and 3 . . . . . . . . . . . . . . . . . . . . 64 5 Discussion 67 6 Conclusion 71 7 Further work 73 Bibliography 75 A Tables I B Figures V xiv List of Figures 1.1 Energy supply and its CO2 emissions by sector . . . . . . . . . . . . . . . . 2 1.2 CC cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.3 System Boundaries for CCS supply chain under study . . . . . . . . . . . . 6 2.1 Offshore pipeline and shipping benchmark . . . . . . . . . . . . . . . . . . 11 2.2 Break-even point for ship and pipeline transport cost . . . . . . . . . . . . 12 2.3 CAPEX for CO2 ships available in literature. . . . . . . . . . . . . . . . . . 14 2.4 Ship construction cost data and corresponding fitting curves for low and medium pressure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.5 Swedish industries by sector and four main clusters . . . . . . . . . . . . . 17 3.1 Techno-Economic Assessment Approach . . . . . . . . . . . . . . . . . . . 19 3.2 Workflow overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 3.3 Workflow for the functioning of the model . . . . . . . . . . . . . . . . . . 21 3.4 Phase diagram of pure CO2 based on the Span and Wagner equation of state. 23 3.5 CO2 shipping cost model. . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.6 Workflow for CAPEX calculation. . . . . . . . . . . . . . . . . . . . . . . . 30 3.7 Depicted workflow of inputs, constraints, variables and steps to follow in order to calculate the OPEX. . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.8 Buffer Storage Investment Cost from Literature Review. . . . . . . . . . . 34 3.9 Workflow for buffer storage costs calculation . . . . . . . . . . . . . . . . . 35 3.10 Workflow for liquefaction costs calculation . . . . . . . . . . . . . . . . . . 37 3.11 Workflow for reconditioning costs calculation . . . . . . . . . . . . . . . . . 39 4.1 CCS projects in Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 4.2 Potential hubs system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 4.3 Route: Gothenburg - Kollsnes. . . . . . . . . . . . . . . . . . . . . . . . . . 46 4.4 Gothenburg - Kollsens breakdown of total cost. . . . . . . . . . . . . . . . 48 4.5 Gothenburg - Kollsnes breakdown liquefaction cost. . . . . . . . . . . . . . 48 4.6 Gothenburg - Kollsnes breakdown of ship cost by fuel type. . . . . . . . . 49 4.7 Route: Stockholm - Kollsnes. . . . . . . . . . . . . . . . . . . . . . . . . . 51 4.8 Stockholm - Kollsnes breakdown total cost. . . . . . . . . . . . . . . . . . . 52 4.9 Route: Västerås - Kollsnes. . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 4.10 Västerås - Kollsnes breakdown total cost with vessel size limitation. . . . . 54 4.11 Södertälje - Kollsnes breakdown total cost with vessel size limitation. . . . 55 4.12 Scenario 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 4.13 Scenario 3: Nynäshamn - Kollsnes Route . . . . . . . . . . . . . . . . . . . 57 4.14 Stockholm - Nynäshamn (hub) - Kollsnes breakdown total cost. . . . . . . 60 4.15 Västerås - Nynäshamn (hub) - Kollsnes breakdown total cost with vessel size limitation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 4.16 Södertälje - Nynäshamn (hub) - Kollsnes breakdown total cost. . . . . . . . 64 4.17 Comparison of costs in relation to the route (direct route or via intermediate storage) considering vessel size restrictions. . . . . . . . . . . . . . . . . . . 65 4.18 Comparison of costs in relation to the route (direct route or via intermediate storage) without considering infrastructure vessel size restrictions. . . . . . 65 xv List of Figures 5.1 Ship capacity and number of vessels vs flow rate. The lines correspond to the ship capacity axis and the bars to the number of vessels axis. . . . . . 67 5.2 Cost of the CCS chain as a function of flow rate for low pressure transport considering two distances and with and without vessel size restrictions. . . 68 B.1 Cost of the CCS chain as a function of flow rate for both transport pressure considering two routes from Stockholm to Kollsnes. . . . . . . . . . . . . . V B.2 Cost of the CCS chain as a function of flow rate for both transport pressure considering two routes from Västerås to Kollsnes. . . . . . . . . . . . . . . V B.3 Cost of the CCS chain as a function of flow rate for both transport pressure considering two routes from Södertälje to Kollsnes. . . . . . . . . . . . . . VI xvi List of Tables 1.1 Condition of CO2 at different stages of the CCS chain. . . . . . . . . . . . 8 3.1 Inputs and constraints applied in the model . . . . . . . . . . . . . . . . . 20 3.2 Pressure condition factors . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 3.3 Ship capacities for low and medium pressure option. . . . . . . . . . . . . . 24 3.4 Ship logistics parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 3.5 Overview of the cost methodology. . . . . . . . . . . . . . . . . . . . . . . 27 3.6 Ship construction cost assumptions used in the model. . . . . . . . . . . . 29 3.7 CO2 tax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 3.8 Fuels costs and GWP forecast. . . . . . . . . . . . . . . . . . . . . . . . . . 32 3.9 Buffer storage cost assumptions used in the model . . . . . . . . . . . . . . 34 3.10 Reference Investment Costs for liquefaction process presented in €m. . . . 36 3.11 Variable operational cost used as reference value for liquefaction process. . 37 3.12 Reference investment costs for reconditioning process . . . . . . . . . . . . 38 3.13 Variable operational costs used as a reference value for reconditioning process. 38 4.1 Validation of the results by comparison . . . . . . . . . . . . . . . . . . . . 41 4.2 Studied scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 4.3 Vessel size limitations for considered ports . . . . . . . . . . . . . . . . . . 46 4.4 Fleet costs and logistics Gothenburg - Kollsnes . . . . . . . . . . . . . . . . 47 4.5 Scenario 2: Emitter’s flow rate. . . . . . . . . . . . . . . . . . . . . . . . . 50 4.6 Fleet costs and logistics Stockholm - Kollsnes . . . . . . . . . . . . . . . . 51 4.7 Fleet costs and logistics Västerås - Kollsnes . . . . . . . . . . . . . . . . . . 53 4.8 Fleet costs and logistics Södertälje - Kollsnes. . . . . . . . . . . . . . . . . 55 4.9 Scenario 3: Distances between emitter and intermediate storage . . . . . . 57 4.10 Fleet costs and logistics Nynäshamn - Kollsnes (second leg) . . . . . . . . . 58 4.11 Fleet cost and logistics Stockholm - Nynäshamn (first leg) . . . . . . . . . 59 4.12 Fleet cost and logistics Stockholm - Nynäshamn - Kollsnes . . . . . . . . . 59 4.13 Fleet cost and logistics Västerås - Nynäshamn (first leg) . . . . . . . . . . 61 4.14 Fleet cost and logistics Västerås - Nynäshamn - Kollsnes . . . . . . . . . . 61 4.15 Fleet cost and logistics Södertälje - Nynäshamn (first leg) . . . . . . . . . . 63 4.16 Fleet cost and logistics Södertälje - Nynäshamn - Kollsnes . . . . . . . . . 63 A.1 System boundaries and methodology from the literature review . . . . . . . I A.2 CCS Projects in Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . . . II A.3 Ships main dimensions relation . . . . . . . . . . . . . . . . . . . . . . . . III xvii List of Tables xviii 1 Introduction Although weather deviations can be provoked naturally (i.e., via variations in solar cycle), it is observable that since the 1800s, humankind is the main contributor of climate change due to use of fossil fuels, such as coal, oil or gas (United Nations, n.d.). According to Hausfather and Friedlingstein (2022), atmospheric carbon dioxide (CO2) concentrations are 51% above pre-industrial levels and is the largest contributor to green- house gas emissions. In order to achieve global climate goals, emissions need to decrease rapidly rather than just stabilise. Therefore, to fight against climate change and thrust the transition towards a sustainable development, Carbon Capture and Storage (CCS) and Carbon Capture and Utilisation (CCU) are increasingly gaining importance among industries and political organisms. Although Carbon Capture (CC) technology is already available, further improvements in terms of processes optimisation, new infrastructure (or adaptation of the existing one) and supply chain logistics need to be carried out. This master’s thesis presents an study and optimisation of a maritime logistic supply chain of Carbon Capture, Utilisation and Storage (CCUS) in Sweden. 1.1 Background and problem description Different definitions for climate change are provided by organisations: The World Health Organisation (WHO) in turn defines climate change as long lasting variation in time on the climate, due to natural reasons or caused by humankind activity (WHO, 2016). It is considered by United Nations (UN) as the deviation of the weather patterns and temperature in a long-term basis (United Nations, n.d.). Climate change threatens people with food and water scarcity, increased flooding, extreme heat, more diseases, and economic losses. Human migration and conflict can be a result too. Communities may adapt to climate change through efforts like coastline protection or expanding access to air conditioning, but some impacts are unavoidable (WHO, 2016). Poorer countries are responsible for a small share of global emissions, yet they have the least ability to adapt and are most vulnerable to climate change (United Nations, n.d.). The shipping industry contributes in an important extent to global climate change, dete- riorating the atmosphere with CO2 emissions. Nevertheless, due to its own international nature, it is not an option to reduce it or stop it, since world economy would be deeply affected (Karin Andersson, 2016). Therefore, new solutions must be found in order to maintain maritime trade at current levels (or increasing it) while reducing its climate footprint. In section 1.1.2 the shipping strategy for a decarbonization path is presented. 1 1. Introduction In order to have a general picture of the current situation, in Fig.1.1 it can be seen the (a) global energy supply and (b) the CO2 by sector. Specifically, the two figures illustrate the distribution of the total energy and the contribution of CO2 emissions by each sector, respectively. (a) (b) Figure 1.1: (a) Energy supply and (b) its CO2 emissions by sector. (Adopted from Ishaq et al., 2022) It can be seen in Fig. 1.1 how the global dependence in non-renewable source of energy is high and thus the emissions related to it. The goal is to change, within the near future, the source from which the needed energy is harvested focusing on renewable energies, such as wind, solar or hydro-power. Nevertheless, this is not an easy transition that can be carried out in an immediate way, since energy demand is higher than what, currently, renewable energies can provide. Therefore, some transitional solution must be found to start reducing CO2 concentration in the atmosphere and avoid keep emitting more as the global economy, industry and transportation processes keeps operating. A solution found is to capture and store this CO2, thus avoiding emissions into the atmosphere. To do so, a new market needs to raise in order to be able to deal with such big gas amounts and transported it from emitters to receivers. CO2 can not only be stored, but used to help in different industrial processes, as it has been used in oil or food industry (Al Baroudi, Awoyomi, Patchigolla, Jonnalagadda, & Anthony, 2021). Moreover, currently, there are ongoing projects that requires CO2 to produce new greener fuels as it can be e-Methanol (Liquid Wind, 2023) with a clear perspective of growing in the next following years. 