PROSPECTIVE LIFE-CYCLE ASSESSMENT OF SECOND-LIFE ELECTRIC VEHICLE BATTERIES AND URANIUM EX TRACTION IN THE US By Dipti Kamath A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Environmental Engineering Œ Doctor of Philosophy Environmental Science and Policy Œ Dual Major 2020 ABSTRACT PROSPECTIVE LIFE-CYCLE ASSESSMENT OF SECOND-LIFE ELECTRIC VEHICLE BATTERIES AND URANIUM EX TRACTION IN THE US By Dipti Kamath Large-scale integration of renewable energy in the electricity grid creates issues such as intermittency and lack of load and peak matching. Battery stor age and nuclear energy are both low-carbon options that can supplement variable el ectricity generation and will be necessary for large-scale renewables deploymen t. However, with changes in resources and technology, it is important to anticipate future issues that might arise with these energy options to ensure the low cost and carbon footprint of electricity. In this di ssertation, the main aim is to conduct a prospective life-cycle assessment (LCA) of sec ond-life electric vehicle batteries (SLBs) as energy storage and uranium extraction for nuclear energy in the US, to identify and mitigate unintended consequences. Current battery technologies, in particular lithium-ion batteries, are expensive and can increase the carbon footprint of the grid due to charge-discharge losses. A possible cheaper and greener alternative for energy st orage is remanufactured SLBs th at have reached end-of-life (EOL). With the increase in electric vehicle (EV) sales, a large number of batteries are expected to reach EOL in the near future. Although SLBs can have 70 to 80% of remaining capacity, it is unknown whether they will perform at par with new ba tteries in various applications. It is also unknown whether SLBs provide cost and carbon emission reduction compared to new batteries, nor is there any information on SLB demand. A ccelerated life testing was conducted on EOL EV battery cells to assess their performance in residential energy storage, commercial fast-charging, and utility-level peak shaving applications. A LCA and an ec onomic evaluation were conducted to calculate the life-cycle carbon footprint and leveliz ed cost of electricity of using SLBs and new batteries in the above-mentioned applications . A system dynamics mode l was developed to compare the cost, carbon footprint and material requirement of EV battery recycling and remanufacturing in the US from 2017 to 2050. Residential energy storage performed the best during the accelerated life testing of battery cells. SLB use instead of new batteries can reduce the levelized costs and carbon footprint for all three applications. Remanufacturing reduced the life- cycle carbon footprint of batteries by 2 to 16% compared to r ecycling only. The economic value of remanufacturing is expected to decrease over time with the decreasing price of new batteries, necessitating policies that can incentivize SLB uptake, given that the SLB mark et is still emerging. In contrast, nuclear energy is a mature t echnology in use since th e 1950s. Over time, uranium ore grades have decreased globally, and open-pit and underground mining methods are being replaced by in-situ leaching (ISL)Šalkaline and acidic. Alkaline ISL is most prominent in the US and is promising for many recently discov ered uranium deposits. As future uranium mines can be expected to use alkaline ISL, assessing its environmental impacts is necessary to assess those of future nuclear power plants in the wo rld. Currently, there is no environmental LCA for alkaline ISL. We performed the first LCA of alkaline ISL in the US for extracting US-based uranium ores of grade 0.036-0.4% U 3O8. The alkaline ISL carbon footprint was found to be almost twice those reported for acidic ISL, but lower than open-pit and undergro und methods. The results indicate a risk of increasing th e carbon footprint of future nuclear energy generation. Similarly, the results for SLB-based applic ations show lower cost and car bon emissions compared to new batteries. It is important to id entify anticipated future change s in resources or technology and include them in the present-day sustainability an alysis of electricity generation. Prospective life- cycle assessment is, thus, a key analysis tool to analyze various low- carbon energy options to ensure the low cost and environmental impacts of electricity generation today and in the future. iv To mom I hope you are seeing this, mia, wherever you are– cos I did it! v ACKNOWLEDGEMENTS My first thanks goes to my advisor and my s upport system in this Ph.D., Dr. Annick Anctil. She was the reason I came to Michigan State Univer sity given her research interests, and she has always been supportive of my rese arch, especially my interests in policy. My research also was supported wholeheartedly by my committee- Dr. Mehrnaz Ghamami, Dr. Sharlissa Moore, and Dr. Volodymyr Tarabara. Their su ggestions and comments ensure d my research was improving and becoming relevant. Thank you for being available whenever I needed any information and guidance. Shout out to my lab group. They have been great colleagues a nd even better friends. I am grateful for our talks, especi ally the video calls during the pandemic. A special thank you to Eunsang, without whom I would not be the researcher I am today. Thank you also for introducing me to Korean coffee, gamjatang, and the right way to use chopsti cks. Thanks also to the many undergraduate and high school students who join ed our group for internships and projects. I especially thank RohanŠI really enjoyed the long discussions on our project. I would also like to thank all my friends in the US and back at home (as well as a few spread across the world). Whether in person or over video calls, the regular meetings made me happy and kept me sane. I am not mentioning all of you, specifically not me ntioning you, Lakshmi (eye rolls). Thanks, love. I also thank Sixue, my first friend and sister in the New World. I loved all our shopping trips and your guz heng music. You were instrument al (pun intended) in making Lansing a home for me. Karen, who took me under her wing, also made su re I loved Lansing as much as she did, showing me the best of this place. I loved our walks in the dog parks in Lansing with lovely Mayzee. vi My journey to where I am today started much earlier than when I set foot in the US. I have to thank many people, from my teachers in the kindergarten who had to listen to me howl throughout the day, to my professors in my unde rgrad and masters, who saw me grow as an engineer. I also have to thank a select few peopleŠMeera ma™am, Ajis h sir, Safeer sir, Nema sir, Arun sir, and Ramana sirŠwho not only believed that I could do it bu t made sure that I got all the necessary support. During my Ph.D., I found that I lost quite a fe w thingsŠhair, sleep, and a lot of money on e-commerce websites while indul ging in my capitalistic shopping adventures. But I also gained quite a lot of things. Some valuable such as patie nce and a better temper, a nd some not so valuable, like a few grays here and there. Bu t I also found my othe r half and love, Anurag, who has been my constant support and strength. With out his encouragement and belief in me, I would have been left with a less interesting Ph.D. Thanks to him, I ga ined a new set of parents, for whom I am very grateful as they kept me buoyed too. All this would not have been possible without my mom and dad, who supported my decision to come to Michigan State University for my Ph.D. They did not care that I would be going miles away as long as I was doing what I t hought was right. I will forever be grateful for that. My sister has been the best, keeping me ab reast with all the news from home, making sure I was not missing out on anything. Thanks, ba. vii TABLE OF CONTENTS LIST OF TABLES ................................................................................................................ ......... ixLIST OF FIGURES ............................................................................................................... ........ xiKEY TO ABBREVIATIONS ...................................................................................................... xiv Chapter 1 Introduction ........................................................................................................ ............ 11.1 Potential solutions for energy options in a carbon-constrained world ..................... 3 1.2 Assessing the sustainability of potential energy options in a carbon-constrained world ............................................................................................................................... 51.3 Dissertation outline ................................................................................................... 6 Chapter 2 Evaluating the levelized cost and carbon footprint of second-life electric vehicle batteries in stationary applications .......................................................................................... ........ 