UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Cost optimization of hydrogen fuel supply chain with environmental policy integration: the case for British… Talebian, Hoda 2020

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

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

Item Metadata

Download

Media
24-ubc_2020_november_talebian_hoda.pdf [ 2.82MB ]
Metadata
JSON: 24-1.0392425.json
JSON-LD: 24-1.0392425-ld.json
RDF/XML (Pretty): 24-1.0392425-rdf.xml
RDF/JSON: 24-1.0392425-rdf.json
Turtle: 24-1.0392425-turtle.txt
N-Triples: 24-1.0392425-rdf-ntriples.txt
Original Record: 24-1.0392425-source.json
Full Text
24-1.0392425-fulltext.txt
Citation
24-1.0392425.ris

Full Text

COST OPTIMIZATION OF HYDROGEN FUEL SUPPLY CHAIN WITH ENVIRONMENTAL POLICY INTEGRATION: THE CASE FOR BRITISH COLUMBIA  by  Hoda Talebian  B.A.Sc., Ferdowsi University of Mashhad, 2009 M.A.Sc., Ferdowsi University of Mashhad, 2012  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Mechanical Engineering) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   July 2020  © Hoda Talebian, 2020 ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: Cost Optimization of Hydrogen Fuel Supply Chain with Environmental Policy Integration: the case for British Columbia  submitted by Hoda Talebian in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Mechanical Engineering  Examining Committee: Dr. Walter Mérida, Professor, Mechanical Engineering, UBC Supervisor  Dr. Harish Krishnan, Professor, Sauder School of Business, UBC Supervisory Committee Member Dr. David Layzell, Professor, Biological Sciences, University of Calgary External Examiner Dr. Dana Grecov, Associate Professor, Mechanical Engineering, UBC University Examiner Dr. Barbara Lence, Professor, Civil Engineering, UBC University Examiner  Additional Supervisory Committee Members: Dr. Farrokh Sassani, Professor, Mechanical Engineering, UBC Supervisory Committee Member    Dr. Taraneh Sowlati, Professor, Faculty of Forestry, UBC Supervisory Committee Member Dr. Martino Tran, Assistant Professor, Community and Regional Planning, UBC Supervisory Committee Member iii  Abstract     By powering fuel cell electric vehicles hydrogen can contribute to greenhouse gas emissions reduction in British Columbia (B.C.)  The province is well positioned to capitalize on its natural resources and policies towards the development of a hydrogen fueling supply chain (HFSC). However, such development requires significant investment with high risks of negative cash flow for years to decades. A spatially explicit multi-period optimization model was developed to design a minimum-cost HFSC based on a mixed integer linear programming formulation. The model was applied to the light duty passenger vehicle sector in B.C. under three hydrogen demand scenarios. The model considered different capacities for all components of the supply chain, covered the on-site production and capacity expansion options as well as minimum storage requirement for fueling stations. Different combinations of the current and potential environmental mandates and the government economic instruments were integrated in the model explicitly. The model measured the effectiveness of the policies on reducing the cost and greenhouse gas (GHG) emissions of the HFSC for each demand scenario. To this end, the GHG emissions were monetized using the social cost of carbon. The results suggested that hydrogen can be cost competitive with gasoline. However, the cost optimal hydrogen infrastructure relied heavily on steam methane reforming (SMR), with small GHG emissions reduction benefits.  Nonetheless, the monetary benefits of well to wheels (WTW) GHG emissions reduction justified the switch from gasoline to SMR-based hydrogen. It was found that central electrolysis can be financially justified by addition of production tax credits or electricity incentives to the current provincial carbon control policies (i.e., carbon tax and low carbon fuel standard). This study assessed the effectiveness of current policies in emissions mitigation from the road freight transport. Moreover, the WTW energy requirement and GHG emissions reduction potential of the all-electric trucking were measured to meet the provincial emissions reduction targets. The results suggested that the B.C. hydroelectricity will fall short of generating sufficient energy to support all-electric trucking. Thus, B.C. has to undertake policies to incentivize electricity generation from diversified renewable energy resources.  iv  Lay Summary     Hydrogen penetration into the transport sector requires sufficient initial fueling network coverage well in advance of the fuel cell electric vehicle rollout. Considering the significant capital investment which will be followed by underutilization, the hydrogen fueling supply chain may face a long period of negative cash flows.  In this work, a cost optimization framework was developed to design a hydrogen fueling supply chain for the successful deployment of fuel cell electric vehicles in British Columbia. The results suggest the share of distributed and central hydrogen production, number, location, capacity of production plants and storage facilities, the transportation links, and the number and distribution of fueling stations in different periods of market development. Moreover, a range of emissions mitigation policies and incentive plans was integrated explicitly in the model to assess their effectiveness on the accelerated adoption of low-carbon hydrogen in the province. v  Preface The original idea of investigating the hydrogen technology pathways for the transportation sector in B.C. was proposed by Walter Mérida. Hoda Talebian is responsible for identifying the specific topics of investigation documented herein. She is responsible for: defining the knowledge gap and research questions, gathering model inputs (data collection), developing the optimization model from inception, developing policy scenarios, and dissemination of results in the form of conferences and publications. Hoda Talebian is responsible for most of the substantive and editorial preparation of this thesis under the supervision of Walter Mérida.  ▪ Chapters 1 and 2 are composed from final version of Hoda Talebian’s Ph.D. proposal defense document. ▪ A version of Chapter 3 was published as a peer-reviewed conference proceeding. Talebian, H., Herrera, OE., Tran, M., Mérida, W. ‘Potential for Hydrogen as a Transportation Fuel in British Columbia: Resource Assessment and GHG Emissions Analysis’ Transp. Res. Board 97th Annu. Meet., 2018. Hoda Talebian was responsible for: scientific question formulation; method design; analysis of the results and discussion; and   article preparation. The project manager, Omar Herrera, and Martino Tran offered critique in article preparation and manuscript edits. Walter Mérida provided direction and feedback on the work.  ▪ A version of Chapter 3, 4 and 5 has been published as a peer-reviewed full article. Talebian, H., Herrera, OE., Mérida, W. ‘Spatial and temporal optimization of hydrogen fuel supply chain for light duty passenger vehicles in British Columbia’ Int J Hydrogen Energy, 44 (2019) pp. 25939-25956. Hoda Talebian was responsible for: scientific question formulation; method design; analysis of the results and discussion; and   article preparation. The project manager, Omar Herrera, offered critique in article preparation and manuscript edits. Walter Mérida provided direction and feedback on the work.  ▪ A version of Chapter 3 and 5 is submitted for publication as a peer-reviewed full article in April 2019. Hoda Talebian was responsible for: scientific question formulation; method   design; analysis of the results and discussion; and article preparation. The project manager, vi  Omar Herrera, offered critique in article preparation and manuscript edits. Walter Mérida provided direction and feedback on the work.  ▪ A version of Chapter 6 has been published as a peer-reviewed full article. Talebian, H., Herrera, OE., Tran, M., Mérida, W. ‘Electrification of road freight transport: Policy implications in British Columbia’ Energy Policy 115 (2018) pp. 109-118. Hoda Talebian was responsible for: scientific question formulation; method design; analysis of the results and discussion; and article preparation. The project manager, Omar Herrera, and Martino Tran offered critique in article preparation and manuscript edits. Walter Mérida provided direction and feedback on the work.  Permission for reproduction of materials published in the International Journal of Hydrogen energy and Energy Policy Journal is covered under Elsevier B.V.’s author’s rights for personal (scholarly) purposes.  vii  Table of Contents  Abstract .......................................................................................................................................... ii Lay Summary ............................................................................................................................... iv Preface .............................................................................................................................................v Table of Contents ........................................................................................................................ vii List of Tables ............................................................................................................................... xii List of Figures ............................................................................................................................. xiv List of Symbols .......................................................................................................................... xvii List of Abbreviations ............................................................................................................... xxiii Acknowledgements .................................................................................................................. xxiv Dedication ...................................................................................................................................xxv Chapter 1: Hydrogen as an Energy Carrier................................................................................1 1.1 Hydrogen applications ........................................................................................................ 1 1.2 Current status and international targets............................................................................... 2 1.3 Hydrogen potential in British Columbia ............................................................................. 2 1.4 Hydrogen role in B.C.’s road transportation sector ............................................................ 3 Chapter 2: Hydrogen Supply Chain for Mobility .......................................................................5 2.1 Hydrogen supply chain structure ........................................................................................ 5 2.1.1 Production facilities .................................................................................................... 5 2.1.2 Terminals and storage facilities .................................................................................. 6 2.1.3 Hydrogen delivery network ........................................................................................ 6 2.1.4 Hydrogen fueling stations ........................................................................................... 7 2.2 Deployment challenges ....................................................................................................... 7 2.3 Hydrogen supply chain design approaches ......................................................................... 8 2.4 Environmental considerations in the hydrogen supply chain design .................................. 9 2.4.1 Low-carbon hydrogen pathways ............................................................................... 10 2.4.2 Enabling low-carbon hydrogen production .............................................................. 10 2.4.3 Low-carbon hydrogen integration in supply chain optimization .............................. 11 viii  2.5 Hydrogen fuel supply chain design in British Columbia .................................................. 12 2.5.1 Objectives ................................................................................................................. 12 2.5.2 Contributions............................................................................................................. 13 2.5.3 Approach (thesis outline) .......................................................................................... 14 Chapter 3: Hydrogen Supply Chain Cost Optimization Model (H2SCOT): Model Inputs .19 3.1 Assessment of energy sources .......................................................................................... 19 3.1.1 Hydrogen production from renewable energy sources ............................................. 19 3.1.2 Hydrogen production from non-renewable energy sources ...................................... 21 3.1.3 Selected energy sources for hydrogen production .................................................... 22 3.2 Geographic divisions ........................................................................................................ 22 3.3 Techno-economic and environmental data ....................................................................... 24 3.3.1 Derivation of the techno-economic parameters ........................................................ 24 3.3.1.1 SMR plant ......................................................................................................... 24 3.3.1.2 Electrolyzer ....................................................................................................... 27 3.3.1.3 By-product hydrogen purification from the chlor-alkali industry .................... 29 3.3.1.4 Liquefier ............................................................................................................ 30 3.3.1.5 Terminal and central storage ............................................................................. 31 3.3.1.5.1 Gas delivery terminal (GH2 storage) .......................................................... 31 3.3.1.5.2 Liquid delivery terminal (LH2 storage) ...................................................... 33 3.3.1.5.3 Transportation ............................................................................................. 35 3.3.1.5.4 Fueling station ............................................................................................. 36 3.3.2 Derivation of the GHG emissions parameters .......................................................... 44 3.4 Hydrogen demand scenario development ......................................................................... 45 3.4.1 Temporal projection of hydrogen demand ................................................................ 45 3.4.1.1 New passenger vehicle projection .................................................................... 45 3.4.2 FCEV penetration to the market (scenario development) ......................................... 47 3.4.2.1 Passenger vehicle stock projection ................................................................... 47 3.4.2.2 FCEV stock projection ...................................................................................... 48 3.4.2.3 Annual hydrogen demand calculation .............................................................. 49 3.4.3 Spatial projection of hydrogen demand in B.C. ........................................................ 51 ix  3.4.3.1 Hydrogen demand distribution in Metro Vancouver ........................................ 52 3.5 Policy scenario development ............................................................................................ 53 3.5.1 Current provincial policies ........................................................................................ 53 3.5.1.1 B.C. low carbon fuel standard (LCFS) ............................................................. 54 3.5.1.2 B.C. carbon tax ................................................................................................. 55 3.5.2 Potential financial and regulatory policies ................................................................ 55 3.5.2.1 Production tax credit (PTC) .............................................................................. 57 3.5.2.2 Capital subsidy .................................................................................................. 58 3.5.2.3 Utility incentives on electrolytic hydrogen ....................................................... 58 3.5.2.4 Higher rates of carbon tax ................................................................................. 59 3.5.2.5 Ban on the SMR hydrogen production without CCS integration ..................... 61 Chapter 4: Hydrogen Supply Chain Cost Optimization Model (H2SCOT): Formulation...62 4.1 MILP basics ...................................................................................................................... 62 4.2 H2SCOT basic assumptions ............................................................................................. 63 4.3 H2SCOT constraints ......................................................................................................... 63 4.3.1 Production facilities .................................................................................................. 63 4.3.2 Terminals with central storage facilities ................................................................... 65 4.3.3 Transportation and distribution ................................................................................. 66 4.3.4 Hydrogen fueling stations ......................................................................................... 68 4.3.5 Hydrogen demand ..................................................................................................... 69 4.3.6 Building new facilities and lifetime consideration ................................................... 69 4.3.7 Capacity expansion ................................................................................................... 70 4.3.8 Non-negativity constraints ........................................................................................ 72 4.4 H2SCOT objective function ............................................................................................. 72 4.4.1 Discounted cost of technology (DCT) ....................................................................... 72 4.4.2 Discounted cost of environmental policies (DCPolicy) ............................................... 76 4.4.2.1 Discounted cost of carbon tax (DCCT) .............................................................. 77 4.4.2.2 Discounted revenue of LCFS (DRLCFS) ............................................................ 79 4.4.2.3 Discounted cost of complementary policies ..................................................... 80 4.5 Post optimization cash-flow analysis ................................................................................ 81 x  4.6 Potential contribution of FCEVs to GHG emissions reduction ........................................ 82 Chapter 5: Hydrogen Fuel Supply Chain Development in British Columbia: Light Duty Passenger Vehicles .......................................................................................................................83 5.1 HFSC configuration with no environmental policy inclusion .......................................... 83 5.2 HFSC configuration with current provincial policy inclusion .......................................... 91 5.3 Economic and environmental evaluation of the hydrogen supply chain .......................... 93 5.4 Effect of complementary policies in low-carbon hydrogen production ........................... 95 5.4.1 Optimal share of production technologies ................................................................ 96 5.4.2 Efficiency assessment of complementary policies ................................................... 98 Chapter 6: Challenges and Potentials in the Heavy-duty Transport Sector ........................104 6.1 Approaches to reduce GHG emissions from the trucking sector .................................... 105 6.2 All-Electric trucking in B.C. by 2040: feasibility study ................................................. 107 6.2.1 Freight trucks stock forecasting .............................................................................. 108 6.2.2 GHG emissions projections from road freight transport: BAU and CLF scenarios 112 6.2.2.1 Business as usual (BAU) scenario .................................................................. 112 6.2.2.2 Current legislation fulfillment (CLF) scenario ............................................... 112 6.2.2.3 BAU and CLF comparison ............................................................................. 114 6.2.3 GHG emissions projections from road freight transport in 2040: electrification effect... ................................................................................................................................ 115 6.2.4 B.C. resource assessment to support all-electric trucking ...................................... 120 6.2.5 Comparative analysis of emission reductions and energy requirements across scenarios .............................................................................................................................. 123 Chapter 7: Conclusion and Future Work ................................................................................126 7.1 Light duty passenger vehicles (current provincial policies) ........................................... 126 7.2 Light duty passenger vehicles (complementary policies) ............................................... 127 7.3 Fright road transport ....................................................................................................... 128 7.3.1 GHG emissions reduction potential in B.C............................................................. 128 7.3.2 Energy requirement and resource availability ........................................................ 129 7.4 Study limitations ............................................................................................................. 130 7.5 Future work ..................................................................................................................... 130 xi  7.5.1 Extending the frontiers of H2SCOT ....................................................................... 130 7.5.2 Multi criteria decision making ................................................................................ 131 7.5.3 Parametric study on economic factors .................................................................... 131 7.5.4 Introducing non-linearity and uncertainty to the model ......................................... 131 7.5.5 Expansion on policy scenarios and integration mechanisms .................................. 132 Bibliography ...............................................................................................................................133 Appendix A ................................................................................................................................ 151 A.1 Hydrogen demand projection .................................................................................. 151 A.2 Distances between supply and demand regions ...................................................... 152 A.3 Other parameters used in the model ........................................................................ 154  xii  List of Tables  Table 2.1. Modeling details of the previously developed platforms compared with H2SCOT.1 . 16 Table 3.1. Distribution and delivered cost of the wood-based biomass feedstock in B.C. [91] ... 20 Table 3.2. Capital and operating cost of SMR plant (100 tonnes/day) [101] ............................... 24 Table 3.3. Capital and operating cost of a PEM electrolyzer [101]. ............................................. 28 Table 3.4.  Capital and operating cost of a hydrogen capture and purification facility (10 tonnes/day). ................................................................................................................................... 29 Table 3.5. IDCC and operating cost of GH2 and LH2 central storage. ........................................ 35 Table 3.6. Capital cost of gas trucks with different payloads [108]. ............................................ 35 Table 3.7. Capital and operating cost of a fueling station. ........................................................... 36 Table 3.8. GHG emissions associated with each component of the HFSC [103], [113]–[115] ... 44 Table 3.9. Economic instruments to promote renewable energy worldwide. ............................... 56 Table 5.1. On-site and central production plants and storage facilities for three demand scenarios over time (no policy inclusion). .................................................................................................... 85 Table 5.2. Effect of capacity alternatives on the total discounted cost of the HFSC for three demand scenarios. ....................................................................................................................................... 91 Table 5.3. Average hydrogen price and GHG emissions reduction per unit of hydrogen production over 30-year time frame for the base case (LCFS+CT)1 .............................................................. 98 Table 5.4. Net present value of the total cost of subsidies in each demand scenario (all values in Millions C$2013) ........................................................................................................................ 101 Table 6.1 Freight truck classification [9] .................................................................................... 108 Table 6.2. ICE truck characteristics [9] ...................................................................................... 110 Table 6.3. Fuel efficiency improvement of freight trucks from deployment of federal regulations in the current legislation fulfillment (CLF) scenario .................................................................. 113 Table 6.4. WTT energy requirement and GHG emissions for the selected hydrogen pathways [103], [112], [214]–[216], [113]–[115], [209]–[213] ................................................................. 118  Table A.1. Hydrogen demand (kg/day) distribution among municipalities in the final year of   each time step for different demand scenarios……………………………….……………………….151 xiii  Table A.2 Distances between potential production and storage locations (1-14) and distances between potential storage locations and the entrance to Metro Vancouver municipalities (Langley Township). ……………………………………………………………………………………..152 Table A.3 Distances between Langley Township (LT) and different municipalities in Metro Vancouver and distances between storage facilities in North Vancouver (NV) and different municipalities in the Metro Vancouver………………………………………………………….152 Table A.4 Distances between potential storage locations and the demand regions (except Metro Vancouver) ……………………………………………………………………………………..153  xiv  List of Figures  Figure 2.1. Superstructure of the HFSC infrastructure in B.C. (CCS: carbon capture and sequestration) ................................................................................................................................ 17 Figure 2.2.  Schematic of H2SCOT .............................................................................................. 18 Figure 3.1. Hydrogen supply regions in British Columbia ........................................................... 23 Figure 3.2. Schematic of gaseous hydrogen central storage. ........................................................ 31 Figure 3.3. Schematic of liquid hydrogen central storage. ........................................................... 33 Figure 3.4. Schematic of a hydrogen fueling station with on-site production, gas delivery, and liquid delivery components. .......................................................................................................... 37 Figure 3.5  Comparison of the annual growth rate of GDP per capita and new passenger vehicle in BC.: projection vs historical data .................................................................................................. 46 Figure 3.6 Comparison of the moving average of GDP per capita and new passenger vehicle growth rates in BC. (1995 -2016) ................................................................................................. 46 Figure 3.7. Share of passenger trucks from the total passenger vehicles: projection versus historical data [9] .......................................................................................................................................... 48 Figure 3.8. (a) Penetration percentage of the new passenger FCEVs to the B.C. market for different demand scenarios over time. (b) Passenger FCEV stock for different demand scenarios in B.C. over time. ...................................................................................................................................... 49 Figure 3.9. Vehicle average use intensity for cars and passenger trucks in BC: projection versus historical data [9]. ......................................................................................................................... 50 Figure 3.10. Annual hydrogen demand in B.C. for different demand scenarios over time. ......... 51 Figure 3.11. Distribution of population density among 10 municipalities in Metro Vancouver over time. .............................................................................................................................................. 53 Figure 3.12. Environmental policies with various deployment strategies over time (a) production tax credit (PTC), (b) capital subsidy, (c) utility incentives on electrolytic hydrogen, (d) carbon tax rate................................................................................................................................................. 61 Figure 5.1. Contribution of different transportation states (G: gas, L: liquid) and deliverable capacities (100, 500, 900 and 3800 kg per truck) to the total number of transportation units for three demand scenarios over time (no policy inclusion). ............................................................. 86 xv  Figure 5.2. Contribution of different fueling station capacities (150, 500, 1000, and 1500 kg/day) to the total number of stations for three demand scenarios over time. Station types: O: on-site production, G: gas delivery, L: liquid delivery (no policy inclusion). ......................................... 87 Figure 5.3. Optimal distribution of production facilities and transportation network in B.C. for (a) pessimistic (b) moderate (c) optimistic demand scenarios in time step 2045-2050 (no policy inclusion)....................................................................................................................................... 88 Figure 5.4. Optimal distribution of hydrogen fueling stations in Metro Vancouver for (a) pessimistic (b) moderate (c) optimistic demand scenarios in time step 2045-2050 (no policy inclusion)....................................................................................................................................... 90 Figure 5.5. Contribution of different technologies to total hydrogen production and the share of liquefied hydrogen for three demand scenarios and policy inclusions (no policy: base case, LCFS: low-carbon fuel standard, CT: carbon tax, C-electrolysis: Central electrolysis, O-electrolysis: On-site electrolysis). ........................................................................................................................... 92 Figure 5.6. Environmental and economic comparison of hydrogen and gasoline infrastructure in the base case (no environmental policy is included) for three demand scenarios: net present value (NPV) of the reduced GHG emissions (IRR = 3%) versus the difference between the NPV of revenues (IRR = 10%). ................................................................................................................. 94 Figure 5.7. Effect of environmental policies on the GHG emissions and the hydrogen price compared to the no policy case for three demand scenarios: CT: carbon tax (C$45 to C$75 from 2020 to 2050), LCFS: low-carbon fuel standard (C$167 to C$0 from 2020 to 2050). All values in Canadian dollars (2013). ............................................................................................................... 95 Figure 5.8. Contribution of production technologies in a cost optimal hydrogen fuel supply chain for the base case (LCFS+CT) and the potential policy included cases in three demand scenarios....................................................................................................................................................... 97 Figure 5.9. Policy efficiency assessment with respect to hydrogen price decrease and GHG emissions reduction compared to the base case: (a) pessimistic (b) moderate (c) optimistic demand scenarios ...................................................................................................................................... 100 Figure 5.10. Hydrogen price increase and the GHG emissions reduction for the potential policy included cases compared to the base case (LCFS+CT). The base case values are presented in Table 5.4................................................................................................................................................ 103 xvi  Figure 6.1. 2016 GHG emissions in B.C.: (a) GHG emissions by sector (b) GHG emissions from road transport: change from 2007 ............................................................................................... 104 Figure 6.2. Historical data and projections to 2040 for light-duty trucks (LDT), medium-duty trucks (MDT) and heavy-duty trucks (HDT): (a) freight vehicle use-intensity in B.C. - (b) number of new freight vehicles in B.C. market – (c) stock of freight vehicles in B.C. ........................... 111 Figure 6.3. WTW GHG emissions from road freight transportation in B.C. for business as usual (BAU) and current legislation fulfillment (CLF) scenarios - historic data and projections to 2040 (a) light-duty trucks (LDT) (b) medium-duty trucks (MDT) (c) heavy-duty trucks (HDT) ...... 115 Figure 6.4.  Hydrogen production pathways ............................................................................... 116 Figure 6.5. Share of all-electric freight trucks (FCE: fuel cell electric, BE: battery electric) for 64% GHG emissions reduction from road freight transport in 2040 (from 2007 level): business as usual (BAU) and current legislation fulfillment (CLF) scenario ......................................................... 119 Figure 6.6. Electricity requirement for 64% GHG emissions reduction from road freight transport in 2040 (from 2007 level) - FCE: fuel cell electric and BE: battery electric trucks- business as usual (BAU) and current legislation fulfillment (CLF) scenario ......................................................... 121 Figure 6.7. GHG emissions change in 2040 road freight transport compared with 2007 - 100% of freight trucks running on hydrogen produced from central natural gas reforming (NGCR) pathway without carbon capture and sequestration (CCS)- business as usual (BAU) and current legislation fulfillment (CLF) scenario .......................................................................................................... 123 Figure 6.8. 2040 projections on the number of all-electric trucks (FCE: fuel cell electric and BE: battery electric) and total energy required for 1% GHG emissions reduction from road freight transport in B.C. .......................................................................................................................... 125  xvii  List of Symbols Sets 𝐴 Nominal capacity of tube tankers (120,600,1100 kg) 𝐶 Nominal capacity of central plants and storage facilities (2,10,50,100 tonnes/day) 𝐷 Product status (gas hydrogen (onsite production), gas hydrogen(delivered), liquid hydrogen (delivered)) 𝐺 Central production grids 𝑔′  𝑔 ?̅? Central storage grids (Warehouses) 𝐺′ Demand grids (Metro Vancouver)  J Stages of capacity expansion (10, 20, 30% for SMR and 10,25,50% for electrolyzer and storage units) 𝑁𝑉 Demand grids (Kamloops, Kelowna, Prince George, Victoria)  𝑅𝑂 Demand grids (Abbotsford, hope, Whistler, Williams Lake)  𝑇 time periods of the planning horizon (6 time-steps: every five years starting from 2020)  𝑌 Plant type with different production technologies (Electrolyser, SMR+CCS, SMR w/o CCS, By-Product Hydrogen purification plant)   Parameters 𝐴𝑇𝑃𝐷_𝐶𝐹𝑖 After tax post depreciation cash flow in year i 𝐴𝑇_𝑇𝑟 Truck daily availability limit (hours) 𝐶𝑎𝑝𝑖 Capital costs occurred in year i 𝐶𝐴𝑃𝐿_𝑇𝑅 Unloading capacity of tanker trucks delivering liquefied hydrogen 𝐶𝐴𝑃𝐺_𝑇𝑅𝑎 Unloading capacity of tube trailers of size a, delivering gaseous hydrogen 𝐶𝑟𝑒𝑑𝑖𝑡_𝐿𝐶𝐹𝑆𝑡 LCFS credit in time step t ($/tones of CO2 displaced) 𝐶𝑢𝑚_𝐶𝐹𝑖 Cumulative cash flow in year i 𝐷𝑐𝑎𝑝_𝑚𝑖𝑛𝑠 Minimum throughput of a fueling station with nominal capacity s 𝐷𝑐𝑎𝑝_𝑚𝑎𝑥𝑠 Maximum throughput of a fueling station with nominal capacity s 𝐷𝐷𝐶𝐶_𝐶𝑐𝑦 Direct depreciable capital cost of a central plant of type y and capacity c  𝐷𝐷𝐶𝐶_𝑆𝑐̅𝑑 Direct depreciable capital cost of a central storage facility of capacity 𝑐̅ which stores hydrogen at status d 𝐷𝐷𝐶𝐶_𝑇𝑅𝐺𝑎 Direct depreciable capital cost of a tube tanker truck of size a transporting hydrogen at gaseous status  𝐷𝐷𝐶𝐶_𝑇𝑅𝐿 Direct depreciable capital cost of a tanker truck transporting hydrogen at liquid status  𝐷𝐷𝐶𝐶_𝑂𝑠 Direct depreciable capital cost of an onsite plant of capacity s 𝐷𝐷𝐶𝐶_𝐷𝑠𝑑  Direct depreciable capital cost of fueling station of capacity s, which delivers hydrogen at status d 𝐷𝑒𝑝_𝑐ℎ𝑖 Depreciation charge in year i 𝐷𝑒𝑐𝑜𝑚 Decommissioning costs 𝐷𝐺_𝑉𝑔′𝑡 Demand of hydrogen in grid 𝑔′ and time step t 𝐷𝐺_𝑁𝑉𝑛′𝑡 Demand of hydrogen in grid 𝑛′and time step t xviii  𝐷𝐺_𝑁𝑉𝑟′𝑡 Demand of hydrogen in grid  𝑟′and time step t 𝐷𝑇𝑡 Demand of hydrogen in B.C. in time step t 𝐸_𝐶𝑜𝑠𝑡 Emission cost (Carbon tax) ($/tonnes of CO2 displaced) 𝐸𝐸𝑅𝑡 Energy efficiency ratio in time step t 𝐹_𝐶𝑐𝑦 Fixed cost of a central plant of type y and capacity c per year 𝐹_𝐷𝑠𝑑 Fixed cost of a fueling station of capacity s which delivers hydrogen at status d per year 𝐹_𝑂𝑠 Fixed cost of an onsite plant of capacity s per year 𝐹_𝑂𝑝𝑟𝑖 Fixed operating costs in year i  𝐹_𝑆𝑐̅𝑑 Fixed cost of a central storage facility of capacity 𝑐̅  which stores hydrogen at status d per year 𝐹_𝑇𝑅𝑑 Fixed cost of a truck transporting hydrogen at status d per year 𝐹𝑢𝑒𝑙_𝑇𝑅_𝑃𝑆𝑑?̅?𝑔 Cost of diesel to transport hydrogen at status d from production grid g to storage grid ?̅? 𝐹𝑢𝑒𝑙_𝑇𝑅_𝑉𝑑𝑔′?̅? Cost of diesel to deliver hydrogen at status d from storage grid ?