1.1.1 Current and upcoming regulations in shipping industry An environmental friendly operation must be achieved by the fleet in charge of the trans- portation of the CO2, as this carried and sequestrated amount cannot be with counterbal- ance with the vessels emissions. Therefore, some background on what regulations must be fulfilled by the liquefied CO2 carriers is presented. Throughout the history of shipbuilding industry, regulations, standards and laws have been gradually established. Regarding policies related to climate, the most important achievements are listed below: 2 1. Introduction • In 1973 parties of the International Maritime Organization (IMO) agreed to the the International Convention for the Prevention of Pollution from Ships (MARPOL), which currently consists of six Annexes. • In 1992 is created in Rio de Janeiro the United Nations Framework Convention on Climate Change (UNFCCC), who established a diplomatic framework for the negotiation of further protocols regarding Greenhouse Gas (GHG) concentrations in the atmosphere. Nevertheless, did not specifically mentioned maritime GHG emissions. • Amendments to MARPOL Annex VI that carried out the regulation for making the Energy Efficiency Design Index (EEDI) and Ship Energy Efficiency Management Plan (SEEMP) mandatory were first discussed during the 58th Marine Environ- ment Protection Committee (MEPC58) and adopted at the MEPC62 held in 2011, entering into force from 1st of January of 2013. • In 2015, the Paris agreement is developed during the 20th Convention Of the Parties (COP21). It aims to pursue efforts to keep the temperature increase below 1.5◦C compared to preindustrial levels (UNFCCC, 2023). In addition, IMO establish a regulation for new vessels consisting on calculating their ship’s EEDI, which must be less than the prescribed for that type of ship by the IMO. This is applied to vessels with Gross Tonnage1 GT ≥ 400, for which the legislation is put in action on 1st of January of 2023. Some exceptions are contemplated. When it comes to existing (and new) vessels, they shall carry on board a SEEMP Karin Andersson (2016), which consist on: • A specific plan for each ship that establishes a management process for implementing energy efficiency measures for the ship’s operations and a process for continuous improvement. • It should incorporate best practices for the energy efficient operation of ships, such as improvements in speed management throughout a voyage or the management of route optimisation that have been previously analysed. Furthermore, other indexes that must be taken under consideration for existing ships is the Energy Efficiency for Existing Ship Index (EEXI) and the annual operational Carbon Intensity Indicator (CII). The former is related to the technical design of the vessels based on a reduction factor with respect to EEDI and should be predicted from the technical department of a shipping company. The later aims to a continuous improvement in the operational carbon intensity of the ship by monitoring, documenting and verifying against the required annual CII (IMO, 2019). 1.1.2 CCUS and shipping industry It is worth of mention that when waste is transported (i.e., liquefied CO2), it must be assured that the industrial activity (transport in this case) will not emit more CO2 into the atmosphere than what it is removing. Therefore, this environmental regulations are of major importance to be met by the logistic activity. 1Nonlinear measure of a ship’s overall internal volume 3 1. Introduction In the light of this, a shift towards a sustainable lifestyle of humankind based on renewable energies is necessary in order to cope with future regulations. Nevertheless, a transition period is needed before the required infrastructure and technology are available to be able to rely entirely and solely on green energy. Hence, CCSU is a suitable solution since it could be ready to use as a mature technology in the near future. North Europe (mainly Norway and United Kingdom), together with United States of America, Canada, Australia and China are leaders in CCS Readiness Index (CCS-RI), meaning that they lead the capability for creating an environment that enables commercial deployment of CCS (Havercroft & Consoli, 2018). Regarding Scandinavian countries, four Liquefied CO2 (LCO2) tankers under the Nor- wegian flag are currently operating. In addition, the Northern Light project2 is planned to start operating in 2024. Therefore, there is an emerging need for efficient and cost- effective CO2 transportation methods between emitters and receivers or storage facilities. Moreover, according to the Swedish government, a strategy for negative greenhouse gas emission (SOU 2020:4) is required. Therefore, supplementary measures are needed to achieve the net zero emission plan in 2045 reaching negative emissions thereafter (Bier- mann et al., 2022). One of these measures is CCUS, which, as it can be seen in Fig. 1.2, is divided in several phases: Figure 1.2: CC cycle. *Stored CO2 can subsequently be used for industry processes. **In an ideal future, greenfuels will feed transportation methods. First, CO2 is produced as a result of the normal operation of industry and cities. This CO2 is then captured3 and transported from the emitters via pipeline (in a gaseous state) or by means of a train and/or truck logistic chain (in a liquid state). Onshore transportation considerations are further explained in the following chapters. Consequently, once LCO2 is available at the port, vessels in charge of its offshore transportation are loaded and ready to head either to CCS or CCU receivers. CCS receivers use depleted oil reservoirs to permanently store the CO2, while CCU use CO2 as a raw material to achieve new products such as fuels. 2First planned infrastructure designed to build up a cross-border, open-source CO2 transport and storage network. Phase I is designed to handle 1.5 MtCO2/y. Phase II is projected to deal with a total of 5 MtCO2/y. In addition, a 7500 m3 LCO2 carrier will be operating in 2024 (Northern Lights, 2023). 3Currently several technologies exists to capture CO2. Nonetheless, what they are and how they work are out of the scope of this thesis. 4 1. Introduction Therefore, specific plan is needed to achieve profitable logistic results for the vast range different particular situations depending on CO2 amount to be transported, location of sources and sinks, size of needed fleet, buffer storage capacity, etc. 1.2 Scope and objective The purpose of this thesis is to design an optimal maritime logistic chain for water- borne CO2 transportation in Sweden. Hence, given inputs such as amount of CO2 to be transported and distance between harbours, which is the most cost-effective way of trans- portation in terms of number of required vessels, as well as its technical characteristics, such as cargo capacity and sailing speed. The aim is to be able to know the optimal solution for a specific case. Every company or facility have its own characteristics in terms of location, sailing restrictions (e.g., maximum length, beam and/or draft), and amount of CO2 emissions. Therefore, the research questions to be answered by this thesis are: • Which is the optimal maritime logistic solution for a particular case emitter under its unique circumstances in terms of location and amount of CO2? • What is the cost of this solution? To that end, a mathematical model is developed so that it can be applied to any route. The inputs of the model are the distance between two ports, the CO2 flow rate to be transported and the fuel with which the ship operates. These parameters can be ajusted according to the case under study. The output of the model is the optimal maritime logistic solution from an economic point of view. That is to say, the answer would pro- vide the number of vessels that compose the fleet, their technical characteristics, such as cargo capacity and sailing speed and the number of round voyages they have to complete annually. Answering the research question is intended also to provide understandings and insights of which barriers already exist in the CCUS market in Europe in general and in Sweden in particular. 1.3 System boundaries In this study, the CCS chain is divided in three parts: (1) The CO2 liquefaction process, (2) the shipping supply chain, and (3) the CO2 reconditioning process. The shipping supply chain considers buffer storage and shipping and is added to the model through a techno- economic assessment (Sections 3.2 and 3.3). With regards to the CO2 liquefaction and reconditioning process, these are added to the model by means of an economic assessment (Section 3.3.3) based on data obtained from the literature review. It is assumed that the CO2 liquefaction process takes place at the carbon capture facility and then transported by ship either (a) to the final port, where it is reconditioned and injected into final storage (Fig. 1.3a), or (b) to a hub and from there to the final port (Fig. 1.3b). In addition, for both scenarios the shipping supply chain is always considered between two ports. This can be seen in Fig. 1.3 where the system boundaries for both scenarios are shown. 