92.1 Second-life batteries (SLBs) in stationary applications ......................................... 10 2.2 Methods .................................................................................................................. 132.2.1 Levelized cost of electricity for SLB applications ........................................... 14 2.2.2 Life-cycle carbon emissions for SLB applications ........................................... 17 2.2.3 Scenarios Considered ....................................................................................... 19 2.3 Results ................................................................................................................... . 222.3.1 Residential Energy Storage Application .......................................................... 22 2.3.2 Commercial-level Electric Vehicle Fast Charging Application ....................... 25 2.3.3 Utility-level Peak Shaving Application ............................................................ 26 2.4 Conclusions ............................................................................................................ 28 Chapter 3 End-of-life management of electric ve hicle batteries in the US: Weighing the technical feasibility of second-life batteries, and the cost, carbon, and ma terial trade-offs of remanufacturing and recycling ................................................................................................................. ................ 313.1 Background ............................................................................................................ 32 3.2 Methods .................................................................................................................. 353.2.1 Testing end-of-life electric vehicle batteries .................................................... 35 3.2.2 System dynamics modeling of end-of-l ife management of electric vehicle batteries in the US ..................................................................................................... 37 3.3 Results and Discussion ........................................................................................... 45 3.3.1 Testing end-of-life electric vehicle batteries .................................................... 45 3.3.2 System dynamics modeling of end-of-l ife management of electric vehicle batteries in the US ..................................................................................................... 47 3.4 Conclusions ............................................................................................................ 51 Chapter 4 A comprehensive life-cycle assessment of uranium extraction by alkaline in-situ leaching in the US ............................................................................................................ ............. 544.1 Long-term environmental sustainability of uranium extraction ............................. 56 4.2 Methods .................................................................................................................. 59viii 4.3 Results and Discussion ........................................................................................... 63 4.3.1 Comparison with existing literature ................................................................. 65 4.3.2 Implications of the changes in uranium demand, production, and ownership . 71 4.4 Conclusions ............................................................................................................ 73 Chapter 5 Conclusions and major contributions ........................................................................... 75 5.1 Sustainability implications of second-life electric vehicle batteries ...................... 75 5.2 Sustainability implications of uranium fuel for nuclear energy ............................. 77 5.3 Conclusions ............................................................................................................ 78 APPENDICES .................................................................................................................... .......... 79APPENDIX A: Supplementary Information for Chapter 2 .......................................... 80 APPENDIX B: Supplementary Information for Chapter 3 .......................................... 88 APPENDIX C: System Dynamics Model Equations for Chapter 3 ............................. 91 APPENDIX D: Supplementary Information for Chapter 4 ........................................ 104 REFERENCES .................................................................................................................... ....... 114 ix LIST OF TABLES Table 2. 1. Assumptions for efficiency and costs of system components al ong with other project details for the reference year of 2017. .............................................................................. 16 Table 3. 1. Assumed inputs for the end-of-life management model. ............................................ 40 Table 4. 1. Review of nuclear fuel cycle life-cycle assessment studies, show ing the global warming potential of mining and milling stage, as well as the total for 1 kWh of electricity, along with methodological details and assumptions................................................................... 67 Table A 1. Life-cycle inventory for various components considered in the analysis. .................. 81 Table A 2. Yearly residential electricity demand for each location. ............................................ 81 Table A 3. Results of component design, econo mic analysis and life- cycle assessment for residential household appliance energy storage application (R). ...................................... 82 Table A 4. Results of compone nt design, economic analysis, a nd life-cycle assessment for residential energy storage application with EV charging (REV). ..................................... 83 Table A 5. Results of the fast charging appli cation in terms of component design, economic analysis and life-cycle assessment when using new LIB.................................................. 84 Table A 6. Results of the fast charging application in terms of component design, economic analysis and life-cycle assessment when using SLBs. ...................................................... 85 Table A 7. Results of component design, economic analysis, and life-cycle assessment for utility-level Peak Shaving (PS). ................................................................................................... 87 Table B 1. Recycling efficiencies (REs) for different battery components as the baseline assumptions,171 along with their prices175Œ181 in $/kg. ...................................................... 89 Table B 2. Battery components for different battery chemistries 125,172 and the average energy density237 om kWh/kg. ...................................................................................................... 89 Table D 1. Recovery rate s from solution and Recovery rates total in percentage for Cases 1 and 2. ......................................................................................................................................... 105Table D 2. Inventory for Case 1 (from Eq. 1). ............................................................................ 107 x Table D 3. Inventory for Case 2 (from Eq. 2). ............................................................................ 108 Table D 4. LCA results for 0.036% ore grade for Case 1 (Eq. 1), with contributio ns from each process (Functional unit: 1 kgU)..................................................................................... 109 Table D 5. LCA results for all uranium ore grades considered for Case 1 (Eq. 1) (Functional unit: 1 kgU). ....................................................................................................................... ..... 112Table D 6. LCA results for all uranium ore grades considered for Case 2 (Eq. 2) (Functional unit: 1 kgU). ....................................................................................................................... ..... 113 xi LIST OF FIGURES Figure 1. 1. Simplified illustration of the duck cu rve showing the total electricity demand (dotted line), the solar generation (in yellow), and th e non-solar generation (solid line) over a 24- hour period, with ramping up of electricity generation ne eded when solar generation reduces. ............................................................................................................................... 2Figure 2. 1. Summary of scenarios considered a nd energy system component s for (a) residential energy storage, (b) commercial-level electric vehicle fast charging, and (c) utility-level peak shaving, with th e different components considered . The demand + battery storage profiles show the typical load demand, PV production, and battery charge and discharge for the battery storage scenario. ........................................................................................ 14 Figure 2. 