̅? to demand grid 𝑔′ 𝐹𝑢𝑒𝑙_𝑇𝑅_𝑁𝑉𝑑𝑛′?̅? Cost of diesel to deliver hydrogen at status d from storage grid ?̅? to demand grid 𝑛′ 𝐹𝑢𝑒𝑙_𝑇𝑅_𝑅𝑂𝑑𝑟′?̅? Cost of diesel to deliver hydrogen at status d from storage grid ?̅? to demand grid  𝑟′ 𝐺𝑎𝑠_𝐶𝐼 WTW Gasoline Carbon Intensity (g CO2/MJ) 𝐺𝐻𝐺_𝐶𝑦 GHG emission of a central plant of type y (gCO2eq/kg H2) 𝐺𝐻𝐺_𝑆𝑑 GHG emission of a central storage facility which stores hydrogen at status d (gCO2eq/kg H2) 𝐺𝐻𝐺_𝑇𝑅 GHG emission of a truck transporting hydrogen at status d (gCO2eq/kg H2) 𝐺𝐻𝐺_𝐷𝑑 GHG emission of a fueling station delivers hydrogen at status d (gCO2eq/kg H2) 𝐺𝐻𝐺_𝑂 GHG emission of an onsite plant (gCO2eq/kg H2) 𝐻2_𝐷 Hydrogen density (MJ/kg) in 1 atm,298K 𝐼𝐷𝐶𝐶_𝐶𝑐𝑦 Indirect depreciable capital cost of a central plant of type y and capacity c 𝐼𝐷𝐶𝐶_𝑆𝑐̅𝑑 Indirect depreciable capital cost of a central storage facility of capacity 𝑐̅ which stores hydrogen at status d 𝐼𝐷𝐶𝐶_𝑂𝑠 Indirect depreciable capital cost of an onsite plant of capacity s 𝐼𝐷𝐶𝐶_𝐷𝑠𝑑 Indirect depreciable capital cost of fueling station of capacity s, which delivers hydrogen at status d 𝐼𝐷𝑒𝑝_𝐶𝑎𝑝𝑖 Initial equity depreciable capital (for units installed in year i) 𝐿𝑎𝑏𝑜𝑟_𝑇𝑅_𝑃𝑆𝑑?̅?𝑔 Driver wage of a truck delivering hydrogen at status d from production grid g to storage grid ?̅? 𝐿𝑎𝑏𝑜𝑟_𝑇𝑅_𝑉𝑑𝑔′?̅? Driver wage of a truck delivering hydrogen at status d from storage grid ?̅? to demand grid 𝑔′ 𝐿𝑎𝑏𝑜𝑟_𝑇𝑅_𝑁𝑉𝑑𝑛′?̅? Driver wage of a truck delivering hydrogen at status d from storage grid ?̅? to demand grid 𝑛′ 𝐿𝑎𝑏𝑜𝑟_𝑇𝑅_𝑅𝑂𝑑𝑟′?̅? Driver wage of a truck delivering hydrogen at status d from storage grid ?̅? to demand grid  𝑟′ xix  𝐿𝐺𝑔′ Distance from Langley Township to demand grid 𝑔′ 𝐿𝐻_𝑁𝑉?̅?𝑛′ Distance from central storage grid ?̅? to demand grid 𝑛′ 𝐿𝐻_𝑅𝑂?̅?𝑟′ Distance central storage grid ?̅? to demand grid 𝑟′ 𝐿𝐻_𝑃𝑆𝑔?̅? Distance from central production grid g to central storage grid ?̅? 𝐿𝐻_𝑉?̅? Distance from central storage grid ?̅? to Langley Township 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝐶 Lifetime of a central plant 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝑆 Lifetime of a central storage facility (warehouse) 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝑂 Lifetime of an onsite plant 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝐷 Lifetime of a fueling station 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝑇𝑅 Lifetime of a truck 𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑡𝑖𝑚𝑒𝑑 Time to load a truck transporting hydrogen at status d 𝐿𝑅_𝑆𝑡 Learning rate of a central storage facility (warehouse) in time step t 𝐿𝑅_𝐶𝑡 Learning rate of a central plant in time step t 𝐿𝑅_𝑂𝑡 Learning rate of an onsite plant in time step t 𝐿𝑅_𝐷𝑡 Learning rate of a fueling station in time step t 𝑁 Number of years being studied 𝑁𝐷𝐶𝐶_𝐶𝑐𝑦 Non-depreciable capital cost of a central plant of type y and capacity c  𝑁𝐷𝐶𝐶_𝑆𝑐̅𝑑 Non-depreciable capital cost of a central storage facility of capacity 𝑐̅ which stores hydrogen at status d 𝑁𝐷𝑒𝑝_𝐶𝑎𝑝𝑖 Non depreciable capital costs in year i 𝑂𝑃_𝐶𝑐𝑦 Operating cost of a central plant of type y and capacity c ($/kg) 𝑂𝑃_𝑆𝑐̅𝑑 Operating cost of a central storage facility of capacity 𝑐̅  which stores hydrogen at status d ($/kg) 𝑂𝑃_𝑂𝑠 Operating cost of an onsite plant of capacity s ($/kg) 𝑂𝑃_𝐷𝑠𝑑 Operating cost of a fueling station of capacity s which delivers hydrogen at status d ($/kg) 𝑂𝑝𝑟𝑖 Operating costs in year i 𝑃𝑐𝑎𝑝_𝑚𝑖𝑛𝑐 Minimum production rate of a central plant with nominal capacity c 𝑃𝑐𝑎𝑝_𝑚𝑎𝑥_𝑝𝑟𝑖𝑛𝑐 Nominal production capacity of a central plant  𝑃𝐷_𝐼𝑛𝑐𝑖 Pre depreciation income in year i 𝑃𝑒𝑟_𝑡𝑧 Percentage of demand in the last year of each time step 𝑃𝑅_𝑐𝑎𝑝𝑗𝑦 Capacity expansion of a central plant of type y at stage j (%) 𝑟 Discount rate 𝑅𝑒𝑣_𝐻2𝑖 Revenue from hydrogen per year 𝑆𝑎𝑙𝑣 Salvage value 𝑆𝑐𝑎𝑝_𝑚𝑖𝑛𝑐̅ Minimum storage rate of a central storage unit (warehouse) with nominal capacity 𝑐̅ 𝑆𝑐𝑎𝑝_𝑚𝑎𝑥_𝑝𝑟𝑖𝑛𝑐̅ Nominal storage capacity of a central storage unit (warehouse) 𝑆𝑅_𝑐𝑎𝑝𝑗 Capacity expansion of a central storage facility at capacity expansion stage of j (%) 𝑇𝑐 Tax credit 𝑇𝑟 Tax rate 𝑇𝑡𝑖 Total taxes in year i xx  𝑢𝑛𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑡𝑖𝑚𝑒𝑑 Time to unload a truck transporting hydrogen at status d 𝑉_𝑂𝑝𝑟𝑖 Variable operating costs in year i 𝑉𝐺 Speed of a truck in demand regions g 𝑉𝐻 Speed of a truck in highways (from regions g’ to Langley Township) 𝑊_𝐶𝑎𝑝𝑖 Cash from working capital reserves in year i 𝑌_𝑅𝑒𝑝𝑖 Replacement costs per year 𝛼 Wage for truck driver (C$/hour) 𝛼_𝐿𝑅_𝐶 Learning index of a central plant   𝛼_𝐿𝑅_𝐷 Learning index of a fueling station   𝛼_𝐿𝑅_𝑂 Learning index of an onsite plant   𝛼_𝐿𝑅_𝑆 Learning index of a central storage facility (warehouse)  𝛽 Fuel cost (C$/litre) 𝛾 Fuel economy of the truck (litre/km) 𝜀 Small number 𝜔 Percentage of maximum capacity   Continuous Variables 𝐷𝐼_𝑉𝑠𝑑𝑔′𝑡 Dispensing rate of a fueling station with capacity s, delivering hydrogen at status d, in region 𝑔′ and time step t 𝐷𝐼_𝑁𝑉𝑠𝑑𝑛′𝑡 Dispensing rate of a fueling station with capacity s, delivering hydrogen at status d, in region 𝑛′ and time step t 𝐷𝐼_𝑅𝑂𝑠𝑑𝑟′𝑡 Dispensing rate of a fueling station with capacity s, delivering hydrogen at status d, in region 𝑟′ and time step t 𝑃𝐶𝑐𝑦𝑑𝑔𝑡 Production rate of a central plant with capacity c, type y, producing hydrogen at status d, in region g and time step t 𝑃𝑐𝑎𝑝_𝑚𝑎𝑥𝑐𝑦𝑑𝑔𝑡 Production capacity (maximum production rate) of a central plant with capacity c, type y, produces hydrogen at status d, in region g and time step t 𝑃𝑂_𝑉𝑠𝑑𝑔′𝑡 Production rate of an onsite plant with capacity s, producing hydrogen at status d, in region 𝑔′ and time step t 𝑃𝑂_𝑁𝑉𝑠𝑑𝑛′𝑡 Production rate of an onsite plant with capacity s, producing hydrogen at status d, in region 𝑛′  and time step t 𝑃𝑂_𝑅𝑂𝑠𝑑𝑟′𝑡 Production rate of an onsite plant with capacity s, producing hydrogen at status d, in region 𝑟′and time step t 𝑆𝑇𝑅_𝑉𝑠𝑑𝑔′𝑡 Storage rate of a fueling station with capacity s, delivering hydrogen at status d, in region 𝑔′ and time step t 𝑆𝑇𝑅_𝑁𝑉𝑠𝑑𝑛′𝑡 Storage rate of a fueling station with capacity s, delivering hydrogen at status d, in region 𝑛′ and time step t 𝑆𝑇𝑅_𝑅𝑂𝑠𝑑𝑟′𝑡 Storage rate of a fueling station with capacity s, delivering hydrogen at status d, in region 𝑟′  and time step t 𝑇𝑆𝑐̅𝑑?̅?𝑡 Storage rate of a central storage facility (warehouse) with capacity 𝑐̅, storing hydrogen at status d, in region ?̅? and time step t 𝑇𝑆_𝐸𝑐̅𝑑?̅?𝑡 Storage rate of a central storage facility (warehouse) for emergency with capacity 𝑐̅, storing hydrogen at status d, in region ?̅? and time step t   xxi  Integers 𝑁𝑁𝑇𝑅𝐺_𝑃𝑆𝑔?̅?𝑎𝑡 Number of new tube trailers of size a transporting gas hydrogen from a central plant in region g to a central storage in region ?̅? in time step t 𝑁𝑁𝑇𝑅𝐺_𝑉?̅?𝑔′𝑎𝑠𝑡 Number of tube trailers of size a delivering gas hydrogen from a central storage in region ?̅? to demand region 𝑔′ , stations of size s, in time step t 𝑁𝑁𝑇𝑅𝐺_𝑁𝑉?̅?𝑛′𝑎𝑠𝑡 Number of new tube trailers of size a delivering gas hydrogen from a central storage in region ?̅? to demand region 𝑛′ , stations of size s, in time step t 𝑁𝑁𝑇𝑅𝐺_𝑅𝑂?̅?𝑟′𝑎𝑠𝑡 Number of new tube trailers of size a delivering gas hydrogen from a central storage in region ?̅? to demand region 𝑟′ , stations of size s, in time step t 𝑁𝑁𝑇𝑅𝐿_𝑃𝑆𝑔?̅?𝑡 Number of new tanker trucks transporting liquid hydrogen from a central plant in region g to a central storage in region ?̅? in time step t 𝑁𝑁𝑇𝑅𝐿_𝑉?̅?𝑔′𝑡 Number of new tanker trucks delivering liquid hydrogen from a central storage in region ?̅? to demand region 𝑔′ in time step t 𝑁𝑁𝑇𝑅𝐿_𝑁𝑉?̅?𝑛′𝑡 Number of new tanker trucks delivering liquid hydrogen from a central storage in region ?̅? to demand region 𝑛′ in time step t 𝑁𝑁𝑇𝑅𝐿_𝑅𝑂?̅?𝑟′𝑡 Number of new tanker trucks delivering liquid hydrogen from a central storage in region ?̅? to demand region 𝑟′ in time step t 𝑁𝑇𝑅𝐺_𝑃𝑆𝑔?̅?𝑎𝑡 Number of tube trailers of size a transporting gas hydrogen from a central plant in region g to a central storage in region ?̅? in time step t 𝑁𝑇𝑅𝐺_𝑉?̅?𝑔′𝑎𝑠𝑡 Number of tube trailers of size a delivering gas hydrogen from a central storage in region ?̅? to demand region 𝑔′, stations of size s, in time step t 𝑁𝑇𝑅𝐺_𝑁𝑉?̅?𝑛′𝑎𝑠𝑡 Number of tube trailers of size a delivering gas hydrogen from a central storage in region ?̅? to demand region 𝑛′ , stations of size s, in time step t 𝑁𝑇𝑅𝐺_𝑅𝑂?̅?𝑟′𝑎𝑠𝑡 Number of tube trailers of size a delivering gas hydrogen from a central storage in region ?̅? to demand region 𝑟′ , stations of size s, in time step t 𝑁𝑇𝑅𝐿_𝑃𝑆𝑔?̅?𝑡 Number of tanker trucks transporting liquid hydrogen from a central plant in region g to a central storage in region ?̅? in time step t 𝑁𝑇𝑅𝐿_𝑉?̅?𝑔′𝑡 Number of tanker trucks delivering liquid hydrogen from a central storage in region ?̅? to demand region 𝑔′ in time step t 𝑁𝑇𝑅𝐿_𝑁𝑉?̅?𝑛′𝑡 Number of tanker trucks delivering liquid hydrogen from a central storage in region ?̅? to demand region 𝑛′ in time step t 𝑁𝑇𝑅𝐿_𝑅𝑂?̅?𝑟′𝑡 Number of tanker trucks delivering liquid hydrogen from a central storage in region ?̅? to demand region 𝑟′ in time step t 𝑌𝐷_𝑁𝑉𝑠𝑑𝑛′𝑡 Number of fueling stations with capacity s, delivering hydrogen at status d, in region 𝑛′ and time step t 𝑌𝐷_𝑅𝑂𝑠𝑑𝑟′𝑡 Number of fueling stations with capacity s, delivering hydrogen at status d, in region 𝑟′ and time step t 𝑌𝐷_𝑉𝑠𝑑𝑔′𝑡 Number of fueling stations with capacity s, delivering hydrogen at status d, in region 𝑔′ and time step t 𝑌𝑂_𝑁𝑉𝑠𝑑𝑛′𝑡 Number of onsite plants with capacity s, producing hydrogen at status d, in region 𝑛′ and time step t 𝑌𝑂_𝑅𝑂𝑠𝑑𝑟′𝑡 Number of onsite plants with capacity s, producing hydrogen at status d, in region 𝑟′ and time step t xxii  𝑌𝑂_𝑉𝑠𝑑𝑔′𝑡 Number of onsite plants with capacity s, producing hydrogen at status d, in region 𝑔′ and time step t 𝑌𝑃𝑂_𝑁𝑉𝑠𝑑𝑛′𝑡 Number of new onsite plants with capacity s, producing hydrogen at status d, in region 𝑛′ and time step t 𝑌𝑃𝑂_𝑅𝑂𝑠𝑑𝑟′𝑡 Number of new onsite plants with capacity s, producing hydrogen at status d, in region 𝑟′  and time step t 𝑌𝑃𝑂_𝑉𝑠𝑑𝑔′𝑡 Number of new onsite plants with capacity s, producing hydrogen at status d, in region 𝑔′ and time step t 𝑌𝑃𝐷_𝑁𝑉𝑠𝑑𝑛′𝑡 Number of new fueling stations with capacity s, delivering hydrogen at status d, in region 𝑛′  and time step t 𝑌𝑃𝐷_𝑅𝑂𝑠𝑑𝑟′𝑡 Number of new fueling stations with capacity s, delivering hydrogen at status d, in region 𝑟′ and time step t 𝑌𝑃𝐷_𝑉𝑠𝑑𝑔′𝑡 Number of new fueling stations with capacity s, delivering hydrogen at status d, in region 𝑔′ and time step t   Binaries 𝑌𝐶𝑐𝑦𝑑𝑔𝑡 1 if a central plant with capacity c, type y, producing hydrogen at status d exists in region g and time step t, 0 otherwise 𝑌𝑃𝐶𝑐𝑦𝑑𝑔𝑡 1 if a new central plant with capacity c, type y, producing hydrogen at status d is established in region g and time step t, 0 otherwise 𝑌𝑃𝐸𝑐𝑦𝑑𝑔?̅? 1 if electrolyzer undergoes a 10-year replacement, 0 otherwise 𝑌𝑃𝑆𝑐̅𝑑?̅?𝑡 1 if a new central storage facility with capacity 𝑐̅, storing hydrogen at status d, is established in region ?̅? and time step t, 0 otherwise 𝑌𝑆𝑐̅𝑑?̅?𝑡 1 if a central storage facility with capacity 𝑐̅, storing hydrogen at status d, is existing in region ?̅? and time step t, 0 otherwise 𝑌′𝐶𝑗𝑐𝑦𝑑𝑔𝑡 1 if a central plant with capacity c, type y, produces hydrogen at status d in region g and time step t, has an increased capacity with status of j, 0 otherwise 𝑌′𝐶_𝑂𝑐𝑦𝑑𝑔𝑡 1 if a central plant with capacity c, type y, produces hydrogen at status d in region g undergoes no capacity expansion until time step t, 0 otherwise 𝑌′𝑃𝐶𝑗𝑐𝑦𝑑𝑔𝑡 1 if capacity expansion at stage j is implemented at time step t for a central plant with capacity c, type y, producing hydrogen at status d in region g, 0 otherwise 𝑌′𝑃𝑆𝑗𝑐̅𝑑?̅?𝑡 1 if capacity expansion at stage j is implemented at time step t for a central storage facility with capacity 𝑐̅, storing hydrogen at status d, in region ?̅? and time step t, 0 otherwise 𝑌′𝑆𝑗𝑐̅𝑑?̅?𝑡 1 if a central storage facility with capacity 𝑐̅, storing hydrogen at status d, in region ?̅? and time step t, has an increased capacity with status of j, 0 otherwise 𝑌′𝑆_𝑂𝑐̅𝑑?̅?𝑡 1 if central storage facility with capacity 𝑐̅, storing hydrogen at status d in region ?̅? undergoes no capacity expansion until time step t, 0 otherwise xxiii  List of Abbreviations BAU Business as Usual B.C. British Columbia BEV Battery Electric Vehicle CCA Capital Cost Allowance CCS carbon Capture and Storage CHP Combined Heat and Power CLF Current Legislation Fulfillment DDCC Direct Depreciable Capital Cost DP Dynamic Programming FCEV Fuel Cell Electric Vehicle FOC Fixed Operating Cost GDP Gross Domestic Product GHG Greenhouse Gas HDT Heavy Duty Truck HFSC Hydrogen Fueling Supply Chain H2SCOT Hydrogen Supply Chain Cost Optimization Tool ICE Internal Combustion Engine IDCC Indirect Depreciable Capital Cost   IRR Internal Rate of Return LCFS low-carbon Fuel Standard LDT Light Duty Truck LFG Landfill Gas LOHC Liquid Organic Hydrogen Carriers MDT Medium Duty Truck MEA Monoethanolamine MILP Mixed-integer linear Programming NDCC Non-depreciable Capital Cost NEB National Energy Board O&M Operation and Maintenance PEM Proton Exchange Membrane PSA Pressure Swing Adsorption PTC Production Tax Credit REPC Replacement Cost RNG Renewable Natural Gas SMR Steam Methane Reforming TTW Tank-to-Wheel VOC Variable Operating Cost WCSB Western Canadian Sedimentary Basins WTT Well-to-Tank WTW Well-to-Wheels   xxiv  Acknowledgements First, I would like to express my sincere gratitude to my supervisor Dr. Walter Mérida for his support, guidance, and the opportunities he provided to me over the years. I would also like to extend my appreciation to my supervisory committee: Dr. Martino Tran, Dr. Farrokh Sassani, Dr. Taraneh Sowlati, and Dr. Harish Krishnan for their helpful feedback and guidance. I acknowledge the agencies that funded my research: the Natural Sciences and Engineering Research Council of Canada (NSERC) through Vanier CGS, the Pacific Institute for Climate Solutions (PICS) and the University of British Columbia (UBC). To the members of the Mérida lab and clean energy research centre (CERC), past and present, thank you for providing a warm, friendly, fun, and collaborative work environment.  I owe particular thanks to our project manager Dr. Omar Herrera, for all the inspiring discussions and ideas we shared. Special thanks to Nadia Adame. Your continual encouragement and optimism kept me excited about my work. I am particularly indebted to my parents, Vida and Bizhan. You have taught me that anything is possible, and your trust has given me strength and confidence throughout my years of education. Amin, my Love, this thesis would never have been completed without you. Thank you for your endless support, sacrifice, courage, and love, and for always pushing me to keep going. xxv  Dedication To Amin.1  Chapter 1: Hydrogen as an Energy Carrier Hydrogen, as an energy carrier, could help tackle climate change. Like fossil fuels, hydrogen can be stored, transported, combusted and combined in chemical reactions. Hydrogen can be produced from a wide range of energy sources, thus increasing the flexibility and sustainability of the energy system. If produced from renewable energy sources, hydrogen could decouple carbon emissions from the energy supply, while maintaining the same user experience as fossil fuels. Hydrogen, as a low-carbon chemical energy carrier, can deliver significant emissions reduction where direct electrification faces technological or economic obstacles.   1.1 Hydrogen applications To date, hydrogen has mainly been used as a feedstock in the refining and chemical industries (i.e., oil refining (33%), ammonia production (27%), methanol production (11%) and steel production (3%)) [1]. Hydrogen or hydrogen-based fuels (synthetic methane, methanol and ammonia), can be used for industrial purposes, transportation, indoor heating and power generation.  In the transportation sector, light-duty fuel cell electric vehicles (FCEVs) have received significant public attention due to longer driving ranges and refueling processes that are similar to those in gasoline vehicles. FCEVs could complement battery electric vehicles (BEVs) as zero-emission vehicles capable of reducing GHG emissions and local air pollution in cities. On the heavy-duty sector, fuel cells have so far powered forklifts and buses at a commercial scale, and medium- to heavy-duty trucks in demonstration projects.  Hydrogen can be used to provide heating, cooling and on-site electricity generation for buildings or local district energy networks. In the short term, blending hydrogen into existing natural gas networks can reduce emissions from the built environment. Longer-term prospects may include the direct use of hydrogen for heat generation via hydrogen boilers or combined heat and power (CHP) via stationary fuel cells [2].  Ammonia can partially substitute coal in coal-fired power plants. This can reduce emissions if ammonia is produced from low-carbon hydrogen. Ammonia and hydrogen provide flexible and low-carbon power generation options in gas turbines. Fuel cells can provide back-up for power outages and electricity generation for off-grid communities. Hydrogen and hydrogen based-fuels 2  can be used as mediums for large-scale seasonal energy storage to balance renewable electricity supply and demand [1], [2].  1.2 Current status and international targets  At the COP23 meeting in Bonn, the Hydrogen Council estimated that hydrogen could contribute approximately to 20% of the total abatement required by 2050 under the Paris targets [3]. By 2019, there were around 50 global targets, mandates and policy incentives in place that directly support hydrogen deployment in industry, transport, built environment and power generation. National hydrogen roadmaps have been developed in 9 countries among the Group of Twenty (G20) and the European Union [1]. Countries like Germany, Japan, China, Australia, France, Korea, Norway and the United Kingdom have devoted billions to the deployment of hydrogen infrastructure for mobility, cogeneration, and renewable storage. For example, China and the State of California are planning to build more than 1000 hydrogen refueling stations to support 1 million FCEVs by 2030; and Korea is targeting a shift to hydrogen of all conventional commercial vehicles by 2025. Japan launched Japan H2 mobility and targets to build 80 hydrogen fueling stations by 2021. Japan has invested on different large-scale hydrogen storage technologies such as chemical hydrides and is a leader in stationary fuel cell technology for micro-cogeneration. Germany developed H2mobility program to support the development of hydrogen fueling stations in national level and the first commercial hydrogen-powered train. Germany supports hydrogen-based seasonal energy storage projects to get the most benefit from renewable energy integration. United Kingdom is planning to blend up to 20% hydrogen in a regional natural gas network and secured funding for seasonal hydrogen storage including power-to-X [1], [4].   1.3 Hydrogen potential in British Columbia  Canada is one of the world’s largest producers of industrial hydrogen, which is mostly used in the chemical and refinery industries [5]. British Columbia (B.C.) has been a Canadian fuel cell hub for more than three decades. However, the fuel cell market has been focused on exports, with a modest domestic growth [6]. B.C. has a potential to benefit from its world-class fuel cell industry to empower the hydrogen economy in the province. Moreover, B.C. has abundant access to low-3  cost natural gas and hydroelectricity, as well as renewable energy sources (wind, geothermal, biomass) to produce hydrogen. Hydrogen can help B.C. to meet its decarbonization target, which requires 80% greenhouse gas (GHG) emissions reduction by 2050 from 2007 levels [7]. It should be mentioned that from 2007 to 2016, the total GHG emissions reduction was around 3% in B.C. [8]. Thus, the province must accelerate its effort to stay on the targeted carbon reduction path. Hydrogen’s role is critical, especially for road transportation, the hard-to-abate energy sectors (long-range transportation, heating and energy-intensive industries) and off-grid communities in B.C. The injection of renewable hydrogen to the natural gas grid and production of hydrogen-based synthetic fuels are potential short-term enablers for the province to meet its GHG emissions reduction target. Hydrogen export to California, Japan, South Korea and China may also be considered due to B.C.’s coastal access to those emerging markets. Hydrogen export is an opportunity for the province to attract international investment, empower the hydrogen industry in B.C., and decrease the hydrogen price in the domestic market.  1.4 Hydrogen role in B.C.’s road transportation sector Based on 2016 data, the transportation sector accounts for the largest portion of the total GHG emissions in B.C. (39%), and more than two thirds of these emissions originate from on the road vehicles [9]. The GHG emissions from the road transport sector increased by 14 % from 2007 to 2016. It is projected that the transportation demand increases as it is directly driven by the economic and population growth [10]. In 2019, the B.C. government passed the Zero-Emission Vehicles Act, which requires all new light-duty cars and trucks sold in the province to be zero-emission by 2040 [11]. All-electric vehicles are the only available options with zero-tailpipe emissions. Thus, FCEVs can complement BEVs to meet this target.   B.C. is ready to adopt hydrogen in the road transport sector. The province deployed the world’s largest fleet of hydrogen fuel cell buses for the 2010 Winter Olympic Games [12]. The hydrogen fuel infrastructure program started in B.C. in 2015 through the Clean Energy Vehicles for British Columbia [13], a policy initiative that provides incentives at the vehicle’s point of sale and for the development of fueling stations. As a result, the first two fully public hydrogen fueling stations in Canada launched in Vancouver, B.C., in 2018 and 2019 as part of a plan to deploy a 6-station network in the Lower Mainland and Victoria [14]. As of July 2019, certain light duty FCEV models 4  are available for purchase in B.C. (e.g., the Toyota Mirai and the Hyundai Nexo), and the first FCEV fleet was announced recently [15]. The province has not yet announced a plan to deploy fuel cell electric trucks to decarbonize the road freight transport. As a very first attempt to incorporate hydrogen in the freight sector, the hydrogen-diesel co-combustion class 8 trucks are being tested in B.C [16].  In order to expedite the FCEV market growth in B.C., the government must develop favorable policies to support the purchase of these vehicles and the development of the hydrogen fueling supply chain (HFSC). As discussed in the next chapter, this infrastructure precedes vehicle adoption. It requires substantial capital investment and is subjected to a negative cash flow that may last for years to decades. This work is a first attempt to develop the most cost effective HFSC plan for B.C. to insure the successful deployment of FCEVs in the province.  5  Chapter 2: Hydrogen Supply Chain for Mobility 2.1 Hydrogen supply chain structure A supply chain is a network of interlinked facilities, engaged in the consistent flow of goods from production to the end user. The hydrogen fueling supply chain (HFSC) consists of a network of integrated facilities to produce, transport, store, distribute, and dispense hydrogen. This infrastructure is similar to the current petroleum-based supply chain. Unlike the petroleum counterpart, hydrogen can also be produced at the fueling stations to fulfill demand.  The main building blocks of an HFSC are as follows:  2.1.1 Production facilities Hydrogen can be produced via thermochemical, electrolytic, photoelectrochemical and biological processes. The thermochemical processes use thermal energy to extract hydrogen from the hydrocarbon-based fuels. Steam reforming of natural gas, partial oxidation of hydrocarbons and coal and biomass gasification are the thermochemical mature technologies for hydrogen production [17]. The electrolytic process uses electricity to split water into hydrogen and oxygen. Alkaline electrolysis, proton exchange membrane electrolysis, and solid oxide high-temperature electrolysis are the industrial water-electrolysis technologies. The electrolytic process creates an opportunity to utilize renewable energy sources such as hydropower, wind, and solar energy for hydrogen production. In photoelectrochemical processes the solar energy dissociates water using semiconductor materials. In biological processes hydrogen is produced as a by-product of microorganism metabolism using sunlight to breakdown water or organic matter. The photoelectrochemical and biological processes are in the early stages of development [18]. Biomass gasification is a mature technology; however, the capital costs of equipment and biomass feedstocks restricts the commercial adoption of this technology to date. Among the aforementioned technologies, steam methane reforming, oil and naphtha reforming, coal gasification and water electrolysis are commercially viable for large scale hydrogen production. Carbon capture and storage (CCS) may also be integrated to reduce the GHG emissions from the hydro-carbon based production pathways. CCS is a process in which the CO2 generated from industrial activities is 6  separated and transported to storage locations. CO2 is then injected into subterranean geological formations for long-term isolation from the atmosphere [19].  2.1.2 Terminals and storage facilities Hydrogen terminal includes the storage and conditioning facilities to feed hydrogen into the distribution network.  Hydrogen storage technologies can be divided into two groups: physical-based and material-based. Hydrogen can be stored physically as a gas or a liquid. As a compressed gas, hydrogen is stored in high-pressure cylindrical vessels for short-term and low demand and in large underground caverns for seasonal demand coverage [20]. Spherical double isolated cryogenic tanks are used for liquid hydrogen storage. In this case, a liquefaction unit is required to convert gas to liquid hydrogen. Compressors and high-pressure cryogenic pumps are also required at the terminal to load gas and liquid hydrogen onto the tube trailers and tankers, respectively [21].   Material-based storage has two main sub-groups of chemical sorption and physical sorption [22]. Hydrogen can be stored in solid-state at moderate pressures and temperatures. This is achieved by an exothermic process in which hydrogen is absorbed in the interstices of metallic alloys or adsorbed on high surface area materials such as activated carbons. An endothermic process is then required to separate the hydrogen from the metal. This reversible process happens in a metal hydride tank. The tank is loaded with hydrogen storage alloy powder, and consists of heat exchange parts and gas transport components [23]. The organic chemical hydride method uses chemical sorption, in which an aromatic compound like toluene is used to convert hydrogen to a saturated cyclic compound. The aim is to store and transport hydrogen medium in atmospheric pressure and temperature. Pure hydrogen is generated by dehydrogenation reaction at the point of use [24].   2.1.3 Hydrogen delivery network Hydrogen delivery consists of hydrogen transmission from the central production to terminals and hydrogen distribution from terminals to the fueling stations. Gaseous hydrogen is delivered on the road by high pressure tube trailers (long steel tubes or composite storage vessels stacked on a trailer) when low volume of hydrogen is required in short distances.  Pipeline is a suitable option 7  for gaseous hydrogen delivery in large demand sizes and to dense areas. Liquid hydrogen is transported on the road in super-insulated, cryogenic tanker trucks. This mode of delivery is suitable for moderate demand and long-distances (the range of hydrogen flow and transport distance for each delivery mode is presented in  [25]). Other potential hydrogen transport modes are rail, barge, and ship; however, they are not yet at a commercial scale [26].   2.1.4 Hydrogen fueling stations Hydrogen fueling stations dispense hydrogen in a form of compressed gas to vehicles. The dispensers may accommodate both 70 MPa and 35 MPa, depending on the type of vehicle being served. The components of a hydrogen fueling station vary with respect to the state of hydrogen received. Compression unit is required when gaseous hydrogen is delivered via tube trailers or pipelines. Liquid pump and evaporation unit (or evaporation and compression unit) are required when liquid hydrogen is delivered to the station. The fueling station may be equipped with a steam methane reformer or electrolyzer to produce hydrogen at the station in small scales. At very early stages of hydrogen penetration to the market, mobile hydrogen fueling stations can be used to provide self-contained hydrogen dispensing capabilities (on-board compression, storage, dispensing and power) to serve low-demand and remote areas [27].  2.2 Deployment challenges Hydrogen fueling supply chain (HFSC) represents a capital-intensive investment, facing high risks of negative cash flow for years to decades. The network of fueling stations along with the upstream supply infrastructure (i.e., production, storage, transport and distribution facilities) must be developed in advance of the fuel cell vehicle roll-out. This is to assure the hydrogen demand satisfaction for the vehicle manufacturers and potential customers [28]. Even at the early stages of demand growth, HFSC faces underutilization, which threatens its economic viability. Moreover, the hydrogen supply chain pathways are diverse. Each combination of technology, scale and location of the components imposes varying costs on the entire supply chain. The network design process is time-dependent and region-specific [29]. The investment decisions which are not supported by rigorous analysis of the spatial and temporal factors (e.g., available energy sources, 8  demand characteristics, local energy prices and decarbonization policies) may face serious financial consequences.   2.3 Hydrogen supply chain design approaches The supply chain network design, also known as supply chain planning, is the process of modeling a supply chain based on strategic targets of the project and the available resources.  The network design of a HFSC has been studied extensively [30], [31]. These studies have been oriented to the strategic decision phase, aimed to generate spatial and temporal decisions on the configuration of the HFSC.  Simulation and optimization are the formal quantitative approaches to design an HFSC. Typical simulations assess predefined pathways, from production to distribution of hydrogen. These simulations usually target economic or environmental performance metrics [32], [33]. Optimization approaches can be used to scan a superstructure that embeds all the possible configurations of a supply chain in an integrated mathematical framework. The optimization models identify the optimal pathway, with respect to the desired performance measures and a set of technical, spatial, temporal, and environmental constraints. These models can be categorized with respect to the spatial measure in to national, regional and local scale. When embedded in a national or global energy system optimization, the hydrogen supply and demand are endogenously optimized through interactions within all energy sectors [34]. For instance, the Energy Technology Systems Analysis Program-MARKet Allocation Model and its successor, the Integrated MARKAL-EFOM, are popular bottom-up linear optimization tools for entire energy system cost minimization [35].  Hydrogen pathway assessments have been integrated within these tools at national and large regional scale for the UK [36], California [37], Italy [38], Japan [39], Spain [40] and Norway [41]. The regional-scale HFSC models optimize the spatially explicit supply chain configurations, considering the demand as an exogenous parameter. These models cover the spatial dynamics of transitions in more detail, while ignoring the dependency of the hydrogen demand to the techno-economic specifications of the overall energy system [42]. The local scale models have been focused on the hydrogen fueling station siting problem. These models optimize the location of hydrogen fueling stations in a relatively small region (e.g., cities) [43], [44], based on the 9  classical facility location optimization techniques such as generalized approach, the p-median and flow intercepting [45].  The optimization models have been widely developed in the literature using mixed-integer linear programming (MILP) techniques. Only one study has been found using dynamic programming (DP) technique to optimize the HFSC [46].  The HFSC models adopt mono- or multi-objective frameworks. The most desired performance measure for mono-objective models is minimizing the total cost of the system [47]–[50] or maximizing the profit [51], [52]. The multi-objective frameworks assess the cost in conjunction with other performance measures such as safety risk [53], [54] and environmental impact minimization [55], [56]. The ε-constraint method dominates the solution approaches to solve multi-objective HFSC problems. This method generates a full set of trade-off solutions based on optimizing one objective function while considering the other objectives as constraints [57].  The HFSC optimization models are also categorized into deterministic and stochastic (or probabilistic) classes based on the nature of input parameters. All spatial, temporal and operational parameters are fixed in a deterministic setting. In stochastic models, uncertainty is introduced in at least one parameter. Uncertainties are classified in three distinct categories: demand uncertainty, process uncertainty and supply uncertainty. The demand uncertainty is the parameter used most frequently, introduced via scenarios with known probabilities [58], [59]. The two-stage linear stochastic programming technique is used to deal with the scenario-based uncertainty inclusion [60].  The HFSC can also be analyzed as static or multi-period models. The static models optimize the HFSC at a point in time [47], [61], while the multi-period models optimize the evolution of the supply chain over a predefined planning horizon [59], [62].   2.4 Environmental considerations in the hydrogen supply chain design Fuel cell electric vehicles have zero tailpipe emissions; though, the upstream GHG emissions from the hydrogen supply chain may limit its benefits as a low-carbon fuel. Emissions are mostly involved in the production and distribution stages of this supply chain.  10  2.4.1 Low-carbon hydrogen pathways The emission reduction potential of hydrogen can only be exploited fully when it is produced through low-carbon pathways. On the production side, conventional fossil fuel-based technologies must be equipped with carbon capture and utilization or storage (CCU or CCS). Currently, low-carbon hydrogen production technologies such as anaerobic digestion, photo fermentation, bio electrochemical systems, and artificial photosynthesis are at the laboratory scale or demonstration stage [63], [64]. Only water electrolysis has increased its share to 4% of the global hydrogen production in the last decade [1]. Only low- or zero carbon electricity (e.g., from renewable sources) can enable significant emissions reduction in hydrogen production from water electrolysis. On the distribution side, the GHG emissions from the diesel trucks, transporting hydrogen from production facilities to the fueling stations, must be reduced. This is achievable through performance improvement of the diesel trucks in the short term and switching to all-electric trucking in the long term [65]. Moreover, hydrogen pipeline transport is economically and environmentally competitive for concentrated large-scale demand [25].    Because hydrogen production from fossil sources (without CCS integration) is still the most economically viable solution, external incentives are required to empower the low-carbon hydrogen production as discussed in the next section.  2.4.2 Enabling low-carbon hydrogen production The following factors could contribute to enhance the economic viability of low-carbon hydrogen production: - Expansion of the hydrogen market  Learning-by-doing and economies-of-scale can reduce the costs and increase the effectiveness of the low-carbon hydrogen pathways [1]. Market expansion could be achieved by considering applications beyond transportation. A wider energy system could, for example, include hydrogen injection into natural gas grids [66], or hydrogen use as an energy storage medium for heat and power generation [67].   11  - Government financial support and favorable regulations Government policies could accelerate the transition toward green hydrogen, especially if they target hydrogen technologies explicitly and promote the renewable energy capacity installations. The policies can apply economic instruments (fiscal and financial, direct investment or market measures), regulations, standards, long-term targets, and RD&D support [68].  Thus far, national policies on transport decarbonization have focused on energy efficiency improvement for combustion engines, biofuel adoption and modal switches (e.g., public transport, biking, walking, etc.) [69]. So far, the hydrogen policies in transport sector have attempted to decrease the risk and cost of early stage FCEV adoption, without considering low-carbon hydrogen production explicitly. The current policies can be separated into financial and regulatory frameworks, and categorized with respect to consumers, automakers and fuel providers.  Consumer-side policies include vehicle purchase subsidies, vehicle purchase tax exemption, free parking, access to high-occupancy vehicle (HOV) lanes and free fueling. Such policies exist in California, Denmark, Germany, South Korea, and the UK [70]–[72].  Automakers are affected by zero emission vehicle regulation and fuel economy targets [73]. Fuel suppliers are affected by low carbon fuel regulation [74], renewable fuel standard [75], and direct subsidies for infrastructure development. In Japan, Germany and California, subsidies up to $61m, $466m and $100m, respectively, have been allocated for the development of hydrogen fueling stations [76]. Low-carbon hydrogen production regulations and subsidies may encourage fuel suppliers to develop a sustainable hydrogen fueling network.  - Regional energy profile A favorable regional energy profile is critical for the long-term economic feasibility of low carbon hydrogen pathways.  Such profile may include the type and amount of renewable energy available, domestic or imported natural gas, geological suitability for CO2 storage, and access to adequate supplies of water for electrolysis.   2.4.3 Low-carbon hydrogen integration in supply chain optimization With no emissions constraints or incentives in place, the HFSC model favors the fossil-fuel hydrogen production technologies in a cost optimal pathway.  