5 1. Introduction (a) Direct Route. (b) Considering a hub. Figure 1.3: System Boundaries for CCS supply chain under study The liquefaction plant at the emitting source is assumed to be as described in Deng, Roussanaly, and Skaugen (2019). This study performs a techno-economic analysis of the liquefaction process considering ten different pressures as an output. According to the aforementioned study, the liquefaction process can be divided as follows: 1. CO2 compression train, 2. pre-cooler, liquefier and flash tank, 3. recirculation flash and compressor, 4. ammonia refrigeration cycle. Moreover, CO2 enters the CO2 compression train at 1 bar and 40◦C after capture and exits at the desired pressure. The present study, as it is explain in Section 3.2.2, considers 7 bar and 15 bar pressure, therefore, the data for these pressures are used. During this study, the term low pressure is used to refer to 7 bar and -49.1◦C and medium pressure to refer to 15 bar and -28.2◦C, these values can be seen in Table 3.2 and Fig. 3.4. For both scenarios, after the liquefaction process, the CO2 is send to the buffer storage where it is stored in liquid form at the port terminal and loaded onto the ship via cargo handling system. Then, it is transported by ship to the receiving port terminal which for scenario (a) would be the final port terminal while for scenario (b) is an intermediate port terminal, known as hub. In scenario (a), CO2 is unloaded and stored in the terminal port and then reconditioned to the necessary conditions for pipeline injection into the final storage site. For scenario (b), CO2 is unloaded to the intermediate port and then loaded to another ship that sails the route from there to the final port, where the CO2 is unloaded and stored and later reconditioned in the same way as scenario (a). Note that the CO2 capture process is not part of the system. Additionally, the cost of loading and unloading CO2 is not included in the model as it is a small percentage compared to the other costs (Orchard et al., 2020). 6 1. Introduction Furthermore, all scenarios assume that pure CO2 is transported, as impurities vary among emitters, and that CO2 is liquefied after capture at the emitting facility. This implies that the CO2 transport from the liquefaction facility to the buffer storage is in liquid form and therefore, arrives by truck, train or ship. Nonetheless, for simplicity and because this configuration may vary from case to case and transport by ship is not always an option, the transport, if needed, between the liquefaction facility to the buffer storage is not considered in the model. It is worth mentioning that carrying out the liquefaction process at the emitting facility is not always the most optimal or viable option, as for facilities outside the port terminal but sufficiently close to it, pipeline transport may be more optimal. In such cases, the liquefaction takes place at the port terminal and the CO2 would arrive at the terminal at high pressure (∼ 90 bar), so the condition before liquefaction is different from the condition assumed in the model, and hence the liquefaction cost changes. 1.4 Outline of the report This report is divided as follows. The current chapter introduces the research question and the background to the project by providing the motivation and scope of the project. Additionally, the limitations and the boundaries of the model are presented. In order to answer the research question and develop the model, the following approach is followed. Firstly, a literature review (Chapter 2) is performed to give an understanding of the studies conducted to date on CO2 transport. It can be seen that numerous studies focus on CO2 transport by pipeline and that more research is needed on CO2 transport by ship. From there, data is collected and analysed to identify the main inputs required for the model to perform the techno-economic assessment presented in Chapter 3. The technical assessment is presented in Section 3.2, in which a logistics model according to the defined inputs is develop, and the economic assessment is presented in Section 3.3. Then, the mathematical model is applied to a three specific scenarios and the results are shown in Chapter 4. Finally, the discussion, conclusions and further work are presented in Chapter 5, Chapter 6 and Chapter 7, respectively. 1.5 Limitations and assumptions Limitations: • Estimated fuel consumption of main and auxiliary engines by means of similar ex- isting ships and empirical formulas. • No differentiation between biogenic and fossil CO2. • Vessel main particulars estimated by means of empirical formula. • Fuel cost is set to the average between upper and lower bounds. Assumptions: • Assumed that the ship is connected to the shore grid while on harbour. Hence no fuel consumption during docking time. 7 1. Introduction • The amount of CO2 to capture will be stable in the near future. • No difference in fuel consumption for ice class vessels or when sailing across ice. • Both main and auxiliary engines use the same fuel. • Ice class vessels same price as non-ice class ones. • Table 1.1 shows the assumptions made regarding the condition of CO2 at different stages of the CCS chain. Table 1.1: Condition of CO2 at different stages of the CCS chain. Condition before Liquefaction Transport Condition Condition after Reconditioning 1 barg 20◦C Low Pressure 100 barg 5◦C 1 barg 20◦C Medium Pressure 100 barg 5◦C 8 2 Literature review Although multiple studies regarding CCS technology are carried out, much remains to be studied regarding its practical implementation. Studies also need to bring together experience from existing projects and establish logistic chain models. In this chapter, the available information in the literature is gathered in order to have an overview of the cur- rent market and research field situations by presenting the new technologies, specifically in regards to waterborne transportation, and the costs and challenges associated to it. 2.1 Background Over the years, carbon dioxide (CO2) has been transported through pipelines, specially in North America due to the availability of large quantities and clean natural sources of CO2 (Neele, Haugen, & Skagestad, 2014). A pipeline infrastructure to transport CO2 requires a continues flow of compressed gas, has a high cost-distance dependency (Al Baroudi et al., 2021) and requires a high initial capital investment (Element Energy, 2018). Yet another option that is increasingly attractive is the ship transport mode. Both technologies, pipelines and vessels, are affected by the economies-of-scale, of which pipelines does it to a greater extent (Bennæs et al., 2022; Kjärstad, Skagestad, Eldrup, & Johnsson, 2016; Knoope, Ramírez, & Faaij, 2015). This means that the more volume of CO2 available for transportation, the more cost-efficient. Different studies (Decarre, Berthiaud, Butin, & Guillaume-Combecave, 2010; Kjärstad et al., 2016; Neele et al., 2014; Roussanaly, Jakobsen, Hognes, & Brunsvold, 2013; Skagestad, Mathisen, Henrik Eldrup, & Aksel Haugen, 2011) consider pipeline transport of CO2 a more cost-effective option when distances1 are relatively short and volumes are considerably high. Carbon dioxide has been also transported in small quantities, between 800 m3 and 1200 m3, by ship for food industry (Al Baroudi et al., 2021; Decarre et al., 2010; Neele et al., 2014; Sköldberg et al., 2021). Studies on bigger scale of liquid CO2 transported by ship began in the early 2000s (Aspelund, Mølnvik, & De Koeijer, 2006; ?). Aspelund et al. (2006) was one of the firsts to study the viability of ship-based transport of CO2 and the study was carried out for the Nordic countries. The study claims that a feasible econom- ical large-scale ship transport of CO2 could be carry out by means of semi-pressurised vessels near the triple point (-52◦C and 6.5 bar), what would also help to avoid dry ice formation in the loading and unloading processes. The research foresees two main op- tions for the (un)loading of the cargo. Either by means of a Submerged Turret Loading (STL) unloading offshore, a direct terminal to terminal service, or a combination of both 1To take into account the distant ratio which is defined by the geographical context (e.g., derouting of the pipeline to avoid mountains, natural reserve, etc.) 9 2. Literature review technologies. Since then, several studies have been carried out to develop a cost-effective large-scale CCS chain. These studies focus on different areas, such as technical aspects, methodology to be applied and economic and techno-economic assessments of the CCS chain. 2.2 State of the art Technical assessment studies (Element Energy, 2018; Roussanaly, Brunsvold, & Hognes, 2014; Roussanaly et al., 2013) focus on the defined system, i.e., elements of the CCS chain, such as loading and offloading equipment, intermediate storage, ship design or liquefaction. With regard to ship design, ? and Aspelund et al. (2006) presents a 20000 m3 capacity vessel at a pressure around the triple point, i.e., low pressure, while Decarre et al. (2010) presents a 30000 m3 capacity vessel where CO2 is transported at -30◦C and at a pressure of 15 bar, i.e., medium pressure (Table 3.2). More recently, Roussanaly, Deng, Skaugen, and Gundersen (2021) carried out a depth comparison between these two CO2 shipping transport conditions with a volume fixed at 20 MtCO2 and a distance of 2000 km and concluded that the most cost-effective option is to transport CO2 at 7 bar. Nonetheless, it should be noted that all these studies (Aspelund et al., 2006; Decarre et al., 2010; Roussanaly et al., 2021; ?) have been conducted with a fixed capacity/volume at a fixed distance, which are parameters that have a large influence on the model as it will be discussed further in Chapter 3. As for intermediate buffer storage, based on LNG shipping experience, Yoo et al. (2013) suggest a storage capacity 20% above the ship capacity (i.e., 120%). On the other hand, Berger, Kaarstad, and Haugen (2005) assumed the capacity of the temporary storage as 150% of the ship capacity while Bjerketvedt, Tomasgard, and Roussanaly (2020) concludes that taking into consideration uncertainties, such as weather delays the most efficient solution is to have an intermediate storage capacity of 118% since it would avoid the increased cost of recovering from delays by increasing the sailing speed. Several Norwegian shipping companies state that they experience longer delays, like the extreme weather scenario, for short periods of time during a normal winter season. The STAwave-method2 of estimating travelling times may be a source of error in extremely heavy weather. This, in combination with planned and unplanned maintenance, could cause an increase on the buffer capacity (Bjerketvedt et al., 2020). Nevertheless, if the risks of higher future fuel prices and ship breakdowns is taken into account, buffer capacity will rise up to 173%. 2.3 Methodology for logistic analysis Regarding methodology studies, J. Jakobsen, Tangen, Nordbø, and Mølnvik (2008) in- troduces a methodology for CO2 logistic chain analysis to contribute to a cost-efficient development of large-scale CCS logistic chain. Based on that, other studies regarding methodology have been carried out with support from the BIGCCS Centre and performed under the Norwegian research program Centres for Environment-friendly Energy Research 2Is related to the calculation of the added resistance due to waves. STAwave-1 is a correction method for ships that may suffer of a limited pitch and heave. STAwave-2 is a correction method with frequency response achieved by empirical methods (Magnussen, 2017). 