2. Hourly and seasonal residential elec tricity pricing structures for all locations. Electricity prices increase from blue (low) to red (high). The variation in time-of-use price is denoted as hourly or seasonal for each location. ........................................................... 17 Figure 2. 3. Solar insolation and commercial electric ity tariff (both electric ity cost in $/kWh and demand cost in $/kW) for the five study locations. .......................................................... 17 Figure 2. 4. Fast charging demand profile based on EV Project data and the assumed demand profile (in black) for the analysis. ..................................................................................... 21 Figure 2. 5. Levelized cost of elec tricity and global warming potenti al for SLBs in the residential energy storage application in (a ) Detroit, (b) Los Angeles, (c) New York City, (d) Phoenix, and (e) Portland. The LCOE and GWP when usin g SLBs (R2/REV2) are compared to Grid only (R0/REV0), Grid with PV (R1/REV1), and Grid with PV using new LIBs (R2n/REV2n). ................................................................................................................... 23Figure 2. 6. Levelized cost of elec tricity and global warming potenti al for SLBs in the commercial- level electric vehicle fast ch arging application in (a) Detroi t, (b) Los Angeles, (c) New York City, (d) Phoenix, and (e) Portland. Th e LCOE and GWP when using SLBs (FS1, FS2, FS3) are compared to Grid only (FS0), Gr id with new LIBs (FS1n), Grid with PV, and new LIBs (FS2n), and off-grid PV with new LIBs (FS3n). ....................................... 26 Figure 2. 7. Levelized cost of elec tricity and global warming potenti al for peak shaving application for scenarios: PS0 (Natural gas), PS1 (SLB), PS1n (new LIB). ....................................... 27 Figure 2. 8. Average (a) levelized cost of electric ity and (b) global warming potential for baseline, new LIB scenario, and SLB scenario for variou s applications at resi dential (without and with EV charging), commercial (fast charging), and utility (peak shaving) levels. ......... 28 Figure 3. 1. Results from initial capacity tests on cells showing the mean remaining capacity and frequencies. .................................................................................................................. ..... 36Figure 3. 2. Test profiles for each test showing the SOC and c-rate variation over time. ............ 37 xii Figure 3. 3. System dynamics causal loop diagram for end-of-life management of electric vehicle batteries. ............................................................................................................................ 39Figure 3. 4. Changing chemistries of (a) end-of-life lithium-ion battery chemistries and (b) end- of-life NMC battery chemistries over time. ...................................................................... 42 Figure 3. 5. Detailed calculation of end-of-life economic valu e from remanufacturing and recycling pathways for the EOL manager, i.e ., remanufacturer and recycler, respectively. ........................................................................................................................................... 43Figure 3. 6. System boundary c onsidered for this study, including the raw material extraction, electric vehicle battery production, and the end-of-life management pathways of recycling and remanufacturing. ........................................................................................................ 4 4Figure 3. 7. Test results for (a) resi dential energy storage, (b) peak shaving, and (c) fast charging applications with the samples given in the legend. ........................................................... 46 Figure 3. 8. (a) The price of new batteries and second-life batteries with differing health factors (80%, 70%, and 50%). The economic value of remanufacturing for remanufacturer in (a) million $/year and in (b) $/kWh/year when second-life batteries with differing health factors (80%, 70%, and 50%). The economic value of recycling for recycler in (a) million $/year and in (b) $/kWh/year when second-life batteries with differing NMC chemistry changes (Baseline, Case 1 with all NMCs as NMC111, Case 2 with all NMCs as NMC622, and Case 3 with all NMCs as NMC811). ......................................................................... 48 Figure 3. 9. Economic value of recycling and re manufacturing in (a) million $/year and (b) $/kWh/year for recycling only vs. recycli ng with remanufacturing pathways for the baseline. The total end-of-life economic value of both recycling only and recycling with remanufacturing pathways is shown in (c) million $/year and (d) $/kWh/year for the baseline. ............................................................................................................................ 49Figure 3. 10. (a) Total battery capacity needed to be produced equivalent to one kWh capacity demand. The carbon footprint in kgCO 2 eq. per kWh of battery capacity, given that (b) only recycling is considered and (b) recycling with remanufacturing is considered. ....... 50 Figure 4. 1. World nuclea r capacity from 1970 to 2017 with projections of nuc lear capacity until 2050 as per International At omic Energy Agency (IAEA), 188 along with uranium production and demand190 between 1970-2018. ............................................................... 55 Figure 4. 2. Global warming potential (gCO 2 eq./ kWh) of the nuclear fu el cycle from various life- cycle assessment studies. .................................................................................................. 56 Figure 4. 3. (a) Schematic of in-situ leaching with a hexagonal well st ructure with a central recovery well passing through an aquifer and an ore body; (b) fraction of world uranium production by extraction method. ..................................................................................... 57 Figure 4. 4. System boundary for this study, shown as a dotted line, includes the uranium extraction process and ion exchange process..................................................................................... 60 xiii Figure 4. 5. Global warming potential (GWP), terrestrial acidificat ion (Ter. Acid.), and land use of uranium extraction using alkaline ISL in the US for ores of grade 0.036% U3O8. ...... 63 Figure 4. 6. (a) GWP (kgCO2 eq./ kgU), (b) terrestrial acidification (kg SO 2 eq./ kgU), and (c) land use (m2a /kgU) of uranium extraction using ISL in the US as a function of ore grade. ... 65 Figure 4. 7. GWP (gCO2 eq./ kWh) of uranium extr action using alkaline ISL in the US in this study in yellow (Cases 1 and 2, full system boun dary) compared with previous literature (in gray). The half-filled circles denote ISL. The studies that did not specify/report the ore grades are shown separately in the shaded area in gray. ................................................... 70 Figure 4. 8. The demand, production, and owne rship of uranium produced in 1988, 2017, and 2025. All units are in tU. ................................................................................................... 72 Figure A 1. Electricity generation data for the Midwest region in grey, 2017 Michigan electricity demand in red, and the net peak demand target in blue. ................................................... 86 Figure A 2. Electricity generation data for the Northwest re gion in grey, 2017 Oregon electricity demand in red, and the net peak demand target in blue. ................................................... 86 Figure B 1. Percentage of batterie s reaching end-of-life at differe nt years for EV batteries and second-life batteries. ........................................................................................................ . 89Figure B 2. Change in lithium-ion battery demand due to the availability of SLBs. ................... 90 Figure B 3. Change in (a) cobalt and (b) nickel demand with recycling and remanufacturing. ... 90 xiv KEY TO ABBREVIATIONS BNEF: Bloomberg New Energy Finance DOD: Depth of discharge EIA: U. S. Energy Information Administration EOL: End-of-life EV: Electric vehicle GWP: Global warming potential IAEA: International Atomic Energy Agency IEC: International Electrotechnical Commission ISL: in-situ leaching LCA: Life-cycle assessment LCI: Life-cycle inventory LIB: Lithium-ion battery LCO: Lithium cobalt oxide LCOE: Levelized cost of electricity LFP: Lithium ferrous phosphate LMO: Lithium manganese oxide LMO/NMC: lithium manganese oxide/ lith ium nickel-cobalt manganese oxide NMC: Nickel manganese cobalt NMC111: LiNi0.33 Mn0.33 Co0.33 O2 NMC622: LiNi0.6 Mn0.2 Co0.2 O2 NMC811: LiNi0.8 Mn0.1 Co0.1 O2 NREL: National Renewable Energy Laboratory xv OEM: Original equipment manufacturer PV: Photovoltaic PWR: Pressurized light water reactor RE: Recycling efficiency RPT: Reference performance test SD: System dynamics SOC: State-of-charge SOH: State-of-health SLB: Second-life battery USGS: United States Geological Survey 1 Chapter 1 Introduction a While energy use across the globe is increasi ng, the energy sector, especially the electric power sector, is entering a new phase of reducing greenhouse gas emissions. 