So far, the optimization models included emissions reduction targets as constraints on the HFSC operation or added the carbon tax 12  as a cost parameter to the cost minimization objective function.  Almansoori and Betancourt-Torcat [61] developed a mono-objective optimization framework to minimize the total cost of HFSC in Germany by 2030. The effect of carbon tax and CO2 emissions target scenarios was investigated on the configuration of the optimal supply chain. In a study by Moreno-Benito et al [77], the carbon tax was included in the economic objective function of a multi- period model to optimize the HFSC in the UK. Yang and Ogden [37] used the TIMES modeling framework to assess the long-term development of HFSC for California. The model was subjected to carbon tax as well as a number of emissions reduction constraints, including various scenarios on the regulatory part of the low carbon fuel policy (as a carbon intensity constraint), the renewable hydrogen mandate, which requires a minimum contribution of renewably produced hydrogen to the total hydrogen supply, and prohibition on coal gasification without CCS inclusion.  A number of other studies justified the cost optimal inclusion of low-carbon hydrogen pathways by assuming large hydrogen demand penetration into different energy sectors [20], [78], [79].    2.5 Hydrogen fuel supply chain design in British Columbia The HFSC planning in Canada is still in its infancy and has not yet been supported by formal optimization modeling. The only regional-based study was performed by Liu et al. [80] for the province of Ontario. Three FCEV market penetration scenarios were projected, and the cost of hydrogen production, storage, and distribution was calculated for a distinct pathway in each demand scenario. As discussed in chapter 1, among Canada’s provinces and territories, British Columbia is well positioned to take advantage of its abundant natural resources and carbon policies to develop a hydrogen fueling network.  2.5.1 Objectives The objectives of the current work are listed as follows: - Development of a comprehensive hydrogen supply chain cost optimization tool (H2SCOT) for the long-term investment planning of hydrogen fuel supply chain (HFSC) at low demand. This model was applied to a case study of light duty passenger vehicles in British Columbia [81]. 13  - Explicit integration of a range of emissions mitigation policies to the HFSC optimization model. - Efficiency assessment of the current policies in road freight transport and the potential contribution of zero-emissions trucks to meet the provincial GHG emissions reduction targets [65].  2.5.2 Contributions - This study is a first attempt to develop a hydrogen supply chain cost optimization tool (H2SCOT) in Canada and British Columbia.  - From the modeling perspective, H2SCOT includes a more comprehensive representation of the HFSC components compared to previous models, as summarized in Table 2.1. The multi-period, spatial-explicit MILP model by Moreno-Benito et al. [82] has the closest superstructure to the current model. Moreover, H2SCOT supports fueling stations and on-site hydrogen production with varying capacities, considers three alternative capacities for gaseous delivery, and includes a capacity expansion option (capacity expansion) for central production and storage facilities. H2SCOT deals with the low hydrogen demand in B.C., as opposed to large demands reported previously (Table 2.1). The aforementioned features enabled proper facility sizing to avoid underutilization costs. Moreover, H2SCOT supports storage facilities for fueling stations and ensures minimum storage requirements will be met. This option was included to cover hourly demand fluctuations at the fueling station. H2SCOT considers the lifetime of all components and the yearly replacement cost of facilities. - Policies aimed at reducing greenhouse gas emissions and improving air quality are often designed to promote the adoption of low-carbon fuels, or zero emission technologies. Hydrogen and its related technologies are often included indirectly, ignored, or excluded explicitly. The lack of specificity in generic policies implies that their impact on hydrogen adoption can be masked by financial or technological artefacts. In this study, a wide range of economic instruments and regulatory measures was included explicitly in H2SCOT. Compared to the previously developed models, H2SCOT is the first attempt to quantify the 14  effectiveness of existing and potential policies on the accelerated adoption of low-carbon hydrogen in the transport sector. - This work is the first contribution to measure the effectiveness of current policies in road freight transport and the potential of zero-emission trucking to meet the provincial GHG emissions reduction targets in B.C.  2.5.3 Approach (thesis outline) The superstructure of the HFSC, considered in this work, is presented in Figure 2.1. This diagram incorporates the potential pathways to produce, transport, store, distribute, and dispense hydrogen for the province of B.C. These pathways were developed based on the availability of local energy sources to produce hydrogen, the commercially available technologies and the projected level of hydrogen demand in the province.  The HFSC superstructure was used to develop the optimization model, as shown in Figure 2.2. The inputs of H2SCOT were defined and formulated in Chapter 3. These inputs are the capital and operational costs and the fuel-side GHG emissions of all alternative components of this supply chain; Hydrogen demand which is exogenously determined by a sub-model for each region over the studied time frame; and the potential supply and demand regions and corresponding distances. Moreover, a number of economic and regulatory instruments (emissions mitigation policies) with various stringencies were defined and formulated in Chapter 3.  The formulation of H2SCOT is presented in Chapter 4. H2SCOT has been developed based on a MILP formulation and is subjected to a number of constraints including mass balance, demand satisfaction, technology capacity limits, and non-negativity. The objective function is to minimize the discounted total cost of infrastructure, which includes the discounted cost of technology and the discounted cost of environmental policies. H2SCOT incorporates 6 equal time steps for the development of an HFSC from 2020 to 2050. The model output comprises of the optimal configuration of HFSC including location, number, type of technology, and capacity of the supply chain’s production, storage, and dispensing components, the average annual hydrogen production, storage, and dispensing rates, as well as the number and type of transportation and distribution trucks between the supply and demand regions.  15  Chapter 5 compares the cost optimal configuration of the HFSC for three demand scenarios in case of light duty passenger FCEVs penetration in B.C.  Chapter 5 also includes the efficiency assessment of current and potential financial and regulatory policies on the environmental and economic performance of the HFSC.   Chapter 6 assesses the potential contribution of battery electric and fuel cell electric trucks to meet GHG emissions reduction targets in road freight transport in B.C. The analysis was based on the efficiency assessment of current policies and the availability of regional resources to support all-electric trucking in B.C.  Chapter 7 provides the main conclusions and limitation of this study along with recommendations for a future work. 16   Table 2.1. Modeling details of the previously developed platforms compared with H2SCOT.1 Study Supply chain components2 Time evolution On-site production Capacity alternative Storage levels of fueling station Capacity expansion Assess Component Lifetime2 Yearly replacement cost Emission policy  Fueling station On-site production Gaseous delivery Guillén-Gosálbez et al. [55]  Sabio et al. [83] CP, CS, TN Multi-period No No No No No Yes3 No No No Almansoori & Shah [59] De-León Almaraz et al. [54] CP, CS, TN, FS Multi-period No No No No No No CP, CS No No Han et al. [84] CP, CS, TN Time-invariant No No No No No No No No Emissions trading4 Dayhim et al. [85] CP, CS, TN Multi-period No No No No No No No No Carbon tax Almansoori & Betancourt-Torcat [61] CP, CS, TN Time-invariant No No No No No No No No Carbon tax5 Moreno-Benito et al. [82] CP, CS, TN, FS Multi-period Yes Yes No No No No CP, CS, TN, FS No Carbon tax, Carbon intensity constraint Yang & Ogden [37] CP, CS, TN, FS Multi-period Yes No No No No No No No Carbon tax, Carbon intensity constraint, Technology ban H2SCOT (this study) CP, CS, TN, FS Multi-period Yes Yes Yes Yes Yes Yes CP, CS, TN, FS Yes Carbon tax, Credit trading, Production tax credit, Capital subsidy, Accelerated depreciation, Utility subsidy, Technology ban 1 This table excludes studies that only contain a qualitative description of the model and present very limited or no information on the formulation, such as Kamarudin et al. [86], Ball et al. [62], Hugo et al. [87], Konda et al. [88], and Stiller et al. [89]). 2 CP: central production, CS: central storage, TN: transportation network, FS: fueling station. 3 The continuous capacity expansion of facilities over time and within certain limits was considered. The shortfall is that constant capacity cannot be maintained for the successive time steps (the capacity expands at each time step or new facilities will be built). 4 Production emissions only. 5 The CO2 emission target was enforced to the constraints of the model.17  HydroelectricityNatural GasBy-product HydrogenH2ElectrolysisPurification PlantCompressionLiquefactionTube TrailerTanker TruckPressurized Cylindrical VesselsSuper-insulated Spherical TanksTube TrailerTanker TruckSteam Methane ReformingSteam Methane Reforming + CCSGas-Gas DispenserOn-site Electrolysis+Gas-Gas DispenserLiquid-Gas DispenserEnergy Source ProductionProduct Conditioning (Terminal)TransportWarehouse ( Storage)Distribution Dispensing Figure 2.1. Superstructure of the HFSC infrastructure in B.C. (CCS: carbon capture and sequestration)            18                     Figure 2.2.  Schematic of H2SCOTMinimize    𝒛 = ∑ 𝑐𝑗𝑇𝑥𝑗 + 𝑑𝑗𝑇𝑦𝑗𝑛𝑗=1  hydrogen demandYear2020 2050 Subject to    ∑ 𝐴𝑗𝑥𝑗 + 𝐾𝑗𝑦𝑗𝑛𝑗=1 ൝≤=≥ൡ 𝑏𝑖     for 𝑖 = 1. . 𝑚                          0 ≤ 𝑥𝑗 ≤ 𝑢𝑗   & 𝑦𝑗 ∈ 𝑍    for 𝑗 = 1. . 𝑛     Capital and operational  costs and the fuel side  GHG emissions of the  potential components  Hydrogen  Demand  Scenarios  Number, Type, Capacity, Storage rate, Location Number, capacity, Dispensing and storage rates, Location Number, Capacity, Production and storage rates, Location Share of central Production Share of  on-site Production Mode, Flow Number, Type, Capacity, production rate, Location Energy Source Production  Conditioning   & Storage Transport Fueling Station Geographic data: potential supply and  demand regions and  corresponding distances  Environmental policy strategies 19  Chapter 3: Hydrogen Supply Chain Cost Optimization Model (H2SCOT): Model Inputs 3.1  Assessment of energy sources A wide range of energy sources can be used to produce hydrogen. In this work, hydrogen pathways were developed based on the availability assessment and economic advantages of the local energy sources in B.C.   3.1.1 Hydrogen production from renewable energy sources Electricity generated from renewable energy sources can be used for electrolytic hydrogen production. Hydropower is responsible for around 92% of the total electricity generation in B.C [10]. The National Energy Board (NEB) projections [10] stated that the total electricity generation in B.C. will be around 81.1 TWh in 2040, of which 86% will be generated from large-scale hydroelectric dams. As B.C. is expected to rely heavily on the affordable hydroelectric power for a long time, harvesting other renewable resources for electricity generation is dependent on their economic viability.  The latest BC hydro integrated resource plan [90] assessed the long-term electricity generation potential of several renewable resources like wind, geothermal, biomass, solar, tidal and wave energy based on the technical and cost attributes. The results indicated that the wind, geothermal, and biomass resources have the least levelized1 energy costs. The total technical onshore and offshore wind potential in B.C. was estimated at 102 TWh, of which 43% can be harvested for less than $200 per MWh. The geothermal resource potential was estimated at around 12 TWh, of which 50% is below $200 per MWh. It is worth mentioning that only conventional hydrothermal resources using flash or binary technologies are considered within BC Hydro’s resource assessment. The wood-based biomass resources available for bioenergy production were estimated at 3.22 million tonnes of dry wood [91] which translates to the technical electricity generation potential of 4.5 TWh, mostly below $200 per MWh. It should be mentioned  1 The levelized cost of a unit of energy ($/MWh) from a resource is the ratio of the present value of the total annual cost of an energy resource to the present value of its annual average energy benefit. The levelized cost is dependent on the accessibility of the generation sites to powerlines.  20  that available biomass for bioenergy production is referred to as the part of wood waste supply that are surplus to the demand of existing forest industry. From the NEB projection database [10], 1.4% of B.C.’s wind resource potential and 40% of the combined biomass and geothermal potential will be used for electricity generation in 2040. These projections also stated that the electricity generation in B.C. will surpass the demand by 14% in 2040. This potentially translates to 226,000 tonnes of electrolytic hydrogen (32 PJHHV) produced at a rate of 50.2 kWh/kgH2. To put this into perspective, the total energy demand from the light duty passenger vehicles in B.C. was around 258 PJ in 2017 [9].  Biomass can be used directly to produce hydrogen through a gasification process. The availability of standing timber, pulp logs, roadside wood waste and sawmill wood waste in B.C. was forecasted to 2040 [91] using B.C.’s fibre model [92]. Different types of available wood-based biomass feedstock in B.C are categorized in Table 3.1 based on the percentage distribution and the average delivered fibre cost. The road-side logging residues may contribute to a larger share of hydrogen production compared to the other wood-based feedstock considering both the distribution percentage and the delivered cost. The available wood-based biomass in B.C. for bioenergy production translates to the technical hydrogen production potential of 334 and 233 kilotons per year (47 and 32 PJHHV/ year), considering 13.8 kg dried wood biomass is required to produce 1 kg of hydrogen.   Table 3.1. Distribution and delivered cost of the wood-based biomass feedstock in B.C. [91] Type  Distribution Average delivered fibre cost (C$/tonne of dry wood)  2016-2025 2026-2040 Standing sawlog timber 72.5% 62% 170.7 Pulp logs 8% 8% 129 Road-side logging residues 15.5% 23% 77 Sawmill hog fuel 4% 7% 30.3  Biogas is another source of potential hydrogen production made up primarily of 50%–70% bio-methane. Feedstocks for renewable natural gas (RNG) production are organic wastes from farms, forests, landfills, and water treatment plants. The landfill gas (LFG) and was considered as a source of biogas in this work. In order to avoid double counting the resource potential, the available biomass in B.C. was not considered as a source of RNG. A report from Golder Associates [93] is the only available resource for LFG assessment in the province. This report considered all the 21  operating municipal solid waste landfills in 2006 in B.C. with a minimum disposal rate of 10,000 tonnes per year and projected methane generation potential from the landfills to 2020. In this work, the “business as usual” projection was used to calculate the methane generation potential in 2050.  It should be noted that only the landfills with minimum methane flow rate of 200 cfm were selected as they provide sufficient economic incentives for developing LFG projects. The steam methane reforming (SMR) technology was adopted to assess the potential hydrogen production from the bio-methane captured from LFG recovery units with 75% recovery factor [94]. The bio methane recovery potential was estimated at 128000 tonnes in 2050, which translates to around 38000 tonnes of hydrogen (5.4 PJHHV) using the conversion rate of 3.4 kg methane per kg of hydrogen. It should be noted that there is a target of 5 percent RNG-blended natural gas in the pipeline distribution system by 2025 and 10 percent by 2030 [10].  This is equivalent to 50 PJ in 2030 as the projected natural gas demand in B.C. is around 500 PJ in 2030.  Thus, it is unclear if the limited resource potential of RNG could practically contribute to hydrogen production. By-product hydrogen vented from chemical plants can also be considered renewable if renewable electricity is used for the electrolytic process in the plant. In B.C, a sodium chlorate plant in Prince George and a chlor alkali plant in North-Vancouver use grid-connected hydroelectricity and vent 18500 kg/ day hydrogen [95]. By-product hydrogen can be captured and purified for a range of applications.   3.1.2 Hydrogen production from non-renewable energy sources B.C. is Canada’s second largest natural gas producer [96]. In this work, B.C.’s raw gas established reserve potential was targeted to assess hydrogen production using SMR technology. The ultimate potential for marketable natural gas (NG) in B.C. is estimated at 15547 billion m³. This is equivalent to 3.3 billion tonnes of hydrogen (468000 PJHHV) using the conversion factor of 4.74 m3 NG per kg of hydrogen in an SMR unit. Based on the NEB projection [10], local demand of natural gas in B.C. will be around 20% of the total production between year 2020 and 2040 from both conventional and unconventional deposits.  NEB presented different scenarios for natural gas production in B.C. The available natural gas for hydrogen production in 2040 was calculated at 76 billion m3 based on the average value of different natural gas production scenarios, and the local NG demand projection (roughly equivalent to 16 million tonnes of hydrogen (2270 PJHHV)).  22  Coal is B.C.’s most valuable mined commodity in terms of annual sale. The coal mines in the province mostly produce a metallurgical grade coal, which is exported to Asia, Europe and South for steel manufacturing. The demonstrated mineable coal resource is around 8400 million tonnes [97], which is equivalent to 853 million tonnes of hydrogen (121040 PJHHV), using the conversion factor of 10 kg of coal per kg of hydrogen in a gasification process. Using the “business as usual” projections, the coal production reaches 30.6 million tonnes annually by 2040. This is equivalent to 3 million tonnes of annual hydrogen production (426 PJHHV /year). It should be noted that the thermal coal with 5% hydrogen content is preferable for hydrogen production compared to the metallurgical coal with 2% hydrogen content. Only 20% of the total coal resources in B.C. are thermal grade coal.  3.1.3 Selected energy sources for hydrogen production In this work, the grid-connected renewable electricity and natural gas was selected along with the available by-product hydrogen from the chlor alkali plant in North-Vancouver. As the hydrogen demand in this work was restricted to the light duty passenger vehicles (section 3.4), it can be fulfilled with widely available natural gas and extra hydroelectricity generation in the next decades. In case of a wider market penetration, the resource assessment showed that wind, geothermal power and biomass can also be harvested to fulfill the hydrogen demand. It should be noted that the Clean BC plan set GHG intensity limit for gasoline and diesel by 2030. This may restrict biomass availability for hydrogen production, as biomass-based fuels (corn ethanol, methanol and biodiesel) are required to increase the renewable content of the fossil fuels. Moreover, coal gasification has a narrow window of opportunity in B.C., due to the small share of mineable thermal coal and the dependency on CCS integration.   3.2 Geographic divisions Fourteen supply regions in B.C. were selected as potential locations for central production facilities and central storage facilities as shown in Figure 3.1. The type of production technology in each region depends on the accessibility to major natural gas pipelines, BC Hydro power transmission lines, by-product hydrogen, and potential carbon sequestration sites. Accordingly, the electrolysis and SMR option were not considered for regions 1 and 5, respectively. The western Canadian 23  sedimentary basins (WCSB), regions 1 and 2, are considered as the potential carbon storages sites. The WCSB composed of depleted gas reservoirs and saline aquifers with the aggregate storage potential of more than 3000 Mt CO2 per year [98]. Thus, SMR plants with CCS integration could potentially be built in these regions. Demand regions in this study are confined to the major metropolitan areas (early adopters are more likely to live in urban areas where the first fueling stations will be built due to a higher population density and per-capita income [99]). Based on the population size, 10 municipalities in Metro Vancouver, Victoria on Vancouver Island, Kelowna and Kamloops in the Southern Interior, Prince George in the North Central area were selected as demand regions. Abbotsford, Hope, Whistler, and Williams Lake were also added because they are located on the busiest roads. The distribution of hydrogen demand among those municipalities over time is discussed in section 3.5.       1: Fort Nelson 2: Fort St John 3: Prince George 4: Williams Lake 5: Mica Creek 6: Prince Rupert 7: Kamloops 8: Merritt 9: Kelowna 10: Nelson 11: Kimberley 12: Hope 13: Victoria 14: North Vancouver 610 1141 2 3 57891213  14Figure 3.1. Hydrogen supply regions in British Columbia 24  3.3 Techno-economic and environmental data In this section, the capital and operating costs of each potential component of the hydrogen supply chain is derived alongside the GHG emissions associated with the flow of hydrogen through the supply chain.  3.3.1 Derivation of the techno-economic parameters  3.3.1.1 SMR plant A steam methane reforming (SMR) plant consists of a steam reforming furnace, shift reactors and a pressure swing adsorption (PSA) unit. The furnace converts the mixture of steam and desulfurized natural gas to syngas (mainly H2 and CO) over a nickel-based catalyst. The syngas then passes through a heat recovery step and is fed into a water gas shift reactor, where it converts to H2 and CO2 over promoted iron oxide catalyst. The final hydrogen purification is accomplished via a PSA system, where the impurities are adsorbed on the surface of adsorbents at relatively high pressure [100].  Table 3.2 details the capital and operating cost of an SMR plant with a capacity of 100 tonnes/day. The direct depreciable capital cost (DDCC) of an SMR plant without carbon capture and sequestration (CCS), consists of the cost of the reformer and the balance of plant and off-sites.   Table 3.2. Capital and operating cost of SMR plant (100 tonnes/day) [101] Capital Expenses   Annual expenses  DDCC* (100 tonnes/day) USD (2013) REPC*  0.5% of DDCC Reformer  28,726,000 FOC* 5% of DDCC Balance of plant and off-sites 11,477,000 VOC*   Process CO2 removal 3,491,000 plant non-fuel O&M  4% of DDCC Stack CO2 removal 3,070,000 O&M CO2 compressor  4% of Eq. 3.4 CO2 compressor (Eq. 3.4 ) 21,282,000 O&M CO2 injection  Eq. 3.8 to 3.11 CO2 injection equipment (Eq. 3.6 ) 103,000 O&M CO2 pipeline  2.5% of Eq. 3.6 Drilling capital cost (Eq. 3.7 ) 436,000 Natural gas  Eq. 3.2 Site screening and evaluation 2,177,000 Electricity  Eq. 3.3 CO2 pipeline cost (Eq. 3.5) 51,144,000 Water  Eq. 3.2 with modification IDCC*  % of DDCC   Site preparation  2%   Engineering & design  10%   Project contingency  15%   Up-front permitting costs  15%   NDCC* 3,200,000   *DDCC: direct depreciable capital costs, IDCC: indirect depreciable capital costs, NDCC: non-depreciable capital costs, REPC: replacement costs, FOC: fixed operating costs, VOC: variable operating costs 25  The installation cost factor of 1.92 was applied to all direct depreciable capital costs except for the CO2 compressor (1.2) and CO2 injection equipment (1.5).  As the cost parameters were reported for large-size (LS) plants (200 to 400 tonnes/day), the scaling factor (𝛼) of 0.88 was used to derive the cost parameters applicable to the medium-size (MS) plants (10 to 100 tonnes/day):  𝐶𝑜𝑠𝑡𝑀𝑆 = 𝐶𝑜𝑠𝑡𝐿𝑆 (𝑆𝑖𝑧𝑒𝑀𝑆𝑆𝑖𝑧𝑒𝐿𝑆)𝛼 3.1  The scaling factor of 0.7 was used for a plant of size 10 to 50 tonnes/day. The annual cost of feedstock (i.e., natural gas (NG), water (W), and electricity (Elec)) for the SMR plant and the CCS facilities were calculated as follows:   𝐶𝑜𝑠𝑡𝑁𝐺 = 365𝑃𝑟𝑖𝑐𝑒𝑁𝐺𝑁𝑈𝑠𝑎𝑔𝑒𝑅𝑒𝑓𝐴𝑣𝑃𝑙𝑎𝑛𝑡𝑃𝑟𝑜_𝑟𝑎𝑡𝑒𝑃𝑙𝑎𝑛𝑡 3.2  In which 𝑁𝑈𝑠𝑎𝑔𝑒𝑅𝑒𝑓 = 0.164𝐺𝐽𝑘𝑔𝐻2 is the NG consumption in the reformer and 𝐴𝑣𝑃𝑙𝑎𝑛𝑡 = 0.98. is the plant availability.   Equation 3.2 was used to calculate the annual cost of water by substituting 𝑁𝑈𝑠𝑎𝑔𝑒𝑅𝑒𝑓with 𝑊𝑈𝑠𝑎𝑔𝑒𝑅𝑒𝑓 = 4.8𝑔𝑎𝑙𝑙𝑜𝑛𝑘𝑔𝐻2 and 𝑃𝑟𝑖𝑐𝑒𝑁𝐺 with 𝑃𝑟𝑖𝑐𝑒𝑤.  𝐶𝑜𝑠𝑡𝐸𝑙𝑒𝑐 = 365𝑃𝑟𝑖𝑐𝑒𝐸𝑙𝑒𝑐 (𝐸𝑈𝑠𝑎𝑔𝑒𝐶𝑜𝑚𝑝𝑟 + 𝐸𝑈𝑠𝑎𝑔𝑒𝑅𝑒𝑓) 𝐴𝑣𝑃𝑙𝑎𝑛𝑡𝑃𝑟𝑜_ 𝑟𝑎𝑡𝑒𝑃𝑙𝑎𝑛𝑡 3.3  In which the energy consumption of the compressor and reformer are 𝐸𝑈𝑠𝑎𝑔𝑒𝑐𝑜𝑚𝑝𝑟 = 0.81 𝐾𝑊ℎ𝑘𝑔𝐻2 and 𝐸𝑈𝑠𝑎𝑔𝑒𝑟𝑒𝑓 = 0.6 𝐾𝑊ℎ𝑘𝑔𝐻2 , respectively. 𝐸𝑈𝑠𝑎𝑔𝑒𝑐𝑜𝑚𝑝𝑟 was not considered for plants without carbon capture technology.  CO2 capture and sequestration (CCS) CO2, which is present in the syngas and flue gas, is captured by different technologies including: PSA, absorption technologies, membranes and cryogenic processes [102]. In this study, it was assumed that the monoethanolamine (MEA) absorption unit was installed on the syngas stream, 26  following the shift reactor, and a secondary MEA treatment unit was installed on the reformer stack to capture CO2 from the flue gas. The CO2 capture efficiency was considered at 90% [103]. The CO2 sequestration is accomplished in three stages: CO2 compression, CO2 transportation to the sequestration site, and CO2 injection into the geological reservoir. At the compression stage, CO2 is compressed from atmospheric pressure to 15 MPa, which is suitable for pipeline transport. This could be accomplished via 9-stage compression. The capital cost of the compressor in U.S. dollars (2005) was calculated using the following formula:  𝐶𝑐𝑜𝑚𝑝𝑟 = 𝑚𝑡𝑟𝑎𝑖𝑛𝑁𝑡𝑟𝑎𝑖𝑛 [0.13 × 106(𝑚𝑡𝑟𝑎𝑖𝑛)−0.71 + 1.4 × 106(𝑚𝑡𝑟𝑎𝑖𝑛)−0.6𝑙𝑛 (𝑃𝑓𝑖𝑛𝑎𝑙𝑃𝑖𝑛𝑖𝑡𝑖𝑎𝑙)] 3.4  In which 𝑚𝑡𝑟𝑎𝑖𝑛 (kg/s) is the CO2 mass flow rate through each compressor train, and 𝑁𝑡𝑟𝑎𝑖𝑛 is the number of compressor trains. 𝑚𝑡𝑟𝑎𝑖𝑛 was calculated by multiplying the CO2 produced from the SMR process (kg/s) by the carbon capture efficiency of the plant. 𝑁𝑡𝑟𝑎𝑖𝑛 equals to 1 in this study, as the compressor power was less than the maximum size of each compressor train, i.e., 40,000 KW.  The operation and maintenance (O&M) cost of the compressors was calculated by multiplying the O&M factor of 0.04 by the capital cost of the compressor. The capital cost of pipelines to transport the captured CO2 to the injection wells was calculated as follows (USD 2005):  𝐶𝑝𝑙 = 9970𝐹𝑙𝑜𝑐𝐹𝑡𝑒𝑟𝐿𝑒𝑛𝑔𝑡ℎ𝑝𝑙1.13𝑚𝑝𝑙0.35 3.5  In which 𝐹𝑙𝑜𝑐 and 𝐹𝑡𝑒𝑟 are the location and terrain factors, with values of 1 and 1.3, respectively. 𝐿𝑒𝑛𝑔𝑡ℎ𝑝𝑙 and 𝑚𝑝𝑙 are the length of the pipeline and the CO2 mass flow rate (tonne/day) through the pipeline, respectively. The pipeline length was assumed at 60 km in this study [98].  The O&M cost of the pipelines was calculated by multiplying the O&M factor of 0.025 by the capital cost of the pipeline. The capital cost of CO2 injection is composed of site screening and evaluation, equipment, and drilling per well. Each well is needed for the injection of 10,000 metric tonnes per day or less. The injection equipment cost includes supply wells, distribution lines, headers, and electrical services, calculated as follows (USD 2005): 27   𝐶𝑖𝑛𝑗𝑒𝑐𝑡 = 49433 × 𝑁𝑤𝑒𝑙𝑙 × (𝑚𝑤𝑒𝑙𝑙280 × 𝑁𝑤𝑒𝑙𝑙)0.5 3.6   In which 𝑁𝑤𝑒𝑙𝑙 is the number of wells, and 𝑚𝑤𝑒𝑙𝑙 is the CO2 mass flow rate delivered to each injection site (tonnes/day).  The drilling cost of an onshore injection well was estimated as follows (USD 2005):  𝐶𝑑𝑟𝑖𝑙𝑙 = 𝑁𝑤𝑒𝑙𝑙 × 106 × 0.1063𝑒0.0008𝑑𝑤𝑒𝑙𝑙  3.7  In which 𝑑𝑤𝑒𝑙𝑙 is the well depth, which was assumed at 1524 m [101].  The O&M cost of injection is composed of normal daily expenses (O&Mdaily), consumables (O&Mcons), surface maintenance (O&Msur), and subsurface maintenance (O&Msubsur), calculated as follows (USD 2005):  𝑂&𝑀𝑑𝑎𝑖𝑙𝑦 = 7596 × 𝑁𝑤𝑒𝑙𝑙 3.8 𝑂&𝑀𝑐𝑜𝑛𝑠 = 20295 × 𝑁𝑤𝑒𝑙𝑙 3.9 𝑂&𝑀𝑠𝑢𝑟 = 15420 × 𝑁𝑤𝑒𝑙𝑙 × (𝑚𝑤𝑒𝑙𝑙280 × 𝑁𝑤𝑒𝑙𝑙)0.5 3.10 𝑂&𝑀𝑠𝑢𝑏𝑠𝑢𝑟 = 5669 × 𝑁𝑤𝑒𝑙𝑙 × (𝑑𝑤𝑒𝑙𝑙1219) 3.11  3.3.1.2 Electrolyzer Water electrolysis is an electro-chemical process for splitting water into hydrogen and oxygen. Currently there are three types of electrolyzers available: alkaline, polymer electrolyte membrane and high temperature solid oxide electrolyzers. In alkaline electrolyzers, the electrolysis cell consists of two electrodes separated by a gas‐tight diaphragm, which is immersed in a liquid electrolyte. The solid oxide electrolyzers are based on a ceramic electrolyte sandwiched in between two electrically connected porous electrodes. In proton exchange membrane (PEM) electrolyzers, the electrolyte is a solid ion conducting membrane which allows protons to be transferred from the anode side of the membrane to the cathode side, where it forms hydrogen. In this study the PEM electrolyzers were considered as they have higher flexibility and better coupling with a limited 28  industry experience compared to solid oxide electrolyzers [104]. Table 3.3 details the capital and operating cost of a PEM electrolyzer.  Table 3.3. Capital and operating cost of a PEM electrolyzer [101]. Capital expenses  Central On-site Annual expenses Central/on-site DDCC*  % of total DDCC REPC* 0.5% of DDCC Stacks (PEM) 37% 38% FOC*  5% of DCC* Hydrogen gas management system—cathode system side 1% 6% VOC*   Oxygen gas management system anode system side 1% 2% non-fuel O&M  1% of DCC* Water reactant delivery management system 1% 5% Electricity Eq. 3.13 Thermal management system 7% 5% Water Eq. 3.13 with modification Power electronics 44% 26%   Controls & sensors 1% 6%   Mechanical balance of plant 2% 5%   Item breakdown—other 3% 2%   Item breakdown—assembly labor 3% 5%   PEM replacement (every 10 years) 12% N/A   IDCC* % of total DDCC   Site preparation  2% 18%   Engineering & design  8% 50,000   Project contingency  15% 15%   Up-front permitting costs  15% 30,000   NDCC 1,200,000** N/A   *DDCC: direct depreciable capital costs, IDCC: indirect depreciable capital costs, DCC: depreciable capital cost (DDCC+IDCC), NDCC: non-depreciable capital costs, REPC: replacement costs, FOC: fixed operating costs, VOC: variable operating costs **For the central plant with capacity of 50t/day (6 acres)  The total direct depreciable capital cost is calculated as follows:  𝐷𝐷𝐶𝐶𝐸𝑙𝑒𝑐 = 𝑃𝑜𝑤𝑒𝑟𝐸𝑙𝑒𝑐𝑈𝑐𝑜𝑠𝑡𝐸𝑙𝑒𝑐𝐶𝐹𝐸𝑙𝑒𝑐 3.12  In which 𝑃𝑜𝑤𝑒𝑟𝐸𝑙𝑒𝑐 is the electrolyzer power (kW), which was calculated by multiplying the capacity of the plant by the electricity usage (𝐸𝑈𝑠𝑎𝑔𝑒𝐸𝑙𝑒𝑐= 50.2 𝑘𝑊ℎ𝑘𝑔 𝐻2 for central and 50.3 𝑘𝑊ℎ𝑘𝑔 𝐻2 for on-site production), 𝑈𝑐𝑜𝑠𝑡𝐸𝑙𝑒𝑐 is the uninstalled cost of the plant (i.e., $400/kW for central and $450/kW for on-site production), and 𝐶𝐹𝐸𝑙𝑒𝑐 is the installation cost factor (i.e., 1.1).  The annual cost of electricity usage is calculated as follows:  𝐶𝑜𝑠𝑡𝐸𝑙𝑒𝑐 = 365𝑃𝑟𝑖𝑐𝑒𝐸𝑙𝑒𝑐𝐸𝑈𝑠𝑎𝑔𝑒𝐸𝑙𝑒𝑐𝐴𝑣𝑃𝑙𝑎𝑛𝑡𝑃𝑟𝑜_𝑟𝑎𝑡𝑒𝑃𝑙𝑎𝑛𝑡 3.13 29   In which 𝐴𝑣𝑝𝑙𝑎𝑛𝑡 = 97%. Equation 3.13 was used to calculate the annual cost of water by substituting 𝐸𝑈𝑠𝑎𝑔𝑒𝐸𝑙𝑒𝑐with 𝑊𝑈𝑠𝑎𝑔𝑒𝐸𝑙𝑒𝑐 = 15 𝑙𝑖𝑡𝑘𝑔𝐻2 and 𝑃𝑟𝑖𝑐𝑒𝐸𝑙𝑒𝑐 with 𝑃𝑟𝑖𝑐𝑒𝑤𝑎𝑡𝑒𝑟. In DDCC calculation, the scaling factor of 0.9 and 0.85 was used for central (10 to 100 tonnes/day) and on-site (100 to 1500 kg/day) electrolyzers, respectively.  3.3.1.3 By-product hydrogen purification from the chlor-alkali industry In the chlor-alkali industry, chlorine, sodium hydroxide and hydrogen are produced via the electrolysis of a concentrated solution of sodium chloride. Hydrogen as a by-product can be captured and purified. In the purification plant, different adsorbents are filled in classification and heated to separate hydrogen from the main impurity gases such as chlorine, oxygen, nitrogen and water. The PSA is also adopted to strengthen the regeneration effect [105].  In the district of North Vancouver, ERCO WorldWide’s sodium chlorate plant and Chemtrade Electrochem’s chlor-alkali facility produce by-product hydrogen streams. The total by-product hydrogen generated by those two operations exceeds 1000 kg/h, with over 600 kg/h being vented.   Table 3.4.  Capital and operating cost of a hydrogen capture and purification facility (10 tonnes/day). Capital expenses   Annual expenses  DDCC*  USD (2013) REPC* 0.5% of DDCC Liquid ring compressor 2,600,000 FOC*  5% of DCC* Contaminant removal system 2,800,000 VOC*   PSA 1,300,000 Non-fuel O&M  1% of DCC* IDCC* % of DDCC Electricity Eq. 3.14 Site preparation  2%   Engineering & design  8%   Project contingency  15%   Up-front permitting costs  15%   NDCC* 1,000,000   *DDCC: direct depreciable capital costs, IDCC: indirect depreciable capital costs, DCC: depreciable capital cost (DDCC + IDCC), NDCC: non-depreciable capital costs, REPC: replacement costs, FOC: fixed operating costs, VOC: variable operating costs  30  The reported capital investment was based on vendor quotes for a plant size of 2 tonnes/day. The scaling factor of 0.6 was used to derive the numbers for a plant size of 10 tonnes/day. The annual cost of electricity for the purification plant is calculated as follows:   𝐶𝑜𝑠𝑡𝐸𝑙𝑒𝑐 = 365𝑃𝑟𝑖𝑐𝑒𝐸𝑙𝑒𝑐𝐸𝑈𝑠𝑎𝑔𝑒𝑃𝑙𝑎𝑛𝑡𝐴𝑣𝑃𝑙𝑎𝑛𝑡𝑃𝑟𝑜_𝑟𝑎𝑡𝑒𝑃𝑙𝑎𝑛𝑡 3.14  In which 𝐴𝑣𝑝𝑙𝑎𝑛𝑡 = 98%. The electricity usage of the hydrogen purification plant (𝐸𝑈𝑠𝑎𝑔𝑒𝑃𝑙𝑎𝑛𝑡) was considered at 3 kWh/kgH2, which includes the electricity usage of the PSA unit, the containment removal system, and the liquid ring compressor.   3.3.1.4 Liquefier Series of refrigerants and a sequence of compression and expansion processes (Joule-Thompson liquefaction cycle) are used in a hydrogen liquefier to convert the gaseous hydrogen to the liquid state. The liquefaction facilities use the ortho-para conversion reactors to convert hydrogen to the para form via a series of catalyst beds. The energy required for liquefaction is around 30 percent of the heating value of hydrogen, which is mainly consumed by the ortho/para conversion process. The capital cost of the liquefier is as follows [106]:  𝐶𝐿𝑖𝑞 = 𝑈𝐶𝐿𝑖𝑞𝑁𝐿𝑖𝑞(𝐶𝑎𝑝𝐿𝑖𝑞)0.8 3.15  In which 𝑈𝐶𝐿𝑖𝑞 is the unit cost of a liquefier (6,655,000 USD (2013)), 𝑁𝐿𝑖𝑞 is the number of equally sized liquefiers in operation, and 𝐶𝑎𝑝𝐿𝑖𝑞 is the capacity of a liquefier.   𝑁𝐿𝑖𝑞 = ⌈𝐶𝑎𝑝𝐿𝑖𝑞𝐶𝑎𝑝𝑚𝑎𝑥,𝐿𝑖𝑔⌉ 3.16  𝐶𝑎𝑝𝑚𝑎𝑥,𝐿𝑖𝑔 is the largest practical size of a liquefaction plant, i.e., 200 tonnes/day.  The owner’s cost provides the funds necessary for engineering studies, permits, training, licensing fees, etc., and was assumed at 12% of the capital cost of the liquefier plant [106]. The average electricity requirement of the plant was assumed at 9.05 kWh/kgH2.  31  3.3.1.5 Terminal and central storage  3.3.1.5.1 Gas delivery terminal (GH2 storage) Storage compressorTruck loading  compressorsLoading baysHydrogen delivered to terminalLow-pressure storage Figure 3.2. Schematic of gaseous hydrogen central storage.  Truck loading and storage compressors The reciprocating compressors are suitable for medium and large flow of hydrogen, as opposed to centrifugal machines which are commonly used for natural gas [107]. The installed capital cost of the reciprocating compressors was calculated as follows (USD 2013) [106]:  𝐶𝐶𝑜𝑚𝑝𝑟 = 44402 × 𝑁𝐶𝑜𝑚𝑝𝑟 (𝑇𝑃𝑜𝑤𝑒𝑟𝑐𝑜𝑚𝑝𝑟𝐼𝑠𝑜𝐸𝑓𝑓𝑐𝑜𝑚𝑝𝑟)0.6038 3.17  𝑁𝐶𝑜𝑚𝑝𝑟,𝑡𝑟𝑢𝑐𝑘 =𝑈𝐶𝑎𝑝𝑡𝑢𝑏𝑒𝑠𝑁𝐵𝑎𝑦𝑠𝐿𝑇𝑖𝑚𝑒𝑡𝑢𝑏𝑒𝑠?̅̇?𝑐𝑜𝑚𝑝𝑟 3.18  𝑁𝐶𝑜𝑚𝑝𝑟,𝑠𝑡𝑜𝑟𝑎𝑔𝑒 =𝐶𝑎𝑝𝑡𝑒𝑟𝑚𝑙24 × 35 3.19  In which 𝑈𝐶𝑎𝑝𝑡𝑢𝑏𝑒𝑠 is the usable capacity of the tubes, 𝐿𝑇𝑖𝑚𝑒𝑡𝑢𝑏𝑒𝑠 is the loading time of the tubes, 𝐶𝑎𝑝𝑡𝑒𝑟𝑚𝑙 is the terminal capacity, 𝑁𝐵𝑎𝑦𝑠 is the number of filling bays, and ?̅̇?𝑐𝑜𝑚𝑝𝑟 is the compressor flow rate at the average storage pressure.  𝑁𝐵𝑎𝑦𝑠 = ⌈⌈𝐶𝑎𝑝𝑡𝑒𝑟𝑚𝑙𝑈𝐶𝑎𝑝𝑡𝑢𝑏𝑒𝑠⌉ × (𝐿𝑇𝑖𝑚𝑒𝑡𝑢𝑏𝑒𝑠 + 𝐿𝑖𝑛𝑇𝑖𝑚𝑒)/24⌉ 3.20 32   ?̅̇?𝑐𝑜𝑚𝑝𝑟 = 85 (𝑃𝑚𝑖𝑛,𝑇 + 0.25(𝑃𝑚𝑎𝑥,𝑇 − 𝑃𝑚𝑖𝑛,𝑇)) /250 3.21  𝐿𝑖𝑛𝑇𝑖𝑚𝑒 is the lingering time of the truck. 𝑇𝑃𝑜𝑤𝑒𝑟𝑐𝑜𝑚𝑝 was calculated using Eq. 3.36, with the number of compression stages calculated as follows:  𝑁𝑠𝑡 = ⌈log 𝑃𝑚𝑎𝑥 − log 𝑃𝑚𝑖𝑛log 𝐶𝑅⌉ 3.22  In which 𝑃𝑚𝑎𝑥 and 𝑃𝑚𝑖𝑛 are the maximum and minimum pressure for the truck loading compressors (55 MPa and 20 MPa), and the maximum and minimum pressure for storage compressors (40 MPa and 5 MPa). 𝐶𝑅 is the allowable compression ratio per stage, set at 2.1.   Compressed gas storage The installed capital cost of the short-term storage unit is calculated as follows [106]:  𝐶𝐿𝑃𝑆 = 𝑈𝐶𝐿𝑃𝑆 ⌈𝑇𝑆𝑐𝑎𝑝𝐿𝑃𝑆𝐶𝑎𝑝𝐿𝑃𝑆⌉ 𝐶𝑎𝑝𝐿𝑃𝑆 3.23  In which 𝑈𝐶𝐿𝑃𝑆 is the unit capital cost of storage, set at 1220 (USD 2013), 𝐶𝑎𝑝𝐿𝑃𝑆 is the capacity of a storage cylinder (Eq. 3.32), and 𝑇𝑆𝑐𝑎𝑝𝐿𝑃𝑆 is the design storage capacity, calculated as follows:  𝑇𝑆𝑐𝑎𝑝𝐿𝑃𝑆 =𝐶𝑎𝑝𝑡𝑒𝑟𝑚𝑙𝐷𝑎𝑦𝑠𝑠𝑡𝑟𝑈𝐶𝑎𝑝𝑐𝑙𝑑 3.24  In which 𝐷𝑎𝑦𝑠𝑠𝑡𝑜𝑟𝑎𝑔𝑒 is the days of storage, set at 0.25, and 𝑈𝐶𝑎𝑝𝑐𝑙𝑑 is the usable cylinder capacity, i.e., 46% of the central storage capacity. The capital cost of the other components of the GH2 central storage, including piping, supply, discharge and headers, plumbing, electrical and instrumentation, building and structure, and truck scale was considered at 1% of the total DDCC.      33  3.3.1.5.2 Liquid delivery terminal (LH2 storage) Storage tank Cryogenic pumpsLoading baysLiquid hydrogen delivered to terminal Figure 3.3. Schematic of liquid hydrogen central storage.  Liquid hydrogen storage tank Most cryogenic tanks are spherical, which minimizes the heat transfer surface area per unit of storage volume. The installed capital cost of a spherical tank is calculated as follows (USD 2013) [106]:  𝐶𝑆,𝑙𝑖𝑞 = 𝑁𝑆,𝑙𝑖𝑞 (5646600 + 3100 ∗𝑉𝑆,𝑙𝑖𝑞𝑁𝑆,𝑙𝑖𝑞) 3.25  𝑁𝑆,𝑙𝑖𝑞 =𝑉𝑆,𝑙𝑖𝑞𝑉𝑚𝑎𝑥,𝑙𝑖𝑞    In which 𝑁𝑆,𝑙𝑖𝑞 is the number of storage spheres, 𝑉𝑚𝑎𝑥,𝑙𝑖𝑞 is the maximum volume of single storage sphere, i.e., 1000 m3, and 𝑉𝑆,𝑙𝑖𝑞 is the total volume of storage, calculated as follows:  𝑉𝑆,𝑙𝑖𝑞 =1𝐷𝐻2(𝑈𝐶𝑎𝑝𝑢𝑠𝑡𝑟𝑈𝑃𝑒𝑟𝑠𝑡𝑟+ 0.0028 × 𝐶𝑎𝑝𝑡𝑒𝑟𝑚𝑙) 3.26   34  In which 𝑈𝑃𝑒𝑟𝑠𝑡𝑟 is the usable percent of liquid storage, i.e., 95%. 𝑈𝐶𝑎𝑝𝑠𝑡𝑟 is the storage usable capacity:  𝑈𝐶𝑎𝑝𝑠𝑡𝑟 = (1 − 𝑠𝑢𝑟𝑔𝑒)𝐶𝑎𝑝𝑡𝑒𝑟𝑚𝑙𝑂𝑢𝑡𝑎𝑔𝑒𝑝𝑙𝑎𝑛𝑡 3.27  The storage must be large enough to handle plant outages and peak demand.  Low-head liquid pump The installed capital cost of a low-head cryogenic pump is calculated as follows (USD 2013) [106]:  𝐶𝑃𝑢𝑚𝑝,𝐿𝐻 = 4423 × 𝑁𝑃𝑢𝑚𝑝,𝐿𝐻𝐶𝑎𝑝𝑃𝑢𝑚𝑝,𝐿𝐻0.3431 3.28  𝐶𝑎𝑝𝑃𝑢𝑚𝑝,𝐿𝐻 is the design capacity of each low-head pump:  𝐶𝑎𝑝𝑃𝑢𝑚𝑝,𝐿𝐻 =1.5 × 𝐶𝑎𝑝𝑡𝑒𝑟𝑚𝑖𝑛𝑎𝑙24 3.29  𝑁𝑃𝑢𝑚𝑝,𝐿𝐻 is the number of required low-head pumps:  𝑁𝑃𝑢𝑚𝑝,𝐿𝐻 = ⌈𝐶𝑎𝑝𝑃𝑢𝑚𝑝,𝐿𝐻?̇?𝑚𝑎𝑥,𝑃𝑢𝑚𝑝⌉ 3.30  ?̇?𝑚𝑎𝑥,𝑃𝑢𝑚𝑝 is the maximum pump throughput, considered at 12000 kg/hr. IDCC and FOC were calculated as a percentage of DDCC for GH2 and LH2 central storages (Table 3.5). The annual cost of electricity for compressors and low-head pump was calculated using Eq. 3.38 and Eq. 3.48, respectively.          