10 2. Literature review (FME) (J. Jakobsen et al., 2011; J. P. Jakobsen, Roussanaly, Brunsvold, & Ananthara- man, 2014; J. P. Jakobsen, Roussanaly, Mølnvik, & Tangen, 2013). In J. P. Jakobsen et al. (2014), the methodology is called iCCS and is intended to provide a tool in which a multiple techno-economic and environmental criteria assessment is carried out to evaluate different CCS chains. Studies such as Roussanaly et al. (2013), Roussanaly et al. (2014) and Roussanaly et al. (2021) use the mentioned methodology to carry out the studies. Moreover, Nam, Lee, Lee, and Chung (2013) uses a mixed integer program (MIP) to op- timise the cost of CO2 transport by ship by optimally locating the liquefaction plant and determining the optimal fleet size, ship type and service frequency of the route to cover the annual CO2 to be transported between the established sources and sinks. Bennæs et al. (2022) also uses a MIP model to determine the optimal design of a European supply chain for CCS presenting a new MIP model for the Ship-Based CCS Logistics Problem (SCLP). This new MIP model assumes that the CO2 arrives to the ports through pipeline in a pressurised state so the CO2 is liquefied and stored at the port before being transported. 2.4 Techno-economic assessment Additionally, economic and techno-economic assessment studies (Coussy, Roussanaly, Bureau-Cauchois, & Wildenborg, 2013; Decarre et al., 2010; Mathisen, Skagestad, Eldrup, & Haugen, 2013; Roussanaly et al., 2014, 2013) focus on the development of an economic and efficient model by comparing the costs3 between transporting CO2 by pipeline and by ship often as a function of distance and cargo capacity. Kjärstad et al. (2016) present in his study the break-even point as a function of distance and volume between both means of transport, while (Roussanaly et al., 2014) shows the benchmark between off- shore pipeline and shipping as transport means to an offshore reservoir. The results for both studies can be seen respectively in Fig. 2.1 and Fig. 2.2. Figure 2.1: Offshore pipeline and shipping benchmark to an offshore reservoir. (Adopted from Roussanaly et al., 2014). 3Costs for ships can be divided in five main parts: 1) Cost of the chosen fleet (crew and sailing costs); 2) storage costs (investment and operating costs); 3) liquefaction; 4) CO2 reconditioning (i.e., process to meet up the conditions required for its transport, and/or injection, such as dehydration, cooling, etc); and 5) loading/unloading. 11 2. Literature review Figure 2.2: Break-even point for ship and pipeline transport cost, €/tCO2, as a function of yearly transport volume and distance. (Adopted from Kjärstad et al., 2016). The figures above show which is the best transport option for a given capacity-distance ratio, as well as where the boundary between the two options lies. Models in literature often follow a Net Present Value (NPV) approach for the economic calculations and thus disregarding the flexibility and uncertainties4 present in the pro- cess. NPV model assumes that volumes, costs, and revenues of a project are known over 4According to Knoope et al. (2015), typical key uncertainties are: 1) Storage volume of the reservoir; 2) coal price; 3) electricity price; 4) fuel oil price; 5) CO2 price; 6) utilisation rate of the emissions source plant (may decrease in the future due to renewable energies). 12 2. Literature review the whole project duration and adaptation after the investment decision is not required. Although the project costs and revenues are the same, different political nature, leads to different discount rates and thus different NPV. For example, a national authority use a lower discount rate than an oil company, since the former one is dealing with higher risks (Roussanaly et al., 2014). In reality there are uncertainties and companies will need adapt to changing situations. Therefore, Knoope et al. (2015) (in agreement with Roussanaly et al. (2014) and Coussy et al. (2013), recommends to take into account uncertainties in the investment costs in order to reach more accurate results) carried out a study applying a Real Option Approach (ROA), where flexibility is taken into account to calculate a more accurate cost- effectiveness of the project. As a result, the study shows that the certain specific cases would be profitable when a ROA is applied, while NPV approach states the opposite. This is of major importance since adding the value of flexibility can prevent a profitable project from being dismissed if the NPV gives a negative result. A general conclusion is that, unlike pipeline infrastructure, CO2 transport by ship does not require a large initial investment. Hence, it could be an attractive option for early CCS logistic chain during the ramp-up phase (Kjärstad, Skagestad, Eldrup, & Johnsson, 2015; Kjärstad et al., 2016; Knoope et al., 2015; Nilsson, 2014). Therefore, ship-based transportation mode could simplify and speed up the CCS infrastructure deployment in the decision making process phase by offering a lower investment threshold. Hence, stakeholders could be more willing to invest despite the high initial financial risk5 (even in the absence of government involvement). The main reason is the lower CAPEX and the characteristic flexibility of ship transport compared to pipelines (Nilsson, 2014). CAPEX for CO2 ships available in literature is gathered in Fig. 2.3 by Knoope et al. (2015). Such study, also affirms that, on the contrary, OPEX are around 50% larger for ship transport compared to pipelines. Orchard et al. (2020) also collects the investment costs from the literature (see Fig. 2.4) and, as it is more recent, it is the one used for the estimation of CAPEX in this research, as explained in Chapter 3. Decarre et al. (2010) states that in the case of offshore storage, i.e., from port to offshore storage, transport by ship is a more economical option than pipeline for distances over 350 km, and in the case of port-to- port for distances over 1100 km. Nevertheless, Yoo et al. (2013) carries out an economic assessment showing that CO2 shipping can play an important role in transporting large volumes of CO2 from various sources, even in distances between 200 and 300 km. 5The first-of-a-kind effect, which is unavoidably attached to uncertainties along the whole investment process. This could lead to higher investment costs due to, for instance, sub-optimal design, additional construction costs, delays, underutilised pipelines over the first years, etc. (Roussanaly et al., 2014) 13 2. Literature review Figure 2.3: CAPEX for CO2 ships available in literature. (Adopted from Knoope et al., 2015). Figure 2.4: Ship construction cost data and corresponding fitting curves for low and medium pressure. (Adopted from Orchard et al., 2020). The literature shows the costs of each process within the whole logistic chain. According to d’Amore, Romano, and Bezzo (2021), the total cost of the transportation phase stands for the 6-18%, while Bennæs et al. (2022) present a 35-40% of the total costs for liquefaction6. Within transportation costs, Knoope et al. (2015) present that a 20-30% is directly related to the ship and 70-80% is in relation to harbour facilities, liquefaction unit, offloading and conditioning equipment. It is worth to mention that the liquefaction costs are barely dependent of the mass flow, since the energy consumption per tonne of CO2 is nearly constant, while the remaining costs decrease with increasing mass flow. Furthermore, 6Liquefaction is directly related to the cost of the electricity. Thus, a rise in electricity price results in an increase of the total costs of the ship-based CCS logistics system 14 2. Literature review Bjerketvedt et al. (2020) claims that the conditioning and shipping are the main costs of the transportation and gives some quantitative data for a particular case: Conditioning has a cost of 15.4 €/tCO2, where the cost of electricity reach half of it with 7.8 €/tCO2. When it comes to the shipping, the cost is of 13.8 €/tCO2, from which 4.1 €/tCO2 corresponds to the fuel cost. The total energy cost for these processes is 11.8 €/tCO2. Finally, the remaining costs of the transport chain are due to the buffer storage (concept introduce before in this chapter) with 2.9 €/tCO2 and the reconditioning at 1.7 €/tCO2. The optimal size and capacity of vessels are closely related to the expected volume of CO2 to transport and this is linked to a seasonal variation. Hence, the question arises as to what will be the most economically optimal; whether to have a reduced number of vessels but with a larger cargo capacity, or instead to have a larger number of vessels with a more modest cargo capacity. Bjerketvedt et al. (2020) present the answer to this question: based on the achieved results, he states that the variation in CO2 emissions due to the change of seasons lead to the option of larger ship, since the increased fuel consumption caused by its size is offset with the fact that smaller ships would need to increase their power and speed in order to comply with the larger CO2 volumes during peak season. Furthermore, the study claims that the seasonal storage of the CO2 is not a cost-efficient strategy in any case compared to increasing the vessel capacity. Therefore, it is proved that the seasonal variations in emissions in the source has a relevant impact on the optimal design of the transport chain. Weather conditions is also directly related to seasons i.e., harsh sailing conditions are more often during winter. For instance, larger wave heights are expected during these periods. As a result, the ship speed would decrease, being small ships more sensitive to these events compared to larger ones. Thus, yet another reason in favour of big vessels (without forgetting to include to this list the aforementioned concept of economies-of- scale). One further matter to have into account when studying a new infrastructure, is always the new risks that these new technologies and processes may cause on personnel dealing with them in particular and society in general. Aspelund et al. (2006) carried out a hazard identification (HAZID) and a preliminary hazard and operability (HAZOP) analyses. As an example, shows that since CO2 is heavier than air, the later one would be displace to upper layers. Hence, gas detectors must be installed to avoid potential suffocation of crew members in a leakage event. Regarding the societal risk7, mitigation procedures are not required in order to obtain an acceptable risk level (d’Amore, Mocellin, Vianello, Maschio, & Bezzo, 2018). The costs for maintaining societal risk levels below dangerous levels is lower than 0.41€/tCO2, that is the 11% of the overall transport costs. Therefore, the implementation of mitigation measures within an European CCS SC shall not be an economic obstacle when planning a safe transport infrastructure. 