1 A 20-50% reduction of the 2010 greenhouse gas levels would require reducing per capit a emissions from ~20 tonnes to 2-3.4 tonnes in the US.1 Future electricity generation sources are expected to have a lower carbon footprint due to this shift in energy a nd environmental policies across the globe.2 While, on the one hand, the focus can be on improvi ng the overall efficiency of the entire electric power system, the focus can also be on replacing fossil fu el generation with renewable energy sources. 3 The challenge of reducing the carbon foot print of energy sources is that the cheap alternatives, such as natural gas, are still carbon-intensive. In certain regions, such as in New York, coal power is expected to be replaced by natural gas, coupled with a planned phase-out of nuclear energy. This can lead to an increase in th e electricity carbon f ootprint, which is c ounterproductive to the decarbonization goal. There is also a focus on low emission vehicl es from research and policy to decarbonize the transportation sector. 4 The most successful low e mission vehicles have been electric vehicles, given their continuously increasing sales. 5 The addition of renewable energy sources can incl ude photovoltaics (centralized and distributed), solar thermal, wind, geothermal, hydro, and biomass energy.3 With the large-scale deployment of renewable energy sources associated with deep d ecarbonization of the electric power sector, there comes the issue of intermittency. The intermittent nature of renewable energy sources, such as a Parts of this chapter have been reproduced in part with permission from: (1) Kamath, D.; Arsenault, R.; Kim, H. C.; Anctil, A. Economic and Environmental Feasibility of Second-Life Lithium-Ion Batteries as Fast Charging Energy Stor age. Environ. Sci. Technol. 2020, acs.est.9b05883. https://doi.org/10.1021/acs.est.9b05883. Copyright 2020 American Chemical Society. (2) Kamath, D.; Shukla, S.; Arsenault, R.; Kim, H. C.; Anc til, A. Evaluating the Cost and Carbon Footprint of Second- Life Electric Vehicle Batteries in Residential an d Utility-Level Applications. Waste Manag. 2020. https://doi.org/10.1016/j.wasman.2020.05.034. Copyright 2020 Elsevier Ltd. 2 photovoltaics and wind, increases variability and un certainty, which can impact the electric power system operations. 6 Various studies have addressed th ese issues by deve loping optimization approaches that take intermittency into acc ount when planning for m eeting the electricity demand.6,7 A common phenomenon with the deployment of large capacities of photovoltaics on the electricity grid is the fi duck curve,fl first reported in California in the early 2010s. 8 The duck curve presents the total electricity demand, al ong with the solar and the non-solar generation needed to meet the demand over a 24-hour period (as shown in a simplified manner in Figure 1. 1). Two main challenges arise from high solar deployment: (a) the need to quickly ramp up electricity generation when the sun sets and sola r generation reduces, and (b) the need to store excess electricity produced by the solar generation, which would otherwise be curtailed and lost.8 Figure 1. 1. Simplified illustration of the duck curve showing the total electricity demand (dotted line), the solar generation (in yellow), and the non-solar generation (solid line) over a 24-hour period, with ramping up of el ectricity generation needed when solar generation reduces. Currently, natural gas peaker plants are comm only used to provide quick ramping up of electricity generation in the evenings. 9 However, with the decarbonization of the electricity grid, low-carbon solutions are needed. Battery storage is one solution that can store the excess solar generation, which can be used at high dema nd times, thereby smoothening the demand curve. Nuclear energy is another solution that can supple ment energy storage with electricity generation that is low-cost, low-carbon, mature, and flexible. However, while batteries can be cost- 3 prohibitive, the technological changes in the nuc lear fuel cycle can change the environmental impacts of nuclear energy. One example is the changes in uranium ex traction methods with changing resource availability. 1.1 Potential solutions for energy options in a carbon-constrained world Energy storage for residential, commercial, and utility-levels is gaining interest since it can store excess electricity generation to be used when the demand for it arises. 10 A recent report on the global energy storage market estimated an annual combined deployment of 3,046 GWh of battery storage (~19 times the curre nt levels) over the next 15 years across mobility applications, electronic devices, and stationary storage. 11 The US Energy Information Administration recently reported that the average cost of utility-level ba ttery storage costs have rapidly decreased in the US from $2,152/kWh in 2015 to $625/kWh in 2018. 12 However, the capital cost of battery storage can still limit its wide-scale deployment. 13,14 An alternative is to use second-life electric vehicle batteries or SLBs . SLBs are end-of-life electric vehicle (EV) lithium-ion batteries that have been remanufactured for use in stationary applications after state of health (SOH) asse ssments from original equipment manufacturers. 15 While the price of new lithium-ion ba tteries (LIBs) was about $209/kWh in 2017, 16 the price of remanufactured SLBs is 30% of ne w batteries and vary from $38-147/kWh. 17,18 While the most recent price estimate for SLBs from a re manufacturing company was $139/kWh in 2018, 19 a consulting firm valued the price of SLBs at $50/kWh in 2019. 20 SLB prices are expected to decline due to increased market competition from other SLB manufacturers 19 and availability of end-of- life EV batteries. 21 Using SLBs can not only support the deca rbonization of the energy sector but also manage the expected large volumes of end-of-life EV batteries. 4 The technical feasibility of using SLBs depe nds on the battery che mistry and in-vehicle use of the EV battery. 22Œ26 Battery chemistries such as lithium iron phosphate have slower capacity degradation24 compared to lithium manganese oxide and lithium nickel-cobalt manganese oxide (LMO/NMC).23 However, with moderate cycling, LM O/NMC can be used for second-life applications. 24,27 A growing number of SLB de monstration projects, such as the Sunbatt Project, Pampus Project, and SASLAB, are in various de ployment stages, and many start-up companies and automotive OEMs have emerged across the world to capitalize on the SLB market by catering to residential users. 5,19,28 However, there is a wide variatio n in the health a nd quality of the available end-of-life EV batter ies, impacting the demand for SL Bs. Apart from changes in SLB supply and demand, other market barri ers also need to be considered in SLB research to support their wide-spread adoption. Some market barriers include remanufacturi ng processes, quality assurance, policy, and public opinion. 29Œ31 Depending on these factors, SLB use in different stationary applications can be st rategically prioritized to help in the decarbonization of the energy sector. The second suitable energy option to support the deep decarbonization of the electric grid is nuclear energy, especially wi th the flexible operation of future nuclear power plants. 32 However, although a low-carbon energy source compared to coal and natural gas, nucle ar energy still has a wide variation in its carbon footprint, ranging from 3.2 to 172.5 gCO 2 eq./ kWh. 33,34,43Œ51,35Œ42 One reason for this variation is the uranium extr action phase, which, depending on the ore grade, technology, and process efficiency , contributes to 3-47% of th e nuclear fuel cycle carbon footprint.33,34,52Œ54,35Œ42 With changes in uran ium extraction technology55 and decreases in uranium ore grades,40 the associated carbon footprin t of uranium extraction and, by extension, that of future 5 nuclear energy can change, leadi ng to the question of whether fu ture nuclear generation will remain a low carbon energy source. 