35  Table 3.5. IDCC and operating cost of GH2 and LH2 central storage. IDCC*  % of DDCC* Site preparation  5% Engineering & design  10% Project contingency  10% Up-front permitting costs  3% Owner’s costs 12% FOC * 5% of DCC* VOC *(electricity)  *DDCC: direct depreciable capital costs, IDCC: indirect depreciable capital costs, DCC: depreciable capital cost (DDCC + IDCC), NDCC: non-depreciable capital costs, REPC: replacement costs, FOC: fixed operating costs, VOC: variable operating costs   3.3.1.5.3 Transportation  Tube trailers The hydrogen payload of a tube trailer is greater than the off-loaded amount. The tubes cannot be completely depressurized (hydrogen at 5 MPa or lower remains in the tubes), so that the actual usable capacity of tubes is less than the rated capacity. Also, there are losses associated with dropping the trailers and removing the empty ones. Gas losses during these operations was assumed to be 3%.  Table 3.6. Capital cost of gas trucks with different payloads [108]. Hydrogen payload (off-loaded + 3% losses + remained hydrogen in vessels below 5 MPa) Trailer + vessels cost (USD 2013) Tractor cost (USD 2013) 120 kg (steel vessels) 180,000 100,000 600 kg (composite vessels) 700,000 100,000 1000 kg (composite vessels) 1,300,000 100,000  Tanker trucks The total amount of hydrogen discharged to the storage tanks is less than the payload of the truck. When all the liquid is discharged from the tank, the saturated hydrogen vapor, which weighs 2% of the total payload, remains in the tank. Also, there are losses associated with loading and offloading the trailer, which may amount to 6–10% of the total payload.  The capital cost of the tanker with the rated capacity of 4100 kg and the tractor was estimated at 1,000,000 and 100,000 (USD 2013), respectively [106].   36  3.3.1.5.4 Fueling station  The components of a fueling station and the associated DDCC depend on whether hydrogen is produced on-site or delivered in gas or liquid form (Figure 3.4) [106], [109]. Other cost parameters, except the annual electricity use, were calculated from DDCC, as mentioned in Table 3.7.  Table 3.7. Capital and operating cost of a fueling station. IDCC* % of DDCC* Site preparation  5% Engineering & design  10% Project contingency  5%  Up-front permitting costs  3% FOC* 5% of DCC* VOC*  Non-fuel O&M  1% of DCC* Electricity  *DDCC: direct depreciable capital costs, IDCC: indirect depreciable capital costs, DCC: depreciable capital cost (DDCC + IDCC), NDCC: non-depreciable capital costs, REPC: replacement costs, FOC: fixed operating costs, VOC: variable operating costs  In an on-site production and dispensing facility, hydrogen is produced with an on-site electrolyzer at a low pressure of 2 MPa and stored in low-pressure storage tank. When needed, hydrogen is compressed in the cascade storage system via a high-pressure compressor and pre-cooled by a refrigeration unit before being dispensed into the vehicle tank. For stations with gaseous hydrogen delivery, the trailer acts as the low-pressure storage tank and the succeeding components are similar to the on-site production facility. For stations with liquid hydrogen delivery, hydrogen is stored in cryogenic tanks at −252°C. When hydrogen is needed, a high-pressure cryogenic pump is used to pass it through a vaporizer. Hydrogen is stored in a high-pressure cascade system before being dispensed into the vehicle tank.   37  Cryogenic storagePump VaporizerCascade storageCompressorPre-cooling unitDispenserLow-pressure storageOn-site electrolyzerTube trailerTanker truck Figure 3.4. Schematic of a hydrogen fueling station with on-site production, gas delivery, and liquid delivery components.   The capital cost of the components of a fueling station is described as follows: Low-pressure storage for hourly surge Hydrogen that is produced on-site is stored at 2 MPa in a low-pressure storage unit. The capital cost of the storage unit was calculated as follows:  𝐶𝐿𝑃𝑆 = 𝑈𝐶𝐿𝑃𝑆 × [𝑃𝐷𝐿𝑃𝑆𝐶𝑎𝑝𝐿𝑃𝑆] × 𝐶𝑎𝑝𝐿𝑃𝑆 × 𝐶𝐹𝐿𝑃𝑆 3.31  𝑈𝐶𝐿𝑃𝑆 is the unit cost of the low-pressure storage unit per kilogram of hydrogen stored, i.e., 1252 (USD 2013), 𝑃𝐷𝐿𝑃𝑆 is the amount of hydrogen needed at a refueling station for peak hours, which was set at 30% of the of the station capacity. 𝐶𝐹𝐸𝑙𝑒𝑐 is the installation cost factor, set at 1.3. 𝐶𝑎𝑝𝐿𝑃𝑆 is the low-pressure storage vessel capacity, calculated as follows:  𝐶𝑎𝑝𝐿𝑃𝑆 = [𝜋4(𝐷𝐿𝑃𝑉 − 2𝑇𝐿𝑃𝑉)(𝐿𝐿𝑃𝑉 − 2𝑇𝐿𝑃𝑉)− 0.083𝜋(𝐷𝐿𝑃𝑉 − 2𝑇𝐿𝑃𝑉)3]2 × 101325 𝑃𝑚𝑎𝑥,𝐿𝑃𝑉𝑍 × 8314 × 𝑇𝑜𝑝𝑟 3.32 38  𝐷𝐿𝑃𝑉, 𝑇𝐿𝑃𝑉, and 𝐿𝐿𝑃𝑉 are the outer diameter, thickness, and length of the low-pressure storage vessel, 𝑃𝑚𝑎𝑥,𝐿𝑃𝑉 is the maximum storage pressure, i.e., 25 MPa, 𝑇𝑜𝑝𝑟 is the hydrogen temperature at operating condition, and 𝑍 is the hydrogen compressibility factor at 𝑃𝑚𝑎𝑥,𝐿𝑃𝑉 and 𝑇𝑜𝑝𝑟.  Compressor  The installed capital cost of the compressor is calculated as follows (USD 2013) [106], [110]:  𝐶𝐶𝑜𝑚𝑝𝑟 = 44402 × 𝑁𝐶𝑜𝑚𝑝𝑟 × (𝑀𝑅𝐶𝑜𝑚𝑝𝑟𝑁𝐶𝑜𝑚𝑝𝑟)0.6038 3.33  In which 𝑁𝐶𝑜𝑚𝑝𝑟 is the number of compressors in operation at any time and 𝑀𝑅𝐶𝑜𝑚𝑝𝑟 is the motor rating:  𝑀𝑅𝐶𝑜𝑚𝑝𝑟 =𝑇𝑃𝑜𝑤𝑒𝑟𝐶𝑜𝑚𝑝𝑟𝐼𝑠𝑜𝐸𝑓𝑓𝐶𝑜𝑚𝑝𝑟×𝑆𝐹𝑀𝑜𝑡𝑜𝑟𝐸𝑓𝑓𝑀𝑜𝑡𝑜𝑟 3.34  𝑁𝐶𝑜𝑚𝑝 = ⌈?̇?𝐶𝑜𝑚𝑝𝑟35⌉ 3.35  In which ?̇?𝐶𝑜𝑚𝑝𝑟 is the compressor flow rate in peak demand, selected at 7% of the fueling station maximum capacity. The maximum compressor capacity was fixed at 35 kg/hr at a pressure ratio of 45 (2–95 MPa); for higher flow rates, multiple compressors were used. 𝐼𝑠𝑜𝐸𝑓𝑓𝐶𝑜𝑚𝑝𝑟 is the isentropic efficiency of the compressor (i.e., 75%), 𝑆𝐹𝑀𝑜𝑡𝑜𝑟 is the sizing factor of the motor (i.e., 110%), 𝑇𝑃𝑜𝑤𝑒𝑟𝐶𝑜𝑚𝑝𝑟 and 𝐸𝑓𝑓𝑀𝑜𝑡𝑜𝑟  are the theoretical power of the compressor and the motor efficiency, respectively. The theoretical power of the compressor was calculated considering equal work by all stages and intercooling back to the original feed temperature:  𝑇𝑃𝑜𝑤𝑒𝑟𝐶𝑜𝑚𝑝𝑟 =?̇?𝐶𝑜𝑚𝑝𝑟3600 × 2.0158× 𝑍 × 𝑅 × 𝑇𝑚𝑎𝑥 × 𝑁𝑠𝑡 × (𝑘𝑘 − 1)× [(𝑃𝑜𝑢𝑡𝑙𝑒𝑡𝑃𝑖𝑛𝑙𝑒𝑡)𝑘−1𝑘×𝑁𝑠𝑡− 1] 3.36   39  The motor efficiency is calculated as follows:  𝐸𝑓𝑓𝑀𝑜𝑡𝑜𝑟 = 0.0008(ln 𝑋)4 − 0.0015(ln 𝑋)3 + 0.0061(ln 𝑋)2 + 0.0311 ln 𝑋 + 0.7617  3.37 𝑋 =𝑇𝑃𝑜𝑤𝑒𝑟𝐶𝑜𝑚𝑝𝑟𝐼𝑠𝑜𝐸𝑓𝑓𝐶𝑜𝑚𝑝𝑟 × 𝑁𝐶𝑜𝑚𝑝𝑟   In which 𝑍 is the compressibility factor (i.e., 1.253 for 25 MPa and 1.282 for 54 MPa, which are the maximum pressures in tube tankers), 𝑅 is the gas constant (i.e., 8.314), 𝑇𝑚𝑎𝑥 is the maximum hydrogen temperature at the station (i.e., 40°C), 𝑁𝑠𝑡 is the number of compression stages (i.e., 2), 𝑘 is the ratio of specific heats (i.e., cp/cv=1.42), and 𝑃𝑜𝑢𝑡𝑙𝑒𝑡 and 𝑃𝑖𝑛𝑙𝑒𝑡 are the outlet and inlet pressure of the compressor, respectively, i.e., 97 and 5 MPa.  The annual energy requirement of the compressor is given as follows:  𝐸𝐶𝑜𝑚𝑝 = 365 ×1𝐸𝑓𝑓𝑀𝑜𝑡𝑜𝑟 × 𝐼𝑠𝑜𝐸𝑓𝑓𝐶𝑜𝑚𝑝𝑟×𝑓?̇? × ?̇?𝐶𝑜𝑚𝑝𝑟3600 × 2.0158× 𝑍 × 𝑅 × 𝑇𝑚𝑎𝑥 × 𝑁𝑠𝑡× (𝑘𝑘 − 1) × [(𝑃𝑜𝑢𝑡𝑙𝑒𝑡𝑃𝑎𝑣𝑒𝑖𝑛𝑙𝑒𝑡)𝑘−1𝑘×𝑁𝑠𝑡− 1] 3.38  In which 𝑓?̇? is the percentage of maximum capacity that is used to calculate the average annual flow rate of hydrogen, set at 0.8. The inlet average pressure of the compressor, 𝑃𝑎𝑣𝑒𝑖𝑛𝑙𝑒𝑡 was calculated as follows:  𝑃𝑎𝑣𝑒𝑖𝑛𝑙𝑒𝑡 =(𝑃𝑚𝑎𝑥,𝑡𝑢𝑏𝑒 − 𝑃𝑚𝑖𝑛,𝑡𝑢𝑏𝑒)ln (𝑃𝑚𝑎𝑥,𝑡𝑢𝑏𝑒𝑃𝑚𝑖𝑛,𝑡𝑢𝑏𝑒) 3.39  In which 𝑃𝑚𝑎𝑥,𝑡𝑢𝑏𝑒 and 𝑃𝑚𝑖𝑛,𝑡𝑢𝑏𝑒 are the maximum and minimum pressure of the tube trailer, i.e., 55 and 5 MPa, respectively.      40  In case of on-site hydrogen production, the following formula is used:  𝐸𝐶𝑜𝑚𝑝𝑟 = 365 (1 + 𝑃𝐷𝐿𝑃𝑆𝐶𝑎𝑝𝑠𝑡𝑎𝑡𝑖𝑜𝑛 )1𝐸𝑓𝑓𝑀𝑜𝑡𝑜𝑟 × 𝐼𝑠𝑜𝐸𝑓𝑓𝐶𝑜𝑚𝑝𝑟𝑓?̇? × ?̇?𝐶𝑜𝑚𝑝𝑟3600 × 2.0158× 𝑍 × 𝑅× 𝑇𝑚𝑎𝑥𝑁𝑠𝑡 (𝑘𝑘 − 1) [(𝑃𝑜𝑢𝑡𝑙𝑒𝑡𝑃𝑖𝑛𝑙𝑒𝑡)𝑘−1𝑘×𝑁𝑠𝑡− 1] 3.40  The capital cost of the power transmission system to the compressor is calculated as follows:  𝐶𝑃𝑇 = 𝐶𝐹𝑃𝑇 (−0.0051816 (𝑀𝑅𝐶𝑜𝑚𝑝𝑟0.746)2+ 55.416 (𝑀𝑅𝐶𝑜𝑚𝑝𝑟0.746) + 24868.8) 3.41  In which 𝐶𝐹𝑃𝑇 is the installation cost factor of the power transmission system, set at 2.24.   Cascade storage Cascade storage includes banks of storage vessels at different pressures, individually controlled by valves that are switched in sequence. When the dispenser is connected to the on-board tank, hydrogen starts flowing from the lowest-pressure bank. When the mass flow rate drops to a pre-set level, the valves sequentially switch to the medium and finally high-pressure bank until the fill is completed. The capital cost of the cascade storage vessel was calculated as follows [110]:  𝐶𝑎𝑝𝑣𝑒𝑠𝑠𝑒𝑙 =0.028317 × 𝑉𝑣𝑠𝑠𝑁𝑐𝑠𝑐(6894.757 × 𝑃𝑚𝑎𝑥)𝑁𝑣𝑠𝑠𝑍 × 4124.86 × 𝑇𝑐𝑠𝑐 3.42   𝑉𝑣𝑠𝑠 is the volume of the cascade storage vessel (9.9 ft3), 𝑁𝑐𝑠𝑐 is the optimum number of banks, set at 5, 2, and 1 for stations with maximum capacity of 1000–1500, 500, and 150 kg/day, respectively. 𝑃𝑚𝑎𝑥 is the maximum pressure in the cascade storage vessels, set at 95 MPa, 𝑁𝑣𝑠𝑠 is the number of vessels in each bank, set at 1, 2, and 2 for high-, medium-, and low-pressure vessels. 𝑇𝑐𝑠𝑐 is the operating storage temperature (K), 𝑍 is the compressibility factor for hydrogen at 𝑇𝑐𝑠𝑐 and 𝑃𝑚𝑎𝑥.    41  The cascade storage includes banks of storage vessels at different pressures. The capital cost of the storage system is calculated as follows:  𝐶𝑐𝑠𝑐 = 𝐶𝐹𝑐𝑠𝑐(𝑈𝐶𝑐𝑠𝑐[(𝐶𝑎𝑝𝑣𝑠𝑠)𝐿𝑃 + (𝐶𝑎𝑝𝑣𝑠𝑠)𝑀𝑃 + (𝐶𝑎𝑝𝑣𝑠𝑠)𝐻𝑃]) 3.43  In which 𝑈𝐶𝑐𝑠𝑐 is the unit cost of the cascade storage system per kg of hydrogen stored, set at 1800 USD/kgH2, 𝐶𝐹𝑐𝑠𝑐 is the installation cost factor with a value of 1.3, and (𝐶𝑎𝑝𝑣𝑠𝑠)𝐿𝑃, (𝐶𝑎𝑝𝑣𝑠𝑠)𝑀𝑃, and (𝐶𝑎𝑝𝑣𝑠𝑠)𝐻𝑃 are the capacity of low-, medium-, and high-pressure vessels, respectively.  Pre-cooling unit The pre-cooling unit is placed between the cascade storage and the dispenser to chill the hydrogen during a fast fill at 70 MPa and keep the on-board tank temperature below 85°C. For this study, the precooling unit consists of a large cooling block with a low cooling capacity and a refrigeration unit to maintain the temperature of the block below -40 C.  The total capital cost of the pre-cooling system is calculates as follows [111]:   𝐷𝐷𝐶𝐶𝑃𝐶𝑆 = 𝑐𝑓𝑅𝑒𝑓 [𝐶𝑎𝑝𝑅𝑒𝑓 × 𝑁𝐻𝑜𝑠𝑒𝑇𝑜𝑢𝑡𝐻2]𝛼+ 𝑁𝐻𝑜𝑠𝑒𝑐𝑓𝐻𝑋 [𝑚𝐻𝑋𝑀𝐻𝑋]𝛽 3.44  In which 𝑐𝑓𝑅𝑒𝑓=13865 is a constant factor. 𝐶𝑎𝑝𝑅𝑒𝑓 is the capacity of refrigeration unit per hose. At four back-to-back fills per hose, 𝐶𝑎𝑝𝑅𝑒𝑓 is 3.4 tonnes for refueling stations of size 500–1000 kg/day and 3.1 tonnes for fueling station of size 150 kg/day. 𝑁𝐻𝑜𝑠𝑒 is the number of hoses, with a value of 4 for a 1000 kg/day station, assuming 16 fills during peak hours, with an average filling time of 7 min per vehicle and an average hose occupied fraction of 50% during peak hours. 𝑇𝑜𝑢𝑡𝐻2 is the hydrogen outlet temperature from the system (i.e., -40°C) and 𝛼 is the power sizing exponent, with a value of 0.8579. For the cooling block, 𝑐𝑓𝐻𝑋 is 35,500 USD (2013) for a reference 1000 kg aluminum block (𝑀𝐻𝑋), and 𝑚𝐻𝑋 is the actual aluminum mass. The power sizing exponent is given by 𝛽 with a value of 0.9 for a cooling block of 1330 kg. The installation cost factor of 2 was applied to the total cost calculated. All costs are in USD (2013).  42  The annual electricity requirement of the pre-cooling unit was calculated, using the refrigeration-specific energy use of 0.325 kWh/kgH2 and the overhead pre-cooling energy use of 0.305 kWh/kgH2.   Dispenser The capital cost of a dispenser is calculated by multiplying the number of hoses by the cost of one hose (104,000 USD 2013), assuming each dispenser has one hose. The result is multiplied by the installation cost factor of 1.3.  Hydrogen cryogenic storage tank The cryogenic storage tanks are sized to satisfy the station average daily demand, with the capital cost as follows:  𝐶𝑆𝑐𝑟𝑦𝑜 = 𝐶𝐹𝑐𝑟𝑦𝑜(𝑈𝐶𝑆𝑐𝑟𝑦𝑜𝐶𝑎𝑝𝑐𝑟𝑦𝑜0.6929) 3.45  In which 𝑈𝐶𝑆𝑐𝑟𝑦𝑜 is the unit cost of a storage tank per kg of hydrogen, i.e., 992 USD (2013), 𝐶𝑎𝑝𝑐𝑟𝑦𝑜 is the capacity of the cryogenic storage tank, i.e., 4020 kg for the station capacity of 1000–1500 kg/day. 𝐶𝐹𝑐𝑟𝑦𝑜 is the installation cost factor with a value of 1.3.  Pump The high-pressure low-temperature pump raises the pressure and transfers the liquid hydrogen from the low-pressure cryogenic storage tank to the high-pressure cascade storage system.  The capital cost of the pump was calculated as follows:  𝐶𝑝𝑢𝑚𝑝 = 𝐶𝐹𝑝𝑢𝑚𝑝(𝑈𝐶𝑝𝑢𝑚𝑝𝑁𝑝𝑢𝑚𝑝) 3.46  In which 𝑈𝐶𝑝𝑢𝑚𝑝 is the unit cost of the pump per kilogram hydrogen, i.e., 712,000 USD (2013), 𝐶𝐹𝑝𝑢𝑚𝑝 is the installation cost factor of the pump with a value of 1.3, and 𝑁𝑝𝑢𝑚𝑝 is the number of pumps required.    43  𝑁𝑝𝑢𝑚𝑝 = ⌈?̇?𝑚𝑎𝑥,𝑒𝑣𝑎𝑝𝑂𝑢𝑡𝑝𝑢𝑡𝑚𝑎𝑥,𝑝𝑢𝑚𝑝⌉ 3.47  In which ?̇?𝑚𝑎𝑥,𝑒𝑣𝑎𝑝 is the required evaporator flow rate for peak hours, i.e., 65 kg/hr for a station with 1000 kg/day capacity, and 𝑂𝑢𝑡𝑝𝑢𝑡𝑚𝑎𝑥,𝑝𝑢𝑚𝑝 is the maximum pump output, i.e., 120 kg/hr.  The annual electricity consumption of the high-pressure pump was calculated as follows:  𝐸𝑝𝑢𝑚𝑝 =365 × 𝑇𝑃𝑜𝑤𝑒𝑟𝑝𝑢𝑚𝑝𝐶𝑎𝑝𝑠𝑡𝑎𝑡𝑖𝑜𝑛𝐼𝑠𝑜𝐸𝑓𝑓𝑝𝑢𝑚𝑝𝐸𝑓𝑓𝑚𝑜𝑡𝑜𝑟?̇?𝑚𝑎𝑥,𝑒𝑣𝑎𝑝 3.48  The capital cost of the power transmission system was calculated as follows:  𝐶𝐸𝑉 = 𝐶𝐹𝑃𝑇 (−0.0051816 (𝑀𝑅𝑝𝑢𝑚𝑝0.746)2+ 55.416 (𝑀𝑅𝑝𝑢𝑚𝑝0.746) + 24868.8) 3.49  In which 𝐶𝐹𝑃𝑇 is the installation cost factor of the transmission system with a value of 2.24 and 𝑀𝑅𝑝𝑢𝑚𝑝 is the motor rating of the pump, calculated as follows:  𝑀𝑅𝑝𝑢𝑚𝑝 =𝑇𝑃𝑜𝑤𝑒𝑟𝑝𝑢𝑚𝑝𝑆𝐹𝑚𝑜𝑡𝑜𝑟𝐼𝑠𝑜𝐸𝑓𝑓𝑝𝑢𝑚𝑝𝐸𝑓𝑓𝑚𝑜𝑡𝑜𝑟 3.50  𝐼𝑠𝑜𝐸𝑓𝑓𝑝𝑢𝑚𝑝 and 𝑆𝐹𝑚𝑜𝑡𝑜𝑟 are the isentropic efficiency of the pump, i.e., 75%, and the size factor of the pump motor, i.e., 110%. 𝑇𝑃𝑜𝑤𝑒𝑟𝑝𝑢𝑚𝑝 is the theoretical power of the pump, calculated as follows:  𝑇𝑃𝑜𝑤𝑒𝑟𝑝𝑢𝑚𝑝 = ?̇?𝑚𝑎𝑥,𝑒𝑣𝑎𝑝 ×(𝑃𝑚𝑎𝑥 + 25 − 𝑃𝑠𝑢𝑝𝑝𝑙𝑦 × 14.696)522 × 𝐷𝐻2 3.51  𝑃𝑚𝑎𝑥 is the maximum pressure in the cascade storage, i.e., 13,688 psi, 𝑃𝑠𝑢𝑝𝑝𝑙𝑦 is the supply pressure from Dewar, i.e., 6 MPa, and 𝐷𝐻2  is the liquid hydrogen density, i.e., 70.8 g/l.  𝐸𝑓𝑓𝑝𝑢𝑚𝑝is the efficiency of the pump motor, which was calculated using Eq. 3.37 with a corrected value of X.  𝑋 =𝑇𝑃𝑜𝑤𝑒𝑟𝑝𝑢𝑚𝑝𝐼𝑠𝑜𝐸𝑓𝑓𝑝𝑢𝑚𝑝 3.52 44  Evaporator Evaporator is placed after the high-pressure pump to gasify the liquid hydrogen and to heat it to the cascade operating temperature. The capital cost of the evaporator was calculated in USD 2013:  𝐶𝑒𝑣𝑎𝑝 = 𝐶𝐹𝑒𝑣𝑎𝑝 (𝑁𝑒𝑣𝑎𝑝(1000 × ?̇?𝑚𝑎𝑥,𝑒𝑣𝑎𝑝 + 15000)) 3.53  In which 𝑁𝑒𝑣𝑎𝑝 is the number of evaporators, set to 1 for capacities below 250 kg/hr. 𝐶𝐹𝑒𝑣𝑎𝑝 is the installation cost factor of the evaporator, with a value of 1.3.  3.3.2 Derivation of the GHG emissions parameters The fuel-side Well-to-Wheels (WTW) GHG emissions were analyzed from the primary energy source extraction to the point of fuel utilization. The life-cycle effects of vehicle manufacturing and infrastructure construction/decommissioning were not covered in this analysis. The unit GHG emissions associated with hydrogen production, CCS, storage, transport and dispensing is dependent on the electricity consumption of the facilities, the natural gas consumed in SMR plants and GHG emissions associated with the diesel exhaust products from the trucks.  The CCS efficiency was considered at 90% [103]. The GHG emissions from the flare system of a hydrogen liquefaction plant was not considered, as the only vented by-product is water vapor. The GHG emissions of hydroelectricity production in B.C. was considered at 11 gCO2eq per kWh [112]. As the share of hydroelectricity is projected to stay above 86% of total electricity generation in B.C.[10], the GHG intensity was assumed to stay constant for the study time-frame.  Table 3.8. GHG emissions associated with each component of the HFSC [103], [113]–[115] Central Production  By product purification SMR SMR+CCS Electrolysis 𝐺𝐻𝐺_𝐶𝑦  (g CO2/kg H2) 33 11400 1140 552 On-site production  On-site electrolysis    𝐺𝐻𝐺_𝑂  (g CO2/kg H2) 601    Central storage  Compression  Liquefaction+  cryo-pumping  𝐺𝐻𝐺_𝑆𝑑 (g CO2/kg H2) 17        99  Transport      𝐺𝐻𝐺_𝑇𝑅 (g CO2/km) 1000    Dispensing  On-site H2 production   Compression+ Refrigeration  Cryo-pumping+ Refrigeration   𝐺𝐻𝐺_𝐷𝑑  (g CO2/kg H2) 22 22 10  45  3.4 Hydrogen demand scenario development The development of a hydrogen fuel supply chain in a region is subjected to the spatial and temporal projection of hydrogen demand. In the road transport sector, hydrogen demand relies on the market share projection of fuel cell electric vehicles. A method is introduced in this section to project the allocation of hydrogen demand for the passenger light duty sector in British Columbia for the time period of 2020–2050.  3.4.1 Temporal projection of hydrogen demand Hydrogen demand scenarios was developed based on the projection on the number of new passenger vehicles in the market and the logistic demand diffusion model, as discussed below.  3.4.1.1 New passenger vehicle projection The number of new passenger vehicles was projected based on the variation of the gross domestic product (GDP) per capita. As Figure 3.5 shows, the historical data on the annual growth rate of real GDP per capita may not be quantitatively correlated to the annual growth rate of new passenger vehicles in B.C. However, a qualitative relationship could be found between their moving averages. Figure 3.6 shows that 80% of all points fall on the first and the third quadrants, where the growth rate of GDP has a positive relationship with the growth rate of new vehicles. Thus, we could strongly argue that the purchasing power drives the number of new passenger vehicles in B.C. The annual GDP growth rate is projected to stay relatively constant [10] with an average value of 0.0075% from 2017 to 2050 in B.C. (Figure 3.5). This is around half of the historical average value from 1995 to 2016 (0.0152%). Accordingly, we assumed that the growth rate of new passenger vehicles will also remain constant at an average value of 0.0027%, which is half of its historical value. This growth rate was then used to project the annual number of new passenger vehicles from 2017 to 2050.  46   Figure 3.5  Comparison of the annual growth rate of GDP per capita and new passenger vehicle in BC.: projection vs historical data   Figure 3.6 Comparison of the moving average of GDP per capita and new passenger vehicle growth rates in BC. (1995 -2016)  -0.2-0.15-0.1-0.0500.050.10.15199520002005201020152020202520302035204020452050Annual growth rate (%)YearReal GDP per capita—historical dataReal GDP per capita—ProjectionNew light duty passenger vehicles—historical dataNew light duty passenger vehicles—projection-0.04-0.0200.020.04-0.12 -0.06 0 0.06 0.12real GDP per capita growth rate (%) New light duty passenger vehicle growth rate (%)2-year moving average3-year moving average4-year moving averageIIIIII IV47  3.4.2 FCEV penetration to the market (scenario development) The penetration percentage of FCEVs for the study period (2020–2050) was calculated using the logistic demand diffusion model [116]:   (𝑛𝑡𝑁)𝐹𝐶𝐸𝑉=  11 + 𝑒𝑥𝑝[−(𝛼𝐹𝐶𝐸𝑉 + 𝛽𝐹𝐶𝐸𝑉(𝑡 − 𝑡0))] 3.54  In which 𝑛𝑡 is the fraction of the new vehicle market that the FCEVs possess at time t, N is the fraction of the new vehicle market that the passenger FCEVs can potentially capture, and α characterizes the time it takes for a diffusion process to start ramping up. β characterizes the steepness of the central portion of the curve, as shown in Figure 3.8 (a). Assuming that by 2035 half of the ultimate market will be reached, α and β were calculated as −9.21 and 0.34, respectively.  The demand scenarios were developed by assigning different values to the maximum penetration percentage of new passenger FCEVs in B.C. (N in Eq. 3.54). The contribution of passenger FCEVs to the new vehicle market in 2050 was assumed to be 10%, 30%, and 50% for pessimistic, moderate, and optimistic scenarios, respectively. The annual number of new passenger FCEVs in each demand scenario was calculated using the annual penetration percentage (𝑛𝑡), calculated from Eq. 3.54, and the projected total new passenger vehicles in the market (Figure 3.8 (a)).  3.4.2.1 Passenger vehicle stock projection The stock of passenger vehicles was calculated each year (t) based on the number of new light duty passenger vehicles and the average vehicle lifetime. The average passenger vehicle lifetime in B.C. was calculated annually based on the maximum vehicle age (Y) and total kilometers traveled (K) [65]:  𝑆𝑡𝑜𝑐𝑘𝑡 = 𝑆𝑡𝑜𝑐𝑘𝑡−1 + 𝑁𝑒𝑤 𝑉𝑒ℎ𝑖𝑐𝑙𝑒𝑡 −  𝑁𝑒𝑤 𝑉𝑒ℎ𝑖𝑐𝑙𝑒𝑡−𝑖   ൝𝑖|0 ≤ 𝑖 ≤ 𝑌, ∑ 𝑈𝑘 ≥ 𝐾𝑡𝑘=𝑖‖ 𝑡 − 𝑖 ≥ 𝑌ൡ 3.55  48  In which Uk is the annual average vehicle use intensity. This method was validated in B.C. by back calculating the historical stock of passenger vehicles [9], and the error was less than 4%.  The share of passenger trucks from the stock of light duty passenger vehicles was projected, assuming that the future trend follows the business as usual scenario as shown in Figure 3.7. It was also assumed that the market share of the new passenger trucks follows the same trend as the stock of passenger trucks.   Figure 3.7. Share of passenger trucks from the total passenger vehicles: projection versus historical data [9]  3.4.2.2 FCEV stock projection The stock of passenger FCEVs for each demand scenario was projected with the same method as discussed for total passenger vehicles, using the scenario-based FCEV market share and the lifetime of 18 years (Figure 3.8 (b)). The share of passenger FCE trucks from the total passenger FCEVs was assumed to be similar to Figure 3.7.  0%10%20%30%40%50%60%199520002005201020152020202520302035204020452050Share of passenger trucks  YearHistorical dataProjection49    (a) (b) Figure 3.8. (a) Penetration percentage of the new passenger FCEVs to the B.C. market for different demand scenarios over time. (b) Passenger FCEV stock for different demand scenarios in B.C. over time.  3.4.2.3 Annual hydrogen demand calculation The annual hydrogen demand in year t for passenger vehicles was calculated by:  𝐻2𝐷𝑒𝑚𝑎𝑛𝑑𝑡= 𝑉𝑈𝑠𝑒_𝐼𝑛𝑡𝑡̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅|𝐹𝐶𝐸_𝐶𝑎𝑟 ×  𝐹_𝐸𝑓𝑓𝑡̅̅ ̅̅ ̅̅ ̅̅ ̅|𝐹𝐶𝐸_𝐶𝑎𝑟 × 𝑆𝑡𝑜𝑐𝑘𝑡|𝐹𝐶𝐸_𝐶𝑎𝑟+ 𝑉𝑈𝑠𝑒_𝐼𝑛𝑡𝑡̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅|𝐹𝐶𝐸_𝑃𝑡𝑟𝑢𝑐𝑘 × 𝐹_𝐸𝑓𝑓𝑡̅̅ ̅̅ ̅̅ ̅̅ ̅|𝐹𝐶𝐸_𝑃𝑡𝑟𝑢𝑐𝑘 × 𝑆𝑡𝑜𝑐𝑘𝑡|𝐹𝐶𝐸_𝑃𝑡𝑟𝑢𝑐𝑘 3.56  In which 𝑆𝑡𝑜𝑐𝑘𝑡|𝐹𝐶𝐸_𝐶𝑎𝑟 and 𝑆𝑡𝑜𝑐𝑘𝑡|𝐹𝐶𝐸_𝑃𝑡𝑟𝑢𝑐𝑘 are the stock of FCE cars and passenger trucks in year t, respectively. The 𝑉𝑈𝑠𝑒_𝐼𝑛𝑡𝑡̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅ (km/year) stands for the vehicle average use intensity and is projected using a quadratic polynomial regression with the minimum mileage value extending over the studied time frame (Figure 3.9).     0%10%20%30%40%50%60%2020 2030 2040 2050Penetration of FCEVs  in the marketYearPessimistic scenarioModerate scenarioOptimistic scenario0200 K400 K600 K800 K1.0 M1.2 M2020 2030 2040 2050FCEV stockYearPessimistic scenarioModerate scenarioOptimistic scenario50  The average fuel efficiency of passenger vehicles 𝐹_𝐸𝑓𝑓̅̅ ̅̅ ̅̅ ̅̅ 𝑡(MJ/km) in each year was calculated based on the historical data [9] and a projected annual 1.5% fuel efficiency improvement (impv) in the new vehicles of the previous year (𝑁𝑒𝑤𝐹_𝐸𝑓𝑓𝑡−1) [117]:  𝐹_𝐸𝑓𝑓̅̅ ̅̅ ̅̅ ̅̅ 𝑡+1 = 𝐹_𝐸𝑓𝑓̅̅ ̅̅ ̅̅ ̅̅ 𝑡 × (1 −𝑁𝑒𝑤_𝑉𝑡+1𝑆𝑡𝑜𝑐𝑘𝑡+1) + (𝑁𝑒𝑤_𝑉𝑡+1𝑆𝑡𝑜𝑐𝑘𝑡+1) × 𝑖𝑚𝑝𝑣 × 𝑁𝑒𝑤𝐹_𝐸𝑓𝑓𝑡−1 3.57   Figure 3.9. Vehicle average use intensity for cars and passenger trucks in BC: projection versus historical data [9]1.  It should be noted that in all calculations, FCEVs were assumed to have the same vehicle use intensity as conventional vehicles, driving on the same road and climatic conditions, and with comparable loads. Figure 3.10 shows the calculated annual hydrogen demand from the stock of light duty FCEVs for three demand scenarios.  1 According to NRCan data [9] the annual vehicle use intensity in BC decreased by 36% and 43% for cars and light duty passenger trucks respectively, from 2000 to 2015. The stock of cars and light duty passenger trucks increased by 35% and 93%, respectively. Considering fuel efficiency improvement, the GHG emissions decreased by 24% from cars, however emissions increased by 8% from light duty passenger trucks from 2000 to 2015. 05,00010,00015,00020,00025,000199520002005201020152020202520302035204020452050Vehicle average use intensity (km/year)YearCars—historical dataCars—projectionPassenger trucks—historical dataPassenger trucks—projection51   Figure 3.10. Annual hydrogen demand in B.C. for different demand scenarios over time.  3.4.3 Spatial projection of hydrogen demand in B.C. From 2013 to 2017, the distribution of internal combustion engine (ICE) passenger vehicles in different regions of B.C. remained relatively unchanged with a mean value of 56% in Metro Vancouver, 17% on Vancouver Island, 16% in the Southern Interior, 6% in the North Central area, and 5% in all remaining areas [118]. In this study, it was assumed that the FCEVs in B.C. have the same distribution as ICE vehicles and are confined to metropolitan areas. For Metro Vancouver, 10 municipalities were selected as hydrogen demand regions. Based on the population size, Victoria on Vancouver Island, Kelowna and Kamloops in the Southern Interior, and Prince George in the North Central area were also selected as demand regions. Additionally, 4 municipalities (Abbotsford, Hope, Whistler, and Williams Lake) located on the busiest roads connecting Metro Vancouver to the other metropolitan areas were selected as demand regions. The demand in these regions was calculated based on the traffic volume. In 2016, the average daily traffic volume passing through Abbotsford, Hope, Whistler, and Williams Lake was approximately 2%, 0.8%, 0.6%, and 0.7% of the total passenger vehicles, respectively [119]. These shares were adopted to account for the hydrogen demand on the major roads in B.C.  010,00020,00030,00040,00050,00060,00070,00080,00090,0002020 2025 2030 2035 2040 2045 2050Total hydrogen demand  (tonnes/year)YearOptimistic scenarioModerate scenarioPessimistic scenario52  The temporal variation of hydrogen demand distribution among the 10 municipalities in Metro Vancouver was calculated based on the projected population density. For other regions, the hydrogen demand distribution was assumed to remain constant over the time frame of this study.  3.4.3.1 Hydrogen demand distribution in Metro Vancouver Metro Vancouver was divided into 10 major municipalities: Surrey and White rock, Vancouver, Burnaby and New Westminster, City of Langley and Langley Township, Coquitlam, North Vancouver, West Vancouver, Maple Ridge and Pitt Meadows, Richmond, and Delta. The term PD was introduced as an indicator of population density for each municipality g in time t, as follows:  𝑃𝐷𝑔|𝑡 =𝐷𝑤𝑔 + 𝐸𝑚𝑝𝑔𝑈𝐶𝐵𝑔 3.58 In which Dwg and Empg stand for the total dwelling units and total employment, respectively, in municipality g in time t. UCBg stands for the urban containment boundary for municipality g. Dwg and Empg were projected to 2040 by [120], and were extrapolated to 2050 to cover the time frame of this study. UCBg was considered constant over time, as stated by [120]. The distribution of hydrogen demand among these municipalities was assumed to be consistent with the distribution of PDg in each t, as shown in Figure 3.11. 53   Figure 3.11. Distribution of population density among 10 municipalities in Metro Vancouver over time.  3.5 Policy scenario development  This work incorporates different combinations of the current and potential provincial environmental mandates to assess their effect on the hydrogen price and environmental performance of the HFSC.   3.5.1 Current provincial policies  The low-carbon fuel standard (LCFS) and carbon tax are the environmental mandates that influence the evolution of the low-carbon fuel infrastructures by monetizing CO2 emissions. The HFSC optimization was performed for three cases in which carbon tax, BC-LCFS, and both policies were included in the model. The carbon tax fees and LCFS revenues were modelled as part of the objective function, as described in chapter 4.  It was also assumed that the hydrogen infrastructure is eligible for accelerated depreciation deduction. To this end, the capital cost allowance (CCA) deduction was calculated using the 0%5%10%15%20%25%30%Population density distribution 20112015202020252030203520402045205054  declining balance method, based on 30% CCA rate for production plants, storage facilities and dispensers and 40% CCA rate for tube trailer and tanker trucks [121].  3.5.1.1 B.C. low carbon fuel standard (LCFS) In 2010, the government of B.C. included the LCFS as part of its Renewable and Low Carbon Fuel Requirements [122]. The LCFS is both a regulatory and a market-based policy to reduce the transportation fuel carbon intensity. The regulatory part enforces a carbon intensity target with which all the producers and importers of transportation fuels must comply. The market-based policy permits the fuel providers to trade and bank emission credits to remain compliant with the regulation. Hydrogen fuel suppliers can benefit from selling the emission credits on an open market to suppliers who incur debits from providing fuels with carbon intensities beyond the limit. Each credit accounts for the tonnes of CO2 avoided on a WTW basis by substituting hydrogen for gasoline. Thus, HFSCs, which have lower emissions, generate more revenue while imposing higher capital and operating costs. As a result, the HFSC incorporates lower-emitting components if the revenues surpass the investment. No long-term projection for carbon credit price was found in the literature nor in the governmental resources. Thus, we assumed that the credit price starts at C$167 (the average price from 2015 to 2017 [123]) in 2020 and decreases over time until it reaches 0 at the final time step (167, 154, 112, 47, 11, and 0 for the 1st to 6th time steps, respectively). The logistic demand diffusion model was used with the same α and β values calculated for the adoption rate of new FCEVs in the market (section 3.4). This assumption was based on the reasoning that the increasing rate of adoption of low-emission technologies increases the number of credits in the market, which decreases the credit price accordingly. Currently, hydrogen producers in B.C. do not receive LCFS credits for using renewable content from the grid. However, this model was developed with the assumption that the electrolyzer pathway allowances were increased to include renewable electricity. Moreover, the credits that may be awarded through a Part 3 Agreement [124] were not considered in this study. This is mainly due to the limited total credits available annually, which must be distributed among all Part 3 fuel suppliers on a case-by-case basis.   55  3.5.1.2 B.C. carbon tax The B.C. carbon tax, introduced in 2008, covers greenhouse gas emissions resulting from the combustion of all fossil fuels used within the province [125]. The tax started at C$10 per tonne of carbon dioxide equivalent when introduced in 2008. It then rose C$5 per tonne each year until it reached C$30 per tonne in 2012. On April 1, 2018, B.C.’s carbon tax rate was C$35 per tonne of carbon dioxide equivalent emissions. This tax rate increases each year by C$5 per tonne until it reaches C$50 per tonne in 2021. At the time of this study, no carbon pricing scheme has been announced in B.C. beyond 2021. It was assumed that the carbon tax will start at C$45 in 2020 and increase annually until 2050 with a value equal to the estimated social cost of carbon [126]. As the carbon tax imposes a cost on a WTW basis to the hydrogen infrastructure, the lower-emissions components may become economically viable despite their higher investment cost.   3.5.2 Potential financial and regulatory policies  No study was found in literature to devise targeted fiscal and financial policies for the accelerated adoption of low-carbon hydrogen in the transport sector. For the purpose of this study, the policies were adopted from a range of economic instruments that has been used to promote renewable energy worldwide. Based on Table 3.9, production tax credit (PTC), capital subsidy (grant) and utility incentive for electrolytic hydrogen were adopted with varying stringency over time. These policies were added to the objective function of H2SCOT in the base case. The base case already includes the BC-LCFS and carbon tax in the objective function. The HFSC was also optimized for cases with higher carbon tax rates (compared to the current policy) as well as for the case in which SMR based hydrogen production without CCS integration is banned for transportation sector.        56  Table 3.9. Economic instruments to promote renewable energy worldwide. Category Policy Type Description Example Fiscal /financial incentives Carbon Tax A carbon tax is a fee levied on each tonne of carbon dioxide emitted from burning carbon-based fuels. Manitoba Emissions Tax on Coal Act (Canada) [127] Carbon tax (Japan [128], British Columbia (Canada) [125] Capital subsidy, Grants One-time payment to cover a fraction of the capital cost of the investment Wood-to-Energy Grants (USA) [129] Smart Grid Investment Grants (USA) [130] California Solar Initiative [131] Ontario saveONenergy (Canada) [132] Feed-in Tariff /premiums A long-term purchase agreement for the sale of renewable electricity. Feed-in tariff sets a minimum price guaranteed which is above the standard market price. Feed-in premiums establishes a constant or sliding premium on existing market price. Ontario Feed-in Tariff Programme (Canada) [133] Residential Net Feed-in Tariff for Western Australia [134] Renewable Energy Act (Germany) [135] Feed-in premium tariffs for renewable power (Denmark) [136] Renewable Energy Feed-In Tariff (France) [137] Loan guarantees A guarantee that allows a lender (financial institutions or utilities) to recover a fraction of the principal and accrued interest on a loan that may go into default from the government Future Fuels Initiative (Canada) [138] Green Loan Guarantee Program (Alberta-Canada) [139] RenovAr program (Argentina) [140] Production tax credit PTC provides an income tax rebate based on the amount of production by a qualified business for a specified period of time Cellulosic Biofuel Producer Tax Credit (USA) [141] The Renewable Electricity Production Tax Credit (USA) [142] Investment tax credit ITC provides an income tax rebate based on the capital investment volume in a qualified business Clean Coal Facility Tax Credit (USA) [143] Solar Investment Tax Credit (USA) [144] San Francisco Solar Energy Incentive Program [145] Tax credit for energy transition (France) [146] Capital Investment Tax Credit (Alberta-Canada) [147] Reduced excise taxes It provides consumption tax exemptions or reduction on the sale of qualified products  Preferential Tax Regimes for Biofuels (UK) [148] Renewable Energy Tax Excise (Poland) [149] Accelerated depreciation any method of depreciation used for income tax purposes that allows greater deductions in the earlier years of the life of an asset Enhanced Capital Allowances (UK) [150] Accelerated Capital Cost Allowance (Canada) [121] Modified Accelerated Cost Recovery System (USA) [151] 57  Policy Type Description Example Soft loans Provided by governments with preferential terms such as lower interest rates, longer loan terms, etc. to reduce the costs of capital. Energy Provisioning (Germany) [152] Preferential loans for energy saving measures (France) [153] Green Municipal Fund (Canada) [154] Advanced Technology Vehicle Manufacturing Loan Program (USA) [155] Utility rebates and incentives Financial incentives from utilities to customers to replace inefficient products or improve the energy efficiency of the existing systems  BC Hydro’s Power Smart Partners Express (PSP) (Canada) [156] Efficiency Incentives for Large Electricity Consumers (Greece) [157] Market-based GHG emissions trading (Cap-and-Trade & Baseline-and-Credit) A central authority creates tradable pollution permits which can be bought and sold in an open market to meet emission reduction objectives Quebec Cap & Trade System for Greenhouse Gas Emissions Allowances [158] Regional Greenhouse Gas Initiative (USA) [159] EU Emissions Trading System [160] Renewable Energy Certificates Renewable energy certificates are proof that energy has been generated from renewable sources. These certificates are classified as a commodity and allow the renewable attributes of energy to be sold or traded separately from the physical unit of energy.  White Certificate Scheme & Obligation (France) [161] Renewable Energy Green Certificate and Trading Mechanism (China) [162] Norway-Sweden Green Certificate Scheme for electricity production [163] Low carbon fuel standard (LCFS) The LCFS includes a credit trading system that allows providers to generate tradable credits through the use of low carbon fuels, and in turn imposes deficits for the use of higher-carbon fuels. Renewable and Low Carbon Fuel Requirements Regulation (B.C-Canada) [122] ARB's Low Carbon Fuel Standard (California-USA) [164]  3.5.2.1 Production tax credit (PTC) Production tax credit (PTC) is a preferential tax treatment that was included in the US Energy Policy Act of 1992 [142]. PTC provides an inflation-adjusted tax credit on every kilowatt-hour of electricity generated by the qualified energy sources for a limited time period. PTC has been one of the major drivers of wind power development in the US which resulted in quadrupling capacity, and the 40 % of cost reduction between 2007 and 2014 [165]. In this study, PTC has been applied to hydrogen produced from i) water electrolysis, ii) SMR equipped with CCS and iii) by-product hydrogen purification. The credits were allocated through different multi-stage settings over time, as shown in Figure 3.12 (a). The maximum tax credit was assumed at C$2/kg adjusted downwards over time in PTC_Step. This setting is designed to assure the overall transition proceeds from policy-driven to market-driven. The same strategy was also 58  considered for PTC_Delay, with an exception that the effective date has a 10-year delay.  Constant tax credit at C$2/kg and C$1/kg was assumed for PTC_$2 and PTC_$1, respectively. The maximum C$2/kg was adopted such that it is less than the applicable tax on the optimal hydrogen price in in the optimistic scenario for the base case. It is likely that the amount of depreciation deductions exceeds the taxable income in the first years of HFSC operation. Thus, HFSC may owe no income tax, which means that earned PTCs potentially go unused. It was assumed that the value of the production tax credits was captured by monetizing tax benefits.  This could happen by applying the deductions and credits against outside income, carrying  the tax benefits forward over time, or through a third-party tax equity investor [166].   3.5.2.2 Capital subsidy  A capital subsidy is a one-time lump-sum payment that covers a portion of the upfront capital cost of an asset. This subsidy is not repayable and aims to enhance the financial viability of an investment. In this study, water electrolysis and CCS technologies were assumed to be eligible for cash subsidy. The upfront cash payment was allocated in different multi-stage settings over time, as shown in Figure 3.12 (b). In Grant_Step, it was assumed that the grant covers the total capital cost of the eligible facilities and steps down in value over the years, until it phases out completely in 2040. Grant_100% provides the maximum support in which the total capital cost of facilities was offset by the government over the 30-year period. Grant_Delay has a stepwise fund allocation similar to Grant_Step, except for the 10-year delay on the effective date of policy.  