2.5 Legal aspect Regarding the legal aspects of CCS, Nilsson (2014) introduced the main points to have into account: 7Societal risk is defined as the health risk that a number of people (population in this case) are exposed to that is triggered by certain hazardous incident in a defined region. 15 2. Literature review • The European Union Emissions Trading System (EU ETS) does not mention specif- ically transport of CO2 by ship. The current solution is an integration of vessels into the ETS on a case-by-case basis, which is a cumbersome, high time consuming and costly process for individual project operators. • Sequestrate CO2 in a sub-seabed geological reservoir is prohibited for the Parties to the London Dumping Protocol. • The CCS Directive prohibits the storage of CO2 8 in any reservoir that extends beyond the territory of an EU member. • Currently, biogenic CO2 is not covered by the EU ETS, which is counter-productive since this biogenic CO2 would contribute to the financial viability by increasing the total CO2 volume available and thus benefit from the economies-of-scale. • Swedish government position regarding biogenic CO2 is that it should receive equal treatment as fossil fuels CO2. Moreover, Swedish government claimed that “ship transport of CO2 is a likely prerequisite, at least initially, for making CCS commer- cially interesting”. Thus, the author recommendations are to 1) give a clearer definition for the term “cap- tured CO2”; 2) more consideration regarding potential market failures shall be shown; and 3) define and constrain accurately the role and competence of each authority in the CCS infrastructure building up process. 2.6 CCS potential in Sweden As for the Nordic countries, in particular Sweden, CO2 emissions come from paper, steel, cement, refineries and chemical industries, which are considered small-medium sized indus- tries as the emissions are typically between 0.1-1.0 MtCO2/year. In addition, the distances between sources to potential storage sites are relatively long, 300 km or, in many cases, considerably more (Kjärstad et al., 2016). Thus, CO2 shipping is often a more suitable transport mode. Moreover, many industries are located on the coast line and the volumes of CO2 that need to be transported are not high. Therefore, hubs are needed in order to form clusters of near small emitters so that enough CO2 could be gathered in order to carry out a profitable process (Kjärstad et al., 2015, 2016; Nilsson, 2014). Sköldberg et al. (2021) takes into consideration Swedish emissions from different industries and identifies, based on location and emissions size, four main clusters for local collaboration. Clusters such as Västkusten/Vänern, Skåne/Danmark, Ostkusten/Mälardalen and Gävle area. It could be less costly to transport the CO2 by ship 800-1300 km further to the west than sequestrating it in a nearest reservoir (e.g., within the Baltic Sea) due, for instance, to bad injectivity9, which is believed to be limited as it is the storage capacity in the Baltic sea (Kjärstad et al., 2016; Nilsson, 2014). Notice that higher injectivity capacity means less costs €/tCO2. 8CO2 emitted by the combustion of organic material. 9Defined as the pressure differential of certain reservoir pressure that is required in order to be able to inject a unit volume of fluid in a given unit of time. Although it can be expressed by means of any combination of pressure, volume and time, is typically defined by psi per barrel per day (psi/bbl/day) (Law Insider, n.d.). 16 2. Literature review Additionally, Karlsson, Normann, Odenberger, and Johnsson (2023) presents an study of the current situation in Sweden in terms of emitters location, and optimal potential hubs sites so that the total emissions within Sweden could be gathered for its marine transportation in a cost-efficient way. In this regard, a future scenario is also studied to account with the market evolution, where more emitters will join to the logistic chain. It takes into account both types of CO2 emissions, i.e., fossil and biogenic, and how both could be handled, either separately or jointly. This matter is further discussed in Section 4.2 and used to support the selected approach in this thesis. Figure 2.5: Swedish industries by sector and four main clusters. (Adopted from Kjärstad et al., 2016). On the one hand, as seen throughout the literature review, several studies have focused on optimising transportation models based on techno-economic assessments comparing transportation by ship and transportation by pipeline. On the other hand, studies to date that consider CO2 transport by ship focus on the technical aspects and optimisation of the whole chain. That is, as instance, how to optimise the liquefaction process or at what temperature and pressure it is most feasible and economical to transport CO2 by ship, however, the transport itself is set as a fixed parameter to the model. Hence, there is no a consistent optimised model for the transportation of CO2 by ship from sources to sinks. As mentioned, CCS projects in Sweden are at an early stage, industries are considered to be small to medium size and thus the CO2 production would not be available in large quantities if single emitters are considered, at least for the time being, and many of them are located along the coast. In addition, initiatives that aim to use CO2 are emerging, like Liquid Wind (2023) who is interested in using the captured CO2 for methanol production. Although CO2 transport by pipeline is, as stated by the International Energy Agency (2020), a mature and established technology, and large-scale CO2 transport by sea is in 17 2. Literature review the demonstration phase, looking into the near future and taking into account the facts that have been established throughout the literature review and presented before in this chapter, it can be concluded that CO2 transport by ship is the most favourable option for Sweden in the near future. Thus, the next step in taking CCS to large scale is to optimise a logistics model to trans- port CO2 by ship between different points of interest (sources and sinks), taking into account the volume to be transported, distances, ship capacities and hubs, producing the best transport model for different situations. Hence, according to the available litera- ture the two main parameters that define each source characteristics for carrying out an economic and feasibility studio are the rate of CO2 volume produced and available for transport at the source, and the distance from it to the injection site. At the view of the gathered information, it can be found some interesting gaps in the literature that motivate this thesis topic: • Not enough research on a cost-effective optimised solution for a maritime generic- ship based CO2 transport can be found. • Lack of mathematical models to study any potential scenario in terms of sea route, source and sink location and volume of CO2 to be transported. 18 3 Methodology In the current chapter, the methodology followed to develop the model in order to answer the research questions found in Chapter 1 is presented. The purpose of the model is to determine the optimal maritime logistics of a CCS chain in Sweden. To that end, a techno-economic assessment is needed. First, the technical components considered for the development of the model are presented (Sections 3.1 and 3.2). The following is the cost evaluation assessment (Section 3.3) for each stage of the CCS chain according to the system boundaries shown in Fig. 1.3. Figure 3.1 shows the methodology followed for the techno-economic assessment, the first half being the technical part and the second half the economical part. Figure 3.1: Techno-Economic Assessment Approach 3.1 Model overview In this section is presented how the model (i.e., logiCO2) is built up, and how all pa- rameters, variables, inputs and constraints are intertwined. Figure 3.2 depicts a general overview of the followed workflow. Figure 3.2: Workflow overview It shall be remarked that, as it can be seen in Fig. 3.2, the inputs that the mathematical 19 3. Methodology model use are the loading and unloading rate, buffer cost, CAPEX and OPEX. It is of highly importance to note for a full understanding of the system under discussion that, liquefaction and reconditioning costs are not parameters that the model takes into account for defining the optimum fleet. However, such costs must be added in order to get to know the total cost of the construction and operation of the required infrastructure. Each considered scenario have its own characteristics. These are, primarily, its location, which will further define the sailing distance and the amount of CO2 gathered, identify as I.01 and I.02 respectively. Additionally, a potential constraints for certain location is the ship size limitations, represented by C.04. Moreover, the type of fuel used can be defined by means of C.03. Hence, under certain conditions and specifications, inherent to a specific scenario, inputs for the model are defined for running the calculation to achieve the most cost-effective fleet solution. Finally, as already mentioned, the liquefaction and reconditioning costs are added to the fleet cost to get to know the total cost. The model box is explained and illustrated in detail in the following section, so that the complex relation between all components, parameters, inputs and constraints can be presented. The built model is shown in Fig. 3.3. For full understanding of the meaning of the numerous inputs and constraints that can be seen in the figure, refer to Table 3.1. Table 3.1: Inputs and constraints applied in the model Inputs (I) 1. Sailing distance d km 2. Amount of CO2 to be transported a t/y 3. Ship operational time To h/y 4. Ship lifetime - y 5. Planning horizon - y 6. Loading time tl h 7. Unloading time tu h 8. Port entry time tpin h 9. Port exit time tpout h 10. Mooring time tmoor h 11. Unmooring time tumoor h 12. Cargo density ρLP [t/m3] 13. Cargo transportation temperature Tc [◦C] 14. Cargo transportation pressure Pc [bar] Constraints (C) 1. Vessel cargo capacity c t 2. Vessel speed v kn 3. Fuel type - - 4. L, B and T limitations - - 5. Cargo pressure - bar 6. Cargo temperature - ◦C 7. Buffer storage capacity cb m3 20 3. Methodology Figure 3.3: Workflow for the functioning of the model As shown previously in Fig. 3.2, the chosen scenario define the main inputs (i.e., dis- tance between ports and amount of CO2 to be transported). Hence, according to these parameters, the cost for every single ship cargo capacity listed in Tab. 3.3, with a range of three speeds (i.e., 10, 14 and 16 knots) for each capacity, is calculated so that the most cost-effective solution can be identified among them. First step is to calculate the round trip time, for which distance between harbours, port entry and exit time, mooring and unmooring operations are taken into account as well as vessel cargo capacity and speed. In addition, the previously calculated loading