1.2 Assessing the sustainability of potential en ergy options in a carbon-constrained world This dissertation evaluated the potential environmental and economic impact of second- life electric vehicle batt eries and the life cycle carbon footprint of uranium fuel fo r nuclear energy. Second-life electric vehicle ba tteries (SLBs) are a nascent technology with currently only demonstration projects and pilot-scale deployments, while nuclear energy using uranium fuel is a mature technology in use since the 1950s. SLB use needs to be studied for different potential applications to evaluate the relative benefits in co st and carbon footprint. An additional aspect that needs to be considered for SLBs due to their nascency is the im pact of remanufacturing on battery material demand, cost, and carbon footprint, compared to recy cling. For nuclear energy, newer methods of uranium extraction have been develo ped for the lower uranium ore grades in the world but have not been include d in the mainstream sust ainability analysis of nuclear energy. Addressing these changes is necessary to evaluate the carb on footprint and environmental impacts of future nuclear power plants. In the case of SLBs, three different applications were considered. The relative economic benefit, in terms of the levelized cost of electric ity, of SLBs was compared to new LIBs to answer whether to use SLBs or not for the considered applications. A life-cycle assessment (LCA) was used to evaluate the environmental impacts of SLBs, focusing on th eir carbon footprint. Additionally, EOL battery cells obtained after in -vehicle use were tested to evaluate their performance in second-life applications to assess their technical feasibility. A system dynamics model was also developed to evaluate the changes in the cost, material, an d carbon trade-offs with the addition of remanufacturing of EOL batteries along with recycling. For uranium fuel for 6 nuclear energy, we identified a lack of environmental assessment for uranium extraction based on alkaline in-situ leaching, which is a prominent me thod in the US and the world. With more deposits being found where alkaline in-sit u leaching provides promising uranium yields, more future mines can be expected to use this me thod for uranium extraction. Therefore, approximating this life-cycle stage can lead to underes timating the carbon footpr int and other environmental impacts of future nuclear energy. To address this issue, a life-cycle assessment of alkaline ISL based uranium extraction was also conducted for varying ore grades, representative of US-based ores. 1.3 Dissertation outline The aim of this dissertation was conducting a prospective life-cycle assessment of two energy options, i.e., second-life electric vehicle batteries and uranium fuel for nuclear energy, by: 1. Evaluating the levelized cost and carbon f ootprint of second-life electric vehicle batteries in stationary applications (Chapter 2) 2. Weighing the technical feasibility of second-life batte ries based on the cost, carbon, and material requirements of remanufacturing and recycling (Chapter 3) 3. Conducting a comprehensive life-cycle assessment of uranium extraction using alkaline in-situ leaching (Chapter 4) To achieve these objectives, a combina tion of methods, incl uding techno-economic analysis (TEA), life-cycle assessment, and syst em dynamics, along with scenario analysis and empirical approaches, were used. The first step was to evaluate whether second-li fe electric vehicle batte ries can be used in stationary applications instead of new lithium-ion batteries. Chapter 2 compares the levelized cost of electricity and life-cycle carbon emissions associated with using SLBs and new LIBs in the US 7 for three energy storage applications: 1) resident ial energy storage with rooftop photovoltaics, 2) commercial-level electric vehicle fast charging, and 3) utility-level peak-shaving, leading to a total of 54 scenarios. For the residential and commercia l applications, five locations were chosen to account for differences in climate, solar irradiance, electricity pricing, and electricity generation mixŠnamely, Detroit, Los Angele s, New York City, Phoenix, a nd Portland. Since peak shaving occurs at a regional level, the states of Michigan and Oregon were chosen for analysis. Chapter 3 presents the result fr om the accelerated testing at th e cell-level that simulated the three applications considered in Chapter 2. Th e cells™ performance was assessed based on capacity degradation. The time duration to reach 80% of initial capacity was found and converted to equivalent time in real-life a pplication. Additionally, a system dynamics model was developed to model the SLB availability in th e US from 2017-2050. EOL electric ve hicle batteries can be either recycled directly or remanufactured into SLBs and used for stationary applications. While remanufacturing can utilize the rema ining capacity of EOL batteries, it also extends the in-use time of the recyclable material. The de lay in the recycling can lead to changes in battery raw material requirementsŠpossibly, leading to increased mining, cost, and carbon footprint. Decisions on the appropriate EOL pathway need to take these considerations into account. A system dynamics model was developed to compare recycling and remanufacturing in terms of economic value, carbon footprint, and material re quirement changes over time while considering changes in EOL battery health and chemistry. The results were used to provide insights relevant to the successful recycling and remanufacturing of EOL EV batteries. In the final chapter, Chapter 4, we cons idered energy generation, focusing on uranium extraction for nuclear energy. As the uranium ore grades gradually decrease across the world, in- situ leaching is the most suitable method to meet the uranium demand. The alkaline in-situ leaching 8 (ISL) method for uranium extraction is most promin ent in the US, and more ISL sites are proposed. There is a lack of a comprehens ive LCA of alkaline ISL for uran ium extraction. To this end, the carbon footprint and other environmental impacts of alkaline ISL were assessed for US-based uranium ores of grade 0.036-0.4% U 3O8. The existing LCA literature on uranium extraction was also systematically revi ewed and compared with these results. We also tracked the changes in uranium demand, production, and ownership in 1985 and 2017 and esti mated these values for 2025 to arrive at implications for the future supply of uranium. Overall, we conducted a prospective life-cycle assessment of second-li fe electric vehicle batteries and uranium fuel for nuclear energy, consider ing them as compleme ntaryŠtwo sides of the same coinŠfor supporting th e decarbonization of the energy se ctor. While the methods used to assess these options differ in certain respects, it is still imperative to include the differences in the energy options in their assessment to inform policy and the broader research community. 9 Chapter 2 Evaluating the levelized cost and carbon footprint of second-life electric vehicle batteries in stationary applications b With growing electric vehicle adoption, the vol ume of end-of-life automotive batteries is increasing rapidly in the US 56 and globally. 5 Most electric vehicles (EVs) are powered by rechargeable lithium-ion battery (LIB) systems, which are typically replaced when their capacity is reduced by 20 to 30%. 19,57 EV LIBs are responsible for approximately 30% of the EV manufacturing cost58 and 37% of the EV manufacturing carbon footprint.59 Recycling is the main EOL management method for automotive LIBs, and once fully established, it could provide revenue and reduce energy cons umption and carbon emissions compared to battery production with virgin materials. 60,61 However, LIB recycling is still in a developmental stage, and some of the challeng es are the low volumes of collected LIBs, low governmental support in certain pa rts of the world, and uncertainti es associated with the full recycling costs. 62 Additionally, LIB chemistries are movi ng towards cheaper low-cobalt and high-nickel chemistries, which can further reduce recycling revenues. 62,63 Including other EV parts, such as power electronics and magnets, in the recycling stream can stil l increase the recycling profit. 64 Previous studies have suggested combini ng recycling with remanufacturing and reuse, 64,65 which can help utilize the remaining 70-80% capacity in EOL batteries. Remanufacturing EOL batteries as second-life batteries (SLBs) involves disassembling batteries to the module- or cell- b Parts of this chapter have been reproduced in part with permission from: (1) Kamath, D.; Arsenault, R.; Kim, H. C.; Anctil, A. Economic and Environmental Feasibility of Second-Life Lithium-Ion Batteries as Fast Charging Energy Stor age. Environ. Sci. Technol. 2020, acs.est.9b05883. https://doi.org/10.1021/acs.est.9b05883. Copyright 2020 American Chemical Society. (2) Kamath, D.; Shukla, S.; Arsenault, R.; Kim, H. C.; Anc til, A. Evaluating the Cost and Carbon Footprint of Second- Life Electric Vehicle Batteries in Residential an d Utility-Level Applications. Waste Manag. 2020. https://doi.org/10.1016/j.wasman.2020.05.034. Copyright 2020 Elsevier Ltd. 10 level, inspecting, repairing, testing, and repacking of cells or modules into ba ttery packs that meet original equipment manufacturers™ quality requirements. Using SLBs can increase EV LIB life while reducing primary raw mate rial extraction, cumulative en ergy demand, and carbon emissions compared to new battery manufacturing. 19,24,25,66Œ68 The high capital costs associated with new LIBs can hinder the large-scale deployment of energy storage, 69 which could also be overcome with SLBs. 17,18 While the purchase pric e of new LIBs was repor ted to be $209/kWh in 2017, 16 those for SLBs are expected to be lower. Earlie r studies and reports have estimated SLB price to be $38-147/kWh,17Œ20 which are expected to fall in the future. 