3.5.2.3 Utility incentives on electrolytic hydrogen Favorable utility rates could encourage certain investments to develop or to pursue energy efficiency. The electricity generation in B.C. is hydro dominated. Thus, the power subsidies could potentially encourage low-carbon electrolytic hydrogen. In this study, it was assumed that the utility subsidizes the electricity rate for central and on-site electrolyzers in four different multi-stage settings as shown in Figure 3.12 (c). In EC_Step, the subsidy covers the total electricity cost of electrolyzers and steps down in value over the years, 59  until it disappears in 2040. EC_100% provides the maximum support in which the total electricity cost was offset by the utility over the 30-year period. EC_25% has the minimum support in which 25% of the electricity rate is covered by the utility over the entire study period. EC_Delay has a stepwise subsidy allocation similar to EC_Step, except for the 10-year delay on the effective date of policy.  3.5.2.4 Higher rates of carbon tax  In the base case and all the policy cases, the carbon tax rate for the time period of 2020 to 2050 was adopted from values of the social cost of carbon (SCC), estimated by the Canada Treasury Board Secretariat’s Analysis Guide [126]. However, it is likely that current estimates of SSC are biased downwards [167], [168]. Thus, Environment Canada suggested to use higher SCC values for sensitivity analysis and updated the “95th percentile” SCC value to C$167/tonne [126]. In this study, two cases were assessed with the annual carbon tax increase at C$2.04/tonne (CT_2X) and C$4.08/tonne (CT_4X), which is two and four times the annual increase of the base case, as shown in Figure 3.12 (d).        60   (a)  (b)  (c) 00.511.522020 2025 2030 2035 2040 2045 2050Tax credit (C$/kg H2)YearPTC_Step PTC_Delay PTC_$2 PTC_$10%25%50%75%100%2020 2025 2030 2035 2040 2045 2050Captital Subsidy (% of capital cost)YearGrant_Step Grant_100% Grant_Delay0%25%50%75%100%2020 2025 2030 2035 2040 2045 2050utility Subsidy (% of electricity rate)YearEC_Step EC_100% EC_25% EC_Delay61   (d) Figure 3.12. Environmental policies with various deployment strategies over time (a) production tax credit (PTC), (b) capital subsidy, (c) utility incentives on electrolytic hydrogen, (d) carbon tax rate  3.5.2.5  Ban on the SMR hydrogen production without CCS integration  Because SMR is the most cost-efficient technology to produce hydrogen, it may delay the adoption of lower-carbon production technologies. Thus, a scenario was developed in which the SMR plants which were not equipped with CCS were banned from hydrogen production for the transportation sector. The HFSC was optimized to assess the extra cost imposed on the infrastructure as well as the GHG emissions reduction benefits.   040801201602002020 2025 2030 2035 2040 2045 2050C$/tonnes CO2Year Base Case CT_2X CT_4X62  Chapter 4: Hydrogen Supply Chain Cost Optimization Model (H2SCOT): Formulation This chapter represents the model’s assumptions, constraints, and objective function. H2SCOT was developed based on mixed integer linear programming (MILP), implemented in AMPL and solved using CPLEX v12.7.0.  4.1 MILP basics A MILP model minimizes (maximizes) a linear function over all n-dimensional vectors x (continuous) and y (integer) subject to a set of linear equality and inequality constraints as well as integrality restrictions on the variables in y.  Minimize (Maximize)  𝑧 = ∑ 𝑐𝑗𝑇𝑥𝑗 + 𝑑𝑗𝑇𝑦𝑗𝑛𝑗=1                                                                               4.1 Subject to                    ∑ 𝐴𝑗𝑥𝑗 + 𝐾𝑗𝑦𝑗𝑛𝑗=1 ൝≤=≥ൡ 𝑏𝑖     for 𝑖 = 1. . 𝑚                                       0 ≤ 𝑥𝑗 ≤ 𝑢𝑗   & 𝑦𝑗 ∈ 𝑍    for 𝑗 = 1. . 𝑛  MILP is popular in capital budgeting where investment selection is made based on cash-flow constraints, warehouse location optimization which involves the decision on the number of facilities and scheduling problems which involve the sequencing and routing decisions.  MILP problems are generally solved using a linear-programming based branch-and-bound algorithm. First, the integer constraints are dropped to find the optimal solution to the "relaxation" of the problem via a standard linear optimization method. If the integer values are found for the decision variables with integer constraints, then those values are selected as optimal.  If non-integral solutions are found for integer variables, this method branches on one of such variables and creates two new sub problems where the value of that variable is more tightly constrained.  These sub problems are solved, and the process is repeated, until a solution is found which satisfies all the integer constraints. 63  4.2 H2SCOT basic assumptions • H2SCOT did not include the possibility of industrial hydrogen sources nor the import of hydrogen into the province. • The number of HFSC components was initialized at a null value. • No lead time was considered for building plants or storage facilities and fueling stations. • The natural gas resources and hydro-electricity generation could support the maximum demand in this study [65], [169]. • The projected price of electricity and natural gas was adopted from the National Energy Board projections [10] for the operating cost calculation. • Large-scale storage options such as salt caverns, liquid organic hydrogen carriers (LOHC) and pipeline distribution networks were not considered in this analysis. The salt cavern (100s of GWh of capacity) is suitable for high demand, especially when seasonal storage is required to deal with the intermittent energy resources (wind and solar energy) [50], [78]. The LOHC pathways are also suitable candidates in case of intermittent renewable electricity generation [20]. The electricity generation in B.C. is almost exclusively from hydro power. The hydrogen delivery via pipelines is a low cost option for large hydrogen demand in dense cities as discussed by Yang and Ogden [170]. Unlike many regions in Europe, B.C represents a 944,735 km2 jurisdiction (2.6 times larger than Germany) with a population of 5 million people (similar to Ireland). Gas and liquid truck delivery were selected in this study because the hydrogen flow rates and the population density of the demand regions were much lower than the suitable limits for pipeline distribution [170].    4.3 H2SCOT constraints 4.3.1 Production facilities Hydrogen can be produced at central plants or on-site at the fueling stations. As discussed in chapter 3, SMR with carbon capture and sequestration (CCS), SMR without CCS, water electrolysis, and by-product hydrogen purification were considered for central production. Water electrolysis was used for on-site production. The maximum primary capacity was selected at 50 and 100 tonnes/day for SMR plants and 10, 50, and 100 tonnes/day for central electrolyzers. The maximum capacity for on-site electrolyzers was selected at 150, 500, 1000, and 1500 kg/day. The 64  capacity of a plant for by-product hydrogen purification was at 10 tonnes/day. In case of liquid hydrogen delivery, the liquefaction plant was attached to the central production plant. The liquefaction plant was not considered for on-site production facilities. The following constraints were considered for the production facilities: • The production rate of a central plant is constrained by the maximum and minimum production capacities:  𝑃𝑐𝑎𝑝_𝑚𝑖𝑛𝑐𝑌𝐶𝑐𝑦𝑑𝑔𝑡 ≤ 𝑃𝐶𝑐𝑦𝑑𝑔𝑡 ≤ 𝑃𝑐𝑎𝑝_𝑚𝑎𝑥𝑐𝑦𝑑𝑔𝑡𝑌𝐶𝑐𝑦𝑑𝑔𝑡          4.2 ∀ 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇   In which 𝑃𝑐𝑎𝑝_𝑚𝑖𝑛𝑐 = 0.1 × 𝑃𝑐𝑎𝑝_𝑚𝑎𝑥_𝑝𝑟𝑖𝑛𝑐. • The production rate of an on-site electrolyzer is bound within certain limits and is equal to the sum of dispensing and storage rates for the corresponding fueling station. For Metro Vancouver municipalities:  𝐷𝑐𝑎𝑝_𝑚𝑖𝑛𝑠𝑌𝑂_𝑉𝑠𝑑𝑔′𝑡 ≤ 𝑃𝑂_𝑉𝑠𝑑𝑔′𝑡 ≤ 𝐷𝑐𝑎𝑝_𝑚𝑎𝑥𝑠𝑌𝑂_𝑉𝑠𝑑𝑔′𝑡      4.3 (a) 𝑃𝑂_𝑉𝑠𝑑𝑔′𝑡 = 𝐷𝐼_𝑉𝑠𝑑𝑔′𝑡 + 𝑆𝑇𝑅_𝑉𝑠𝑑𝑔′𝑡         4.4 (a) ∀ 𝑠 ∈ 𝑆, 𝑑 ∈ 𝐷, 𝑔′ ∈ 𝐺′, 𝑡 ∈ 𝑇      • For major municipalities except Metro Vancouver (Kamloops, Kelowna, Prince George and Victoria):  𝐷𝑐𝑎𝑝_𝑚𝑖𝑛𝑠𝑌𝑂_𝑁𝑉𝑠𝑑𝑛′𝑡 ≤ 𝑃𝑂_𝑁𝑉𝑠𝑑𝑛′𝑡 ≤ 𝐷𝑐𝑎𝑝_𝑚𝑎𝑥𝑠𝑌𝑂_𝑁𝑉𝑠𝑑𝑛′𝑡        4.3 (b) 𝑃𝑂_𝑁𝑉𝑠𝑑𝑛′𝑡 = 𝐷𝐼_𝑁𝑉𝑠𝑑𝑛′𝑡 + 𝑆𝑇𝑅_𝑁𝑉𝑠𝑑𝑛′𝑡         ∀ 𝑠 ∈ 𝑆, 𝑑 ∈ 𝐷, 𝑛′ ∈ 𝑁𝑉, 𝑡 ∈ 𝑇 4.4 (b)  • For major municipalities on the connecting roads (Abbotsford, Hope, Whistler and Williams Lake):  𝐷𝑐𝑎𝑝_𝑚𝑖𝑛𝑠𝑌𝑂_𝑅𝑂𝑠𝑑𝑟′𝑡 ≤ 𝑃𝑂_𝑅𝑂𝑠𝑑𝑟′𝑡 ≤ 𝐷𝑐𝑎𝑝_𝑚𝑎𝑥𝑠𝑌𝑂_𝑅𝑂𝑠𝑑𝑟′𝑡        4.3 (c) 𝑃𝑂_𝑅𝑂𝑠𝑑𝑟′𝑡 = 𝐷𝐼_𝑅𝑂𝑠𝑑𝑟′𝑡 + 𝑆𝑇𝑅_𝑅𝑂𝑠𝑑𝑟′𝑡         ∀ 𝑠 ∈ 𝑆, 𝑑 ∈ 𝐷, 𝑟′ ∈ 𝑅𝑂, 𝑡 ∈ 𝑇 4.4 (c) 65  • Depending on the state of hydrogen (d), the following constraints are also applied:  𝑌𝐶𝑐𝑦𝑑𝑔𝑡 = 0          4.5  ∀ 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇, 𝑑 ∈ 𝐷: 𝑑 = 1  𝑌𝑂_𝑉𝑠𝑑𝑔′𝑡 = 0, 𝑌𝑂_𝑁𝑉𝑠𝑑𝑛′𝑡 = 0,           𝑌𝑂_𝑅𝑂𝑠𝑑𝑟′𝑡 = 0       4.6 ∀ 𝑠 ∈ 𝑆, 𝑔′ ∈ 𝐺′, 𝑛′ ∈ 𝑁𝑉, 𝑟′ ∈ 𝑅𝑂, 𝑡 ∈ 𝑇 , 𝑑 ∈ 𝐷: 𝑑 ≠ 1   • The continuous variables of 𝑍𝐶𝑐𝑦𝑑𝑔𝑡 and M (sufficiently large) was introduced to linearize 𝑃𝑐𝑎𝑝_𝑚𝑎𝑥𝑐𝑦𝑑𝑔𝑡𝑌𝐶𝑐𝑦𝑑𝑔𝑡  , as follows:  𝑍𝐶𝑐𝑦𝑑𝑔𝑡 ≥ 0 𝑍𝐶𝑐𝑦𝑑𝑔𝑡 ≤ 𝑀 ∗ 𝑌𝐶𝑐𝑦𝑑𝑔𝑡 𝑍𝐶𝑐𝑦𝑑𝑔𝑡 ≤ 𝑃𝑐𝑎𝑝_𝑚𝑎𝑥𝑐𝑦𝑑𝑔𝑡 𝑃𝑐𝑎𝑝_𝑚𝑎𝑥𝑐𝑦𝑑𝑔𝑡 − 𝑍𝐶𝑐𝑦𝑑𝑔𝑡 ≤ 𝑀 ∗ (1 − 𝑌𝐶𝑐𝑦𝑑𝑔𝑡)  4.3.2 Terminals with central storage facilities Terminals are geographically dispersed and can be located in any supply regions. The inventory of the central storage facilities supports both the daily demand of hydrogen and provides a backup for demand fluctuation. The maximum primary capacity options of 10, 50, and 100 tonnes/day were selected for central storage facilities. The following constraints were considered for the terminals: • The average inventory of hydrogen falls within the maximum and minimum capacity of the facility:  𝑇𝑆𝑐̅𝑑?̅?𝑡 ≤ 𝑆𝑐𝑎𝑝_𝑚𝑎𝑥𝑐̅𝑑?̅?𝑡𝑌𝑆𝑐̅𝑑?̅?𝑡 4.7 𝑆𝑐𝑎𝑝_𝑚𝑖𝑛𝑐̅𝑌𝑆𝑐̅𝑑?̅?𝑡 ≤ 𝑇𝑆𝐸𝑐̅𝑑?̅?𝑡        ∀ 𝑐̅ ∈ 𝐶̅, 𝑑 ∈ 𝐷, ?̅? ∈ ?̅?, 𝑡 ∈ 𝑇              In which 𝑆𝑐𝑎𝑝_𝑚𝑖𝑛𝑐̅ = 0.1 × 𝑆𝑐𝑎𝑝_𝑚𝑎𝑥_𝑝𝑟𝑖𝑛𝑐̅. 𝑇𝑆𝑐̅𝑑?̅?𝑡 is the total inventory of the storage facility (daily demand plus backup), and 𝑇𝑆_𝐸𝑐̅𝑑?̅?𝑡 is the backup inventory.  66  • Terminals receive all the gaseous/liquefied hydrogen produced by central plants:   ∑ 𝑃𝐶𝑐𝑦𝑑𝑔𝑡𝑐𝑦𝑔− ∑ 𝑇𝑆𝑐̅𝑑?̅?𝑡𝑐̅?̅?= 0               ∀ 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑐̅ ∈ 𝐶̅, 𝑑 ∈ 𝐷, ?̅? ∈ ?̅?, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇 4.8  4.3.3 Transportation and distribution Two hydrogen transport modes were included in the model: liquid hydrogen tanker truck with the deliverable capacity of 3800 kg and compressed-gaseous hydrogen tube trailers with three delivery capacities of 100, 500, and 900 kg. The following constraints were considered for the transportation and distribution network: • All the hydrogen produced in each supply region was stored within the warehouses of the same region (next to the plant) or transported to the warehouses in other regions. In the former case, no transportation was considered between the plant and the warehouse. For gaseous hydrogen:  ∑ 𝑃𝐶𝑐𝑦𝑑𝑔𝑡𝑐𝑦− ∑ 𝑇𝑆𝑐̅𝑑?̅?𝑡𝑐̅− ∑ 𝐶𝐴𝑃_𝐺𝑎 𝑁𝐺_𝑃𝑆𝑔?̅?𝑎𝑡 = 0?̅?𝑎 ∀𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑔 ∈ 𝐺, 𝑐̅ ∈ 𝐶̅, ?̅? ∈ 𝐺 ̅: ?̅? = 𝑔, 𝑎 ∈ 𝐴, 𝑡 ∈ 𝑇, 𝑑 ∈ 𝐷: 𝑑 = 2 4.9  The notations were adjusted to develop the mass balance equation for liquid hydrogen transport. • Hydrogen is distributed from central warehouses to the demand regions by tube tankers/tanker trucks. For gaseous hydrogen distribution:  ∑ (𝑇𝑆𝑐̅𝑑?̅?𝑡 − 𝑇𝑆𝐸𝑐̅𝑑?̅?𝑡)𝑐̅= ∑ 𝐶𝐴𝑃_𝐺𝑎 (∑ 𝑁𝐺_𝑉?̅?𝑔′𝑎𝑠𝑡  +𝑔′𝑠∑ 𝑁𝐺_𝑁𝑉?̅?𝑛′𝑎𝑠𝑡  𝑛′+ ∑ 𝑁𝐺_𝑅𝑂?̅?𝑟′𝑎𝑠𝑡      𝑟′)𝑎 ∀ 𝑐̅ ∈ 𝐶̅, ?̅? ∈ ?̅?, 𝑛′ ∈ 𝑁𝑉, 𝑟′ ∈ 𝑅𝑂, 𝑔′ ∈ 𝐺′, 𝑎 ∈ 𝐴, 𝑠 ∈ 𝑆, 𝑡 ∈ 𝑇, 𝑑 ∈ 𝐷: 𝑑 = 2 4.10   67  For liquid hydrogen distribution:  ∑ (𝑇𝑆𝑐̅𝑑?̅?𝑡 − 𝑇𝑆𝐸𝑐̅𝑑?̅?𝑡)𝑐̅= 𝐶𝐴𝑃𝐿_𝑇𝑅 (∑ 𝑁𝑇𝑅𝐿_𝑉?̅?𝑔′𝑡𝑔′ + ∑ 𝑁𝑇𝑅𝐿_𝑁𝑉?̅?𝑛′𝑎𝑡𝑛′+ ∑ 𝑁𝑇𝑅𝐿_𝑅𝑂?̅?𝑟′𝑎𝑡 𝑟′) 4.11 ∀ 𝑐̅ ∈ 𝐶̅, ?̅? ∈ ?̅?, 𝑛′ ∈ 𝑁𝑉, 𝑟′ ∈ 𝑅𝑂, 𝑔′ ∈ 𝐺′, 𝑡 ∈ 𝑇, 𝑑 ∈ 𝐷: 𝑑 = 3   • All trucks entering a demand region serve the fueling stations of that region. It was assumed that no product transfer could occur between demand regions. The mass balance equation for gas and liquid hydrogen delivery to Metro Vancouver municipalities is as follows:  ∑(𝐷𝐼_𝑉𝑠𝑑𝑔′𝑡 + 𝑆𝑇𝑅_𝑉𝑠𝑑𝑔′𝑡)𝑠= ∑ 𝐶𝐴𝑃_𝐺𝑎𝑁𝐺_𝑉?̅?𝑔′𝑎𝑠𝑡 ?̅?𝑎𝑠 ∀ ?̅? ∈ ?̅?, 𝑔′ ∈ 𝐺′, 𝑠 ∈ 𝑆, 𝑎 ∈ 𝐴, 𝑡 ∈ 𝑇, 𝑑 ∈ 𝐷: 𝑑 = 2 4.12  ∑(𝐷𝐼_𝑉𝑠𝑑𝑔′𝑡 + 𝑆𝑇𝑅_𝑉𝑠𝑑𝑔′𝑡)𝑠= 𝐶𝐴𝑃𝐿_𝑇𝑅 ∑ 𝑁𝑇𝑅𝐿_𝑉?̅?𝑔′𝑡?̅? 4.13  ∀ ?̅? ∈ ?̅?, 𝑔′ ∈ 𝐺′, 𝑠 ∈ 𝑆, 𝑡 ∈ 𝑇, 𝑑 ∈ 𝐷: 𝑑 = 3 The notation was adjusted to develop the mass balance equation for gas and liquid hydrogen distribution to other municipalities. • Trucks serve the plants and fueling stations within their daily availability limit. For transportation network: 𝐴𝑇_𝑇𝑟 − (2𝐿𝐻_𝑃𝑆𝑔?̅?𝑉𝐻+ 𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑡𝑖𝑚𝑒𝑑|𝑑=2) < 0   :   𝑁𝑇𝑅𝐺_𝑃𝑆𝑔?̅?𝑎𝑡 = 0 𝐴𝑇_𝑇𝑟 − (2𝐿𝐻_𝑃𝑆𝑔?̅?𝑉𝐻+ 𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑡𝑖𝑚𝑒𝑑|𝑑=3) < 0   :  𝑁𝑇𝑅𝐿_𝑃𝑆𝑔?̅?𝑡 = 0  4.14 For distribution to Metro-Vancouver:  𝐴𝑇_𝑇𝑟 − (2𝐿𝐻_𝑉?̅?𝑉𝐻+2𝐿𝐺𝑔′𝑉𝐺+ 𝑢𝑛𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑡𝑖𝑚𝑒𝑑|𝑑=2) < 0   : 𝑁𝑇𝑅𝐺_𝑉?̅?𝑔′𝑎𝑡 = 0 4.15 68   𝐴𝑇_𝑇𝑟 − (2𝐿𝐻_𝑉?̅?𝑉𝐻+2𝐿𝐺𝑔′𝑉𝐺+ 𝑢𝑛𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑡𝑖𝑚𝑒𝑑|𝑑=3) < 0   : 𝑁𝑇𝑅𝐿_𝑉?̅?𝑔′𝑡 = 0  The notation was adjusted for gas and liquid hydrogen distribution to other municipalities.  4.3.4 Hydrogen fueling stations Three types of fueling stations were considered in this study: stations that receive gaseous hydrogen, those that receive liquid hydrogen, and stations with on-site production. The maximum potential capacities were selected at 150, 500, 1000 and 1500 kg/day, and the minimum capacity was fixed at 10% of the maximum capacity.  The following constraints were considered for the fueling station: • The dispensing rate of a fueling station is constrained between the maximum and minimum capacity of the station. In case of Metro Vancouver municipalities:  𝐷𝑐𝑎𝑝_𝑚𝑖𝑛𝑠𝑌𝐷𝑉𝑠𝑑𝑔′𝑡 ≤ 𝐷𝐼𝑉𝑠𝑑𝑔′𝑡 ≤ 𝐷𝑐𝑎𝑝_𝑚𝑎𝑥𝑠𝑌𝐷𝑉𝑠𝑑𝑔′𝑡        ∀ 𝑠 ∈ 𝑆, 𝑑 ∈ 𝐷, 𝑔′ ∈ 𝐺′, 𝑡 ∈ 𝑇  4.16  In which, 𝐷𝑐𝑎𝑝_𝑚𝑖𝑛𝑠 = 0.1 × 𝐷𝑐𝑎𝑝_𝑚𝑎𝑥𝑠. • The average backup inventory of a fueling station is constrained by the maximum and minimum capacity of that station. The exception is for gaseous delivery, where a tube tanker truck leaves the full tubes at the station and collects the empty tubes. Thus, the remaining gas in the tube trailers acts as the storage for the station. For Metro Vancouver municipalities:  𝑆𝑇𝑅𝑉𝑠𝑑𝑔′𝑡 ≤ 𝐷𝑐𝑎𝑝_𝑚𝑎𝑥𝑠𝑌𝐷𝑉𝑠𝑑𝑔′𝑡               ∀ 𝑠 ∈ 𝑆, 𝑑 ∈ 𝐷: 𝑑 ≠ 2, 𝑔′ ∈ 𝐺′, 𝑡 ∈ 𝑇 4.17 𝐷𝑐𝑎𝑝_𝑚𝑖𝑛𝑠𝑌𝐷𝑉𝑠𝑑𝑔′𝑡 ≤ 𝑆𝑇𝑅𝑉𝑠𝑑𝑔′𝑡                ∀ 𝑠 ∈ 𝑆, 𝑑 ∈ 𝐷, 𝑔′ ∈ 𝐺′, 𝑡 ∈ 𝑇   The notation was adjusted to develop the dispensing rate and backup inventory constraints for other municipalities.  69  4.3.5 Hydrogen demand The total hydrogen demand for light duty passenger vehicles in B.C. is equal to the demand of Metro Vancouver municipalities, major municipalities outside Metro Vancouver (Kamloops, Kelowna, Prince George, and Victoria), and the municipalities on the connecting roads (Whistler, Hope, Abbotsford, and Williams Lake):  𝐷𝑇𝑡 = ∑ 𝐷𝐺𝑉𝑔′𝑡𝑔′+ ∑ 𝐷𝐺𝑁𝑉𝑛′𝑡𝑛′+ ∑ 𝐷𝐺𝑅𝑂𝑟′𝑡𝑟′     ∀𝑔′ ∈ 𝐺′, 𝑛′ ∈ 𝑁𝑉, 𝑟′ ∈ 𝑅𝑂, 𝑡 ∈ 𝑇 4.18  The terms on the right are the hydrogen demand at the final year of each time step.  • The total hydrogen demand in each region is equal to the dispensing rate of all fueling stations in that region. In case of Metro-Vancouver:  𝐷𝐺𝑉𝑔′𝑡 = ∑ 𝐷𝐼𝑉𝑠𝑑𝑔′𝑡𝑠𝑑             ∀ 𝑠 ∈ 𝑆, 𝑑 ∈ 𝐷,   𝑔′ ∈ 𝐺′, 𝑡 ∈ 𝑇  4.19 • The total hydrogen demand (𝐷𝑇𝑡) is equal to the difference between total hydrogen production (central and on-site) and total backup storage (central and on-site):  𝐷𝑇𝑡 = ∑ (𝑃𝑂_𝑉𝑠𝑑𝑔′𝑡 − 𝑆𝑇𝑅_𝑉𝑠𝑑𝑔′𝑡)𝑠𝑑𝑔′+ ∑ (𝑃𝑂_𝑁𝑉𝑠𝑑𝑛′𝑡 − 𝑆𝑇𝑅_𝑁𝑉𝑠𝑑𝑛′𝑡)𝑠𝑑𝑛′+ ∑(𝑃𝑂_𝑉𝑠𝑑𝑟′𝑡 − 𝑆𝑇𝑅_𝑉𝑠𝑑𝑟′𝑡)𝑠𝑑𝑟′+ ∑ 𝑃𝐶𝑐𝑦𝑑𝑔𝑡𝑐𝑦𝑑𝑔−  ∑ 𝑇𝑆_𝐸𝑐̅𝑑?̅?𝑡𝑐̅𝑑?̅?       ∀ 𝑐̅ ∈ 𝐶̅, ?̅? ∈ ?̅?, 𝑠 ∈ 𝑆, 𝑐 ∈ 𝐶, 𝑑 ∈ 𝐷, 𝑔 ∈ 𝐺, 𝑛′ ∈ 𝑁𝑉, 𝑟′ ∈ 𝑅𝑂, 𝑦 ∈ 𝑌, 𝑔′ ∈ 𝐺′, 𝑡 ∈ 𝑇  4.20  4.3.6 Building new facilities and lifetime consideration • Plants or terminals with the same technology, state of product, and capacity could not co-exist in each region, as they were considered binaries.  70  • A central plant existed in time step (t), since it was established in that time step, or it was existing from the previous time steps and was working within its lifetime.    𝑌𝐶𝑐𝑦𝑑𝑔𝑡 = 𝑌𝑃𝐶𝑐𝑦𝑑𝑔𝑡 + 𝑌𝐶𝑐𝑦𝑑𝑔(𝑡−1) 4.21 ∀ 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇: 2 ≤ 𝑡 ≤ 𝐿𝑇, 𝐿𝑇 = 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝐶/𝑡𝑖𝑚𝑒𝑠𝑡𝑒𝑝  𝑌𝐶𝑐𝑦𝑑𝑔𝑡 = 𝑌𝑃𝐶𝑐𝑦𝑑𝑔𝑡 + 𝑌𝐶𝑐𝑦𝑑𝑔(𝑡−1) − 𝑌𝑃𝐶𝑐𝑦𝑑𝑔(𝑡−𝐿𝑇)  ∀ 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇 ∶ 𝑡 ≥ 𝐿𝑇 + 1, 𝐿𝑇 = 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝐶/𝑡𝑖𝑚𝑒𝑠𝑡𝑒𝑝   • All plants were new in the first time step:  𝑌𝑃𝐶𝑐𝑦𝑑𝑔𝑡 = 𝑌𝐶𝑐𝑦𝑑𝑔𝑡                               ∀ 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇: 𝑡 = 1 4.22  The notation was adjusted to develop the time evolution constraints for warehouses, trucks, on-site production plants, and refueling stations.  4.3.7 Capacity expansion  In case the demand exceeds the maximum primary capacity, a capacity expansion could be considered as an alternative to building a new facility. Three stages of capacity expansion (j) were considered for the plants and storage facilities. For SMR plants, the capacity expansion stages were fixed at 10%, 20%, and 30%, whereas they were 10%, 25%, and 50% for electrolyzers and storage facilities. The capacity expansion option was not considered for the by-product hydrogen purification plant. The following capacity expansion constraints were considered in the model: • Each stage of capacity expansion could take place once during the lifetime of the facility. In case of a central production plant:  𝑌′𝑃𝐶𝑗𝑐𝑦𝑑𝑔𝑡 ≥ 𝑌′𝐶𝑗𝑐𝑦𝑑𝑔𝑡 − 𝑌′𝐶𝑗𝑐𝑦𝑑𝑔(𝑡−1)                                                                                    4.23 𝑌′𝑃𝐶𝑗𝑐𝑦𝑑𝑔𝑡 ≤ 𝑌′𝐶𝑗𝑐𝑦𝑑𝑔𝑡 𝑌′𝑃𝐶𝑗𝑐𝑦𝑑𝑔𝑡 ≤ 1 − 𝑌′𝐶𝑗𝑐𝑦𝑑𝑔(𝑡−1) ∀ 𝑗 ∈ 𝐽, 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇: 𝑡 ≥ 2 71   • The capacity expansion could not happen in the same time step as the facility establishment. In case of a central production plant:  𝑌𝑃𝐶𝑐𝑦𝑑𝑔𝑡 + 𝑌′𝑃𝐶𝑗𝑐𝑦𝑑𝑔𝑡 ≤ 1                                                                                                   4.24 ∀ 𝑗 ∈ 𝐽, 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇  • Each stage of capacity expansion (𝑌′𝐶𝑗𝑐𝑦𝑑𝑔𝑡) took place when the rate of product flow (𝑃𝐶𝑦𝑑𝑔𝑡) exceeded the maximum primary capacity (𝑃𝑐𝑎𝑝_𝑚𝑎𝑥_𝑝𝑟𝑖𝑛𝑐), as follows:  (𝑃𝑐𝑎𝑝_max _𝑝𝑟𝑖𝑛𝑐 − 𝜀)𝑌′𝐶𝑗𝑐𝑦𝑑𝑔𝑡|𝑗=1 + ∑ ((1 + 𝑅𝑒𝑣𝑗𝑦)𝑃𝑐𝑎𝑝_max _𝑝𝑟𝑖𝑛𝑐 − 𝜀)𝑌′𝐶(𝑗+1)𝑐𝑦𝑑𝑔𝑡3𝑗=1≤ 𝑃𝐶𝑦𝑑𝑔𝑡 ≤ 𝑃𝑐𝑎𝑝_𝑚𝑎𝑥_𝑝𝑟𝑖𝑛𝑐 (𝑌′𝐶_𝑂𝑐𝑦𝑑𝑔𝑡 + ∑(1 + 𝑅𝑒𝑣𝑗𝑦)𝑌′𝐶𝑗𝑐𝑦𝑑𝑔𝑡3𝑗=1) ∀ 𝑗 ∈ 𝐽, 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇  4.25 • The maximum capacity of a facility in each time step was determined based on the maximum primary capacity and the total stages of capacity expansion that happened before that time step. In case of a central production plant:  𝑃𝑐𝑎𝑝_𝑚𝑎𝑥𝑐𝑦𝑑𝑔𝑡 = 𝑃𝑐𝑎𝑝_𝑚𝑎𝑥_𝑝𝑟𝑖𝑛𝑐 (1 + (∑ 𝑅𝑒𝑣𝑗𝑦𝑌′𝐶𝑗𝑐𝑦𝑑𝑔𝑡3𝑗=1) + 𝑅𝑒𝑣𝑗𝑦|𝑗=3𝑌′𝐶𝑗𝑐𝑦𝑑𝑔𝑡|𝑗=4) ∀ 𝑗 ∈ 𝐽, 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇  4.26 • A maximum of one stage of capacity expansion was possible for each facility in each time step. In case of a central production plant:  𝑌′𝐶_𝑂𝑐𝑦𝑑𝑔𝑡 + ∑ 𝑌′𝐶𝑗𝑐𝑦𝑑𝑔𝑡3𝑗=1 ≤ 1                                                                                             4.27 ∀ 𝑗 ∈ 𝐽, 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑔 ∈ 𝐺, 𝑡 ∈ 𝑇 72   The capacity expansion constraints for the central storage facilities follow the same logics as plants.   4.3.8 Non-negativity constraints Non-negativity constraints must be represented to ensure that the variables are continuous, positive integers and binaries.  4.4 H2SCOT objective function The objective function is to minimize the discounted total cost of infrastructure (𝐷𝐶𝐼𝑁𝐹), which includes the discounted cost of technology (𝐷𝐶𝑇) and the discounted cost of environmental policies (𝐷𝐶𝑃𝑜𝑙𝑖𝑐𝑦). It was assumed that the supply chain components were built in the first year with the maximum capacity to fulfill the demand in the last year of each time step.   𝐷𝐶𝐼𝑁𝐹 = 𝐷𝐶𝑇 + 𝐷𝐶𝑃𝑜𝑙𝑖𝑐𝑦 4.28  4.4.1 Discounted cost of technology (DCT)   𝐷𝐶𝑇 =  𝐷𝐶𝑇𝑒𝑠𝑡 + 𝐷𝐶𝑇𝑟𝑒𝑣 + 𝐷𝐶𝑇𝑜𝑝𝑟 4.29  𝐷𝐶𝑇𝑒𝑠𝑡 is the capital cost of establishing an HFSC, which is composed of direct depreciable capital cost, indirect depreciable capital cost, and non-depreciable capital cost. The learning rate of the technology was calculated based on the demand growth in each time step and was applied to the unit capital cost of the facilities. For instance, the learning rate of the central plants is calculated as follows:  𝐿𝑅_𝐶𝑡 = (𝐷𝑇𝑡=1𝐷𝑇𝑡)𝛼_𝐿𝑅_𝐶 4.30  73   𝐷𝐶𝑇𝑒𝑠𝑡 is calculated as follows:  ∑1(1 + 𝑟)𝑡𝑖𝑚𝑒𝑠𝑡𝑒𝑝(𝑡−1)[ ∑ [𝐸𝑆𝑇_𝐶𝑐𝑦𝑑𝑔𝑡 + 0.15𝐿𝑅_𝐶𝑡𝐷𝐷𝐶_𝐶𝑐𝑦𝑌𝑃𝐸𝑐𝑦𝑑𝑔?̅?]𝑐𝑦𝑑𝑔𝑡+ ∑ 𝐸𝑆𝑇_𝑆𝑐̅𝑑?̅?𝑡𝑐̅𝑑?̅?+ ∑ 𝐸𝑆𝑇_𝑉𝑠𝑑𝑔′𝑡𝑠𝑑𝑔′+ ∑ 𝐸𝑆𝑇_𝑁𝑉𝑠𝑑𝑛′𝑡𝑠𝑑𝑛′+ ∑ 𝐸𝑆𝑇_𝑅𝑂𝑠𝑑𝑟′𝑡𝑠𝑑𝑟′+ ∑ 𝐷𝐷𝐶_𝐺𝑎 (∑ 𝑁𝑁𝐺_𝑃𝑆𝑔?̅?𝑎𝑡𝑔?̅?+ ∑ 𝑁𝑁𝐺_𝑉?̅?𝑔′𝑎𝑠𝑡?̅?𝑔′𝑠𝑎+ ∑ 𝑁𝑁𝐺_𝑁𝑉?̅?𝑛′𝑎𝑠𝑡?̅?𝑛′𝑠+ ∑ 𝑁𝑁𝐺_𝑅𝑂?̅?𝑟′𝑎𝑠𝑡?̅?𝑟′𝑠)+ 𝐷𝐷𝐶_𝐿 (∑ 𝑁𝑁𝐿_𝑃𝑆𝑔?̅?𝑡𝑔?̅?+ ∑ 𝑁𝑁𝐿_𝑉?̅?𝑔′𝑡?̅?𝑔′+ ∑ 𝑁𝑁𝐿_𝑁𝑉?̅?𝑛′𝑡?̅?𝑛′+ ∑ 𝑁𝑁𝐿_𝑅𝑂?̅?𝑟′?̅?𝑟′)] 4.31 ∀ 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑐̅ ∈ 𝐶̅, ?̅? ∈ ?̅?, 𝑔′ ∈ 𝐺′, 𝑔 ∈ 𝐺, 𝑛′ ∈ 𝑁𝑉, 𝑟′ ∈ 𝑅𝑂, 𝑔′ ∈ 𝐺′, 𝑠 ∈ 𝑆, 𝑎 ∈ 𝐴,  𝑡 ∈ 𝑇  In which: 𝐸𝑆𝑇_𝐶𝑐𝑦𝑑𝑔𝑡 = (𝐿𝑅_𝐶𝑡(𝐷𝐷𝐶_𝐶𝑐𝑦 + 𝐼𝐷𝐶_𝐶𝑐𝑦) + 𝑁𝐶_𝐶𝑐𝑦)𝑌𝑃𝐶𝑐𝑦𝑑𝑔𝑡 𝐸𝑆𝑇_𝑆𝑐̅𝑑?̅?𝑡 = (𝐿𝑅_𝑆𝑡(𝐷𝐷𝐶_𝑆𝑐̅𝑑 + 𝐼𝐷𝐶_𝑆𝑐̅𝑑) + 𝑁𝐶_𝑆𝑐̅𝑑)𝑌𝑃𝑆𝑐̅𝑑?̅?𝑡 𝐸𝑆𝑇_𝑉𝑠𝑑𝑔′𝑡 = 𝐿𝑅_𝑂𝑡𝐷𝐶_𝑂𝑠𝑌𝑃𝑂_𝑉𝑠𝑑𝑔′𝑡 + 𝐿𝑅_𝐷𝑡𝐷𝐶_𝐷𝑠𝑑𝑌𝑃𝐷_𝑉𝑠𝑑𝑔′𝑡 𝐸𝑆𝑇_𝑁𝑉𝑠𝑑𝑛′𝑡 and 𝐸𝑆𝑇_𝑅𝑂𝑠𝑑𝑟′𝑡 were developed with proper notation for other municipalities. 𝑌𝑃𝐸𝑐𝑦𝑑𝑔?̅? in Eq. 4.31 considers a 10-year replacement cost of the stacks for central electrolyzers.  74  𝐷𝐶𝑇𝑟𝑒𝑣 in Eq. 4.29 is the capital cost of capacity expansion for the central plants and warehouses, calculated as follows:  ∑1(1 + 𝑟)𝑡𝑖𝑚𝑒𝑠𝑡𝑒𝑝(𝑡−1)[𝐿𝑅_𝐶𝑡 ∑ 𝑃𝑅𝑐𝑎𝑝𝑗𝑦𝐷𝐷𝐶𝐶𝑐𝑦 (𝑌′𝑃𝐶𝑗𝑐𝑦𝑑𝑔𝑡𝑗𝑐𝑦𝑑𝑔𝑡− ∑ 𝑌′𝑃𝐶𝑗𝑐𝑦𝑑𝑔𝑝𝑡−1 (𝑡>1)𝑝=1∑ 𝑌′𝑃𝐶𝑟𝑐𝑦𝑑𝑔𝑡3𝑟=1)+ 𝐿𝑅_𝑆𝑡 ∑ 𝑆𝑅_𝑐𝑎𝑝𝑗𝐷𝐷𝐶_𝑆𝑐̅𝑑 (𝑌′𝑃𝑆𝑗𝑐̅𝑑?̅?𝑡𝑗𝑐̅𝑑?̅?− ∑ 𝑌′𝑃𝑆𝑗𝑐̅𝑑?̅?𝑝𝑡−1 (𝑡>1)𝑝=1∑ 𝑌′𝑃𝑆𝑟𝑐̅𝑑?̅?𝑡3𝑟=1)] ∀ 𝑗 ∈ 𝐽, 𝑟 ∈ 𝐽, 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑐̅ ∈ 𝐶̅, ?̅? ∈ ?̅?, 𝑡 ∈ 𝑇, 𝑝 ∈ 𝑇 4.32  𝑃𝑅_𝑐𝑎𝑝𝑗𝑦 is the percentage of direct depreciable capital cost of plants, for each stage of capacity expansion. 𝑌′𝑃𝐶𝑗𝑐𝑦𝑑𝑔𝑡 − ∑ 𝑌′𝑃𝐶𝑗𝑐𝑦𝑑𝑔𝑝𝑡−1 (𝑡>1)𝑝=1 ∑ 𝑌′𝑃𝐶𝑟𝑐𝑦𝑑𝑔𝑡3𝑟=1  demonstrates the stage of capacity expansion with respect to the history of capacity expansion for the specific plant. The same logic applies to the storage facilities. 𝐷𝐶𝑇𝑜𝑝𝑟 in Eq. 4.29 is the sum of operating and yearly replacement costs. The operating cost of each facility consists of fixed costs and variable costs. The fixed cost is calculated based on a fixed percentage of the depreciable capital cost, while the variable costs depend on the hydrogen flow.          75  𝐷𝐶𝑇𝑜𝑝𝑟 is calculated as follows:  ∑1(1 + 𝑟)(𝑧−1)[∑ (∑ 𝑂𝑃𝑅_𝐶𝑐𝑦𝑑𝑔𝑡𝑐𝑦𝑔+ ∑ 𝑂𝑃𝑅_𝑆𝑐̅𝑑?̅?𝑡𝑐̅+ ∑ 𝑂𝑃𝑅_𝑉𝑠𝑑𝑔′𝑡𝑠𝑔′𝑑𝑡𝑡′+ ∑ 𝑂𝑃𝑅_𝑁𝑉𝑠𝑑𝑛′𝑡𝑠𝑛′+ ∑ 𝑂𝑃𝑅_𝑅𝑂𝑠𝑑𝑟′𝑡𝑠𝑟′)+ 𝑃𝑒𝑟_𝑡𝑧 ∑ (∑ 𝑂𝑃𝑅𝐺_𝑃𝑆𝑔?̅?𝑎𝑡𝑔𝑎+ ∑ 𝑂𝑃𝑅𝐺_𝑉?̅?𝑔′𝑎𝑠𝑡𝑔′𝑎𝑠?̅?+ ∑ 𝑂𝑃𝑅𝐺_𝑁𝑉?̅?𝑛′𝑎𝑠𝑡𝑛′𝑎+ ∑ 𝑂𝑃𝑅𝐺_𝑅𝑂?̅?𝑟′𝑎𝑠𝑡𝑟′𝑠+ ∑ 𝑂𝑃𝑅𝐿_𝑃𝑆𝑔?̅?𝑡𝑔+ ∑ 𝑂𝑃𝑅𝐿_𝑉?̅?𝑔′𝑡 + ∑ 𝑂𝑃𝑅𝐿_𝑁𝑉?̅?𝑛′𝑡 +𝑛′𝑔′∑ 𝑂𝑃𝑅𝐿_𝑅𝑂?̅?𝑟′𝑡𝑟′)]        4.33 𝑧 = 𝑡𝑖𝑚𝑒𝑠𝑡𝑒𝑝(𝑡 − 1) + 𝑡′ ∀ 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑐̅ ∈ 𝐶̅, ?̅? ∈ ?̅?, 𝑔 ∈ 𝐺, 𝑔′ ∈ 𝐺′, 𝑛′ ∈ 𝑁𝑉, 𝑟′ ∈ 𝑅𝑂, 𝑠 ∈ 𝑆, 𝑡 ∈ 𝑇,  𝑡′ = 1 … 𝑡𝑖𝑚𝑒𝑠𝑡𝑒𝑝    In which: 𝑂𝑃𝑅_𝐶𝑐𝑦𝑑𝑔𝑡 = 𝑃𝑒𝑟_𝑡𝑧𝑃𝐶𝑐𝑦𝑑𝑔𝑡𝑂𝑃_𝐶𝑐𝑦 + (𝐹_𝐶𝑐𝑦 + 𝜔𝐿𝑅_𝐶𝑡𝐷𝐶_𝐶𝑐𝑦)𝑌𝐶𝑐𝑦𝑑𝑔𝑡 𝑂𝑃𝑅_𝑆𝑐̅𝑑?̅?𝑡 = 𝑃𝑒𝑟_𝑡𝑧𝑇𝑆𝑐?̅??̅?𝑡𝑂𝑃_𝑆𝑐̅𝑑 + (𝐹_𝑆𝑐̅𝑑 + 𝜔𝐿𝑅_𝑆𝑡𝐷𝐶_𝑆𝑐̅𝑑)𝑌𝑆𝑐̅𝑑?̅?𝑡 𝑂𝑃𝑅_𝑉𝑠𝑑𝑔′𝑡 = 𝑃𝑒𝑟_𝑡𝑧𝑂𝑃_𝑂𝑠𝑃𝑂_𝑉𝑠𝑑𝑔′𝑡 + (𝐹_𝑂𝑠 + 𝜔𝐿𝑅_𝑂𝑡𝐷𝐶_𝑂𝑠)𝑌𝑂_𝑉𝑠𝑑𝑔′𝑡+ 𝑃𝑒𝑟_𝑡𝑧𝑂𝑃_𝐷𝑠𝑑(𝐷𝐼_𝑉𝑠𝑑𝑔′𝑡 + 𝑆𝑇𝑅_𝑉𝑠𝑑𝑔′𝑡)+ (𝐹_𝐷𝑠𝑑 + 𝜔𝐿𝑅_𝐷𝑡𝐷𝐶_𝐷𝑠𝑑)𝑌𝐷_𝑉𝑠𝑑𝑔′𝑡 𝑂𝑃𝑅𝐺_𝑃𝑆𝑔?̅?𝑎𝑡 = 𝑁𝐺_𝑃𝑆𝑔?̅?𝑎𝑡 ((𝐿𝑇𝑅_𝑃𝑆𝑑?̅?𝑔|𝑑=2 + 𝐹𝑇𝑅_𝑃𝑆𝑑?̅?𝑔|𝑑=2) + 𝐹_𝐺𝑎) 𝑂𝑃𝑅𝐿_𝑃𝑆𝑔?̅?𝑡 = 𝑁𝐿_𝑃𝑆𝑔?̅?𝑡 ((𝐿𝑇𝑅_𝑃𝑆𝑑?̅?𝑔|𝑑=3 + 𝐹𝑇𝑅_𝑃𝑆𝑑?̅?𝑔|𝑑=3) + 𝐹_𝐿) 76  𝑃𝑒𝑟_𝑡𝑧 accounts for the ratio of annual hydrogen flow rate to the maximum flow rate in each time step. The yearly replacement cost of the plants, warehouses, and fueling stations (ω) accounts for 0.5% of the total depreciable capital cost of the corresponding facility. The cost of fuel and driver wages for hydrogen transport was calculated based on the transport time, the hourly wage of the driver, the fuel cost, and the fuel economy of the truck.  For transportation network:  𝐿𝑇𝑅_𝑃𝑆𝑑?̅?𝑔 = 𝛼 (2𝐿𝐻_𝑃𝑆𝑔?̅?𝑉𝐻+ 𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑡𝑖𝑚𝑒𝑑)                                                                          4.34 𝐹𝑇𝑅_𝑃𝑆𝑑?̅?𝑔 = 2𝛾𝛽𝐿𝐻_𝑃𝑆𝑔?̅?                                                                                                       4.35  For distribution network (to Metro-Vancouver):  𝐿𝑎𝑏𝑜𝑟_𝑇𝑅_𝑉𝑑𝑔′?̅? = 𝛼 (2𝐿𝐻_𝑉?̅?𝑉𝐻+2𝐿𝐺𝑔′𝑉𝐺+ 𝑢𝑛𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑡𝑖𝑚𝑒𝑑)                                                    4.36 𝐹𝑢𝑒𝑙_𝑇𝑅_𝑉𝑑𝑔′?̅? = 2𝛾𝛽(𝐿𝐻_𝑉?̅? + 𝐿𝐺𝑔)                                                                                      4.37  The notation was adjusted to calculate the cost of fuel and driver wages for the trucks distributing hydrogen to other municipalities.  4.4.2 Discounted cost of environmental policies (DCPolicy) The discounted cost of environmental policies consists of the cost and revenue of currently deployed policies in B.C. (carbon tax and LCFS) and the discounted revenue of the complementary subsidy-based policies, as follows:  𝐷𝐶𝑃𝑜𝑙𝑖𝑐𝑦 =  𝐷𝐶𝐶𝑇 − 𝐷𝑅𝐿𝐶𝐹𝑆 − 𝐷𝑅𝑠𝑢𝑏 4.38  77  4.4.2.1 Discounted cost of carbon tax (DCCT) 𝐷𝐶𝐶𝑇 was calculated on a yearly basis and discounted over the entire time frame. The emission from the unit hydrogen flow in each component was multiplied by the hydrogen flow rate, and the result is multiplied by the emission cost per tonne of CO2 dispersed (i.e., the carbon tax):    ∑365𝑃𝑒𝑟_𝑡𝑧𝐸_𝐶𝑜𝑠𝑡(1 + 𝑟)(𝑧−1)[ ∑ 𝐺𝐻𝐺_𝐶𝑦𝑃𝐶𝑐𝑦𝑑𝑔𝑡𝑐𝑦𝑑𝑔+ ∑   𝐺𝐻𝐺_𝑆𝑑𝑇𝑆𝑐̅𝑑?̅?𝑡 +𝑐̅𝑑?̅?𝑡𝑡′+ 𝐺𝐻𝐺_𝑇𝑅 (∑ ((2𝐿𝐻_𝑃𝑆𝑔?̅?)𝑁𝑇𝑅𝐺_𝑃𝑆𝑔?̅?𝑎𝑡)𝑔?̅?𝑎+ ∑ ((2𝐿𝐻_𝑉?̅? + 2𝐿𝐺𝑔′)𝑁𝐺_𝑉?̅?𝑔′𝑎𝑡)?̅?𝑔′𝑎+ ∑ ((2𝐿𝐻_𝑁𝑉?̅?𝑛′)𝑁𝐺_𝑁𝑉?̅?𝑛′𝑎𝑠𝑡)?̅?𝑛′𝑎𝑠+ ∑ ((2𝐿𝐻_𝑅𝑂?̅?𝑟′)𝑁𝐺_𝑅𝑂?̅?𝑟′𝑎𝑠𝑡)?̅?𝑟′𝑎𝑠+ ∑ ((2𝐿𝐻_𝑃𝑆𝑔?̅?)𝑁𝐿_𝑃𝑆𝑔?̅?𝑡)𝑔?̅?+ ∑((2𝐿𝐻_𝑉?̅? + 2𝐿𝐺𝑔′)𝑁𝐿_𝑉?̅?𝑔′𝑡)?̅?𝑔′+ ∑ ((2𝐿𝐻_𝑁𝑉?̅?𝑛′)𝑁𝐿_𝑁𝑉?̅?𝑛′𝑡)?̅?𝑟′+ ∑((2𝐿𝐻_𝑅𝑂?̅?𝑟′)𝑁𝐿_𝑅𝑂?̅?𝑟′𝑡)?̅?𝑟′) + ∑ 𝐸𝑀𝑁_𝑉𝑠𝑑𝑔′𝑡𝑠𝑑𝑔′+ ∑ 𝐸𝑀𝑁_𝑁𝑉𝑠𝑑𝑛′𝑡𝑠𝑑𝑛′+ ∑ 𝐸𝑀𝑁_𝑅𝑂𝑠𝑑𝑟′𝑡𝑠𝑑𝑟′] 4.39 𝑧 = 𝑡𝑖𝑚𝑒𝑠𝑡𝑒𝑝(𝑡 − 1) + 𝑡′ ∀ 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑐̅ ∈ 𝐶̅, ?̅? ∈ ?̅?, 𝑔 ∈ 𝐺, 𝑔′ ∈ 𝐺′, 𝑛′ ∈ 𝑁𝑉, 𝑟′ ∈ 𝑅𝑂, 𝑠 ∈ 𝑆, 𝑡 ∈ 𝑇,  𝑡′ = 1 … 𝑡𝑖𝑚𝑒𝑠𝑡𝑒𝑝 In which: 𝐸𝑀𝑁_𝑉𝑠𝑑𝑔′𝑡 = 𝐺𝐻𝐺_𝐷𝑑(𝐷𝐼_𝑉𝑠𝑑𝑔′𝑡 + 𝑆𝑇𝑅_𝑉𝑠𝑑𝑔′𝑡) + 𝐺𝐻𝐺_𝑂 × 𝑃𝑂_𝑉𝑠𝑑𝑔′𝑡 78  The notations were adjusted to account for the GHG emissions in other municipalities (i.e., 𝐸𝑀𝑁_𝑁𝑉𝑠𝑑𝑛′𝑡 and 𝐸𝑀𝑁_𝑅𝑂𝑠𝑑𝑟′𝑡). The vessels transporting hydrogen from the Port of Vancouver to Victoria were assumed to produce 13% of the GHG emissions of road transportation [171] per tonne-kilometer. This assumption was used for the distance of 47 km between the ports of Tsawwassen, B.C., and Swartz Bay, B.C.                       79  4.4.2.2 Discounted revenue of LCFS (DRLCFS) The LCFS revenue was calculated based on the difference between the carbon intensity of gasoline and hydrogen and their energy efficiency ratio, multiplied by the LCFS credit price per tonnes of CO2 displaced. The carbon intensity of hydrogen was calculated on WTW basis by considering the share of each component on the final fuel-side GHG emissions of the supply chain.   ∑365𝑃𝑒𝑟_𝑡𝑧 × 1.2𝐷𝑇𝑡 × 𝐶𝑟𝑒𝑑𝑖𝑡_𝐿𝐶𝐹𝑆𝑡 × 1𝐸 − 6 × 𝐻2_𝐷(1 + 𝑟)(𝑧−1)𝑡𝑡′[𝐺𝑎𝑠_𝐶𝐼 × 𝐸𝐸𝑅𝑡−1120 × 1.2𝐷𝑇𝑡[ ∑ 𝐺𝐻𝐺_𝐶𝑦𝑃𝐶𝑐𝑦𝑑𝑔𝑡𝑐𝑦𝑑𝑔+ ∑   𝐺𝐻𝐺_𝑆𝑑𝑇𝑆𝑐̅𝑑?̅?𝑡 +𝑐̅𝑑?̅?+ 𝐺𝐻𝐺_𝑇𝑅 (∑ ((2𝐿𝐻_𝑃𝑆𝑔?̅?)𝑁𝑇𝑅𝐺_𝑃𝑆𝑔?̅?𝑎𝑡)𝑔?̅?𝑎+ ∑ ((2𝐿𝐻_𝑉?̅? + 2𝐿𝐺𝑔′)𝑁𝐺_𝑉?̅?𝑔′𝑎𝑠𝑡)?̅?𝑔′𝑎𝑠+ ∑ ((2𝐿𝐻_𝑁𝑉?̅?𝑛′)𝑁𝐺_𝑁𝑉?̅?𝑛′𝑎𝑠𝑡)?̅?𝑛′𝑎𝑠+ ∑ ((2𝐿𝐻_𝑅𝑂?̅?𝑟′)𝑁𝐺_𝑅𝑂?̅?𝑟′𝑎𝑠𝑡)?̅?𝑟′𝑎𝑠+ ∑ ((2𝐿𝐻_𝑃𝑆𝑔?̅?)𝑁𝐿_𝑃𝑆𝑔?̅?𝑡)𝑔?̅?+ ∑((2𝐿𝐻_𝑉?̅? + 2𝐿𝐺𝑔′)𝑁𝐿_𝑉?̅?𝑔′𝑡)?̅?𝑔′+ ∑ ((2𝐿𝐻_𝑁𝑉?̅?𝑛′)𝑁𝐿_𝑁𝑉?̅?𝑛′𝑡)?̅?𝑟′+ ∑((2𝐿𝐻_𝑅𝑂?̅?𝑟′)𝑁𝐿_𝑅𝑂?̅?𝑟′𝑡)?̅?𝑟′) + ∑ 𝐸𝑀𝑁_𝑉𝑠𝑑𝑔′𝑡𝑠𝑑𝑔′+ ∑ 𝐸𝑀𝑁_𝑁𝑉𝑠𝑑𝑛′𝑡𝑠𝑑𝑛′+ ∑ 𝐸𝑀𝑁_𝑅𝑂𝑠𝑑𝑟′𝑡𝑠𝑑𝑟′]] 4.40 𝑧 = 𝑡𝑖𝑚𝑒𝑠𝑡𝑒𝑝(𝑡 − 1) + 𝑡′ ∀ 𝑐 ∈ 𝐶, 𝑦 ∈ 𝑌, 𝑑 ∈ 𝐷, 𝑐̅ ∈ 𝐶̅, ?̅? ∈ ?̅?, 𝑔 ∈ 𝐺, 𝑔′ ∈ 𝐺′, 𝑛′ ∈ 𝑁𝑉, 𝑟′ ∈ 𝑅𝑂, 𝑠 ∈ 𝑆, 𝑡 ∈ 𝑇,  𝑡′ = 1 … 𝑡𝑖𝑚𝑒𝑠𝑡𝑒𝑝 80  4.4.2.3 Discounted cost of complementary policies The revenues from the incentive-based policies were added to the objective function of the model, as follows: • The PTC is calculated by multiplying the tax credits in each time step by the production rate of the eligible facilities and added as a revenue term to the policy term of the objective function:  𝐷𝐶𝑃𝑜𝑙𝑖𝑐𝑦 =   𝐷𝐶𝐶𝑇 − 𝐷𝑅𝐿𝐶𝐹𝑆 − 𝐷𝑅𝑃𝑇𝐶 4.41  The PTC is deducted from the total annual tax in the post optimization cash flow analysis (section 4.4) • The capital subsidy was included in the discounted cost of technology ( 𝐷𝐶𝑇) of the objective function as follows:  𝐷𝐶𝑇 =  (𝐷𝐶𝑇𝑒𝑠𝑡)𝑁 + ((1 − 𝐺𝑟𝑎𝑛𝑡) × 𝐷𝐶𝑇𝑒𝑠𝑡)𝐸 + 𝑁𝐶𝑇𝑒𝑠𝑡 + 𝐷𝐶𝑇𝑟𝑒𝑣 + 𝐷𝐶𝑇𝑜𝑝𝑟 4.42  In which (𝐷𝐶𝑇𝑒𝑠𝑡)𝑁 is the depreciable capital cost of the non-eligible facilities, (𝐷𝐶𝑇𝑒𝑠𝑡)𝐸 is the depreciable capital cost of eligible facilities, 𝑁𝐶𝑇𝑒𝑠𝑡 is the non-depreciable capital cost, 𝐷𝐶𝑇𝑟𝑒𝑣 is the cost of capacity expansion, and 𝐷𝐶𝑇𝑜𝑝𝑟 is the operating cost of the facilities. In calculating the accelerated depreciation allowances, the amount of the grant was subtracted from the property's capital cost.  • The utility subsidy (𝑈𝐼𝑁𝐶) is included in the operating cost of objective function of H2SCOT as follows:  𝐷𝐶𝑇 =  𝐷𝐶𝑇𝑒𝑠𝑡 + 𝑁𝐶𝑇𝑒𝑠𝑡 + 𝐷𝐶𝑇𝑟𝑒𝑣 + (𝐹𝐶𝑜𝑝𝑟 + 𝑉𝐶𝑜𝑝𝑟)𝑁+ (𝐹𝐶𝑜𝑝𝑟 + (1 − 𝑈𝐼𝑁𝐶)𝑉𝐶𝑜𝑝𝑟)𝐸 4.43  In which, (𝐹𝐶𝑜𝑝𝑟 + 𝑉𝐶𝑜𝑝𝑟)𝑁and (𝐹𝐶𝑜𝑝𝑟 + (1 − 𝑈𝐼𝑁𝐶)𝑉𝐶𝑜𝑝𝑟)𝐸 are the fixed (𝐹𝐶𝑜𝑝𝑟) and variable (𝑉𝐶𝑜𝑝𝑟) operating cost of non-eligible and eligible facilities for this policy, respectively.   81  4.5 Post optimization cash-flow analysis Based on the optimal infrastructure in each demand scenario, the hydrogen price trends were examined over time to meet a target internal rate of return (IRR) of the investment (10%). To this end, the annual after-tax post depreciation cash flow was calculated by subtracting the annual pre-depreciation income from the total taxes and the capital cost of the infrastructure.   𝐴𝑇𝑃𝐷_𝐶𝐹𝑖 = 𝑃𝐷_𝐼𝑛𝑐𝑖 − 𝑇𝑡𝑖 − 𝐶𝑎𝑝𝑖                      ∀𝑖: 𝑖 = 1 … 𝑁                                             4.44  The annual capital costs include the yearly direct and indirect depreciable capital costs, non-depreciable capital costs and yearly replacement costs. The first three terms could be nonzero only at the first year of each time step, when the establishment of infrastructure is planned. The working capital is not considered in the calculation.  𝐶𝑎𝑝𝑖 = 𝐷𝐷𝑒𝑝_𝐶𝑎𝑝𝑖 + 𝐼𝐷𝑒𝑝_𝐶𝑎𝑝𝑖 + 𝑁𝐷𝑒𝑝_𝐶𝑎𝑝𝑖 + 𝑌_𝑅𝑒𝑝𝑖              ∀𝑖: 𝑖 = 1 … 𝑁                   4.45  The annual operational cost of the supply chain is the sum of fixed and variable operating costs. It was assumed that the salvage value of the facilities and the decommissioning cost cancel each other out.   𝑂𝑝𝑟𝑖 = 𝐹_𝑂𝑝𝑟𝑖 + 𝑉_𝑂𝑝𝑟𝑖                                        ∀𝑖: 𝑖 = 1 … 𝑁                                              4.46  The annual pre-depreciation income was calculated by subtracting hydrogen revenue from the operating cost.   𝑃𝐷_𝐼𝑛𝑐𝑖 = 𝑅𝑒𝑣_𝐻2𝑖 − 𝑂𝑝𝑟𝑖                                        ∀𝑖: 𝑖 = 1 … 𝑁                                                    4.47  The total annual tax was calculated by multiplying the tax rate by the taxable income and subtracting the result from the tax credit.   𝑇𝑡𝑖 = 𝑇𝑟 × (𝑃𝐷_𝐼𝑛𝑐𝑖 − 𝐷𝑒𝑝_𝑐ℎ𝑖) − 𝑇𝑐                  ∀𝑖: 𝑖 = 1 … 𝑁                                               4.48  To assess the taxable income, the depreciation of the supply chain facilities has to be determined. To this end, the capital cost allowance (CCA) deduction was calculated using the declining balance 82  method, based on 30% CCA rate for production plants, storage facilities and dispensers and 40% CCA rate for tube trailer and tanker trucks.   4.6 Potential contribution of FCEVs to GHG emissions reduction The annual GHG emissions reduction was calculated by subtracting the WTW GHG emissions of the gasoline cars that were replaced by the FCEVs from the total GHG emissions of the HFSC in each demand scenario:  𝐺𝐻𝐺𝑅𝑛𝑖 = 𝐺𝐻𝐺_𝐸𝑖|𝐺𝑎𝑠𝑜𝑙𝑖𝑛𝑒_𝐶𝑎𝑟𝑆𝑡𝑜𝑐𝑘𝑖|𝐹𝐶𝐸_𝐶𝑎𝑟+ 𝐺𝐻𝐺_𝐸𝑖|𝐺𝑎𝑠𝑜𝑙𝑖𝑛𝑒_𝑃𝑇𝑟𝑢𝑐𝑘𝑆𝑡𝑜𝑐𝑘𝑖|𝐹𝐶𝐸_𝑃𝑇𝑟𝑢𝑐𝑘 − 𝐺𝐻𝐺𝑖|𝐻𝐹𝑆𝐶 ∀𝑖: 𝑖 = 1 … 𝑁 4.49  In which 𝐺𝐻𝐺𝑖|𝐻𝐹𝑆𝐶  represents the annual GHG emissions from the HFSC infrastructure (calculated by the optimization model), and 𝐺𝐻𝐺_𝐸 is the WTW annual GHG emissions per vehicle, which was calculated for gasoline passenger cars and trucks up to 2050, as follows:   𝐺𝐻𝐺_𝐸𝑖|𝑉𝑒ℎ𝑖𝑐𝑙𝑒_𝑡𝑦𝑝𝑒 = 𝐹_𝐸𝑚𝑚𝑅𝑎𝑡𝑒̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ × 𝐹_𝐸𝑓𝑓𝑖̅̅ ̅̅ ̅̅ ̅̅ ̅ × 𝑉𝑈𝑠𝑒_𝐼𝑛𝑡̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ 𝑖 ∀𝑖: 𝑖 = 1 … 𝑁 4.50  𝐹_𝐸𝑚𝑚𝑅𝑎𝑡𝑒̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅  (g/MJ) accounts for the gasoline average WTW GHG emissions rate, which was set at 79.33 g/MJ in compliance with the low-carbon fuel standard for 2020. 𝐹_𝐸𝑓𝑓𝑖̅̅ ̅̅ ̅̅ ̅̅ ̅ (MJ/km) was calculated from Eq. 3.57 and 𝑉𝑈𝑠𝑒_𝐼𝑛𝑡̅̅ ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ 𝑖 (km/year) was derived from Figure 3.9 . 83  Chapter 5: Hydrogen Fuel Supply Chain Development in British Columbia: Light Duty Passenger Vehicles This chapter focuses on the results of the optimization model for three FCEV penetration scenarios in B.C. from 2020 to 2050. The results are presented in terms of the configuration of the supply chain in each time step, the average hydrogen price and the WTW GHG emissions from this supply chain. In each demand scenario, the environmental and economic trade-offs were measured for the case where no environmental policy is included. The results were then compared with the cases where various realizations of the current provincial policies (carbon tax and LCFS) are integrated to the model. The most suitable set of current policies (in terms of economic and environmental performance) was then selected to serve as a base case for the adoption of further environmental subsidies and regulations. The aim is to identify potential financial and regulatory tools to increase low-carbon hydrogen production in the cost optimal HFSC. The effectiveness of potential policies was measured with respect to the reduction in hydrogen price and GHG emissions per unit of subsidy alongside the contribution of low-carbon hydrogen in the resultant supply chain.  The validation of supply chain optimization models at the strategic decision phase is challenging. The HFSC optimization models are predictive with limited available historical data. Thus, the validation of these models is a longitudinal activity. Furthermore, the HFSC optimization models are regionally specific. The assumptions on the network topology, demand characteristics (temporal and spatial patterns), the resource availability, and the regulatory environment makes the optimization results unique and incomparable to the models which have been developed for a different region, even with a similar modeling structure.    5.1 HFSC configuration with no environmental policy inclusion The HFSC cost minimization was performed without considering an emission policy term in the objective function. Table 5.1 shows the development of on-site and central production plants and central storage facilities for the three demand scenarios over time. On-site electrolysis was responsible for 100% of the hydrogen production in the first time step for all demand scenarios. The addition of central production facilities, especially SMR plants, decreases the contribution of on-site production in all demand scenarios. The hydrogen purification plant in the district of North 84  Vancouver was selected by the model as the least capital-intensive investment in the second time step for all demand scenarios. As the hydrogen demand grew over time, SMR facilities were added close to demand regions, to reduce the transportation cost. More expensive production technologies (central electrolysis and CCS) were not selected for any demand scenario. The storage facilities were built in the same region as the production sites (attached to the production plant) to avoid the transportation cost. A capacity expansion was applied to the SMR plant and the storage unit in the optimistic scenario. This option fulfilled the growing demand by increasing the rated capacity of the existing facilities and imposed less capital expenditure than building new ones. The hydrogen state was only gaseous for pessimistic and moderate scenarios. Due to the deployment of high-capacity composite gas vessels, liquid hydrogen played a role in the optimistic scenario as higher regional demand justified the liquefaction cost.                    85  Table 5.1. On-site and central production plants and storage facilities for three demand scenarios over time (no policy inclusion). Demand scenario HFSC component1 Supply region Time step 1 2 3 4 5 6 Pessimistic Central production 14  H2P, G, 10t H2P, G, 10t H2P, G, 10t H2P, G, 10t H2P, G, 10t 8     SMR, G, 50t SMR, G, 50t Central storage 14  G, 10t G, 10t G, 10t G, 10t G, 10t 8     G, 50t G, 50t On-site electrolysis2 V 62.5%  32.4% 6.7% 26.3% 8% 0.5% NV 37.5% 19.6% 15.1% 28.7% 9.7% 0 % R 0% 2.7% 3% 3.2% 1.4% 1% Moderate Central production  14  H2P, G, 10t H2P, G, 10t H2P, G, 10t H2P, G, 10t H2P, G, 10t 8    SMR, G, 50t  SMR, G, 50t  SMR, G, 50t  8     SMR, G, 100t SMR, G, 100t Central storage 14  G, 10t G, 10t G, 10t G, 10t G, 10t 8    G, 50t  G, 50t  G, 50t  8     G, 100t G, 100t  On-site electrolysis V 59% 11.4% 32.7% 11.7% 0.1% 2.9% NV 38.7% 14.2% 29.4% 9.9% 0% 1.4% R 2.3% 2.5% 2.8% 0.3% 0.2% 0.3% Optimistic Central production  14  H2P, G, 10t H2P, G, 10t H2P, G, 10t H2P, G, 10t H2P, G, 10t 8      SMR, Gas, 50t  12   SMR, G, 100t SMR, G, 100t SMR, G, 100t SMR, G, 100t 12    SMR, L, 50t SMR, L, 50t SMR, L, 50t (10%) 12     SMR, L, 100t SMR, L, 100t (20%) 3    SMR, L, 50t SMR, L, 50t SMR, L, 50t Central storage 14  G, 10t G, 10t G, 10t G, 10t G, 10t 8      G, 10t 12   G, 50t G, 50t (25%) G, 50t (50%) G, 50t (50%) 12    L, 50t L, 50t L, 50t 12     L, 100t L, 100t (25%) 3    L, 10t L, 10t (25%) L, 10t (25%) On-site electrolysis V 57.6% 8.3% 2.6% 2.9% 3.8% 0.0% NV 39.2% 21.0% 2.6% 6.4% 0.2% 0.1% R 3.2% 2.7% 0.5% 0.5% 0.4% 0.0% 1 H2P: hydrogen purification plant, SMR: steam methane reforming, C-Elec: central electrolyzer, hydrogen status (G: gas, L: liquid), maximum capacity in tonnes (% capacity expansion). 2 Percent of on-site production to total production: V: Greater Vancouver regions, NV: Kelowna, Kamloops, Victoria, Prince George; R: Abbotsford, Whistler, Hope, Williams Lake. 86  As Figure 5.1 shows, the gas trailers with the lowest capacity have the highest contribution at the initial time step. As the demand grew over time, the transportation network developed toward higher capacity units. The medium-capacity delivery trucks (500 kg per load) were the dominant mode of transportation in all time steps (except the first time step), for all demand scenarios.  The transition from low to high capacity was also observed for the fueling stations over time. As Figure 5.2 shows, each capacity category is the aggregation of all fueling stations that received hydrogen in gaseous or liquid form and those with on-site production. While the lowest-capacity stations (150 kg/day) dominated the dispensing network in the first two time steps, the largest contribution in succeeding time steps was 500 kg/day for the pessimistic scenario and 1500 kg/day for moderate and optimistic scenarios.    Figure 5.1. Contribution of different transportation states (G: gas, L: liquid) and deliverable capacities (100, 500, 900 and 3800 kg per truck) to the total number of transportation units for three demand scenarios over time (no policy inclusion). 0%10%20%30%40%50%60%70%80%90%100%1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6Pessimistic scenario Moderate scenario Optimistic scenarioPercentage of total transport Time stepG100 G500 G900 L380087   Figure 5.2. Contribution of different fueling station capacities (150, 500, 1000, and 1500 kg/day) to the total number of stations for three demand scenarios over time. Station types: O: on-site production, G: gas delivery, L: liquid delivery (no policy inclusion).  Figure 5.3 (a) to (c) show the optimal geographical distribution of production facilities and the transportation network in B.C. The central storage facilities are attached to the production units. Figure 5.4 (a) to (c) shows the fueling station network for the Metro Vancouver municipalities. These maps illustrate the last time step (2050) of the pessimistic (a), moderate (b) and optimistic (c) scenarios for the base case.   (a) 0%10%20%30%40%50%60%70%80%90%100%1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6Pessimistic scenario Moderate scenario Optimistic scenarioPercentage of toal stationsTime stepOGL150 OGL500 OGL1000 OGL150088   (b)  (c) Figure 5.3. Optimal distribution of production facilities and transportation network in B.C. for (a) pessimistic (b) moderate (c) optimistic demand scenarios in time step 2045-2050 (no policy inclusion).  89   (a)  (b) 90   (c) Figure 5.4. Optimal distribution of hydrogen fueling stations in Metro Vancouver for (a) pessimistic (b) moderate (c) optimistic demand scenarios in time step 2045-2050 (no policy inclusion).  In contrast to previous studies, various capacity options of the transportation network and fueling stations were incorporated into the model. As shown in Table 5.2, a fixed capacity delivers a suboptimal cost of the HFSC. The impact of assuming a fixed capacity was the largest in the pessimistic scenario (21%), while it was smaller for the moderate (13%) and the optimistic scenarios (6%). The assumption of a fixed capacity increased the chance of facility underutilization and decreased the cost advantage of economies of scale. The total cost of the  supply chain was more susceptible to this extra cost at low demand.        91  Table 5.2. Effect of capacity alternatives on the total discounted cost of the HFSC for three demand scenarios.   Demand scenario   Pessimistic  Moderate  Optimistic  Multiple capacities Total discounted cost of HFSC (C$ million) 260.1 572.6 875.3 Fixed capacity 315.1 647.9 930.4 Multiple capacities: transportation capacities: gaseous: 100, 500, and 900 kg; liquid: 3800 kg. Fueling station capacities: 150, 500, 1000, and 1500 kg/day. Fixed capacity: transportation capacity: gaseous: 180 kg, liquid: 3800 kg. Fueling station capacity: on-site production and gaseous delivery: 500 kg/day, liquid delivery: 1500 kg/day. All values in Canadian dollars (2013).  5.2 HFSC configuration with current provincial policy inclusion The low-carbon fuel standard (LCFS) and carbon tax are the environmental mandates that influence the evolution of the low-carbon fuel infrastructures by monetizing CO2 emissions.  The HFSC optimization was performed for three cases in which carbon tax, BC-LCFS, and both policies were included in the model. Figure 5.5 shows the share of production technologies and the share of liquid hydrogen in policy-included cases alongside with the base case. In all cases, the share of SMR-based hydrogen production increased from the pessimistic to the optimistic scenario, which is mainly due to the decreasing share of by-product hydrogen and the on-site production. The decreasing share of by-product hydrogen is justified by the constant maximum capacity of the purification plant at the district of North Vancouver. The decreasing share of on-site electrolysis is associated with the dominant effect of economies of scale in reducing the cost of central production over the on-site production, which avoids the transportation cost.  The inclusion of the LCFS in the base case decreased the share of SMR technology in favor of on-site production by 1% for the pessimistic scenario and around 2% for the moderate and optimistic scenario. This technology shift shows that the higher investment cost of electrolyzers was justified by the extra revenues from the avoided GHG emissions. A slightly greater technology shift was observed when the carbon tax was included in the base case for pessimistic and moderate scenarios. However, as the demand grew (i.e., pessimistic to optimistic scenario), the transition toward low-carbon investments became more expensive than the base cost of carbon pricing. When the BC-LCFS was layered in addition to the carbon tax, the share of SMR technology was reduced in favor of on-site electrolysis for all demand scenarios. Thus, coupling the LCFS with 92  the carbon tax was effective in encouraging providers to avoid taxes by directing the LCFS’s revenues toward establishment of lower-emissions technologies.     Figure 5.5. Contribution of different technologies to total hydrogen production and the share of liquefied hydrogen for three demand scenarios and policy inclusions (no policy: base case, LCFS: low-carbon fuel standard, CT: carbon tax, C-electrolysis: Central electrolysis, O-electrolysis: On-site electrolysis).  Figure 5.5 shows that in cases that included policy, liquid hydrogen delivery was required in the moderate scenario. The demand in the moderate scenario was sufficiently large in some regions to support the liquid delivery. Liquid delivery has a lower GHG emissions footprint than gas delivery, as fewer trucks are needed to transport the same amount of hydrogen but has cost as a barrier. When the policies were added to the base case, the liquefaction cost was justified by less CO2 avoidance cost and higher revenue from LCFS. Higher demand in the optimistic scenario made the liquefaction affordable in the base case and even more attractive in policy-included cases compared to the moderate scenario. It should be noted that the percentage of liquefied hydrogen to the total production (as presented in Figure 5.5) was also affected by the variation of on-site production, which was not paired with liquefaction.  0% 0% 0% 0%0%29%27% 27%55% 51% 48%52%0%10%20%30%40%50%60%70%80%90%100%No policyLCFSCTCT+LCFSNo policyLCFSCTCT+LCFSNo policyLCFSCTCT+LCFSPessimistic scenario Moderate scenario Optimistic scenarioPercentage of total productionPurification SMR C-Electrolysis O-Electrolysis Liquefied hydrogen93  5.3 Economic and environmental evaluation of the hydrogen supply chain The post-optimization cash flow analysis (section 4.5) was performed with a real IRR of 10%. The resulting average hydrogen price was C$12 per kg in the pessimistic scenario, while it decreased to C$9 per kg and C$8/kg in the moderate and optimistic scenarios, respectively. While similar demand gaps existed between the three scenarios (Figure 3.10), the effect of demand growth on the hydrogen price was less noticeable in a moderate-to-optimistic transition compared to a pessimistic-to-moderate transition.  The key question is whether the hydrogen fueling infrastructure is economically and environmentally competitive with its gasoline counterpart. For the economic comparison, the difference between the discounted revenue of hydrogen and gasoline infrastructure in a 30-year time frame was compared for the same vehicle stock and an IRR of 10%. The revenue of the gasoline infrastructure was calculated based on the fuel efficiency of the gasoline vehicles, mileage, number of vehicles, and the gasoline price. The trajectory of gasoline price was adopted from the National energy Board [10] as the average projected price of various scenarios presented for British Columbia. The sales tax and excise tax were not considered for hydrogen at the point of sale except for carbon tax in the corresponding scenario. For the environmental comparison, the annual WTW GHG emissions avoidance was calculated from section 4.6. The reduction in GHG emissions was monetized by adopting the social cost of carbon (SCC) for the 2020–2050 time-frame with the 3% social discount rate, as recommended by the Canada Treasury Board Secretariat’s Analysis Guide [126].  As Figure 5.6 shows, for a 30-year time frame, the discounted revenue of an HFSC was less than the gasoline supply chain by C$22 million in the pessimistic scenario. However, the extra investment was justified by the avoided GHG emissions, at a discounted value of C$80 million. For moderate and optimistic scenarios, higher demand drove the hydrogen price below that of gasoline (1 kg hydrogen replaces 8 liters of gasoline). Thus, more emissions avoidance benefits are expected at the lower discounted total cost for HFSC compared to the gasoline counterpart.   94   Figure 5.6. Environmental and economic comparison of hydrogen and gasoline infrastructure in the base case (no environmental policy is included) for three demand scenarios: net present value (NPV) of the reduced GHG emissions (IRR = 3%) versus the difference between the NPV of revenues (IRR = 10%).  The addition of carbon tax and LCFS affected the hydrogen price and the avoided WTW GHG emissions as shown in Figure 5.7. The inclusion of carbon tax in the pessimistic increased GHG emissions reduction value, compared to the base case, while increasing the hydrogen price by C$0.48. The inclusion of LCFS in the base case generated extra revenues that decreased the hydrogen final price. However, a negligible contribution was observed to the emissions reduction. In other words, the LCFS revenue from the lower-emissions pathways could not pay off the cost of adoption. The LCFS had a stronger additive impact on emissions reduction when implemented with carbon tax and sets the hydrogen price below the carbon tax and above the LCFS-only cases for all demand scenarios.  It is observed that the environmental policy inclusion became less effective at higher demand scenarios. This shows that the increasing credits from LCFS and the growing fees from the carbon tax cannot keep up with the growing cost of investment in lower-emissions pathways.      80 M181 M275 M- 22 M134 M314 M- 100 M- 50 M 0 M 50 M 100 M 150 M 200 M 250 M 300 M 350 MPessimistic scenario Moderate scenario Optimistic scenarioCanadian dollars (2013)NPV (GHG emissions reduction ) NPV (hydrogen revenue minus gasoline revenue)95   Figure 5.7. Effect of environmental policies on the GHG emissions and the hydrogen price compared to the no policy case for three demand scenarios: CT: carbon tax (C$45 to C$75 from 2020 to 2050), LCFS: low-carbon fuel standard (C$167 to C$0 from 2020 to 2050). All values in Canadian dollars (2013).  5.4 Effect of complementary policies in low-carbon hydrogen production The results in section 5.3 showed that the combination of LCFS and carbon tax (LCFS+CT case) is a suitable option when hydrogen price and emissions reduction is equally important. The costs and revenues associated with these policies increased the contribution of the on-site electrolysis, which decreased the GHG emissions from this supply chain. However, the stringency of those policies was not sufficient to support large-scale low-carbon hydrogen production technologies (central electrolysis or CCS) for the considered demand scenarios. In this section, a set of potential economic instruments (explained in section 3.5.2), was layered on top of the LCFS+CT case. The aim was to measure effectiveness of these instruments to boost low-carbon hydrogen production in a cost optimal HFSC.  -1.03kg-1.16kg-0.19kg-0.62kg-1.04kg-0.16kg-0.38kg-0.66kg-0.11kg$0.48-$0.31-$0.82$0.79$0.12-$0.73$0.82$0.39-$0.61CTLCFS+CTLCFSCTLCFS+CTLCFSCTLCFS+CTLCFSPessimistic scenario Moderate scenario Optimistic scenarioΔGHG emissions reduction (kgCO2eq/kgH2) ΔH2 Price (C$)96  5.4.1 Optimal share of production technologies Figure 5.8 compares the cost optimal share of production technologies for the LCFS+CT case alongside the complementary policy included cases in three hydrogen demand scenarios. The percentages represent the cumulative share of each technology over the time period of 2020-2050.  All potential policies enhanced the production share of on-site electrolysis; however, not all policies provided enough support to make the central electrolysis affordable. Moreover, all the complementary policies failed to financially justify CCS adoption in the SMR-production pathway. In the optimistic scenario, the 100% electricity rate subsidy (EC_100%) eliminated the SMR production in favor of central electrolysis. In pessimistic and moderate demand scenarios, similar technology shift was also achieved through the stepwise utility incentives with a 10-year delay (EC_Delay). This is mainly due to the larger portion of the demand that was fulfilled by the purified by-product hydrogen (compared to the optimistic scenario). The expiration of utility subsidies in the last two time steps in the EC_Step case, and the small subsidy size in the EC_25% case prevented the integration of central electrolysis in the HFSC. Thus, the size and duration of the utility subsidy are financially crucial for the technology shift. The production tax credit at the constant rate of 2$/kg H2 (PTC_$2) resulted in similar share of production technologies as 100% electricity rate subsidy (EC_100%). A stepwise PTC, with a 10-year delay (PTC_Delay), eliminated the viability of SMR production in favor of water electrolysis in all demand scenarios. The inclusion of all PTC schemes in pessimistic scenario resulted in complete independency from SMR production; however, higher demand scenarios were still reliant on SMR in PTC_$1 and PTC_Step cases.  It was observed that different schemes of capital subsidy increased the share of on-site hydrogen production, however, they failed to justify central electrolysis, except for the Grant_100% in the optimistic scenario. In all demand scenarios, SMR was found to be the dominant technology, even with the extensive capital subsidy allocation for low-carbon hydrogen production. In the LCFS+CT case, the carbon tax grew annually by C$1.02 per tonne.  In CT_2X case, the annual tax growth was set at C$2.04 per tonne, which resulted in larger share of onsite electrolysis in all demand scenarios; however, no contribution was observed from central electrolysis nor CCS. By increasing the annual tax growth to C$4.08 per tonne (CT_4X), SMR production was 97  eliminated from the supply chain in the pessimistic scenario. However, in moderate and optimistic scenarios, the contribution of central electrolysis was small and SMR remained the dominant hydrogen production technology.   Figure 5.8. Contribution of production technologies in a cost optimal hydrogen fuel supply chain for the base case (LCFS+CT) and the potential policy included cases in three demand scenarios  Figure 5.8 shows the effect of an SMR production ban (without CCS integration) on the configuration of the cost optimal production technologies (NSMR_CCS).  In all demand scenarios the production was shared between by-product hydrogen purification and electrolysis with no contribution from SMR with CCS integration.  It should be noted that the geological formations suitable for CO2 storage are located at Northeast B.C.  Hydrogen transport via trucks and tube tankers to the demand regions which are largely located in the South West of B.C. (more than1200 km distance) may become cost restrictive. The low hydrogen demand in this study did not justify the pipeline transport [25].  0%10%20%30%40%50%60%70%80%90%100%Base caseEC_25%EC_StepEC_100%EC_DelayPTC_$1PTC_StepPTC_$2PTC_DelayGrant_StepGrant_100%Grant_DelayCT_2XCT_4XNSMR_CCSBase caseEC_25%EC_StepEC_100%EC_DelayPTC_$1PTC_StepPTC_$2PTC_DelayGrant_StepGrant_100%Grant_DelayCT_2XCT_4XNSMR_CCSBase caseEC_25%EC_StepEC_100%EC_DelayPTC_$1PTC_StepPTC_$2PTC_DelayGrant_StepGrant_100%Grant_DelayCT_2XCT_4XNSMR_CCSPessimistic Moderate OptimisticBy-Product purification SMR Cenral Electrolysis Onsite Electrolysis98  5.4.2 Efficiency assessment of complementary policies It was observed that the contribution of low-carbon production technologies was directly affected by the timing and stringency of the policy schemes. However, a cost-benefit analysis is required to measure the efficiency of environmental policies in each demand scenario.  Table 5.3 compares the average hydrogen price and the GHG emissions reduction per unit of hydrogen production for three demand scenarios in the base case (LCFS+CT). The emissions reductions were calculated by replacing ICEVs with FCEVs on the well to wheels basis, divided by the total hydrogen production over a 30-year time frame.  The subsidy mechanisms were expected to decrease hydrogen price and GHG emissions from the base case. The subsidy effectiveness was defined in terms of three effectiveness indicators:  - Hydrogen price change per unit of subsidy cost. This indicator measures the effectiveness of every unit of subsidy. - GHG emissions reduction per unit of subsidy cost. This indicator measures the effectiveness of every unit of subsidy. - GHG emissions reduction per unit of hydrogen produced. This indicator measures the effectiveness of the full size of subsidy to shift the production technology. In this study, the effectiveness of each policy was presented with respect to each effectiveness indicator (Figure 5.9). The relative importance of each indicator will depend on local, regional, national or sectorial goals and priorities.    Table 5.3. Average hydrogen price and GHG emissions reduction per unit of hydrogen production over 30-year time frame for the base case (LCFS+CT)1  Demand scenario H2 Price (C$/kg) WTW GHG emissions reduction (kg CO2eq/kg H2) Pessimistic 11.22 14.77 Moderate 8.59 11.84 Optimistic 8.12 9.97  1 A discrepancy between the numbers in Table 5.3 and Figure 5.7 is due to the aggregation of demand regions for Metro-Vancouver area (from 10 regions to 5) to improve the optimization speed.   99  Figure 5.9 shows that every unit of subsidies was more effective in GHG emissions reduction as the demand grew from pessimistic to optimistic scenario. This is consistent with the growing share of low-carbon hydrogen production in larger demand scenarios (bubble size in Figure 5.9). It was also observed that for production tax credit (PTC) and utility incentives, the effectiveness of subsidies in hydrogen price reduction improved from pessimistic to optimistic scenarios. However, capital subsidies resulted in smaller hydrogen price decline in larger demand scenarios. This is partly attributed to the economies of scale and learning by doing which weakens the effect of grants compared to the operational-based incentives.   (a)    (b)  0.290.174.463.114.334.344.334.350.170.520.30-$1.2-$1.0-$0.8-$0.6-$0.4-$0.2$0.00 1 2 3 4 5 6 7 8 9 10 11 12H2price /C$ supportkg CO2 reduction /C$ supportBubble size: kg CO2 reduction/kg H20.320.27 6.814.394.424.406.746.750.260.610.51-$1.2-$1.0-$0.8-$0.6-$0.4-$0.2$0.00 1 2 3 4 5 6 7 8 9 10 11 12H2price /C$ supportkg CO2 reduction /C$ support100   (c)   Figure 5.9. Policy efficiency assessment with respect to hydrogen price decrease and GHG emissions reduction compared to the base case: (a) pessimistic (b) moderate (c) optimistic demand scenarios  The results show that PTC in general is the most promising policy when all three efficiency factors are equally taken into consideration. This is in part a result of credit generation potential from by-product hydrogen purification, which was not the case for other policy schemes. PTC_$1 resulted in the largest GHG emissions reduction per unit of subsidy in all demand scenarios. In the pessimistic scenario, the effect of this policy on the total emissions reduction was similar to the more cost intensive PTC schemes (Table 5.3). However, a smaller contribution was observed in hydrogen price reduction. This result shows that the low capacity of production infrastructure restricted the utilization of tax credits in favor of technology shift in pessimistic scenario. Thus, as the size of subsidy grows, the hydrogen price decreases without more investment in low-carbon hydrogen production. In moderate and optimistic scenarios, deploying PTC_$1 resulted in similar total GHG emissions reduction as PTC_Step. This shows that the duration of PTC_1$ over the last two time steps, offsets its smaller size with respect to emissions reduction. 0.540.508.746.264.634.618.718.790.48 0.750.55-$1.2-$1.0-$0.8-$0.6-$0.4-$0.2$0.00 1 2 3 4 5 6 7 8 9 10 11 12H2price /C$ supportkg CO2 reduction /C$ supportEC_25% EC_Step EC_100% EC_Delay PTC_$1 PTC_StepPTC_$2 PTC_Delay Grant_Step Grant_100% Grant_Delay101  Table 5.4 shows that the subsidy size of PTC_Delay was smaller than PTC_$2. However, larger emissions reduction was observed due to the production capacity limit. The EC_100% resulted in a slightly higher GHG emissions reduction than PTC_$2 in all demand scenarios. However, PTC_2$ had a higher per unit effectiveness in hydrogen price and emissions reduction with a noticeably smaller subsidy size (Table 5.4). In all demand scenarios, the EC_Delay corresponds to the highest and the lowest per unit effectiveness in GHG emissions and hydrogen price reduction, respectively, compared to the other electricity incentive schemes. The per unit contribution of EC_25% and EC_Step in hydrogen price reduction is very close or even larger than EC_100%. However, the small size of these policies failed to justify a noticeable technology switch. The capital subsidy had a competitive per unit effectiveness in emissions reduction with other policies. However, the small contribution in total emissions reduction indicates that the capital expenditure is not as restrictive as the operational costs.  Table 5.4. Net present value of the total cost of subsidies in each demand scenario (all values in Millions C$2013)  Policy case Demand scenario PTC_$1 PTC_Step PTC_$2 PTC_Delay EC_25% EC_Step EC_100% EC_Delay Grant_Step Grant_100% Grant_Delay Pessimistic 12.0 15.4 24.0  18.9 3.8 4.0 34.6 16.2 2.4 4.0 1.4  Moderate 28.6 37.9  72.4  57.0 10.3 12.2 138.0 58.4 4.4 8.1  4.1  Optimistic 40.3 55.4  124.4  98.5 11.6  16.4  251.2  109.1  6.7  9.8 4.9   Figure 5.10 compares the hydrogen price increase and the GHG emissions reduction for the cases with higher carbon tax compared with the base case. The hydrogen price and emissions reduction were compared to banning SMR production without the CCS adoption (NSMR_CCS). Moreover, the aforementioned policy schemes were compared with the case which SMR with CCS is the only central production option alongside by-product hydrogen purification (NELEC_CCS). It should be noted that the CCS option was never selected by the optimization model in this study. Hence, NELEC_CCS was introduced to assess the extent of extra cost and the emission reduction contribution, upon its potential adoption. 102  Higher tax rates or a restriction on the type of production technology increased the hydrogen price compared to the base case. The price difference could be subsidized to consumers at the point of hydrogen sale. Higher carbon tax rates improved the total GHG emissions reduction as well as the ratio of emissions reduction to the price increase, in all demand scenarios. In other words, it was less expensive to reduce a unit of emissions at a higher tax rate. From pessimistic to optimistic demand scenario, the emissions reduction became more expensive. Accordingly, the lower cost of emissions reduction was obtained by imposing higher carbon tax (CT_4X) on the lower demand scenario. As Figure 5.10 shows, the case which bans SMR production without CCS integration (NSMR_CCS) contributed to the largest GHG emissions reduction per unit of hydrogen, compared to other policy cases. The emissions reduction benefit increased from pessimistic to optimistic demand scenario, and larger gap was observed between the hydrogen price increase and the emissions reduction. Thus, NSMR_CCS is more effective in emissions reduction and price control at higher demand scenarios.  NELEC_CCS required the largest amount of subsidy to keep the end-user price of hydrogen at the base case level. Moreover, the emissions reduction level was considerably lower than the NSMR_CCS case. Thus, the cost optimal HFSC based on electrolytic hydrogen (NSMR_CCS) was both economically and environmentally more beneficial than relying on carbon capture and sequestration in British Columbia.  103   Figure 5.10. Hydrogen price increase and the GHG emissions reduction for the potential policy included cases compared to the base case (LCFS+CT). The base case values are presented in Table 5.4.   02468101200.511.522.533.544.5CT_2XCT_4XNSMR_CCSNELEC_CCSCT_2XCT_4XNSMR_CCSNELEC_CCSCT_2XCT_4XNSMR_CCSNELEC_CCSPessimistic Moderate OptimisticGHG emissions avoidance (kg CO2eq/kg H2)H2price (C$/kg) H2 price increase GHG emissions avoidance104  Chapter 6: Challenges and Potentials in the Heavy-duty Transport Sector The 2016 data in B.C. shows that trucking industry had a similar contribution in GHG emissions as the light duty vehicles with a round 6% GHG emissions increase from 2007 [172]. Freight trucks are also significant sources of criteria air contaminants (CAC), like Particulate Matter (PM 10, PM 2.5), Nitrogen Oxides (NOx), Carbon Monoxide (CO) Sulphur Oxides (SOx) and Volatile Organic Compounds (VOC), which adversely affect air quality and human health.    (a)  (b) Figure 6.1. 2016 GHG emissions in B.C.: (a) GHG emissions by sector (b) GHG emissions from road transport: change from 2007 34%4%6%4%6%7% 12%13%13%1%27%Stationary Combustion SourcesAfforestation and DeforestationIndustrial ProcessesAgricultureWasteFugitive SourcesAviation, Railway,Marine and off-roadTransportationFreight TrucksCars and Passenger TrucksBusesRoad Transportation5.7% 6.3%-32.2%-40%-30%-20%-10%0%10%GHG Emissions Change105  The total WTW GHG emissions of freight trucks was around 8 MtCO2eq in 2016, and the per capita Gross Domestic Product (GDP) is projected to grow in B.C. by 20% in the next 20 years [10]. Due to the direct correlation between the number of freight trucks and GDP, it may be difficult to reduce emissions in this sector while simultaneously ensuring economic growth [173].  6.1 Approaches to reduce GHG emissions from the trucking sector Several options are suggested for reducing GHG emissions from freight trucks. The non-technical options consider the efficiency improvement of freight logistics such as load-matching and maximizing capacity, a modal shift to more energy-efficient means of transportation (e.g., rail) and the standardization of logistics-related facilities and equipment [174]. The technical improvements deal with the efficiency of internal combustion engine (ICE) trucks. In 2013, Canada began regulating on-road GHG emissions from ICE freight trucks with Gross Vehicle Weight Rating (GVWR) above 3856 kg. Under the Canadian Environmental Protection Act, two phases of regulations have been proposed for the deployment of advanced cost-effective technologies to increase the fuel efficiency and GHG emissions standards for new freight trucks. The first phase applies to 2014 and newer model vehicles, which reach full stringency with model year 2018 [175]. The second phase is built upon the first phase and reach full stringency with model year 2027 [176].  It is projected that the full deployment of this legislation will decrease the GHG emissions by 15-50% from freight trucks with model year 2027 compared to the 2010 counterparts depending on the vehicle’s duty cycle.  While the legislation targets the fuel efficiency of conventional gasoline or diesel trucks, some attempts have been focused on alternative fuels. The deep-carbon reduction scenarios for road freight transport often rely on significant amounts of biofuels. The B.C. Low Carbon Fuels Regulations states that by 2020 the life-cycle GHG intensity of all transportation fuels must be reduced by 10% from 2013 levels [122]. This requirement is expected to be met using first generation biofuels in the fuel blends, ethanol from corn and grain, and biodiesel from canola. Ethanol and biodiesel are already being blended into refined petroleum fuels in B.C. and the blending percentage is rising steadily. The Ethanol content increased in the gasoline pool from 5% in 2010 to 6.3% in 2014, and the biodiesel blend reached 5.6% in 2014 [177].  One of the important challenges associated with biofuels is the indirect land use change, which can result in additional 106  GHG emissions and raises concerns around food security and biodiversity maintenance [178]. Moreover, the amount of sustainable biofuels which will be available beyond 2020 is uncertain [179]. Given the uncertainties and difficulties with biofuels, this option is not likely to result in significant GHG emissions reductions of road freight transport required by 2050 in B.C. [180]. The non-renewable low-carbon fuels such as CNG, LNG and propane are now being considered as transition fuels that could serve as cost-competitive, near-term solutions.  The greenhouse gas reduction regulation under the Clean Energy Act offers incentives to diversify and grow the market for natural gas in B.C.’s transportation sector [181]. The incentives target medium- and heavy-duty trucks switching from diesel to natural gas, and decrease the  fuel  costs  on  a  per  kilometer  basis [182]. Natural gas trucks can reduce tailpipe greenhouse gas emissions by as much as 20% over gasoline or diesel trucks [183]. However, climate benefits of natural gas heavily depend on the lifecycle emissions of methane [184], [185]. The hydrogen enriched natural gas (HCNG) engine is another promising technology to enhance fuel economy and decrease emissions compared with CNG counterparts. However, implementing the perfect methane/hydrogen mixture with the current CNG infrastructure and on-board storage are the major challenges facing the adaptation of this technology. Moreover, mitigating the NOx increase as a result of hydrogen enrichment is challenging and needs to be addressed effectively [186].   The large-scale GHG emissions reduction in B.C. requires that the long-term fuel portfolio shifts toward renewable or carbon-neutral fuels. The electrification of road transportation offers zero-tailpipe emission potential. Electrification could result in the large-scale GHG emissions reduction if the energy carrier is generated from renewable resources or the production facilities are equipped with carbon capturing technologies. All-Electric vehicles are classified into battery electric vehicles (BEVs) and fuel cell electric vehicles (FCEVs). To date, electrification has primarily targeted the passenger vehicle market. Commercial all-electric heavy-duty vehicles are limited to urban delivery trucks and buses at the moment [187]–[194].  The BEVs use electricity sourced from the electrical grid to recharge on-board batteries. Current battery electric trucks, using lithium-ion batteries, have a range of 150-400 km, depending on the mass of the battery. These trucks are being developed worldwide for daily based travel on defined routes with low average speeds, high idle times and high frequency of stops and starts [187], [188]. This duty-cycle makes the overnight stationary charging and battery swapping suitable for short-107  haul battery electric trucks. There are a number of demonstration projects for battery electric semi-tractors that target captive truck fleets within the companies’ distribution network [191], [192].  For long-haul applications, the low energy density of batteries is a barrier, as significant weight and volumes are required to address the short vehicle range and long recharging times. Even if the energy density is improved by factors of 5-10, the weight increase of a 40 tonnes GVWR truck would be approximately 2 -4 tonnes [188]. Moreover, an overnight plug-in charging unit of 19 kW can regenerate the 200 kWh battery within 10-hour, which is far beyond the acceptable idle times for long-haul trucks. To make the charging time compatible with the refueling time of a conventional truck (less than 30 minutes), a 400 kW DC charger and upgrades to the transmission network would be required. However, battery electric long-haul trucks are still part of long-term vehicle portfolio when combined with on-the-road charging technology, e.g., overhead catenary wires or dynamic inductive charging [188].  Unlike BEVs, FCEVs are comparable to conventional ICE vehicles in terms of range and fueling time. The toxicity and fire hazard properties of hydrogen rank it as the safest fuel with a safety factor of 1, while the safety factors of methane and gasoline are 0.8 and 0.53, respectively [195]. Fuel cell technology has been deployed with fuel cell buses [190] and it has successfully penetrated the forklift market [196]. Demonstrations for fuel cell trucks such as package delivery vans and semi-tractors used in refuse or drayage service are in early stages [193], [194], [197] Fuel cell durability and the volume and weight of the onboard hydrogen storage are the key technical challenges to the adoption of Fuel cell technology in heavy-duty vehicles.  Moreover, hydrogen fueling stations need to be distributed and available for heavy-duty fuel cell vehicles with suitable fueling protocols. The California Fuel Cell Partnership provided an Action Plan to support the implementation of fuel cell technology in medium-duty and heavy-duty trucks in California [198].  6.2 All-Electric trucking in B.C. by 2040: feasibility study As discussed in chapter 1 and 3, B.C. has several competitive advantages including energy resources, technologies deployments, and policies to pursue opportunities in zero-emission powertrains. In this section, we examined the potential of all-electric freight trucks to achieve 64% GHG emissions reduction by 2040. To this end, the 2040 fuel-side WTW GHG emissions from B.C. trucking sector was projected for two scenarios; named as the business as usual (BAU) and 108  the current legislation fulfillment (CLF). The BAU scenario considers no technology improvement in ICE trucks, while the CLF considers the full deployment of current legislation targeting freight transportation. The potential of battery electric and fuel cell trucks to meet the mid-term GHG emissions reduction targets for 2040 was investigated for both scenarios. Moreover, the total WTW energy requirement for all-electric trucking was quantified and the availability of different energy resources in B.C. to support zero emission trucking was assessed. It should be mentioned that the analysis was based on GDP projections, and forecasts of electricity and natural gas production and demand in B.C. [10], which were available until 2040, at the time of the study.  It was also assumed that the mid-term target for reducing GHG emissions from freight road transportation is 64% by 2040 from the level of 2007.  6.2.1 Freight trucks stock forecasting The first step to project the GHG emissions from road freight transport is to project the stock of freight vehicles. Natural Resources Canada (NRCan) has classified the freight trucks to Light Duty (LD), Medium Duty (MD) and Heavy Duty (HD) based on the GVWR as shown in Table 6.1.  For each truck class, the NRCan comprehensive energy use database for transportation sector in B.C. [9] provided the average vehicle use-intensity (kilometers traveled per vehicle annually), number of new vehicles and the vehicle stock from 2000 to 2014. These historical trends were used to project the stock of each truck class by 2040.   Table 6.1 Freight truck classification [9]  The historical data on the freight vehicle use-intensity in B.C. (Figure 6.2 (a)) show that the average annual distance driven per vehicle has decreased between 28% and 46% over 14 years.  As there are no projections available in the literature for B.C, the vehicle use-intensity was fitted with a quadratic polynomial regression with the minimum mileage value extending over the Truck Class GVWR Category/kg Class Range Icon Light Duty Truck (LDT)  3855 1-2  Medium Duty Truck (MDT) 3856 to 14969 3-7  Heavy Duty Truck (HDT)  14970 8  109  studied time frame. The quadratic regression provides a conservative projection for this study. Linear and exponential regressions produce near zero vehicle use-intensity for year 2030 onward which is unrealistic for a freight vehicle. Due to uncertainties associated with the projections, the maximum positive and negative deviation from the polynomial fit was selected to account for the uncertainty region of the study domain.   The number of new freight vehicles has been projected based on historic trends and the real GDP per capita [199]–[201]. For B.C., the annual increase rate of new trucks per real GDP per capita was calculated from the historic data on the number of new trucks and the real GDP per capita between years 2000 to 2014 [9]. As this historic annual increase rate did not follow a traceable path, the average increase rate (?̅?) is used for the projection. Having the average increase rate of new trucks and the projections on the real GDP per capita to 2040 [10], the new vehicles of each truck class (𝑁𝑒𝑤𝑇) were projected to 2040 as follows:  ?̅? =∑𝑁𝑒𝑤𝑇𝑖+1 − 𝑁𝑒𝑤𝑇𝑖𝐺𝐷𝑃𝑖+1 − 𝐺𝐷𝑃𝑖𝑛𝑖𝑛 − 𝑖 𝑁𝑒𝑤𝑇𝑘 = ?̅?(𝐺𝐷𝑃𝑘 − 𝐺𝐷𝑃𝑘−1) + 𝑁𝑒𝑤𝑇𝑘−1 𝑖 = 2000, 𝑛 = 2013, 𝑁 = 2040, 𝑘 = 𝑛 + 2, … , 𝑁 6.1  Figure 6.2 (b) shows that the number of new freight vehicles entering B.C.’s market will increase due to the projected increase in real GDP per capita in B.C. The model presented here was calibrated to historical data and compared to the projected number of new vehicles, where the average difference was used as the range of uncertainty for the new vehicle projection. The stock of each truck class was projected using the average truck lifetime in B.C., either in years or total kilometers (Table 6.2), and the projections on the number of new trucks and the average vehicle use-intensity, as described in Equation 3.55.      110  Table 6.2. ICE truck characteristics [9] ICE Trucks Average Fuel efficiency (litre/100km) Fuel type Lifetime LDT 11.7 Gasoline 300,000 km or 20 years MDT 22,21.7 Gasoline, Diesel 450,000 km or 15 years HDT 40 Diesel 900,000 km or 17 years  It is worth mentioning that there are several constraints for the future growth of freight movements, such as the congestion and capacity of road networks, sudden change in fuel prices and economic indicators and the availability of trucks and drivers. However, the analysis of those factors was beyond the scope of this study. Figure 6.2 (c) shows that the stock of heavy-duty truck (HDT) grows by 100% in 2040 compared with 2014, while the growth of medium-duty (MDT) and light-duty trucks (LDT) is 34% and 42%, respectively.  The uncertainties associated with the vehicle use-intensity was not reflected in the stock projections, as the vehicle lifetime constraint measured in years was met prior to the lifetime constraint measured in total distance travelled for all vehicle classes (Equation 3.55).    (a)  020,00040,00060,00080,000100,000120,0002000 2010 2020 2030 2040Vehicle Use-Intensity (km)YearLDT- Historical DataMDT- Historical DataHDT- Historical DataLDT- Regression AnalysisMDT- Regression AnalysisHDT- Regression Analysis111      (b)   (c)  Figure 6.2. Historical data and projections to 2040 for light-duty trucks (LDT), medium-duty trucks (MDT) and heavy-duty trucks (HDT): (a) freight vehicle use-intensity in B.C. - (b) number of new freight vehicles in B.C. market – (c) stock of freight vehicles in B.C.  01000020000300004000050000600007000005,00010,00015,00020,00025,00030,00035,00040,0002000 2010 2020 2030 2040Real GDP per capita (chained (2007) dollars)Number of New TrucksYearReal GDP Per Capita LDT- ProjectionLDT- Historical Data MDT- ProjectionMDT- Historical Data HDT- ProjectionHDT- Historical Data0100,000200,000300,000400,000500,000600,000700,0002000 2010 2020 2030 2040StockYearLDT- Historical Data LDT- ProjectionMDT- Historical Data MDT- ProjectionHDT- Historical Data HDT- Projection112  6.2.2 GHG emissions projections from road freight transport: BAU and CLF scenarios The fuel-side WTW GHG emissions are analyzed from the primary energy source extraction to the point of fuel utilization. It should be noted that the life-cycle effects of vehicle manufacturing and infrastructure construction/decommissioning were not covered in the fuel-side GHG emission analysis. For the historical WTW GHG emissions calculation, the tank-to-wheel (TTW) GHG emissions for different truck classes were extracted from NRCan database [9]. The fuel average TTW GHG emissions rate was considered as 2370 and 2734 gCO2eq/litre for gasoline and diesel, respectively [9]. The GHG emissions associated with fuel production (Well-to-Tank (WTT)), were also considered as 690 and 617 gCO2eq/litre for gasoline and diesel, respectively [115]. Two scenarios were considered for the projections, with no alternative fuel or powertrain being added to the market, as follows:  6.2.2.1 Business as usual (BAU) scenario This scenario gives a conservative projection, considering the current technology (Year 2014) remains unchanged. Thus, constant average fuel efficiency (Table 6.2) was used for the entire projection period. The annual WTW GHG emissions (gCO2eq) were calculated for each ICE truck class using the fuel average WTW GHG emissions rate (gCO2eq/litre), average fuel efficiency (litre/km), and the forecast results on the stock and vehicle use-intensity (km), as described in section 4.6.    