and unloading time is considered so that the round trip time is known. Secondly, it is necessary to know the number of ships that make up the fleet, which will be the most restrictive solution among the two approaches defined below: 21 3. Methodology • Dependent on the buffer storage capacity: As mentioned in section 3.3.2 the buffer storage capacity is a 18% above the vessel cargo capacity to account to uncertainties as it can be delays due to weather or port congestion. Thus, the calculated round trip time is used so that relating it with the CO2 volume to transport, and the buffer storage capacity, the minimum number of vessels needed to comply with required port arrivals rate to avoid the overfilling of the buffer storage is achieved. • Dependent on the voyage time: This would be the other possible constriction that will define a minimum number of vessels. It cannot be less that the minimum needed fleet size to cope with the total amount of CO2 to be transport in a defined period. In this study this period is one year, thus, the required quantity of ships are those that can deal with the total amount of CO2 that is emitted in one year at the defined scenario. For this approach, vessel cargo capacity and speed are taken into account, as well as the CO2 volume to be transported and the ship operational time, assumed to be 8400 hours per year. The voyages per ship is then calculated by dividing the number of ships and the fleet voyages per year, which is a function of the CO2 volume, the vessel cargo capacity and the ship operational time. The voyages per ship is then used to get to know the net sailing time (i.e., disregarding mooring time where it is assumed that main and auxiliary engines are not being utilised) by relating it to the loading and unloading time and the ship operational time. Now, the net sailing time (i.e., main and auxiliary engine are running) is used to calculate the sailing costs, which are directly dependent on the ship operational time and the variable OPEX calculated in section 3.3.1.2. By adding the annual CAPEX and the annual fixed OPEX defined by Drewry (2021) and relating them to the sailing costs, and to the number of ships, it is thus calculated the whole fleet cost per year. Finally, to estimate the total cost, the fleet cost shall be related to the buffer cost, lique- faction and reconditioning cost, the CO2 amount and the planning horizon. 3.2 Technical Modelling This section addresses the first two steps of Fig. 3.1, i.e, the technical analysis. The technical modelling in this study takes into consideration the shipping supply chain (the middle part in Fig. 1.3), that is, technical aspects regarding transport pressure, buffer storage and ship characteristics. With regards the CO2 liquefaction and reconditioning process are added into the model just from an economical point of view and are assessed in Section 3.3. Hence, from a technical approach and taking into consideration the mentioned aspects, the model provides an optimised shipping logistics to transport an amount of CO2 between two ports. Therefore, the main inputs of the model are the distance and the CO2 flow rate, as well as the fuel used to operate the ship. Nonetheless, the latter only has an economic impact and is assessed in Section 3.3 22 3. Methodology 3.2.1 Buffer storage In this study the shipping supply chain consists manly of buffer storage and shipping and is located between the liquefaction and the reconditioning processes in the CCS chain under consideration, as can be seen in Fig. 1.3. The model assumes that both processes are continuous, whereas the shipping logistics is considered discontinuous. Therefore, intermediate buffer storage is needed in the supply chain to integrate the whole chain. As shown in Chapter 2, the studies conducted to date consider different capacities for buffer storage. Yoo et al. (2013) and Berger et al. (2005) suggest a storage capacity of 20% and 50% above the ship capacity, respectively, while Element Energy (2018) considers a buffer storage capacity of 20% above the ship fleet. Nonetheless, the present study, following Bjerketvedt et al. (2020), assumes that a total buffer storage of 18% above vessel capacity at each port is sufficient to account for uncertainties such as weather delays. Furthermore, intermediate buffer storage and shipping logistics depend on optimal trans- port pressure, required fleet size (number of CO2 tankers), and their mix (ship types or capacities) to transport a certain amount of CO2 from emitter to permanent storage. The amount of CO2 to be transported and the distance are input to the model, while the vessel capacity, number of vessels needed and speed are outputs. Therefore, as explained in more detail below, different vessel pressures, capacities and speeds are considered. 3.2.2 Pressure transport Figure 3.4 illustrates the phase diagram of pure CO2 and, as can be seen, at atmospheric pressure it exists only in a gaseous or solid state (Orchard et al., 2020). Therefore, to transport CO2 in liquid state, it requires pressurisation and it can be from the triple point to the critical point. Within this range, three main shipping pressures are highlighted: low pressure (LP), which corresponds to pressures close to the triple point, medium pressure (MP), which corresponds to pressures around 15 bar, and high pressure (HP), which corresponds to pressures around the critical point. Figure 3.4: Phase diagram of pure CO2 based on the Span and Wagner equation of state. (Adopted from Deng et al., 2019). 23 3. Methodology As described in Section 2.2, low pressure and medium pressure are the most discussed in the studies carried out, whereas high-pressure condition is not considered a cost-effective option due the high cost and low volumes (Orchard et al., 2020; Roussanaly et al., 2021; Seo, Huh, Lee, & Chang, 2016). With regards low-pressure shipping, it has been studied from a research perspective in studies such as Kather and Engel (2018) and Roussanaly et al. (2021). These studies state that is more cost-efficient than medium pressure, how- ever, the technology is yet to be proven. In contrast, medium-pressure shipping does have application in industry as CO2 is transported at this pressure on a small scale in food industry, and the technology is therefore more mature and ready. Nonetheless, CO2 trans- port by ship at medium pressure poses a constraint on scaling up to accommodate further growth of CCS, whereas low-pressure transport enables significantly greater capacities, even though there is limited practical experience. Thus, the present work considers low-pressure and medium-pressure conditions for the optimisation model. Table 3.2 shows the density and the pressure used in the model for low and medium pressure conditions. As can be seen, CO2 at medium pressure transport condition has a lower density than at low pressure condition, which results in a less efficient storage. According to Brevik engineering AS (2020) a higher density increases the storage efficiency by around 10% compared to medium pressure transport condition. Table 3.2: Pressure Condition factors. (Deng et al., 2019). Factor Low Pressure Medium Pressure CO2 density [kg/m3] 1133 1042 Pressure [bar] 7 15 Temperature [◦C] -49.1 -28.2 3.2.3 Ship characteristics As mentioned above, low-pressure shipping allows bigger ship capacities than medium- pressure shipping. Based on Roussanaly et al. (2021) and Element Energy (2018), the maximum capacity for the medium-pressure ship option is set at 10 ktCO2/ship. As with the current tank configuration available in the industry, a higher capacity is unlikely to be feasible due to pressure limiting the practical CO2 diameter tank. Moreover, for the low-pressure option, the maximum capacity is 50 ktCO2/ship as there is no reliable cost data for ships with bigger capacities (Roussanaly et al., 2021). Table 3.3 shows all the capacities under consideration. Table 3.3: Ship capacities for low and medium pressure option. Low Pressure transport Capacities [ktCO2] Medium Pressure transport Capacities [ktCO2] 2.5, 5.0, 7.5, 10, 12.5, 15, 20, 25, 30, 35, 40, 45, 50 2.5, 5.0, 7.5, 10 Furthermore, a ship carrying CO2 is, according to MARPOL (IMO, 1983), considered to be a tanker, more specifically a NLS (Noxious Liquid Substances) tanker. In general, tanker speeds range from 9 to 17 knots, with 9 to 15 knots being the most typical for oil 24 3. Methodology tankers and 12 to 17 knots for chemical tankers (Shafran, 2022). Thus, three speeds in this range are considered for each ship type: 10, 14 and 16 knots and the speed is assumed to be constant throughout the voyage. 3.2.4 Shipping logistics The main purpose of the model is to optimise the shipping logistics, i.e. vessel capacity, number of vessel and speed, to obtain the most economical cost. And the number of CO2 carriers required and their capacity depend to a large extent on the amount of CO2 to be transported and the distance of the route. Therefore, a logistics profile is developed for each type of ship and scenario. There are two constraints on the number of CO2 carriers required. One is the fleet must be able to transport the amount of CO2/year entered as an input, considering that each ship has 8400 operating hours (To). And the other is that a ship must be at the port terminal before the buffer storage is completely full. So the model uses two different methods to calculate the number of vessels needed. One by considering the total travel time and the other by considering the time it requires filling the buffer storage, selecting the one that meets both conditions. In addition, the model assumes, for all scenarios, that all vessels in the fleet have the same capacity. On the one hand, the travel time or round-trip time (Trt), given by Equation 3.1, considers sailing time which depends on distance from port to port (d) and vessel speed (v), mooring (Tmoor) and unmooring time (Tumoor), port entry (Tpin) and exit (Tpout) time as well as loading (Tl) and unloading (Tu) time. Therefore, the number of ships of type i (xi) is obtained through Equation 3.2. Trt = 2 d v + Tpin + Tpout + Tmoor + Tumoor + Tl + Tu (3.1) xi · axi ≥ a (3.2) In which, a is the amount of LCO2 that needs to be transported during the planning horizon and axi is the amount of LCO2 that a vessel can transport during the planning horizon and it is defined by Equation 3.3 where cxi is the capacity of the vessel. axi = cxi · To Trt (3.3) One the other hand, the model assumes that the amount of LCO2 arrives continuously in the port throughout the year and that the buffer storage is operational all year round. Equation 3.4 gives the number of vessels of type i needed taking into account the time it takes to fill the buffer storage (Tb). xi · Tb ≥ Trt (3.4) Where the Tb is given by Equation 3.5 in which cb is the buffer capacity and Rb is