19,21 2.1 Second-life batteries (SLBs) in stationary applications In residential applications, energy storage is e ither used to leverage time-of-use pricing or increase self-consumption of roof top solar generation and consequently result in cost and carbon savings. Adding lithium-ion batteries to rooftop PV in the US can reduce the electricity cost by increasing PV self-consumption by up to 17%. 70 However, studies have identified the capital cost of adding LIBs as a limitation to the associated electricity cost reduction. 13,14 To address this, many studies have considered SLBs as an alternativ e and have found electricity cost reductions when adding SLBs to residential PV systems in Portugal, 71 UK, 72 and Germany. 73 Typically, SLBs are not economical for leveling residential loads alone without PV. 74,75 Life-cycle assessment (LCA) studies, mostly European case studies, have shown SLB use with rooftop PV to reduce the carbon footprint, 27,76 in some cases subject to the carbon emissions of the electricity generation s ources and battery efficiency. 77 In a previous study, the locational differencesŠPV generation, electricity price, and electricity ge neration sourcesŠ were found to impact the levelized cost of elec tricity and carbon footprint of using SLBs for commercial fast- 11 charging.78 However, the effect of these differences fo r residential applications with SLBs has not been studied widely, nor have SLBs been compared with new LIBs. A consequence of the increasing number of EVs is the growing need for EV charging infrastructure. The cost of charging an EV depends not only on the amount of electricity purchased but also on the power at which energy transfer occurs. In other words, for the same amount of electricity bought, the cost increa ses as the charging ti me decreases. The cost per kW power or demand cost varies widely with location and depends on the maximum power bought by the consumer at any time in a month. It can be as high as $19.98/kW in Phoenix at peak times79 or be non-existent like in New York City 80 for power demands up to 100 kW. The total cost is, therefore, the sum of the cost per kWh elect ricity demand and the cost per kW power demand from the grid. This means that for fast char ging, the consumer may have to pay a significant premium for the higher power levels. Alternately, the station owne r may pay a premium to the utility, which will be passed off to the consumer (in one way or another.) Neither of these scenarios is desirable, and both present an economic impediment to technology adoption and acceleration. Renewable energy sources, such as photovoltaic s (PV), can be added as auxiliary or alternate electricity sources in fast charging st ations to reduce the overa ll power consumption from the grid.81,82 Power consumption at high demand times can also be reduced by adding a battery storage system. 81Œ84 One drawback of this approach is the current high initial battery cost, which could be overcome by using SLBs. No study to date has evaluated both the economic and environmental benefits of SLB-based fast charging systems. Potential utility-level applications of new LIBs, as well as SLBs, include peak shaving. 85 Peak shaving with energy storage reduces the power demand during peak times, reducing the need for peaking generators, which are operated only during high demand times. 86 This, in turn, reduces 12 high peak time electricity costs. Based on a US-b ased study that evaluated the potential value of using SLBs in utility-level applications, SLBs generated revenue for peak shaving. 87 However, the study did not include balance-of -system costs and had considerable uncertainty from assumed depth-of-discharge, discharge duration, and revenue benefit range. Studies on energy storageŠnew LIBs and SLBsŠfor peak shaving showed that using SLBs lowered the carbon emissions co mpared to using lead-acid batteries 24,68 and electricity generated from natural gas.66,88 However, these studies modeled generalized conditions of battery cycling (daily discharges) or electricity generation (coal power plants). SLBs can be cheaper than new LIBs, but they can also have a comparatively shorter life, n eeding more frequent replacement. Therefore, to understand the relative benefits of using SLBs for residential and utility level applications, SLBs need to be compared with ne w LIBs or existing baseli ne technologies, based on their associated levelized cost of electricity and carbon emissions. SLBs can potentially provide inexpensive, low-carbon energy storage. However, it is unknown whether SLBs will be better or worse than new batteries and whether SLBs will provide similar cost and carbon emission redu ction for residential and utility level stationary applications in all locations. Previous studies have determined the cost or life-cycle carbon emissions of SLB use in residential and utility-level applications for specific locations and generalized load demands. It is important to compare the applications u nder different scenarios of electricity pricing, electricity grid carbon intensit y, electricity consumption needs, and solar insolation, as these locational factors can impact the cost and carbon emissions of SLB a pplications. In this chapter, the main aim was to ascertain the relative levelized cost of electricity and life-cycle carbon emissions of SLBs and new LIBs in the US for three applications: residential energy storage, commercial-level electric vehicle fast charging, and utility-level peak shaving SLB use was also 13 compared to the existing conven tional technology (baseline) for the applic ation, and thereby inform strategic prioritization of second-life EV battery use. 2.2 Methods Three applications for new LIBs and SLBs were considered: 1) residential energy storage, 2) commercial-level electric vehicle fast chargi ng, and 3) utility-level peak-shaving in the US, leading to a total of 54 scenarios. Figure 2. 1 sh ows the three applications, along with the scenarios considered, the corresponding energy system components, and the de mand + battery storage profiles. The levelized cost of electricity fo r each application scenario was calculated by minimizing the net present value of system com ponents that met the electricity demand (see Section 2.2.1). The project lifetime was assumed to be ten years due to uncer tainties in the future prices of PV, battery, an d electricity, as well as the future changes in new and second-life battery technologies. Using life-cycle a ssessment, the carbon emissions fo r each scenario were calculated based on a 10-year project lifetime, excluding the first-life and transportation of SLBs, and the end-of-life treatment of the system (see Sec tion 2.2.2). Five locations were chosen for the residential and commercial applicationsŠDetro it, Los Angeles, New York City, Phoenix, and PortlandŠdue to their diversity in solar radiation, climate, electricity pricing, and electricity grid carbon intensity, whereas for the utility-level app lication, the states of Michigan and Oregon are chosen for similar reasons. S ection 2.2.3 provides the details of the load/consumption, the scenarios, and the energy system component s in each scenario for each application. 14 Figure 2. 1. Summary of scenarios considered a nd energy system component s for (a) residential energy storage, (b) commercial-level electric ve hicle fast charging, and (c) utility-level peak shaving, with the different components considere d. The demand + battery storage profiles show the typical load demand, PV production, and batter y charge and discharge for the battery storage scenario. 2.2.1 Levelized cost of electricity for SLB applications A net present value approach was used to determine the economically optimal energy system for each scenario using the Homer Pro software. 89 Multiple simulations of energy systems were run for each scenario, with differing comp onent capacities that met the electricity demand over the project life of 10 years from 2017 to 2027. The energy system with the minimum net present cost was determined along with the component capacities, i.e., the amount of electricity purchased from the grid (kWh), the SLB capacity (kWh), the PV capacity (kW), and the inverter capacity. The net present cost ( NPC ) can be obtained using Eq. 2-1, 89,90 where CAnn is the 15 annualized cost, CRF is the capital recovery factor (given in Eq. 2-2), i is the discount rate, and N is the number of years. –Eq. 2-1 –Eq. 2-2 The annualized cost was used to compare the yearly cost incurred in SLB-based scenarios with those in new LIB-based scenarios and the base line. The levelized cost of electricity (LCOE) was another metric used to compare SLB with th e baseline. The LCOE was calculated by dividing the annualized cost by the annual electricity consumption of the scenario as given in Eq. 2-3.89 –Eq. 2-3 Table 1 summarizes the assumptions used as in puts in Homer Pro for the project life and discount factor, along with the pr ice and performance details of the PV, SLB, and inverter. A 30% rebate on the system purchase (the PV and inverter system) was assumed based on the 2017 Residential Renewable Energy Tax Credit and the 2017 Business Energy Tax Credit, which have been scaled down to 26% in 2020.91,92 The rebate can only be applied to the portion of the battery system charged from PV, not the grid. A battery charged only by PV would receive a 30% rebate, while a battery that is charged 20% of the time by PV would receive only a 6% rebate. The yearly solar radiation profiles and temperature data were obtained from the National Renewable Energy Laboratory™s National Solar Radiation Database 93 and the NASA Surface Meteorology and Solar Energy Database,94 respectively. 