6.2.2.2 Current legislation fulfillment (CLF) scenario  This scenario gives a favorable projection on the efficiency of ICE trucks. It reflects the full deployment of the proposed federal regulations for the GHG emissions reductions from medium and heavy-duty vehicles [175], [176]. These regulations mandate the fuel efficiency improvement of the trucks by considering a combination of engine efficiency improvements, lower rolling resistance tires, aerodynamic drag improvements, mass reduction, axle and transmission efficiency improvements and workday idle reduction systems. The regulatory standards were grouped into 8 categories based on gross vehicle weight, which include combination tractors (class 7 and 8), vocational vehicles (class 2b-8) and heavy-duty pick-ups and vans (class 2b-3). The standards for 113  tractor trucks are classified under 9 subcategories based on weight, roof height and cab configuration. There are also separate standards targeting the engines of these vehicles. However, the available B.C.’s truck statistics are solely based on three weight categories [9] as shown in Table 6.1. In order to use these standards with the available B.C.’s statistics, fuel efficiency improvement of trucks was averaged for three weight categories as shown in Table 6.3. Moreover, as the aforementioned regulations do not cover the GVWR below 3855 kg, the legislation amending the passenger automobiles and light truck GHG emissions [202], [203] was used to represent the light-duty freight trucks.     The annual WTW GHG emissions of freight trucks were calculated for this scenario using equations in section 4.6 and considering the fuel efficiency improvement tabulated in Table 6.3.   Table 6.3. Fuel efficiency improvement of freight trucks from deployment of federal regulations in the current legislation fulfillment (CLF) scenario  Phase 1 Phase 2  2014-2020 2021-2023 2024-2026 2027 onward LDT 10% 20% 25% 30% MDT 10% 15% 20% 25% HDT 10% 20% 30% 35%  The historical data on truck utilization and GHG emissions in B.C. are based on the number of registered trucks in the province. Thus, the share of trucks entering from other provinces or from United States borders that are not registered in B.C. was not considered as the source of GHG emissions (the Weigh2GoBC program does not track vehicles entering the province unless they are registered in the program) [204]. In order to maintain the consistency of the data in the projection, we ignored the effect of incoming trucks on the vehicle-use intensity and GHG emissions of B.C.  114  6.2.2.3 BAU and CLF comparison Figure 6.3 shows the results of the fuel-side WTW GHG emissions analysis for different freight truck classes in B.C. If the current ICE technology persists, the BAU scenario projects that the 2040 GHG emissions of LDTs, MDTs and HDTs will increase by 39%, 53% and 84%, respectively, from 2007 levels (regardless of associated uncertainties). With the fulfilment of the current legislation (CLF scenario), these emissions will increase by 11%, 28% and 50% from LDTs, MDTs and HDTs, respectively (regardless of associated uncertainties). For LDTs the GHG emissions stay unchanged for around 16 years and start to decrease afterwards. For MDTs, the GHG emissions will fall modestly or stay unchanged for around 19 years, then rise gradually afterwards. For HDTs, there are periods of 2-4 years with slight GHG emissions reductions, however, a net rising trend can be observed for studied timeframe. These results suggest that the current legislation, which focuses mainly on fuel efficiency improvement of ICE powertrains, will fail to meet GHG emissions reduction targets by 2040. Thus, switching to zero tailpipe emission powertrains are required as part of the long-term solution.    (a)  0.00.51.01.52.02.53.02000 2010 2020 2030 2040WTW GHG Emissions (MtCO2eq)YearLDT- BAU ScenarioLDT- CLF ScenarioLDT-Target115   (b)  (c) Figure 6.3. WTW GHG emissions from road freight transportation in B.C. for business as usual (BAU) and current legislation fulfillment (CLF) scenarios - historic data and projections to 2040 (a) light-duty trucks (LDT) (b) medium-duty trucks (MDT) (c) heavy-duty trucks (HDT)  6.2.3 GHG emissions projections from road freight transport in 2040: electrification effect  As FC and BE trucks have zero tailpipe emissions, the WTW GHG emissions analysis is equivalent to the well-to-tank (WTT) evaluation. For fuel cell trucks, the WTT GHG emissions are involved in the production, transportation and distribution of hydrogen from the energy source to the on-0123456782000 2010 2020 2030 2040WTW GHG Emissions (MtCO2eq)YearMDT- BAU ScenarioMDT-CLF ScenarioMDT-Target0123456782000 2010 2020 2030 2040WTW GHG Emissions (MtCO2eq)YearHDT- BAU ScenarioHDT- CLF ScenarioHDT-Target116  board tank of vehicle. Figure 6.4 shows the two selected hydrogen production pathways with WTT energy requirement and GHG emissions mentioned in Table 6.4. The pathway including central natural gas reforming (NGCR) was selected as it is the predominant industrial hydrogen production technology worldwide [205], and B.C. has large reserves of commercially available natural gas [96].  The HyCE is a renewable pathway for hydrogen production using central electrolysis which is feasible in B.C. due to the dominance of relatively cheap hydroelectric power.  For the battery electric trucks, the WTT analysis accounts for the emissions associated with electricity generation. The electricity loss from transmission lines was estimated to be 10% [206].   Figure 6.4.  Hydrogen production pathways   The effect of electrification on the GHG emissions of road freight transport in 2040 was investigated by substituting the WTW GHG emissions of ICE trucks with the WTT GHG emissions of battery electric and fuel cell trucks in the BAU and CLF scenarios. For the BAU scenario, the fuel efficiency of all-electric trucks was estimated based on the energy efficiency of the powertrains provided by [207], and the average fuel efficiency of ICE trucks in B.C. driven from Table 6.2. For battery electric trucks, the fuel efficiency was estimated at 2.5, 1.3 and 0.6 km/KWh for light-duty, medium-duty and heavy-duty trucks, respectively. The fuel efficiency of fuel cell trucks was estimated at 62, 35 and 16 km/kg H2 for the aforementioned classes, correspondingly. For the CLF Scenario, some sections of the current legislation which were not dependent on the powertrain were applied to all-electric trucks, e.g., lower rolling resistance tires, aerodynamic drag improvements and speed limiters. These technologies are projected to increase the fuel efficiency by 15% for LDT, 10% for MDT and 20% for HDVs by 2027 [176]. The following assumptions were considered in the WTW GHG emissions calculations of the all-electric freight trucks: Production technology/ ConditioningDistribution DispensingResource PathwayCentral SMR+CCSLiquefaction Compression Natural GasNGCRCentral ElectrolysisTube Tankers (100 km)CompressionHydro ElectricityHyCE Tanker Trucks (300 km)Compression117  - This study considered the effects of uncertainties associated with the projection of new vehicles and the vehicle use-intensity on the stock of all-electric vehicles, energy requirements and the GHG emissions calculations. The uncertainties associated with vehicle average fuel efficiency, vehicle average lifetime and the technology efficiency for different components of fuel supply chain were not covered in this analysis. - As the share of hydroelectricity is projected to stay above 86% of total electricity generation in B.C. [10], the GHG intensity of electricity generation was assumed to stay constant for the studied time-frame. - The charging loss is included in the total fuel efficiency of the battery electric trucks [208].  - The driving range of 120 km was assumed for all classes of battery electric trucks. Based on this assumption, the effect of battery weight on the fuel efficiency of battery electric trucks was not considered in this analysis [188].  - The total electricity required in NGCR and HyCE pathways was assumed to be generated from hydropower. - The ICE trucks were assumed to deliver hydrogen for both NGCR and HyCE pathways. In the BAU scenario the GHG emissions associated with hydrogen delivery was used from Table 6.4. In the CLF scenario, the fuel efficiency improvement of 35% was considered from fully deployment of federal regulations (Table 6.3).           118   Table 6.4. WTT energy requirement and GHG emissions for the selected hydrogen pathways [103], [112], [214]–[216], [113]–[115], [209]–[213]  Feedstock production, conditioning and transportation  Natural gas Hydro power GHG emissions 354 gCO2eq/m3 11 gCO2eq/kWh  Feedstock transformation to hydrogen  Central reforming+ CCS Central electrolysis  NG Electricity 50.2 kWh/kg H2 Energy requirement 4.745 m3/kg H2 1.4 kWh/kg H2 GHG emissions 1140  gCO2eq/kgH2  11 gCO2eq/kWh 11 gCO2eq/kWh  Distribution and dispensing  Gaseous delivery Liquefied delivery  Compression at dispenser to 87.5 MPa Delivery (tube trailer) Compression to 25 MPa Liquid hydrogen to gas compression to 87.5 MPa Delivery (tanker truck) Liquefaction Energy requirement 2 kWh/kg H2  1.5 kWh/kg H2 0.6 kWh/kg H2  9 kWh/kg H2 GHG emissions 11  gCO2eq/kWh 138  gCO2eq/tonne.km 11  gCO2eq/kWh 11  gCO2eq/kWh 138  gCO2eq/tonne.km 11  gCO2eq/kWh  119  Figure 6.5 shows the share of all-electric freight trucks required to reduce 64% GHG emissions from road freight transport in 2040 compared to those from 2007. The results suggest that the share of all-electric freight trucks (either battery electric or fuel cell) has to be more than 65% of the freight stock, regardless of the WTT pathway and the considered scenario. As the annual number of new trucks varies between 5% and 7% of the stock during the projection period, the all-electric new trucks are required to reach 100% market share as early as 2025. Figure 6.5 also indicates that less battery electric trucks are required to meet the target compared to fuel cell trucks. However, the market penetration of the battery technology is dependent on the duty cycle of the vehicle, especially, for long-haul HDTs, battery is a challenging technology to adopt. The same amount of GHG emissions could be reduced by 5-6% more heavy-duty fuel cell trucks, if hydrogen is produced via HyCE pathway. It is also observed that the full deployment of current legislation (CLF scenario) in ICE trucks has the same effect in terms of GHG emissions reduction as 7-10% penetration of all-electric freight trucks, in 2040.    Figure 6.5. Share of all-electric freight trucks (FCE: fuel cell electric, BE: battery electric) for 64% GHG emissions reduction from road freight transport in 2040 (from 2007 level): business as usual (BAU) and current legislation fulfillment (CLF) scenario  0%20%40%60%80%100%LDT MDT HDT LDT MDT HDTBAU Scenario CLF ScenarioShare of All-electric Vechicles from Freight Stock FCE- HyCE Pathway FCE- NGCR Pathway BE-Hydro Electricity120  6.2.4 B.C. resource assessment to support all-electric trucking    Figure 6.6 compares the electricity requirement to support the 2040 all-electric trucking, described in Figure 6.5, for BAU and CLF scenarios. The National energy Board projections [10] stated that the total electricity generation in B.C. will be around 81.1 TWh in 2040, of which 86% will be generated from large-scale hydroelectric dams. These projections also stated that the hydroelectricity production in B.C. will surpass the demand by 12% in 2040. Figure 6.6 shows that the extra electricity generation in B.C. can support up to 33% of the fuel cell trucks (with HyCE pathway) and up to 72% of the battery electric trucks in BAU scenario, regardless of associated uncertainties. These percentages can increase up to 42% and 93%, respectively in CLF scenario. The NGCR pathway also requires electricity in different stages of hydrogen production, transportation and distribution. The total electricity requirement for meeting 2040 targets via this pathway is 69% and 55% of 2040 extra electricity generation in BAU and CLF scenarios, respectively. For illustrative purposes, the total electricity demand of all-electric trucks was compared to the projected capacity of Site C dam, which will be the 4th largest producer of hydroelectricity in B.C. The government of B.C. recently decided to proceed with the Site C project despite opposition from indigenous communities and the mounting construction costs [217]. As shown in Figure 6.6, the required hydroelectric energy for FC HyCE pathway in BAU scenario is around 6.5 times the total electricity generation of Site C [218]. Even supporting the battery electric trucks in the CLF scenario requires around 2.5 times the total electricity generation of Site C.  121   Figure 6.6. Electricity requirement for 64% GHG emissions reduction from road freight transport in 2040 (from 2007 level) - FCE: fuel cell electric and BE: battery electric trucks- business as usual (BAU) and current legislation fulfillment (CLF) scenario  The total electricity generation (e.g., in TWh), may not give a comprehensive picture of the electricity availability to support the mass electrification in the road freight sector. In B.C, the installed generation capacity and the peak load of electricity is projected to be 21000 MW and 16900 MW in 2040, respectively [10]. Assuming that sufficient battery electric trucks penetrate the market to meet the 2040 GHG emissions reduction target in the BAU scenario (Figure 6.5). Even with the total extra generation capacity, up to 10% of all battery electric trucks could use DC fast chargers (50 kW), or up to 25% could use Level 2 AC chargers (19.2 kW) during peak hours. And if all battery electric trucks use off-peak hours for charging using Level 2 AC chargers (between 5 to 8 hours), a 16300 MW load will be added to the system. This means peak hours may extend to midnight and early morning.   The projections on the total electricity generation show that the hydroelectric power can hardly satisfy the large electrification of road freight transport in B.C. Moreover, the projected installed electricity capacity is not ready to support the large percentage of battery electric trucks on the 02468101214161820LDT MDT HDT LDT MDT HDT Site C Extra HydroElectricityGenerated in2040BAU Scenario CLF ScenarioElectricity  (TWh)FCE- HyCE Pathway FCE- NGCR Pathway BE-Hydro ElectricityDemand Supply122  roads. It should also be noted that B.C. can no longer rely on any imported power to meet the forecast requirement. The BC Clean Energy Act called on BC Hydro to become self-sufficient in electricity production and a net exporter of clean electricity starting in 2016. Moreover, the Clean Energy Act banned the future development of large-scale hydro-electric storage dam projects on all rivers in B.C., except for site C. Thus, the diversification of the renewable supply mix seems to be inevitable to support large-scale electrification. As discussed in Chapter 3 (section 3.1) 44 TWh of the wind resource potential and around 6 TWh of the geothermal resource potential in B.C. can be harvested for less than $200 per MWh. Moreover, the wood-based biomass resources available for bioenergy production has the technical electricity production potential of 4.5 TWh, generated mostly below $200 per MWh.  It is evident that there is a huge wind electricity potential in B.C. to support the electricity demand from the transportation sector. However, the intermittency associated with wind-generated electricity poses a challenge with regards to load leveling at large capacities [219]. The power-to-hydrogen pathway is a promising option to mitigate the intermittency of wind energy in a form of stored hydrogen. Hydrogen could be produced via electrolysis during off-peak demand hours at lower price and stored as an electricity back-up or directly used for transportation needs. In  the short term, as the transportation is predominantly reliant on fossil fuels, the electrolytic hydrogen can be used in oil refineries to reduce the carbon intensity of the petroleum fuels [220]. Moreover, electrolytic hydrogen can be injected to the natural gas system and used in hydrogen enriched natural gas (HCNG) engines [186]. Thus, the power-to-hydrogen is helpful to increase the flexibility of the power system and enables the high contribution of wind electricity in a short and long-term perspective [221], [222]. The NGCR pathway opens up the opportunity to partially unburden the renewable electricity generation to reduce GHG emissions from the road freight transport. The natural gas requirement for NGCR pathway is approximately 3×109 m3 and 2.4×109 m3 for BAU and CLF scenarios which is 3% of projected production for 2040 in B.C. [10]. However, the GHG emissions reduction of the NGCR pathway is dependent on the deployment of large-scale carbon capture and sequestration (CCS) facilities. It should be mentioned that CCS technology is yet to be widely deployed. The economic feasibility and potential environmental impacts of CCS may limit its application [223]. Figure 6.7 shows that the NGCR pathway without CCS falls short of meeting 123  the GHG emissions target in road freight transport, even with 100% of truck stock running on hydrogen.   Figure 6.7. GHG emissions change in 2040 road freight transport compared with 2007 - 100% of freight trucks running on hydrogen produced from central natural gas reforming (NGCR) pathway without carbon capture and sequestration (CCS)- business as usual (BAU) and current legislation fulfillment (CLF) scenario  6.2.5 Comparative analysis of emission reductions and energy requirements across scenarios       Converting more than 65% of all freight trucks to electric powertrain by 2040 may be challenging. Currently, there is uncertainty over the cost and lifetime of these vehicles. Moreover, the availability of charging stations and hydrogen refueling infrastructure in neighboring provinces and the United States can affect the all-electric long-haul transportation in B.C. Hence, we consider the penetration requirements for every 1% GHG emissions reduction from the trucking sector in 2040.   According to Figure 6.8, 11,000 to 14,000 all-electric freight trucks are required for every 1% GHG reduction from B.C.’s road freight transport in 2040. As the contribution of HDTs to the -60%-50%-40%-30%-20%-10%0%10%20%30%40%LDT MDT HDT LDT MDT HDTBAU Scenario CLF ScenarioGHG Emissions Change from Road Freight Transport (2040 compared to 2007 ) 124  GHG emissions is higher, a smaller number of all-electric HDTs is necessary to reduce the same amount of GHG emissions compared to the all-electric LDTs or MDTs. It is also observed that a larger number of all-electric trucks is required for every 1% GHG emissions reduction in CLF scenario than BAU scenario. As the energy efficiency of ICE technology in CLF scenario is higher than the BAU scenario, the CLF scenario is more resilient to emissions reduction. In other words, the ICE technology in CLF scenario is competing with all-electric powertrains in GHG emissions reduction.  In terms of well-to-wheels energy requirements (Table 6.4), the hydrogen dependent pathways require more than twice as much energy as the battery electric dependent pathway (i.e., 2.2 for the HyCE and 2.5 for the NGCR pathway). Amongst, the NGCR is the most energy intensive pathway for all-electric trucking. The hydroelectricity requirement for every 1% GHG emissions reduction from road freight transport is 1.5% to 1.8% of the B.C.’s 2040 total extra hydroelectricity generation for battery electric trucks and 3.3% to 3.8% for fuel cell trucks with HyCE pathway, depending on the scenario.  125   Figure 6.8. 2040 projections on the number of all-electric trucks (FCE: fuel cell electric and BE: battery electric) and total energy required for 1% GHG emissions reduction from road freight transport in B.C.   01002003004005006000200040006000800010000120001400016000FCE- HyCE PathwayFCE- NGCR PathwayBE- HydroelectricityFCE- HyCE PathwayFCE- NGCR PathwayBE- HydroelectricityBAU Scenario CLF ScenarioEnergy (GWh)Number of All-elecrtic TrucksLDT MDT HDT Energy126  Chapter 7: Conclusion and Future Work The optimization-based framework (H2SCOT) was developed for the long-term planning of hydrogen fuel supply chains (HFSC) at low demand. The model considered various capacity options for all components of the supply chain, covered the on-site production and capacity expansion options as well as minimum storage requirement for fueling stations. The model also included a range of environmental policies in the formulation of the objective function. The H2SCOT was applied to a case study of light duty passenger in British Columbia, considering three demand scenarios and a 30-year time frame.  Freight road transport has a similar contribution to GHG emissions as the light duty passenger vehicles in B.C. However, there exists no government plans to support zero-emission freight transport. In this study, the WTW energy requirement and GHG emissions reduction potential of the battery electric and fuel cell electric trucks were measured to meet the provincial emissions reduction targets. The results can be used to develop a plan to support the purchase of these vehicles and the infrastructure development.    7.1 Light duty passenger vehicles (current provincial policies) H2SCOT was applied to the case study of light duty passenger vehicles in B.C. The optimization was performed for the case in which no environmental policy was included, as well as for the cases in which carbon tax, low carbon fuel standard (LCFS) and a mix of both policies were included.  - SMR was found to be the dominant hydrogen production technology even with the current carbon control policies in place. While the inclusion of carbon tax and LCFS boosted the adoption of on-site electrolysis, the tax rates and carbon credits were not sufficient to pay off the central electrolysis nor the carbon capture and sequestration (CCS) without significant cost reductions in the relevant technologies.  - The configuration of transportation networks and fueling stations switched gradually from low- to high-capacity units, as the demand grew over time. High-capacity gas delivery delayed the adoption of liquid delivery, so that liquid hydrogen was restricted to the optimistic scenario and to the policy-included cases in the moderate scenario. 127  - A post-optimization economic analysis was conducted to compare hydrogen with the gasoline supply chain based on the potential revenues and the monetary value of GHG emissions avoidance. The results showed that even in the pessimistic scenario, the monetary benefit of emissions reduction was 4 times the extra costs incurred by the HFSC. For moderate and optimistic scenarios, the HFSC was both economically and environmentally cost competitive.  - The effectiveness of environmental policies was found to depend on the demand and decreased from the pessimistic to the optimistic scenario. The inclusion of carbon tax improved the emission reduction contributions; however, the hydrogen price increase became a restriction as the demand increased. The inclusion of LCFS decreased the hydrogen price compared to the base case, while a negligible effect was observed on emissions reduction.  - Coupling the carbon tax with the LCFS was found to work best for the pessimistic scenario in which the emissions reduction was achieved at a lower hydrogen price, compared to the base case. For moderate and optimistic scenarios, the policy coupling reduced the GHG emissions at the expense of hydrogen price increase; yet it is the most suitable policy option when hydrogen price and GHG emissions are weighted equally.   7.2 Light duty passenger vehicles (complementary policies) The effectiveness of complementary policies was assessed on the economic viability of low-carbon hydrogen production. The policies were integrated in the formulation of H2SCOT. Stepwise deployment strategies of each policy were adopted in addition to the current policies in B.C.  (i.e., the carbon tax and LCFS) for light duty passenger FCEVs.   -Production subsidies and electricity incentives were found to be more effective in GHG emissions reduction than grant subsidies, bans on SMR-production or adoption of higher carbon tax rates.  -Every unit of production subsidies and electricity incentives were more effective in hydrogen price decrease as the demand grew from pessimistic to optimistic scenarios. However, the reverse effect was observed when grant subsidies were applied to the supply chain. -The addition of a production tax credit (PTC) to the current policies in B.C., was found to be an effective strategy to boost the low-carbon hydrogen production and decrease the hydrogen price. A PTC can be considered a market subsidy to decouple the financial support systems from government budgets by obtaining the required subsidies from undesirable technologies. Thus, 128  higher tax rates can be imposed on conventional fuels and a share of revenues can be directed as a tax incentive for low-carbon hydrogen production. -The size of the hydrogen supply chain (e.g., hydrogen demand) restricts the potential contribution of subsidies in the technology shift (GHG emissions reduction). Thus, the deployment strategy of policies over time was found to be more effective in GHG emissions reduction than the total subsidies allocated in each demand scenario.  -Higher rates of carbon tax reduced the cost of GHG emissions reduction from the hydrogen supply chain. This effect was more pronounced at lower demand levels.   -Pathways relying on SMR production with carbon capture and sequestration (CCS) were found to be less economically and environmentally favorable compared to the electrolytic hydrogen production. Low demand and long-distance hydrogen transport from Northeast B.C. diminish the financial and environmental benefit of CCS integration into the SMR facility.   - Large subsidies were required to shift from SMR production to electrolytic hydrogen. However, subsidies are essential to avoid locking into SMR technology with a lower environmental benefit and long investment cycle. The subsidy schemes in this study were developed for a 30-year time frame, which made them susceptible to government changes and budget volatility.   7.3 Fright road transport  7.3.1 GHG emissions reduction potential in B.C. The analysis was built based on two scenarios: the business as usual (BAU) with no technology improvement in ICE trucks and the current legislation fulfillment (CLF), which considered the full deployment of current legislation targeting freight transportation. - The analysis showed that the continuity of the current ICE technology (BAU scenario) by 2040 results in 39%, 53% and 84% GHG emissions increase from 2007 levels for LDTs, MDTs and HDTs, respectively. Moreover, the CLF scenario fail to set the GHG emissions on a downward trajectory.   - The projection results showed that all-electric trucking can help B.C. reduce 64% of the emissions from road freight transport by 2040. The WTW energy and GHG emissions analyses indicated that the share of all-electric freight trucks would have to be more than 65% of the stock, regardless of the WTW pathway and the considered scenario. Therefore, the government must enforce strict 129  fleet emission regulations and allocate early-market subsidies for manufacturers, customers and the infrastructure developers to promote all-electric vehicles. Moreover, the partnerships between public authorities to mass-purchase electric vehicles for the public fleets, can provide reliable demand for vehicle manufacturers (Lambert, 2017c).  7.3.2 Energy requirement and resource availability - As the WTW energy efficiency of battery electric trucks is more than two times higher than fuel cell trucks, less battery electric trucks are required to meet the 2040 GHG emissions target.  However, the adaptiveness of the battery technology is dependent on the duty cycle of the vehicle. Battery electric trucks could cover urban delivery with short and well-defined routes. This duty cycle is suited for light-duty and medium-duty classes. The heavy-duty class is suitable for long-haul application which can be satisfied by fuel cell trucks. It is recommended that policy strategies support both fuel cell and battery electric powertrains, as they are complementary solutions to decarbonize road freight transport. - The analysis showed that every 1% GHG emissions reduction from road freight transport requires between 1.5% and 3.8% of 2040 extra hydroelectric generation in B.C. Thus, the B.C. hydroelectricity will fall short of generating sufficient energy to support all-electric trucking required to fulfill the 2040 emissions reduction target. Therefore, B.C. has to undertake policies to incentivize electricity generation from diversified renewable energy resources. Wind energy provides reliability, wide scale resource availability and economic competitiveness with hydro power. The current B.C.’s policies such as 10-year exemption from participation rents for new wind projects has laid the ground for wind energy development. However, more policies may be required to address the economic challenges of wind project developments in the private sector. Along with expanding energy resources, transmission capacity needs to be increased to meet the on-peak demand created by mass adoption of electric vehicles.  - Natural gas may provide a pathway for low-carbon hydrogen production in B.C., but it would require CCS technology development and deployment. This pathway can help B.C. decrease the electricity requirements for all-electric trucking.  130  7.4 Study limitations  - The optimization results were calculated using hydrogen fuel cell penetration in the passenger light duty sector for three demand scenarios. Hydrogen demand from other transportation sectors, industry, energy storage, or for export would modify the optimal HFSC and could lower the total cost of hydrogen.  - The LCFS credit price and carbon tax assumptions are subject to uncertainty. This study aimed to show the links between policy measures to the configuration of the HFSC, the price of hydrogen, and the WTW GHG emissions. A detailed analysis is needed to assess the sensitivity of an HFSC to various levels of revenues and fees generated through those policy measures.    - The subsidy schemes in this study were developed for a 30-year time frame, which made them susceptible to government changes and budget volatility.  - The policy effectiveness was measured for a limited number of scenarios. Moreover, the assessment is sensitive to the assumptions on energy price, discount rate, the rate of technology development and the technological breakthrough. Moreover, there are transaction and program costs and policy implementation challenges that were not considered in this study. Thus, it is recommended to interpret the results of this study as directional estimates rather than exact quantifications.   7.5 Future work 7.5.1 Extending the frontiers of H2SCOT The current optimization model can be expanded to include hydrogen demand from B.C. road freight transport as well as the road transport sector in other provinces in Canada.  Moreover, the model can include hydrogen export to other jurisdictions, as well as the possibility of blending hydrogen to the natural gas network. The model expansion requires a comprehensive techno-economic analysis on the configuration of potential hydrogen supply chain in each province based on resource availability, vehicle stock projections, GHG emissions targets and current environmental policies. Based on the level of hydrogen penetration to the market and the current and planned energy profile in each region, it 131  may be required to include seasonal hydrogen storage facilities and hydrogen pipeline transport to the model.  H2SCOT was developed based on the commercially available technologies. Emerging technologies can be added to the model as they enter into the commercial stage.  7.5.2 Multi criteria decision making H2SCOT was developed based on a mono-objective framework. The aim was to minimize the total cost of the supply chain. The model also dealt with the environmental impact by assigning monetary value to the GHG emissions from the HFSC. A future analysis can include an expansion of the model to a multi-objective setting, with environmental impacts and safety as separate objective functions. As there is not a unique optimal solution to this class of problems, the concept of optimality is replaced with Pareto optimality[224].  A set of Pareto optimal solutions is generated and the sorting and arranging methods (e.g.,  ELECTRE, TOPSIS and M-TOPSIS) are used to determine the best alternative among the available options [225].  7.5.3 Parametric study on economic factors The configuration of a cost optimal HFSC is dependent on several economic factors such as the projection of electricity and natural gas price, the rate of technology cost reduction over time (technology learning rate) and the favorable interest rates for different investors. HFSC can be optimized for different realization of economic factors as well as their interactions. Accordingly, the sensitivity of HFSC to each economic factor can be measured with respect to the variations in hydrogen price and the emissions reduction benefits.  7.5.4 Introducing non-linearity and uncertainty to the model As fuel cell powertrains penetrate to the trucking sector, a portion of the hydrogen transport fleet may also run on hydrogen. This introduces non-linearity to the model, as the demand is also a function of the number of trucks and transportation distance. A future study can deal with the non-linearity using non-linear solvers like MINOS.   132  In this work, the optimization was performed for three demand scenarios. Scenario development is a post-optimization technique to introduce uncertainty to the model. This technique does not provide a single overall optimal solution for all scenarios. Stochastic programming can overcome this issue by incorporating uncertainty in the optimization model during the decision-making process. This method optimizes the expected value of the objective function over all scenarios, while finding solutions that are feasible for all realization of uncertain parameters. The technique is based on capturing the uncertainty in terms of a number of likely scenarios with known probability distributions that are possible to materialize during the lifetime of the supply chain [31], [32]. The future work can be an extension of H2SCOT from the deterministic framework into a multi-stage stochastic model to optimize HFSC planning with respect to hydrogen demand uncertainty.   7.5.5 Expansion on policy scenarios and integration mechanisms  A limited number of environmental policies was studied in this work.  Future analyses can expand on the number of policies, policy integration strategy and the policy switch over time.  Because hydrogen may be penetrated from transportation to the wider energy market, different sets of policies are required to be implemented in each sector. The interaction between policies can be studied to help the sustainable development in all hydrogen markets.        133  Bibliography [1] IEA, “The Future of Hydrogen - Seizing today’s opportunities,” 2019. [2] IEA, “Technology Roadmap: Hydrogen and Fuel Cells,” p. 81, 2015. [3] Hydrogen Council, “Hydrogen scaling up,” 2017. [4] IPHE (International Partnership for Hydrogen and Fuel Cells in the Economy), “Country Updates,” 2019. . [5] IBISWorld, “Oxygen & Hydrogen Gas Manufacturing in Canada,” 2018. [6] Canadian Hydrogen and Fuel Cell Association, “Canadian Hydrogen and Fuel Cell Sector Profile,” 2018. [7] Government of B.C., “British Columbia Climate Action for the 21st Centry,” 2005. [8] Envionment and Climate Change Canada, National Inventory Report 1990-2016: Greenhouse Gas Sources and Sinks in Canada, Annex 10. 2018. [9] NRCan, “Road Transportation Secondary Energy Use and GHG Emissions,British Columbia and Territories,” Natural Resources Canada, 2019. [Online]. Available: http://oee.nrcan.gc.ca/corporate/statistics/neud/dpa/menus/trends/comprehensive/trends_tran_bct.cfm. [Accessed: 15-Dec-2019]. [10] NEB, “Canada’s Energy Future 2019: Energy Supply and Demand Projections to 2040,” 2019. [11] Government of B.C., Zero-Emission Vehicle Act, Bill 28. 2019. [12] Industry Canada, “Canadian Fuel Cell Commercialization Roadmap Update,” 2008. [13] Government of B.C., “British Columbia’s Clean Energy Vehicle Program Phase 1 Review,” 2015. [14] “HTEC, Shell plan Canada’s first retail hydrogen refueling station in Vancouver this year,” Fuel Cells Bull., 2018. [15] “Ballard Power Systems Announces Purchase of B.C.’s First Fleet of Hydrogen-Powered Fuel Cell Electric Vehicles,” 2019. [Online]. Available: http://www.ballard.com/about-ballard/newsroom/news-releases/2019/07/16/ballard-power-systems-announces-purchase-of-b.c.’s-first-fleet-of-hydrogen-powered-fuel-cell-electric-vehicles. [Accessed: 21-Jul-2019]. [16] “Hydra Energy.” [Online]. Available: https://hydraenergy.com/. [Accessed: 10-Dec-2019]. 134  [17] M. Ruth, “Hydrogen Pathways: Cost, Well-to-Wheels Energy Use, and Emissions for the Current Technology Status of Seven Hydrogen Production, Delivery, and Distribution Scenarios,” 2009. [18] I. Dincer and C. Acar, “Review and evaluation of hydrogen production methods for better sustainability,” Int. J. Hydrogen Energy, vol. 40, no. 34, pp. 11094–11111, Sep. 2015. [19] IEA, 20 Years of Carbon Capture and Storage. OECD, 2016. [20] M. Reuß, T. Grube, M. Robinius, P. Preuster, P. Wasserscheid, and D. Stolten, “Seasonal storage and alternative carriers: A flexible hydrogen supply chain model,” Appl. Energy, 2017. [21] IEA, “Hydrogen Production & Storage: R&D Priorities and Gaps,” 2006. [22] S. Niaz, T. Manzoor, and A. H. Pandith, “Hydrogen storage: Materials, methods and perspectives,” Renewable and Sustainable Energy Reviews. 