the flow rate of the buffer storage 25 3. Methodology Tb = cb Rb (3.5) In Table 3.4 can be seen the ship logistics parameters for the model in more detail. Table 3.4: Ship logistics parameters. Input Low Pressure Medium Pressure Units Port entry & Exit Time 2 2 h Mooring/Unmooring time 0.25 0.25 h Loading/Unloading time 12.81 12.81 h Ship operational time 8400 8400 h/y Planning horizon 1 1 y Operational speed 10, 14, 16 10, 14, 16 knots Ship Capacities 2.5, 5.0, 7.5, 10, 12.5, 15, 20, 25, 30, 35, 40, 45, 50 2.5, 5.0, 7.5, 10 ktCO2 The times shown in Table 3.4 are obtained from the literature review. The port entry and exit time as well as the mooring time are established in accordance with Orchard et al. (2020). As for the loading and unloading time, it is obtained by means of literature research and finding the average for different ship sizes. Brownsort, Koornneef, Energy, Gas, and de Kler (2015) considers a fixed loading time of 15 h for four different vessel capacities: 10 kt, 20 kt, 30 kt and 50kt. Whereas Bennæs et al. (2022); ZEP (Zero Emissions Platform) (2011); Bjerketvedt et al. (2020); Roussanaly et al. (2021) consider a fixed loading time of 12 h for different vessel capacities. Equinor (2019) contemplate a loading rate of 800 tCO2/h for a ship that transports 7500 tons of CO2 while Losnegård, Nysæter, Knud- sen, Belgaroui, and Forin (2020) contemplates 1200 tCO2/h for the same ship capacity. Moreover, Orchard et al. (2020) considers a loading rate of 600 tCO2/h for a ship that transports 10000 tons of CO2. Therefore, the model considers 12.81 h as loading and unloading time, regardless of vessel capacity, since it is assumed that vessels with smaller capacities have a smaller flow rate due to the sizing and number of pumps while vessels with larger capacities have a higher flow rate. Loading and unloading time should be further investigated since the time spend at port is highly dependant on it. The required time depends on the installed pumps at port or onboard of the ship that carry out the loading and unloading tasks. Since the fleet designed for this rather new market is expected to follow a shuttle transport system, would be a good option to use port facilities to install the needed pumps, so that bigger ones could be easier to install as size and weight of them would not be an issue. Furthermore, there can be found some advantages regarding the maintenance and/or replacing labours. In favour of on board installed pumps can be mention the higher flexibility for the ship, since can be docked also in ports with lack of enough big or none pumping systems. 26 3. Methodology 3.3 Cost assessment methodology This section shows the methodology followed to obtain the different costs. The mathe- matical model considers the cost of each component of the CO2 shipping supply chain, i.e., liquefaction cost, buffer storage cost, CO2 ship cost and reconditioning cost. More- over, the cost methodology adopted can be divided in three groups: CO2 ship cost, buffer storage cost and the cost of CO2 liquefaction and reconditioning process. Table 3.5 shows an overview of of the method used to scale the cost of each component of the CCS chain under study and Fig. 3.5 shows an overview of the model. Table 3.5: Overview of the cost methodology. CCS chain component. Cost Scaling CO2 Ship Investment costs are scaled with ship capacity using curve fitting established by Orchard et al. (2020). Fixed operational costs are adopted from Drewry (2021). Variable operational costs explained in detail in 3.3.1.2. Buffer Storage Investment costs are scaled linearly with ship capacity. Fixed operational costs are a percentage of investment costs. Liquefaction and Reconditioning Investment costs are scaled using the cost power law Equation. Fixed operational costs are a percentage of investment costs. Variable operational costs are scale linearly with flow rate. Figure 3.5: CO2 shipping cost model. Throughout the whole process of defining and calculating the parameters and variables required to achieve the desired results, certain inputs are necessary as well as the im- position of different constraints in order to achieve the desired result within the defined scope. I.e., time limitations, vessel size restrictions or type of fuel used by the ships, as well as specific requirements to be fulfilled with respect to the pressure and temperature conditions at which the cargo must be transported. These inputs and constraints are shown in Tab. 3.1. The following calculations are carried out for low pressure values. Nevertheless, similarly, it is the process for medium pressure option, changing the input values accordingly. 27 3. Methodology 3.3.1 CO2 ship cost The ship costs are divided into hiring costs, CH , and sailing costs, CS. The hiring costs consist of CAPEX and fixed OPEX, while the sailing costs are represented by variable OPEX. Below it can be seen listed in detail the groups and subgroups considered for building up the total costs in this thesis: 1. Capital expenses (CAPEX): Defined as the costs that enable the shipowner to take possession of the vessel and represent, on the one hand, an item during a considered period, and, on the other hand, the financial costs arising from the use of borrowed capital (loan) to finance the vessel. 2. Operation expenses (OPEX): Refers to those costs which must be paid in order to keep vessel seaworthy at all times and therefore fit for service. Additionally, these costs are divided in two subgroups so that the can be implemented in the model more accurately: (a) Fixed OPEX: Adopted from Drewry (2021). i. Manning: Crew salaries. Note that this cost does directly depend on crew size, different roles onboard as well as its close relation to different nationalities. ii. Insurance: Split in hull and machinery (H&M) and protection and in- demnity insurance (P&I). iii. Stores, spares and lubes: Represent the costs of rental for stores, ware- houses, etc and the purchase of the needed spare items. Lubes account for several required oils onboard designed for different specific tasks. Note that lube it is needed in any mechanical equipment that requires a reduction in friction, heat dissipation and meet certain cleanliness levels. iv. Repair and maintenance (R&M) and dry-docking: Which, in turn, can be divided in: A. Scheduled repairs: Routine maintenance, surveys, etc. B. Unscheduled repairs: Occurs due to accidents on the vessel. C. Hybrid repairs: Due to strategic decision with respect to market situ- ation. D. Retrospective repairs: Imposed changes or updates triggered by manda- tory incoming regulations. v. Management and administration (M&A): Arises from the owner’s management decisions such as tradings and also number and location of the offices since its location is attached to different taxation and regulations according to the local laws. (b) Variable OPEX: Defined as the sailing costs. i. Fuel consumption. 28 3. Methodology ii. CO2 emissions tax. iii. Harbour fees. iv. Channels and/or canals fees (if applicable). 3.3.1.1 Hiring cost The hiring cost depends mainly on the ship size and the number of vessels needed. Al- though costs such as maintenance and insurance usually increase with vessel age, due to their more often maintenance operations, hiring costs are consider constant over the vessel life. Moreover, the hiring costs are assumed linearly proportional to the number of vessels. The total construction cost, i.e CAPEX, of each ship type is estimated by means of a curve fitting from the CO2 ship cost database presented in Orchard et al. (2020). This can be seen in Fig. 2.4 where the orange curve corresponds to ships transporting CO2 at medium pressure and the blue curve at low pressure. The x in the fitting curve corresponds to the CO2 capacity transported by the ship. And the ship construction cost is given in millions of Euros (€m). The numbers of the regression line equations can be found in Table 3.6. Table 3.6: Ship construction cost assumptions used in the model. (Adopted from Or- chard et al., 2020). CO2 transport condition Constant CAPEX €m/tCO2 CAPEX Exponent Low Pressure 0.2849 0.5162 Medium Pressure 0.959 0.4309 Regarding the capital expenditures involved in the vessels construction, there is a potential cost saving when several same-kind ships are built in the same shipyard. These savings depends on a wide range of factors, as are the size and complexity of the vessel, the total number of ships to be built and also the specific shipyard where it is going to be constructed, since its prices are closely related to its location in terms of national economy, materials market and prices, equipment, transportation costs, etc. Nonetheless, studies show an economy of scale in the shipbuilding process linked to savings between 5 to 30% per vessel. Using a conservative approach, the following assumption is taken to calculate the CAPEX when more than one same-kind vessel is needed: CAPEXk = k ∗ CAPEX1 (3.6) where n stands for the number of vessels and CAPEX1 are the capital expenses associated for one vessel and CAPEXk are the total costs for the construction of n vessels. Once the total CAPEX is know for the most optimal solution for the case under study, two methods are proposed to calculate the annual payment of the loan, so that is up to the user to choose which one to use: 29 3. Methodology • French system: Equal annual payments including repayment of the principal and their associated interests. • German system: The total payments over the repayment period of the loan do not remain constant, but are decreasing. As in the calculation of the buffer storage cost, the French Devolution system is applied in order to obtain the CAPEX to be paid over the planning horizon. Applying Eq. 3.7 it can be known the annual payment: α = P j(1 + j)m (1 + j)m − 1 (3.7) Where α stands for the annual payment, j is the interest rate, m is the return period (in years) and P the pending loan1 at the beginning of the year. Therefore, the annual payment consists on a loan payment and an interest payment defined below: Interest payment = Interest rate · Pending loan Loan payment = α - Interest payment The block diagram depicted in Fig. 3.6 show the relation between the previously intro- duced parameters. Figure 3.6: Workflow for CAPEX calculation. Where F/G stands for French or German system. The return period is defined as input I.05 in Table 3.1 (planning horizon). As explained above, it can be seen in a graphical way how all parameters and conditions are consider inputs for a mathematical operation represented by fx box2, which will be used in later calculations, e.g., buffer cost. It must be bear in mind that the return period of the loan3 used as payment for the capital investment is set to ten years. Therefore, the prices shown in the tables of Chapter 4 are for the first ten years of operation. After this period, fleet capital investments are subtracted from the total costs, as its devolution period is completed. 