16 Figure 2. 3 presents the hourly and seasonal va riation in the 2017 resi dential electricity pricing structures from the ma in utilities in each location, 80,95Œ98 which were used in this analysis. The electricity pricing structure in Portland varied only on an hourly basis, while the one in Los Angeles was characterized by seasonal variation with high prices during the summer afternoons. New York City had a pricing structure with a combination of hourly an d seasonal variation. The increase in the residential elec tricity prices over the period of 2017-2027 was based on the US EIA projections.99 Similarly, the commercial electricity pricing varied across the five locations, as shown in Figure 2. 3, and included the demand cost associated with the power demand from the grid, i.e., $ per kW, in addition to those associated with electr icity demand from the grid, i.e., $ per kWh. Figure 2.3 also shows the solar insolation of the locations. Table 2. 1. Assumptions for efficiency and costs of system components al ong with other project details for the reference year of 2017. Parameters Unit Value Ref. SLB New LIB Battery cost $/kWh 65 209 16Œ18 Battery roundtrip efficiency % 91% 95% 26,27,100 Battery lifetime throughput kWh/kWh 2,000 for residential and utility-level; 2,600 for commercial level 3,400 25,26,66,101 PV module cost $/kW 600 for commercial- and utility-level, 650 for residential level 102 PV derating factor % 90.50% PV module lifetime years 30 Inverter cost $/kW 100 for utility level, 130 for residential level Inverter efficiency % 98% Inverter lifetime years 15 Project lifetime years 10 Assumed Discount rate % 6.9% 102 Inflation rate % 2.5% PV rebate % 30% for residential and commercial-level 91 17 Figure 2. 2. Hourly and seasonal residential elec tricity pricing structures for all locations. Electricity prices increase from blue (low) to red (high). The va riation in time-of-use price is denoted as hourly or seasonal for each location. Figure 2. 3. Solar insolation and commercial electric ity tariff (both electric ity cost in $/kWh and demand cost in $/kW) for the five study locations. For peak shaving, overnight cost, fuel cost, and plant efficiency we re based on previous studies on natural gas peaking plants and energy storage. 86,103Œ105 The baseload generation cost for charging the battery was assumed to be $0.1/kWh ba sed on the average cost of electricity over the last ten years. 106 2.2.2 Life-cycle carbon emissions for SLB applications For each scenario, the grid purchase and capacitie s of PV, SLB, and inverter were used to calculate the life-cycle global warming potential (GWP) at each location. The life-cycle GWP of the application scenarios, excluding the SLB first- life and the end-of-life treatment of the system, 18 was calculated over ten years using the TRACI 2.1 method107 in SimaPro v8.5 108 software. The functional unit defines the function or the service provided by the cons idered system or application. Therefore, for this analysis, the functional unit changed with each application. For instance, the functional unit for the residential application was th e delivery of electricity to meet the demand of the house with or without EV char ging over the project lifetime of 10 years. For fast charging, the functional unit was defined as the delivery of el ectricity for 10,000 fast charge events over the course of 10 years, whereas for peak shaving, it wa s the delivery of electric ity to meet one kWh of peak demand over ten years. A second functional un it was also considered for the residential and commercial-level applications to compare them with the utility-level application, where the GWP was calculated for every kWh of elec tricity delivered to meet the demand over the project lifetime. The life-cycle inventories for the PV system and other material inputs were taken from Ecoinvent 3 109 and DATASMART LCI databases. 110 The DATASMART invent ory for electricity was updated based on the me thodology in literature,111 with the mix of fuels representative of the 2016 US EPA eGRID subregion for each location. 112 The EV SLB inventory was based on a previous publication on electric vehi cle lithium-ion battery manufacturing, 59 and the repurposing process was assumed to take place in the US. Comp ared to a new lithium-ion battery, the material required for the enclosure was assumed to be 30% , while the same amount of primary energy for pack assembly was required. 78,113 Details of the life-cycle invent ory are provided in Appendix A. The assessment excluded the first- life and transportation of SLBs and the EOL treatment of the system. 19 2.2.3 Scenarios Considered 2.2.3.1 Residential Energy Storage Application Two energy demand cases were considered fo r the residential energy storage system application. In the first one, the system was desi gned to meet the residential energy storage demand (Case R). In the second case, the electricity required to charge an EV was added to the residential energy storage demand (Case REV). Figure 2. 1 (a) shows the three resi dential energy storage application scenarios, grid only (R0/REV0), grid with PV (R1/RE V1), grid with PV and battery (R2/REV2), where the consumpti on corresponded to residential en ergy storage (R) and with the additional demand of EV charging (REV). The demand and battery storage profiles for the application are also shown for th e scenario where the rooftop PV is complemented by an energy storage system, and the excess genera tion is stored for later use. The residential electricity c onsumption profiles from app liance use for representative buildings were obtained from prev ious work by our collaborator. 114,115 The EV was assumed to have a 30 kWh battery that was charged with a 6.6 kW Level 1 home char ger for four hours per day. The charging was assumed to st art either at 8 pm or when low time-of-use pricing began, whichever was later. 116 For case R, three scenarios were considered with different system components. For R0, only the electr icity grid was considered; for R1, PV was added to the grid; and R2, SLB was added to R1. R2n was a sub-scen ario of R2 with new LIBs instead of SLBs. Similarly, three scenarios were also considered for case RE V with EV charging. For REV0, only the electricity grid was considered; for REV1, a PV was connected to the grid with PV; and for REV2, SLB was added to REV1. Similar to R2n, REV2n was a sub-scenario of REV2 with new LIBs instead of SLBs. The rooftop PV size was assumed to 5 kW. 20 2.2.3.2 Commercial-level Electric Vehicle Fast Charging Application The economic and environmental benefits of using SLBs in 100 kW fast charging systems were compared for four different scenarios. The scenarios, based on whether system components (grid electricity purchase, SLB, and PV) were incl uded or not, were: grid on ly (FS0), grid with battery (FS1), grid with battery and PV (FS2), and off-grid PV with battery (FS3), as shown in Figure 2. 1 (b). For the scenarios that include battery storage, the grid power used to charge the batteries was limited to 8 kW, which corresponds to the typical power demand from a Level 2 charger. FS1n, FS2n, and FS3n were sub-scenarios of FS1, FS2, and FS3 with new LIBs instead of SLBs. The load profile was based on the EV Proj ect data from Idaho National Laboratory, 117 which deployed electric vehicle charging infrastr ucture and studied the charging patterns in selected cities. The demand profile from San Fr ancisco, shown as the red line in Figure S1, was used to find the peak de mand times, and the number of peaks was based on the average number of charges per day. The city of San Francisco was chosen due to th e higher number of EV charging demand relative to other locations considered in the EV project. This was e xpected to offer a more realistic case study than lower usage areas, given the increasing number of EV sales predicted. The typical number of fast charging events in a day was found to be four, and the corresponding peak demand times are also identified. The EV Projec t assumed a fast charging time of 21 minutes, which, based on the average electricity use per char ge, translated to a charging power of 25 kW. To consider the 100 kW power s cenario, the charging times were reduced to 15 minutes per charge. The demand profile for this study is shown as the black line in Figure 2. 4. 21 Figure 2. 4. Fast charging demand profile based on EV Project data and the assumed demand profile (in black) for the analysis. 