2015. [23] N. A. A. Rusman and M. Dahari, “A review on the current progress of metal hydrides material for solid-state hydrogen storage applications,” Int. J. Hydrogen Energy, vol. 41, no. 28, pp. 12108–12126, Jul. 2016. [24] R. B. Biniwale, S. Rayalu, S. Devotta, and M. Ichikawa, “Chemical hydrides: A solution to high capacity hydrogen storage and supply,” Int. J. Hydrogen Energy, vol. 33, no. 1, pp. 360–365, Jan. 2008. [25] C. Yang and J. Ogden, “Determining the lowest-cost hydrogen delivery mode,” Int. J. Hydrogen Energy, 2007. [26] U.S. DRIVE, “Hydrogen Delivery, Technical Team Roadmap,” 2017. [27] J. Alazemi and J. Andrews, “Automotive hydrogen fuelling stations: An international review,” Renew. Sustain. Energy Rev., vol. 48, pp. 483–499, Aug. 2015. [28] M. Weeda and A. Elgowainy, “Large-Scale Hydrogen Delivery Infrastructure,” 2015. [29] J. Ogden and C. Yang, “The Hydrogen Transition: a NextSTEPS white paper,” 2014. [30] H. Dagdougui, “Models, methods and approaches for the planning and design of the future hydrogen supply chain,” Int. J. Hydrogen Energy, vol. 37, no. 6, pp. 5318–5327, Mar. 2012. [31] P. Bolat and C. Thiel, “Hydrogen supply chain architecture for bottom-up energy systems models. Part 1: Developing pathways,” International Journal of Hydrogen Energy. 2014. [32] C. THOMAS, “Affordable hydrogen supply pathways for fuel cell vehicles,” Int. J. 135  Hydrogen Energy, 2002. [33] A. D. Little, “Guidance for transportation technologies:Fuel choice for fuel cell vehicles. Phase II final deliverableto DOE,” 2002. [34] P. Bolat and C. Thiel, “Hydrogen supply chain architecture for bottom-up energy systems models. Part 1: Developing pathways,” International Journal of Hydrogen Energy, vol. 39, no. 17. pp. 8881–8897, 2014. [35] R. Loulou, G. Goldstein, and K. Noble, “Documentation for the MARKAL Family of Models,” Int. Energy Agency, 2004. [36] N. Strachan, N. Balta-Ozkan, D. Joffe, K. McGeevor, and N. Hughes, “Soft-linking energy systems and GIS models to investigate spatial hydrogen infrastructure development in a low-carbon UK energy system,” Int. J. Hydrogen Energy, 2009. [37] C. Yang and J. M. Ogden, “Renewable and low carbon hydrogen for California-Modeling the long term evolution of fuel infrastructure using a quasi-spatial TIMES model,” Int. J. Hydrogen Energy, 2013. [38] M. Contaldi, F. Gracceva, and A. Mattucci, “Hydrogen perspectives in Italy: Analysis of possible deployment scenarios,” Int. J. Hydrogen Energy, 2008. [39] E. Endo, “Market penetration analysis of fuel cell vehicles in Japan by using the energy system model MARKAL,” Int. J. Hydrogen Energy, 2007. [40] A. Contreras, E. Guervós, and F. Posso, “Market penetration analysis of the use of hydrogen in the road transport sector of the Madrid region, using MARKAL,” Int. J. Hydrogen Energy, vol. 34, no. 1, pp. 13–20, Jan. 2009. [41] E. Rosenberg, A. Fidje, K. A. Espegren, C. Stiller, A. M. Svensson, and S. Møller-Holst, “Market penetration analysis of hydrogen vehicles in Norwegian passenger transport towards 2050,” Int. J. Hydrogen Energy, 2010. [42] P. Agnolucci and W. McDowall, “Designing future hydrogen infrastructure: Insights from analysis at different spatial scales,” Int. J. Hydrogen Energy, vol. 38, no. 13, pp. 5181–5191, May 2013. [43] M. Kuby and S. Lim, “The flow-refueling location problem for alternative-fuel vehicles,” Socioecon. Plann. Sci., vol. 39, no. 2, pp. 125–145, Jun. 2005. [44] Z. Lin, J. Ogden, Y. Fan, and C. W. Chen, “The fuel-travel-back approach to hydrogen 136  station siting,” Int. J. Hydrogen Energy, vol. 33, no. 12, pp. 3096–3101, 2008. [45] M. Muratori, B. Bush, C. Hunter, and M. Melaina, “Modeling Hydrogen Refueling Infrastructure to Support Passenger Vehicles †,” Energies, vol. 11, no. 5, p. 1171, May 2018. [46] Z. Lin, C.-W. Chen, J. Ogden, and Y. Fan, “The least-cost hydrogen for Southern California,” Int. J. Hydrogen Energy, vol. 33, no. 12, pp. 3009–3014, 2008. [47] J. Kim, Y. Lee, and I. Moon, “Optimization of a hydrogen supply chain under demand uncertainty,” Int. J. Hydrogen Energy, 2008. [48] A. Almansoori and N. Shah, “Design and operation of a future hydrogen supply chain: Multi-period model,” Int. J. Hydrogen Energy, vol. 34, no. 19, pp. 7883–7897, 2009. [49] S. K. Kamarudin, W. R. W. Daud, Z. Yaakub, Z. Misron, W. Anuar, and N. N. A. N. Yusuf, “Synthesis and optimization of future hydrogen energy infrastructure planning in Peninsular Malaysia,” Int. J. Hydrogen Energy, vol. 34, no. 5, pp. 2077–2088, 2009. [50] S. Samsatli, I. Staffell, and N. J. Samsatli, “Optimal design and operation of integrated wind-hydrogen-electricity networks for decarbonising the domestic transport sector in Great Britain,” Int. J. Hydrogen Energy, 2016. [51] J. H. Han, J. H. Ryu, and I. B. Lee, “Modeling the operation of hydrogen supply networks considering facility location,” Int. J. Hydrogen Energy, 2012. [52] N. Parker, Y. Fan, and J. Ogden, “From waste to hydrogen: An optimal design of energy production and distribution network,” Transp. Res. Part E Logist. Transp. Rev., 2010. [53] J. Kim and I. Moon, “Strategic design of hydrogen infrastructure considering cost and safety using multiobjective optimization,” Int. J. Hydrogen Energy, 2008. [54] S. De-León Almaraz, C. Azzaro-Pantel, L. Montastruc, and M. Boix, “Deployment of a hydrogen supply chain by multi-objective/multi-period optimisation at regional and national scales,” Chem. Eng. Res. Des., vol. 104, pp. 11–31, Dec. 2015. [55] G. Guillén-Gosálbez, F. D. Mele, and I. E. Grossmann, “A bi-criterion optimization approach for the design and planning of hydrogen supply chains for vehicle use,” AIChE J., 2010. [56] Z. Li, D. Gao, L. Chang, P. Liu, and E. N. Pistikopoulos, “Hydrogen infrastructure design and optimization: A case study of China,” Int. J. Hydrogen Energy, vol. 33, no. 20, pp. 137  5275–5286, 2008. [57] K. Deb, K. Sindhya, and J. Hakanen, “Multi-objective optimization,” in Decision Sciences: Theory and Practice, 2016. [58] P. Nunes, F. Oliveira, S. Hamacher, and A. Almansoori, “Design of a hydrogen supply chain with uncertainty,” Int. J. Hydrogen Energy, vol. 40, no. 46, pp. 16408–16418, Dec. 2015. [59] A. Almansoori and N. Shah, “Design and operation of a stochastic hydrogen supply chain network under demand uncertainty,” Int. J. Hydrogen Energy, vol. 37, no. 5, pp. 3965–3977, Mar. 2012. [60] A. J. Kleywegt and A. Shapiro, “Stochastic Optimization,” in Handbook of Industrial Engineering, 2007. [61] A. Almansoori and A. Betancourt-Torcat, “Design of optimization model for a hydrogen supply chain under emission constraints - A case study of Germany,” Energy, vol. 111, pp. 414–429, Sep. 2016. [62] M. Ball, M. Wietschel, and O. Rentz, “Integration of a hydrogen economy into the German energy system: an optimising modelling approach,” Int. J. Hydrogen Energy, vol. 32, no. 10–11, pp. 1355–1368, 2007. [63] V. Artero, F. Chandezon, D. T. Co, and B. Dietzek, “European and International Initiatives in the Field of Artificial Photosynthesis,” Adv. Bot. Res., 2016. [64] B. S. F. Boodhun, A. Mudhoo, G. Kumar, S. H. Kim, and C. Y. Lin, “Research perspectives on constraints, prospects and opportunities in biohydrogen production,” Int. J. Hydrogen Energy, 2017. [65] H. Talebian, O. E. Herrera, M. Tran, and W. Mérida, “Electrification of road freight transport: Policy implications in British Columbia,” Energy Policy, 2018. [66] M. Melaina, O. Antonia, and M. Penev, “Blending Hydrogen into Natural Gas Pipeline Networks: A Review of Key Issues,” Contract, 2013. [67] S. Ould Amrouche, D. Rekioua, T. Rekioua, and S. Bacha, “Overview of energy storage in renewable energy systems,” Int. J. Hydrogen Energy, 2016. [68] A. Bose Styczynski and L. Hughes, “Public policy strategies for next-generation vehicle technologies: An overview of leading markets,” Environ. Innov. Soc. Transitions, 2019. [69] M. Vieweg et al., “Towards Decarbonising Transport: A 2018 Stocktake on Sectoral 138  Ambition in the G20,” 2018. [70] P. Slowik, D. Hall, N. Lutsey, M. Nicholas, and S. Wappelhorst, “Funding the Transition to All Zero-Emission Vehicles,” 2019. [71] Hydrogen Mobility Europe, “Prolonged Tax Exemption for FCEVs in Denmark,” 2019. [Online]. Available: https://h2me.eu/2019/01/03/prolonged-tax-exemption-for-fcevs-in-denmark/. [Accessed: 05-Dec-2019]. [72] J. McKinney, É. Bond, M. Crowell, and E. Odufuwa, Joint Agency Staff Report on Assembly Bill 8: Assessment of Time and Cost Needed to Attain 100 Hydrogen Refueling Stations in California. 2015. [73] M. Nilsson, K. Hillman, A. Rickne, and T. Magnusson, Paving the road to sustainable transport: Governance and innovation in low-carbon vehicles. 2012. [74] S. Yeh, J. Witcover, G. E. Lade, and D. Sperling, “A review of low carbon fuel policies: Principles, program status and future directions,” Energy Policy. 2016. [75] J. H. Stock, “The Renewable Fuel Standard: A Path Forward,” Cent. Glob. Energy Policy, 2015. [76] World Energy Council, “Innovation Insight Brief: New Hydrogen Economy- Hope or Hype?,” 2019. [77] M. Moreno-Benito, P. Agnolucci, W. McDowall, and L. G. Papageorgiou, “Towards a sustainable hydrogen economy: role of carbon price for achieving GHG emission targets,” in Computer Aided Chemical Engineering, 2016. [78] L. Welder, D. S. Ryberg, L. Kotzur, T. Grube, M. Robinius, and D. Stolten, “Spatio-temporal optimization of a future energy system for power-to-hydrogen applications in Germany,” Energy, 2018. [79] S. Samsatli and N. J. Samsatli, “The role of renewable hydrogen and inter-seasonal storage in decarbonising heat – Comprehensive optimisation of future renewable energy value chains,” Appl. Energy, 2019. [80] H. Liu, A. Almansoori, M. Fowler, and A. Elkamel, “Analysis of Ontario’s hydrogen economy demands from hydrogen fuel cell vehicles,” Int. J. Hydrogen Energy, vol. 37, no. 11, pp. 8905–8916, Jun. 2012. [81] H. Talebian, O. E. Herrera, and W. Mérida, “Spatial and temporal optimization of hydrogen 139  fuel supply chain for light duty passenger vehicles in British Columbia,” Int. J. Hydrogen Energy, 2019. [82] M. Moreno-Benito, P. Agnolucci, and L. G. Papageorgiou, “Towards a sustainable hydrogen economy: Optimisation-based framework for hydrogen infrastructure development,” Comput. Chem. Eng., vol. 102, pp. 110–127, Jul. 2017. [83] N. Sabio, A. Kostin, G. Guillén-Gosálbez, and L. Jiménez, “Holistic minimization of the life cycle environmental impact of hydrogen infrastructures using multi-objective optimization and principal component analysis,” Int. J. Hydrogen Energy, vol. 37, no. 6, pp. 5385–5405, 2012. [84] J.-H. Han, J.-H. Ryu, and I.-B. Lee, “Multi-objective optimization design of hydrogen infrastructures simultaneously considering economic cost, safety and CO2 emission,” Chem. Eng. Res. Des., vol. 91, no. 8, pp. 1427–1439, Aug. 2013. [85] M. Dayhim, M. A. Jafari, and M. Mazurek, “Planning sustainable hydrogen supply chain infrastructure with uncertain demand,” Int. J. Hydrogen Energy, vol. 39, no. 13, pp. 6789–6801, Apr. 2014. [86] S. K. Kamarudin, W. R. W. Daud, Z. Yaakub, Z. Misron, W. Anuar, and N. N. A. N. Yusuf, “Synthesis and optimization of future hydrogen energy infrastructure planning in Peninsular Malaysia,” Int. J. Hydrogen Energy, 2009. [87] A. Hugo, P. Rutter, S. Pistikopoulos, A. Amorelli, and G. Zoia, “Hydrogen infrastructure strategic planning using multi-objective optimization,” Int. J. Hydrogen Energy, vol. 30, no. 15, pp. 1523–1534, 2005. [88] N. V. S. N. Murthy Konda, N. Shah, and N. P. Brandon, “Optimal transition towards a large-scale hydrogen infrastructure for the transport sector: The case for the Netherlands,” Int. J. Hydrogen Energy, vol. 36, no. 8, pp. 4619–4635, 2011. [89] C. Stiller, U. Bünger, S. Møller-Holst, A. M. Svensson, K. A. Espegren, and M. Nowak, “Pathways to a hydrogen fuel infrastructure in Norway,” Int. J. Hydrogen Energy, vol. 35, no. 7, pp. 2597–2601, Apr. 2010. [90] BC Hydro, “The Integrated Resource Plan (IRP),” 2013. [Online]. Available: https://www.bchydro.com/energy-in-bc/planning-for-our-future/irp/current-plan/document-centre/reports/november-2013-irp.html. [Accessed: 10-Feb-2017]. 140  [91] Industrial Forestry Service, “Wood Based Biomass in British Columbia and its Potential for New Electricity Generation,” 2015. [92] B. Stennes, K. Niquidet, and G. C. van Kooten, “Modelling Bioenergy Uptake in the British Columbia Fibre Allocation and Transport Model,” 2011. [93] Golder Associates, “Inventory of Greenhouse Gas Generation from Landfills in British Columbia,” 2008. [94] Landfill Gas Energy Project Development Handbook. U.S. Environmental Protection Agency, 2016. [95] Ministry of Energy Mines and Petroleum Resources Government of British Columbia, “British Columbia Hydrogen Study,” 2019. [96] NRCan, “British Columbia’s Shale and Tight Resources,” Natural Resources Canada, 2017. [Online]. Available:  http://www.nrcan.gc.ca/energy/sources/shale-tight-resources/17692. [Accessed: 30-Jul-2017]. [97] Ministry of Energy and Mines, “British Columbia Coal Industry Overview,” 2016. [98] US Department of Energy, “Carbon Storage Atlas- Fifth Edition,” 2015. [99] H2USA, “National Hydrogen scenarios, How many stations, where and when?,” 2017. [100] J. C. Molburg and R. D. Doctor, “Hydrogen from Steam-Methane Reforming with CO 2 Capture,” 20th Annu. Int. Pittsburgh Coal Conf., 2003. [101] NREL, “H2A Production Model, Version 3.” [Online].  Available: http://www.hydrogen.energy.gov/h2a_production.html. [Accessed: 10-Jan-2019]. [102] D. Y. C. Leung, G. Caramanna, and M. M. Maroto-Valer, “An overview of current status of carbon dioxide capture and storage technologies,” Renewable and Sustainable Energy Reviews. 2014. [103] T. Ramsden, M. Ruth, and V. Diakov, “Hydrogen Pathways: Updated Cost, Well-to-Wheels Energy Use, and Emissions for the Current Technology Status of Ten Hydrogen Production, Delivery, and Distribution Scenarios,” 2013. [104] G. Gahleitner, “Hydrogen from renewable electricity: An international review of power-to-gas pilot plants for stationary applications,” International Journal of Hydrogen Energy. 2013. [105] D. Braxenholm, “By-Product Hydrogen to Fuel Cell Vehicles : A Summary of Available 141  Purification Methods for Surplus Hydrogen Gas from Sodium Chlorate Production, and Status of Hydrogen Purity Analyses,” 2016. [106] U.S. Department of Energy, “Hydrogen delivery scenario analysis model (HDSAM 3.1),” 2015. [Online]. Available: https://www.hydrogen.energy.gov/h2a_delivery.html. [107] H. P. Bloch, Compressors and Modern Process Applications. 2006. [108] K. Reddi, A. Elgowainy, N. Rustagi, and E. Gupta, “Techno-economic analysis of conventional and advanced high-pressure tube trailer configurations for compressed hydrogen gas transportation and refueling,” Int. J. Hydrogen Energy, vol. 43, no. 9, pp. 4428–4438, Mar. 2018. [109] NREL, “Hydrogen Station Compression, Storage and Dispensing - Technical Status and Costs,” 2014. [110] K. Reddi, A. Elgowainy, and E. Sutherland, “Hydrogen refueling station compression and storage optimization with tube-trailer deliveries,” Int. J. Hydrogen Energy, 2014. [111] A. Elgowainy, K. Reddi, D. Y. Lee, N. Rustagi, and E. Gupta, “Techno-economic and thermodynamic analysis of pre-cooling systems at gaseous hydrogen refueling stations,” Int. J. Hydrogen Energy, 2017. [112] BC Hydro, “BC Hydro’s Greenhouse Gas Intensities 2004-2014,” 2014. [Online]. Available: https://www.bchydro.com/content/dam/BCHydro/.../ghg-intensities-2004-2014.pdf%0A. [Accessed: 19-Jan-2017]. [113] E. Cetinkaya, I. Dincer, and G. F. Naterer, “Life Cycle Assessment of Various Hydrogen Production Methods,” Int. J. Hydrogen Energy, vol. 37, no. 3, pp. 2071–2080, 2012. [114] M. Melaina, M. Penev, and D. Heimiller, “Resource Assessment for Hydrogen Production, Hydrogen Production Potential from Fossil and Renewable Energy Resources,” 2013. [115] NRCan, “GHGenius Model 4.03.” Natural Resources Canada (NRCan), 2013. [116] P. Lund, “Market penetration rates of new energy technologies,” Energy Policy, 2006. [117] J. Heywood and D. MacKenzie, “On the Road toward 2050:Potential for Substantial Reductionsin Light-Duty Vehicle Energy Useand Greenhouse Gas Emissions,” 2015. [118] ICBC, “Quick Statistics,” 2018. [119] The Ministry of Transportation and Infrastructure, “Traffic Data Program.” [Online]. Available: https://www.th.gov.bc.ca/trafficData/. [Accessed: 15-Jan-2019]. 142  [120] metrovancouver, “Metro Vancouver 2040 Shaping Our Future, Regional Growth StrategyBylaw No.1136.” [121] Government of Canada, “Accelerated Investment Incentive,” 2018. [Online]. Available: https://www.canada.ca/en/revenue-agency/services/tax/businesses/topics/sole-proprietorships-partnerships/report-business-income-expenses/claiming-capital-cost-allowance/accelerated-investment-incentive.html. [Accessed: 10-Oct-2019]. [122] Government of B.C., “Renewable and Low Carbon Fuel Requirements Regulation, B.C. Reg. 394/2008,” Government of British Columbia, 2008. [Online]. Available: http://www.bclaws.ca/EPLibraries/bclaws_new/document/ID/freeside/394_2008. [Accessed: 15-Feb-2017]. [123] Government of B.C., “Credit Transfers,” 2018. [Online]. Available: https://www2.gov.bc.ca/gov/content/industry/electricity-alternative-energy/transportation-energies/renewable-low-carbon-fuels/credits-transfers. [Accessed: 01-Dec-2018]. [124] Ministry of Energy Mines and Petroleum Resources Government of British Columbia, “Part 3 Agreement Programs for 2017 and 2018,” 2017. [125] Government of B.C., “British Columbia’s Revenue-Neutral Carbon Tax,” Government of British Columbia, 2008. [Online]. Available: http://www2.gov.bc.ca/gov/content/environment/climate-change/planning-and-action/carbon-tax. [Accessed: 20-Dec-2016]. [126] Environment and Climate Change Canada, “Technical update to Environment and Climate Change Canada’s social cost of greenhouse gas estimates,” 2016. [Online]. Available: http://www.ec.gc.ca/cc/default.asp?lang=En&n=BE705779-1. [Accessed: 01-Mar-2019]. [127] Government of Manitoba, “The Emissions Tax on Coal and Petroleum Coke Act,” Manitoba Tax Publications, 2015. [Online]. Available: http://web2.gov.mb.ca/laws/statutes/ccsm/e090e.php. [Accessed: 11-Oct-2019]. [128] Ministry of the Environment (Japan), “Details on the Carbon Tax (Tax for Climate Change Mitigation),” 2012. [129] U.S. Department of Agriculture, “USDA Awards Funds to Expand, Accelerate Wood Energy and Wood Products Markets in 19 States,” 2016. [Online]. Available: https://www.usda.gov/media/press-releases/2016/05/13/usda-awards-funds-expand-143  accelerate-wood-energy-and-wood-products. [Accessed: 25-Oct-2019]. [130] US department of Energy, “Smart Grid Investment Grant Program,” 2007. [Online]. Available: https://www.smartgrid.gov/recovery_act/overview/smart_grid_investment_grant_program.html. [Accessed: 20-Nov-2019]. [131] California Public Utilities Commission, “California Solar Initiative Program Handbook,” 2017. [132] Independent Electricity System Operator, “Save on Energy,” 2019. [Online]. Available: https://saveonenergy.ca/. [Accessed: 10-Nov-2019]. [133] Independent Electricity System Operator, “Feed-in Tariff Program.” [Online]. Available: http://www.ieso.ca/sector-participants/feed-in-tariff-program/overview. [Accessed: 10-Nov-2019]. [134] Soalr Choice, “Feed-in Tariff Scheme announced by Western Australia,” 2010. [Online]. Available: https://www.solarchoice.net.au/blog/feed-in-tariff-scheme-announced-by-western-australia/. [Accessed: 20-Nov-2019]. [135] E. Waffenschmidt, “The Renewable Energy Act in Germany: Its basic idea and recent developments,” in 7th International 100% Renewable Energy Conference (IRENEC 2017), 2017. [136] K. Wikberg, “Premium tariff (Law on the Promotion of Renewable Energy),” 2019. [Online]. Available: http://www.res-legal.eu/search-by-country/denmark/single/s/res-e/t/promotion/aid/premium-tariff-law-on-the-promotion-of-renewable-energy/lastp/96/. [Accessed: 01-Nov-2019]. [137] H. Vidalic, “Feed-in tariff (Tarif d´achat),” 2019. [Online]. Available: http://www.res-legal.eu/search-by-country/france/single/s/res-e/t/promotion/aid/feed-in-tariff-tarif-dachat/lastp/131/. [Accessed: 01-Nov-2019]. [138] A. Taylor and M. Raynolds, “Government Policies for Promoting and Managing the Production and Consumption of Bioenergy,” 2003. [139] Energy efficiency Alberta, “Green Loan Guarantee Program,” 2017. [Online]. Available: https://efficiencyalberta.ca/financing/green-loan-guarantee-program. [Accessed: 05-Dec-2019]. 144  [140] International Finance Corporation, “RenovAr (Argentina): Scaling ‘Express Edition,” 2019. [141] B. D. Yacobucci, “Biofuels Incentives: A Summary of Federal Programs,” 2012. [142] M. F. Sherlock, “The Renewable Electricity Production Tax Credit: In Brief,” 2018. [143] P. Folger and M. F. Sherlock, “Clean Coal Loan Guarantees and Tax Incentives: Issues in Brief,” 2014. [144] Solar energy Industries Association, “Solar Investment Tax Credit (ITC),” 2006. [Online]. Available: https://www.seia.org/initiatives/solar-investment-tax-credit-itc. [Accessed: 20-Nov-2019]. [145] San Francisco Water Power Sewer, “GoSolarSF Program: Program Handbook,” 2019. [146] Ministry of the Ecological and Inclusive Transition, “Financial aid for energy renovation,” 2019. [Online]. Available: https://www.ecologique-solidaire.gouv.fr/aides-financieres-renovation-energetique. [147] Government of Alberta, “Capital Investment Tax Credit (CITC),” 2017. [Online]. Available: https://www.alberta.ca/capital-investment-tax-credit.aspx. [Accessed: 25-Nov-2019]. [148] HM Revenue & Customs, “Excise Notice 179e: biofuels and other fuel substitutes,” 2019. [Online]. Available: https://www.gov.uk/government/publications/excise-notice-179e-biofuels-and-other-fuel-substitutes/excise-notice-179e-biofuels-and-other-fuel-substitutes. [Accessed: 12-Nov-2019]. [149] K. Ignaciuk, “Tax regulation mechanism,” 2019. [Online]. Available: http://www.res-legal.eu/search-by-country/poland/single/s/res-e/t/promotion/aid/tax-regulation-mechanism/lastp/175/. [Accessed: 10-Nov-2019]. [150] E. & I. S. Department for Business, “Energy Technology List (ETL),” 2019. [Online]. Available: https://www.gov.uk/guidance/energy-technology-list. [Accessed: 12-Nov-2019]. [151] IRS, “Publication 946: How To Depreciate Property,” 2018. [152] T. Sternkopf, “Loan (KfW Renewable Energy Programme – Standard),” 2019. [Online]. Available: http://www.res-legal.eu/search-by-country/germany/single/s/res-e/t/promotion/aid/loan-kfw-renewable-energy-programme-standard/lastp/135/. [Accessed: 10-Dec-2019]. 145  [153] International Energy Agency (IEA), “Preferential loans for energy saving measures,” 2019. [Online]. Available: https://www.iea.org/policies/1806-preferential-loans-for-energy-saving-measures. [Accessed: 10-Dec-2019]. [154] Federation of Canadian Municipalities (FCM), “Green Municipal Fund: Five-Year Plan,” 2018. [155] U.S. Department of Energy, “Advanced Technology Vehicles Manufacturing (ATVM) direct loan program,” 2007. [Online]. Available: https://www.energy.gov/lpo/advanced-technology-vehicles-manufacturing-atvm-loan-program. [Accessed: 20-Nov-2019]. [156] BC hydro, “Power Smart Partners Express Program,” 2010. [Online]. Available: https://www.bchydro.com/ecatalog/external/doc/A10-163 PSP Express Eligible Products List.hi (april 29).pdf. [Accessed: 20-Oct-2019]. [157] Centre for Renewable Energy Sources and Saving, “Energy Efficiency trends and policies in Greece,” 2018. [158] Gouvernement of Québec, “Quebec Cap-and-Trade System for Greenhouse Gas Emission Allowances (C&T),” 2018. [159] RGGI Inc., “Regional Greenhouse Gas Initiative (RGGI),” 2019. [Online]. Available: https://www.rggi.org/. [Accessed: 05-Dec-2019]. [160] European Commission, “EU Emissions Trading System (EU ETS),” 2005. [Online]. Available: https://ec.europa.eu/clima/policies/ets_en#tab-0-0. [Accessed: 10-Dec-2019]. [161] L.-G. Giraudet, L. Bodineau, and D. Finon., “The Costs and Benefits of White Certificates Schemes,” 2011. [162] China Development and Reform Commission, “Notice of the Three Departments on the Trial Implementation of the Renewable Energy Green Power Certificate Issuance and Voluntary Subscription Trading System,” 2017. [Online]. Available: http://www.gov.cn/xinwen/2017-02/03/content_5164836.htm. [Accessed: 05-Nov-2019]. [163] Norwegian Water Resources and Energy Directorate & and Swedish Energy Agency, “The Norwegian-Swedish Electricity Certificate Market,” 2015. [164] California Air Resource Board, “Low Carbon Fuel Standard,” 2011. [Online]. Available: https://ww3.arb.ca.gov/fuels/lcfs/lcfs.htm. [Accessed: 20-Nov-2019]. [165] IRENA, The Power to Change: Solar and Wind Cost Reduction Potential to 2025. 2016. 146  [166] M. P. Keightley, D. J. Marples, and Molly F. Sherlock, “Tax Equity Financing: An Introduction and Policy Considerations,” 2019. [167] R. S. Pindyck, “The social cost of carbon revisited,” J. Environ. Econ. Manage., 2019. [168] K. Ricke, L. Drouet, K. Caldeira, and M. Tavoni, “Country-level social cost of carbon,” Nat. Clim. Chang., 2018. [169] H. Talebian, O. E. Herrera, M. Tran, and W. Mérida, “Potential for Hydrogen as a Transportation Fuel in British Columbia: Resource Assessment and GHG Emissions Analysis,” in Transportation Research Board 97th Annual Meeting, 2018. [170] C. Yang and J. Ogden, “Determining the Lowest-cost Hydrogen Delivery Mode,” Int. J. Hydrogen Energy, vol. 32, no. 2, pp. 268–286, Feb. 2007. [171] Research and Traffic Group, “Environmental and Social Impacts of Marine Transport in the Great Lakes-St. Lawrence Seaway Region,” 2013. [172] Government of B.C., “Provincial Greenhouse Gas Emissions Inventory,” 2019. [Online]. Available:https://www2.gov.bc.ca/gov/content/environment/climate-change/data/provincial-inventory. [Accessed: 01-Nov-2019]. [173] OECD, “Moving Freight with Better Trucks: Improving Safety, Productivity and Sustainability,” 2011. [174] CalSTA, “California Freight Mobility Plan,” 2014. [175] Canada Gazette Part II, “Canadian Environmental Protection Act,1999: Heavy-duty Vehicle and Engine Greenhouse Gas Emission Regulations,” vol. 147, no. 6, 2013. [176] Canada Gazette Part II, “Regulations Amending the Heavy-duty Vehicle and Engine Greenhouse Gas Emission Regulations and Other Regulations Made Under the Canadian Environmental Protection Act, 1999,” vol. 151, no. 9, 2017. [177] J. Wolinetz and M. Moorhouse, “Biofuels in Canada: Tracking Progress in Tackling Greenhouse Gas Emissions from Transportation Fuels,” 2016. [178] D. Tomes, P. Lakshmanan, and D. Songstad, Eds., Biofuels. Springer-Verlag New York, 2011. [179] M. Girvan, J. , & Hall, “BC Fibre Model User Guide,” 2008. [180] BC Hydro, “Integrated Resource Plan Appendix 6B-1 Greenhouse Gas Reduction Scenarios for the Western Interconnection: (2010-2050),” 2011. 147  [181] Government of B.C., “Clean Energy Act Greenhouse Gas Reduction (Clean energy) Regulation B.C. Reg. 102/2012,” 2012. [Online]. Available: http://www.bclaws.ca/civix/document/id/complete/statreg/102_2012. [182] Natural Gas Use in Transportation Roundtable, “Natural Gas Use in the Canadian Transportation Sector: Deployment Roadmap,” 2010. [183] H. McJeon et al., “Limited Impact on Decadal-scale Climate Change from Increased Use of Natural Gas,” Nature, vol. 514, no. 7523, pp. 482–485, 2014. [184] R. W. Howarth, “A Bridge to Nowhere: Methane Emissions and the Greenhouse Gas Footprint of Natural Gas,” Energy Sci. Eng., vol. 2, no. 2, pp. 47–60, 2014. [185] A. Venkatesh, P. Jaramillo, W. M. Griffin, and H. S. Matthews, “Uncertainty in Life Cycle Greenhouse Gas Emissions from United States Natural Gas End-uses and its Effects on Policy.,” Environ. Sci. Technol., vol. 45, no. 19, pp. 8182–9, 2011. [186] R. K. Mehra, H. Duan, R. Juknelevičius, F. Ma, and J. Li, “Progress in Hydrogen Enriched Compressed Natural Gas (HCNG) Internal Combustion Engines - A Comprehensive Review,” Renewable and Sustainable Energy Reviews, vol. 80. pp. 1458–1498, 2017. [187] NREL, “Clean Cities Guide to Alternative Fuel and Advanced Medium- and Heavy-Duty Vehicles,” National Renewable Energy Laboratory (NREL), Golden, CO., 2013. [188] E. den Boer, J. Pagenkopf, S. Aarnink, and F. Kleiner, “Zero Emissions Trucks An Overview of State-of-the-art Technologies and their Potential,” 2013. [189] T. Hua et al., “Status of hydrogen fuel cell electric buses worldwide,” J. Power Sources, vol. 269, pp. 975–993, 2014. [190] L. Eudy, M. Post, and C. Gikakis, “Fuel Cell Buses in U.S. Transit Fleets: Current Status 2015,” 2015. [191] F. Lambert, “BYD Announces New Electric Truck Assembly Factory in Canada,” 2017. [Online]. Available: https://electrek.co/2017/11/15/byd-new-electric-truck-assembly-factory-canada/. [Accessed: 05-Dec-2017]. [192] F. Lambert, “Tesla Semi,” 2017. [Online]. Available: https://electrek.co/guides/tesla-semi/. [Accessed: 15-Dec-2017]. [193] T. HSU, “Nikola Tweaks Hydrogen Truck Design, Raises Funding,” 2017. [Online]. Available: https://www.trucks.com/2017/06/26/nikola-electric-truck-redesign/. [Accessed: 148  05-Oct-2017]. [194] J. O’Dell, “Toyota’s Heavy-Duty Fuel Cell Truck Finally Hits the Road,” 2017. [Online]. Available: https://www.trucks.com/2017/10/12/toyota-hydrogen-fuel-cell-electric-truck-hits-road/. [Accessed: 05-Nov-2017]. [195] T. N. Veziroǧlu and S. Şahin, “21st Century’s Energy: Hydrogen Energy System,” Energy Convers. Manag., vol. 49, no. 7, pp. 1820–1831, 2008. [196] U.S. Department of Energy, “State of the States: Fuel cells in America 2013,” 2013. [197] ARB, “Technology Assessment: Medium- and Heavy-duty Fuel Cell Electric Vehicles,” 2015. [198] CaFCP, “Medium- & Heavy-duty Fuel Cell Electric Truck Action Plan for California,” 2016. [199] J. Eom, L. Schipper, and L. Thompson, “We Keep on truckin’: Trends in Freight Energy Use and Carbon Emissions in 11 IEA Countries,” Energy Policy, vol. 45, pp. 327–341, 2012. [200] Y.-H. Pai, J.-H. Ke, H.-F. Huang, C.-M. Lee, J.-M. Zen, and F.-S. Shieu, “CF4 plasma treatment for preparing gas diffusion layers in membrane electrode assemblies,” J. Power Sources, vol. 161, no. 1, pp. 275–281, Oct. 2006. [201] T. Limanond, S. Jomnonkwao, and A. Srikaew, “Projection of future transport energy demand of Thailand,” Energy Policy, vol. 39, no. 5, pp. 2754–2763, 2011. [202] Canada Gazette Part II, “Regulations Amending the Passenger Automobile and Light Truck Greenhouse Gas Emission Regulations SOR/2014-207,” vol. 148, no. 21, 2014. [203] Canada Gazette Part II, “Passenger Automobile and Light Truck Greenhouse Gas Emission Regulations (SOR/2010-201),” vol. 144, no. 21, 2010. [204] Government of B.C., “Weigh2GoBC,” 2017. [Online]. Available: https://www2.gov.bc.ca/gov/content/transportation/vehicle-safety-enforcement/services/weigh2gobc-join. [Accessed: 01-Dec-2017]. [205] R. S. Poudyal et al., Compendium of Hydrogen Energy. 2015. [206] BC Hydro, “BC Hydro’s Service Plan 2017/18 – 2019/20,” 2017. [207] E. Helmers and P. Marx, “Electric Cars: Technical Characteristics and Environmental Impacts,” Environ. Sci. Eur., vol. 24, no. 14, pp. 1–15, 2012. 149  [208] C. E. Thomas, “Fuel cell and Battery Electric Vehicles Compared,” Int. J. Hydrogen Energy, vol. 34, no. 15, pp. 6005–6020, 2009. [209] J. I. Levene, M. K. Mann, R. M. Margolis, and A. Milbrandt, “An Analysis of Hydrogen Production from Renewable Electricity Sources,” Sol. Energy, vol. 81, no. 6, pp. 773–780, 2007. [210] U.S. DRIVE Partnership, “Hydrogen Delivery Technical Team Roadmap,” 2013. [211] Argonne National Laboratory, “GREET (The Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation Model)- GREET_1_2016.” 2016. [212] T. Fridriksson, A. Mateos, P. Audinet, and Y. Orucu, “Green house Gases from Geothermal Power Production: Interim Technical Note,” 2016. [213] R. Edwards, J.-F. Larivé, D. Rickeard, and W. Weindorf, “Well-to-Wheels Analysis of Future Automotive Fuels and Powertrains in the European Context, Well-to-Tank (WTT) Report – Appendix 2, Version 4a,” 2014. [214] S. Ramachandran and U. Stimming, “Well to Wheel Analysis of Low Carbon Alternatives for Road Traffic,” R. Soc. Chem., vol. 8, no. 8, pp. 3313–3324, 2015. [215] K. Damen, M. Van Troost, A. Faaij, and W. Turkenburg, “A comparison of electricity and hydrogen production systems with CO 2 capture and storage. Part A: Review and selection of promising conversion and capture technologies,” Prog. Energy Combust. Sci., vol. 32, no. 2, pp. 215–246, 2006. [216] C. He, H. Sun, Y. Xu, and S. Lv, “Hydrogen Refueling Station Siting of Expressway Based on the Optimization of Hydrogen Life Cycle Cost,” Int. J. Hydrogen Energy, vol. 42, no. 26, pp. 16313–16324, 2017. [217] Government of B.C., “Site C Dam Information & Updates,” 2017. [Online]. Available: https://www2.gov.bc.ca/gov/content/industry/electricity-alternative-energy/electricity/site-c. [Accessed: 15-Dec-2017]. [218] BC Hydro, “Site C Clean Energy Project,” 2017. [Online]. Available: https://www.sitecproject.com/. [Accessed: 05-Jul-2017]. [219] APS Physics, “Integrating Renewable Electricity on the Grid,” 2010. [220] A. Al-Subaie, A. Maroufmashat, A. Elkamel, and M. Fowler, “Presenting the Implementation of Power-to-Gas to an Oil Refinery as a Way to Reduce Carbon Intensity 150  of Petroleum Fuels,” Int. J. Hydrogen Energy, vol. 42, no. 30, pp. 19376–19388, 2017. [221] A. Maroufmashat and M. Fowler, “Transition of Future Energy System Infrastructure; Through Power-to-Gas Pathways,” Energies, vol. 10, no. 8. 2017. [222] Tractebel Engineering and Hinicio Consulting, “Study on Early Business Cases for H2 in Energy Storage & More Broadly Power to H2 Applications,” 2017. [223] D. Sawyer, R. Harding, C. Pozlott, and P. Dickey, “Carbon Capture and Storage — The Environmental and Economic Case and Challenges,” 2008. [224] C. A. Coello Coello, “Multi-objective optimization,” in Handbook of Heuristics, 2018. [225] E. Løken, “Use of multicriteria decision analysis methods for energy planning problems,” Renewable and Sustainable Energy Reviews. 2007.  151  Appendix A   A.1 Hydrogen demand projection Table A.1 Hydrogen demand (kg/day) distribution among municipalities in the final year of each time step for different demand scenarios: time 2025 2030 2035 2040 2045 2050 Scenario/Municipality Pes Mod Opt Pes Mod Opt Pes Mod Opt Pes Mod Opt Pes Mod Opt Pes Mod Opt Surrey 17 52 86 97 290 483 388 1165 1942 994 2981 4969 1723 5169 8615 2272 6817 11361 Vancouver 51 152 253 271 814 1356 1049 3148 5246 2594 7781 12968 4357 13071 21786 5581 16744 27906 Burnaby 28 85 141 156 467 778 615 1846 3076 1548 4643 7738 2637 7911 13185 3416 10248 17080 Coquitlam 20 59 99 110 331 551 436 1309 2181 1089 3268 5447 1828 5484 9140 2314 6943 11572 Langley 20 61 101 114 342 571 460 1379 2298 1174 3523 5871 2027 6081 10134 2655 7965 13274 Delta 15 46 77 83 248 414 318 954 1591 777 2330 3884 1283 3848 6414 1620 4859 8098 Maple Ridge 15 42 70 77 231 385 299 898 1497 730 2191 3652 1200 3601 6002 1505 4515 7525 North Vancouver 21 62 103 111 333 554 432 1297 2162 1078 3233 5388 1827 5480 9134 2362 7086 11811 Richmond 35 80 133 145 436 726 574 1723 2872 1450 4351 7252 2488 7465 12442 3255 9765 16276 West Vancouver 0 24 41 44 132 221 172 515 858 424 1271 2119 707 2122 3536 897 2690 4483  Kelowna 39 118 197 216 647 1079 847 2542 4236 2117 6352 10587 3585 10756 17926 4621 13863 23105 Kamloops 24 71 118 129 388 647 508 1525 2542 1270 3811 6352 2151 6454 10756 2773 8318 13863 Prince George 24 71 118 129 388 647 508 1525 2542 1270 3811 6352 2151 6454 10756 2773 8318 13863 Victoria 67 201 335 367 1100 1834 1440 4321 7202 3600 10799 17998 6095 18285 30475 7856 23567 39278  Abbotsford 0 18 30 33 98 164 129 386 643 321 964 1607 544 1632 2720 701 2104 3506 Hope 0 0 15 15 41 68 54 161 268 134 402 669 227 680 1134 292 877 1461 Whistler 0 0 0 0 30 50 39 118 196 98 295 491 166 499 831 214 643 1071 Williams Lake 0 0 0 0 33 55 43 130 217 108 325 542 184 551 918 237 710 1183       152  A.2 Distances between supply and demand regions   Google Maps was used to estimate distance in km, based on the zip code of the regions.  Table A.2 Distances between potential production and storage locations (1-14) and distances between potential storage locations and the entrance to Metro Vancouver municipalities (Langley Township).   Table A.3 Distances between Langley Township (LT) and different municipalities in Metro Vancouver and distances between storage facilities in North Vancouver (NV) and different municipalities in the Metro Vancouver  1: Surrey 2: Vancouver 3: Burnaby 4: Coquitlam 5: Langley 6: Delta 7: Maple Ridge 8: North Vancouver 9: Richmond 10: West Vancouver LT 30 62 44 9 55 27 37 44 60 70 NV 32 18 12 45 27 43 30 48 6 15          1 2 3 4 5 6 7 8 9 10 11 12 13 14 LT 1: Fort Nelson 0 381 809 1048 1572 1526 1333 1360 1487 1712 1471 1439 1680 1593 1541 2: Fort St John  0 437 676 1193 1154 961 988 1115 1334 1093 1067 1308 1221 1169 3: Prince George   0 239 870 718 524 552 679 976 864 631 872 784 733 4: Williams Lake    0 637 952 290 318 445 743 881 397 638 551 499 5: Mica Creek     0 1583 351 436 336 389 524 552 793 705 653 6: Prince Rupert      0 1236 1264 1391 1689 1582 1343 1584 1497 1445 7: Kamloops       0 87 167 457 595 203 444 356 307 8: Merritt        0 127 456 680 120 361 273 220 9: Kelowna         0 346 558 239 479 392 339 10: Nelson          0 259 510 750 662 609 11: Kimberley           0 721 961 874 822 12: Hope            0 242 154 103 13: Victoria             0 126 N/A 14: North Vancouver              0 N/A 153  Table A.4 Distances between potential storage locations and the demand regions (except Metro Vancouver)   Major Municipalities Connecting municipalities  1:  Kelowna 2: Kamloops 3:  Prince George 4: Victoria 1: Abbotsford 2: Hope 3: Whistler 4:  Williams Lake 1: Fort Nelson 1487 1330 810 1711 1522 1439 1440 1048 2: Fort St John 1115 958 1154 1341 1151 1067 1068 676 3: Prince George 687 522 10 900 715 631 632 240 4: Williams Lake 453 288 240 670 481 397 398 10 5: Mica Creek 336 350 871 825 635 553 648 637 6: Prince Rupert 1391 1234 718 1616 1426 1343 1344 952 7: Kamloops 167 10 525 467 289 204 302 290 8: Merritt 127 87 552 400 202 120 291 318 9: Kelowna 10 167 687 511 321 239 417 445 10: Nelson 346 551 1020 781 591 509 754 743 11: Kimberley 580 595 846 1070 879 797 893 881 12: Hope 239 204 631 275 85 10 272 397 13: Victoria N/A N/A N/A 111 N/A N/A N/A N/A 14: North Vancouver 388 352 781 100 70 150 124 547            154  A.3 Other parameters used in the model 𝑃𝑐𝑎𝑝_𝑚𝑎𝑥_𝑝𝑟𝑖𝑛𝑐  100 ,50, 10 (tonnes/day) 𝑆𝑐𝑎𝑝_𝑚𝑎𝑥_𝑝𝑟𝑖𝑛𝑐̅ 100 ,50, 10 (tonnes/day) 𝐷𝑐𝑎𝑝_𝑚𝑎𝑥𝑠 1500, 1000, 500, 150 (kg/day) 𝛼_𝐿𝑅_𝑆 0.07 𝛼_𝐿𝑅_𝐶 0.07 𝛼_𝐿𝑅_𝑂 0.106 𝛼_𝐿𝑅_𝐷 0.106 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝐶 40 years 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝑆 20 years 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝑂 10 years 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝐷 10 years 𝐿𝑖𝑓𝑒𝑡𝑖𝑚𝑒_𝑇𝑅 20 years Truck parameters  𝛼 25 C$/hour 𝛾 0.3    litre/km 𝛽  1.4   C$ /litre 𝑉𝐻 70 km/hour 𝑉𝐺 40 km/hour 𝐶𝐴𝑃𝐿_𝑇𝑅 3800 kg 𝐶𝐴𝑃𝐺_𝑇𝑅𝑎 100,500, 900 kg 𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑡𝑖𝑚𝑒𝑑 d=2: 1.5 hour    d=3: 3 hour 𝑢𝑛𝐿𝑜𝑎𝑑𝑖𝑛𝑔𝑡𝑖𝑚𝑒𝑑 d=2: 1.5 hour    d=3: 3.5 hour 𝐻2_𝐷 120 MJ/kg 𝐺𝑎𝑠_𝐶𝐼 79.33 g/MJ 𝐸𝐸𝑅𝑡 2.5 𝑟 10% 𝑁 30 years 𝑇𝑟 27% timestep 5    

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0392425/manifest

Comment

Related Items