1Corresponds to the ship price. Notice that it depends on the percentage of the total amount covered by the loan. 2Highlighted with dashed perimeter so that it can be identified when is applied in following processes. 3The loan is assumed to be the 80% of the fleet required CAPEX. 30 3. Methodology 3.3.1.2 Sailing cost Sailing costs are the major expenses in the transportation activity4. The main reason for this is the fuel consumption over the whole vessel life. As it can be found in the literature (Kjärstad et al., 2015, 2016; Knoope et al., 2015; Nilsson, 2014), the advantage of waterborne transportation of CO2 is its low CAPEX comparing it to other options, but in the long term, considerably bigger OPEX must be faced. In order to estimate the mentioned fuel consumption for the wide range of vessels considered in this study, in terms of different capacities and velocities, the main engine and auxiliary engines powers are calculated by means of empirical formulas and adopting values for similar existing vessels. When it comes to the auxiliary engines power, the adoption of a valid value is based on the following considerations: Liquid Natural Gas (LNG) vessels are excluded since cargo conditions when it comes to temperature are rather different to LCO2. Liquid Petroleum Gas (LPG) carriers cargo conditions are similar, since it is transported in tanks integrated into the hull, as a tanker, therefore, is not consider a pressurised vessel as it is LCO2 carriers, where CO2 is trans- ported into pressurised and refrigerated tanks. Finally, ethylene carriers are a similar options valid for the purpose when it comes to cargo transportation conditions. Hence, an estimation of different fuel prices for the following next few years is adopted from Lagemann et al. (2023). An upper an lower bound are presented, as well as the Global Warming Potential (GWP) of each fuel in Well-to-Tank (WTT) and Tank-to- Wheel (TTW) phases. For this study only TTW GWP is considered and the average of the upper and lower bounds is selected for the calculations, thus, dealing with a margin of error of ±50% in the fuel cost. GWP is treated as a cost when considering the carbon taxes5 that are applied to the amount of CO2 emitted while sailing. The aim is to incentive consumers to carry out a transition towards greener energy sources and thus base its industrial processes in a sustainable way. Carbon tax price depends on a wide range of parameters. For instance, when they are applied (i.e., current times or in next decades) since it follows an increasing price trend. Location is also of great importance, since different countries have different targets. Hence, the trend differ and so will the carbon price. An important differentiation must be done when it comes to the kind of economy that is involved. Advanced economies with net zero emissions target will have higher price on their carbon emissions than developing economies. Market and industrial sector are further characteristics to have into account. For instance, in the European Union, the EU ETS sets a price on carbon emissions from energy-intensive industries and power plants. The current price of carbon credits under the ETS is around 50 €/tCO2 (European comission, 2022). In contrast, none of the states in United States has a carbon tax, although several states have implemented their own carbon pricing mechanisms (Carbon tax center, 2022). Different carbon prices are tabulated in Tab. 3.7. Additionally, the values taken into consideration for fuel costs6 and GWP can be seen listed in Tab. 3.8. 4Liquefaction costs is often the higher cost, nevertheless, this is consider outside the defined trans- portation activity, as it is a process carried out before the transport itself. 5Environmental tax that aims to reduce greenhouse gas emissions by putting a price on the carbon content of fossil fuels. 6Market effects are not accounted for. 31 3. Methodology Table 3.7: CO2 tax [€/tCO2] using a $ to € conversion factor of 0.91€/$. (Adopted from IEA, 2022). Stated Policies Scenario 2030 2040 2050 Canada 49 56 70 Chile, Colombia 12 19 26 China 25 39 48 European Union 82 89 103 Korea 38 61 81 Announced pledges scenario Advanced economies with net zero emissions pledges 123 159 182 Emerging market and developing economies with net zero emissions pledges 36 100 146 Other emerging market and developing economies - 15 43 Net zero emissions by 2050 scenario Advanced economies with net zero emissions pledges 127 187 228 Emerging market and developing economies with net zero emissions pledges 82 146 182 Other emerging market and developing economies 23 77 164 For this study, the selected carbon tax are the one highlighted in bold corresponding to European Union. To be on the safe side and accounting for fluctuations, the final value used for the carbon tax in this study is 95 $/tCO2, which corresponds to 86 €/tCO2. The cost used from Tab 3.8 for building up the model can be found in the second, third and fourth column. As mentioned before, the TTW GWP is considered as a cost when the carbon tax is applied to it. Furthermore, the value taken for the estimation of the fuel cost due to its consumption while engines are running are set to the average between the upper and lower bound for each type of fuel. Table 3.8: Fuels costs and GWP forecast. (Adopted from Lagemann et al., 2023). Fuel GWP [gCO2eq/kWh] Bounds cost [USD/MWh] WTT TTW Upper Lower VLSFO 47.5 284.1 95 38 bio-Diesel 70 150 128 93 e-Diesel 0 4.5 423 131 LNG 66.6 238.8 81 32 bio-LNG 49.7 6 119 89 e-LNG 0 6 358 115 LPG 30 237.5 98.3 39.3 Methanol 112.7 253.4 210 90 bio-Methanol 112.68 3.24 97 66 e-Methanol 0 3.5 385 116 Ammonia 87.1 19 220 56 e-Ammonia 0 19 220 80 LH2 108.7 0 245 55 e-LH2 0 0 245 79 32 3. Methodology To particularise it to a specific case, the sailing costs are in terms of the distance between source and sink, Hence, depending on the speed, the sailing time and the docked time (loading and unloading) the sailing time differs and so it does the fuel consumption, since it is assumed that the main engine will not be operating while docked at harbour. It is assumed that electric power on board is achieved by means of shore grid, which means that auxiliary engines are also off during this time and therefore no fuel consumed by them either. This assumption is taken at the view of current and even tighten future restrictions when it comes to exhaust gas in areas near to population, where vessels cannot emit contaminants to the atmosphere, being the connection to shore grid the best solution. It shall be bear in mind that port fees would increase due to the energy consumption.. Figure 3.7 shows the process followed to calculate the total fuel costs or variable OPEX. Notice that fixed OPEX are adopted from literature, while, as explained before, the fuel cost is based on the estimated values for different fuels found in Lagemann et al. (2023). Figure 3.7: Depicted workflow of inputs, constraints, variables and steps to follow in order to calculate the OPEX. Where C.01 and C.02 stands for constraints in vessel cargo capacity and speed respectively, I.03 represents the ship operational time and I.13 and I.14 are the inputs to account for the cargo transportation temperature and pressure. These parameters are defined in Tab. 3.1. Notice that fx box represents certain relation of operations and equations with respect to the inputs in order to achieve the desired output/variable. It can be seen how all parameters are intertwined and at which stage of the process are they added into the equation. It can be observed that it mainly depends on vessel speed and cargo capacity (i.e., ship size) for which the hull forms are defined and thus the engine power and hence the fuel consumption. 3.3.2 Buffer storage cost The buffer storage cost is composed by the investment cost (CAPEX) and operational cost (OPEX). The investment cost is proportional to the buffer storage capacity and, as indicated in Section 3.2, is 1.18 times the vessel capacity according to Bjerketvedt et al. (2020). Figure 3.8 shows different buffer storage investment costs found in the literature review. All values are expressed in Euros per tonne of capacity of CO2. When necessary, the following conversions have been used. 1 USD corresponds to 0.91 € and 1 £ corresponds 33 3. Methodology to 1.14 € on Apr. 28th, 20237. The data shown in Table 3.2 is used to convert from cubic metre to tonne of CO2. With regard to the most recent studies, Orchard et al. (2020) uses for the shipping model a cost of 1300 €/tCO2 stored in the tanks for low pressure condition and 2770 €/tCO2 for medium pressure condition. Whereas Roussanaly et al. (2021) considers 550 and 920 €/m3 of CO2 stored for the low and medium pressure shipping conditions, respectively. Additionally, Bennæs et al. (2022), based on the latter data, uses an investment cost of 478 €/tCO2 for the low-pressure transport condition and 800 €/tCO2 for the medium-pressure shipping condition. Figure 3.8: Buffer Storage Investment Cost from Literature Review. For the transport model developed in this study, the CAPEX presented in Roussanaly et al. (2021) is used as it provides values for both the low pressure condition and the medium pressure condition. Additionally, the values are similar to other studies such as Element Energy (2018); Kujanpää, Rauramo, and Arasto (2011); Seo et al. (2016). Also, according to the literature review (Bennæs et al., 2022; Element Energy, 2018; Roussanaly et al., 2021), operational costs range between 5% and 6%. For the present study, the operational cost is set to 6% of the CAPEX. The values used in the shipping model can be seen in Table 3.9. Table 3.9: Buffer storage cost assumptions used in the model Transport CO2 Condition CAPEX OPEX Units Low Pressure 485.44 6 % €/tCO2 Medium Pressure 882.92 6 % €/tCO2 As the planning horizon considered in the present work is one year, the model considers the CAPEX to be paid in the first year. This is calculated through the French Devolution System explained in Section 3.3.1.1 through Equation 3.7 with an interest (annual) rate of 8%, a loan percentage of 80% and a return time of 25 years. 7Assumed as fixed values for current time in this thesis. 34 3. Methodology Hence, the total buffer cost, CB, is given by Equation 3.8 where CBO is the operational cost of the buffer storage and CBCH is the CAPEX cost in the planning horizon for the buffer storage. CB = CBCH + CBO (3.8) In Fig. 3.9 it can be seen the process followed to achieve the desired buffer cost. Notice that the dashed fx box is the same method followed previously to calculate the annual CAPEX so that interest can be taken into account (see Fig. 3.6). Therefore, final buffer cost will be given in €m per yea