2.2.3.3 Utility-level Peak Shaving Application The peak shaving application was modeled to meet the peak demand by using either conventional peaking power from natural gas or energy storage (SLB and new LIB). Since the peak shaving application was modeled for a regi on™s demand and not a location, it was evaluated for two statesŠMichigan and Oregon, chosen due to the differences in th eir electricity demands and the carbon emissions associated with their electricity gene ration. The peak shaving operation depends on the demand profile as the peaking cap acity is determined based on the peak demand. Since state-level demand data was not readily avai lable, the historical ge neration data was used from all of the balancing authorities operating in a state, and the data we re proportionally reduced to match the annual demand of th e corresponding state in the year 2017. Mich igan™s electricity demand was obtained by reducing the Midwest Independent System Operator generation data proportionally to Michigan™s annual demand. 118 For Oregon, the demand was obtained by reducing the generation data from Bonne ville Power Administration and Portland General Electric Company proportionally to Oregon™s annual demand.118 22 Similar to the NREL report on the potenti al of battery storage peaking power, 86 the net peak demand target was determined to be 80% of the annual peak. Any load below this threshold was met by the baseload electricity grid power, and any de mand above it was denoted as peak demand. The system components for the two scenarios considered to meet the peak demand were (1) PS0: Electricity grid with an added natural gas power plant, and (2) PS1: Electricity grid with an added SLB storage for peak shaving. PS1n was a sub-scenario with new LIBs used instead of SLBs in PS1. The baseload generation used to ch arge the battery energy storage was assumed to be the present generation without the peaking capacity of natural gas generation. The natural gas peaking plant was assumed to have a capacity equal to 20% of maximu m demand, and the costs were based on the US EIA assumptions. 105 The peak shaving scenarios, PS0 and PS1, are shown in Figure 2. 1 (c), along with th e modified demand profile with the reduced peak demand when using energy storage. 2.3 Results 2.3.1 Residential Energy Storage Application Based on the electricity demand for each location, the energy system components with the minimum net present cost were determined, along with the associated LCOE and life-cycle GWP for all scenarios. Tables S3 and S4 provide the component capacities, the annualized cost, LCOE, and GWP for all scenarios of the residential ener gy storage application (R) and residential energy storage application with EV ch arging (REV), respectively. 23 Figure 2. 5. Levelized cost of el ectricity and global wa rming potential for SLBs in the residential energy storage application in (a) Detroit, (b) Los Angeles, (c) Ne w York City, (d) Phoenix, and (e) Portland. The LCOE and GWP when using SLBs (R2/REV2) are compared to Grid only (R0/REV0), Grid with PV (R1/REV1), and Grid with PV using new LIBs (R2n/REV2n). The battery capacity required with the 5 kW residential rooftop PV was dependent on whether EV charging was consider ed or not and also on the sola r insolation (higher capacity to store higher PV generation) or the battery type (higher capacity w ith lower roundtrip efficiency of SLBs). The battery roundtrip efficiency is the rati o of the energy output to the energy input, and it increases as the energy loss due to storage decreases. As the roundt rip efficiency decreases, higher capacity will be needed to meet the same demand. For the residential energy storage application (R), a SLB capacity of 15-20 kWh was required with the rooftop PV (R2), while a new LIB capacity of 5-15 kWh was require d (R2n). Locations with higher solar insolation needed the highest battery capacities to store the excess PV generation. For residential energy storage with EV charging (REV), a similar SLB capacity of 15-20 kWh was required (REV2), while the required capacity for new LIBs was 5-20 kWh (REV 2n). The battery capacity requirement did not change drastically with the addition of EV chargi ng since the extra electricity demand occurred at 24 night (low-peak pricing time). The capacity requirement in Los Angeles, however, was 20 kWh for both SLBs and new LIBs since the electricity purchase had to be reduced due to the high low peak pricing. Figure 2. 5 shows the LCOE and GWP for the re sidential energy storage application with and without EV charging. Replaci ng new LIBs with SLBs for resi dential rooftop PV reduced the LCOE by 14-27% for residential energy storage applications (R 2 vs. R2n) and by 7-15% when adding EV charging demand (REV2 vs. REV2n). Irrespective of whether EV charging was added or not, SLBs reduced the LCOE wh en added to rooftop PV compared to grid only scenario (R0/ REV0) as well as rooftop PV only scenario (R1/REV1). Our LCOE results for new LIBs with PV were similar to those obtained fo r the US-based study ($0.11-0.16/kWh). 70 The results presented here showed that adding SLBs to PV provided an average annual co st reduction of $343 in the US, comparable with a study that found a cost reduction of ~$313/ year for an average house in Portugal.71 Another study also found a dding SLBs with PV to be profitable in Stuttgart, Germany.73 The addition of the EV charging demand (R EV) did not increase the LCOE savings substantially, even with an increase in electricity consumption, leading to lower percentage savings in all locations. The LCOE reduction for the resi dential applications (R and REV) depended on the electricity pricing structure. For the reside ntial energy storage application (R), a seasonally varying pricing structure along with high solar in solation led to maximum percentage savings in electricity cost, as seen in Los Angeles (53% w ith rebate). In contrast, when EV charging was added, SLBs with PV (REV2) ha d the maximum percentage savings compared to the baseline (REV0) when pricing structures varied both da ily and seasonally. The hourly variation in the electricity pricing was important since the EV charging occurred at low-peak times, and the 25 locations with hourly variation had lower LCOE when EV charging was added to residential energy storage demand. This effect wa s most prominent in cities with high differences in electricity pricing, such as New York City, where the LCOE halved for all scenarios when EV charging was included, compared to meeting the re sidential energy storage demand only. Compared to new LIBs (R2n), the life-cycle GWP reduced by 10-44% for the residential energy storage demand, as shown by the bar graphs in Figure 2. 5 (a)-(e). When adding the EV charging demand, the GWP reduc tion for SLB (REV2) was lowe r in all locations (2-16%) compared to new LIB use (REV2n). The exception was Los Angeles, where the differences in roundtrip efficiency between the two 20 kWh batter ies increased the grid electricity consumption with SLBs. The lower GWP reduction when adding EV charging (REV) in all scenarios can be explained by the 5 kW capacity limit imposed on the rooftop PV size, which was a reasonable practical constraint. Larger PV capacity would be necessary to replace grid electricity purchases. The GWP also reduced significantly with SLB (R 2/REV2) compared to th e grid only (R0/REV0) or rooftop PV scenarios (R1/REV1) in all loca tions for both residentia l application cases. The detailed results for the residential ap plications are provided in Appendix A1. 2.3.2 Commercial-level Electric Vehicle Fast Charging Application Using SLBs instead of new batteries redu ced the LCOE and GWP for all scenarios considered, i.e., in all cases, SLBs provided co st and GWP benefits compared to new LIBs, as shown in Figure 2. 6. LCOE and GWP were lower for SLB-based off-grid c onfigurations due to the large energy storage needed. This represen ted a 34-41% reduction in LCOE and a 50-77% reduction in GWP for SLB-based of f-grid configurations (FS3) relative to using new LIBs (FS3n). As larger storage was needed in cities with lo wer solar insolation, Detroit and Portland had the 26 lowest LCOE and GWP. For on-grid configurations, the LCOE and GWP were reduced by 12- 27% and 7-15%, respectively, when using SLBs ( FS1, FS2) instead of new batteries (FS1n, FS2n). This result demonstrated economic and environmenta l benefits for the use of SLBs instead of new LIBs for commercial applications such as fast charging systems. The detailed results for the fast charging application are provided in Appendix A2. Figure 2. 6. Levelized cost of el ectricity and global warming potential for SLBs in the commercial- level electric vehicle fast charging application in (a) Detroit, (b) Los Angeles, (c) New York City, (d) Phoenix, and (e) Portland. The LCOE a nd GWP when using SLBs (FS1, FS2, FS3) are compared to Grid only (FS0), Grid with new LI Bs (FS1n), Grid with PV, and new LIBs (FS2n), and off-grid PV with new LIBs (FS3n). 2.3.3 Utility-level Peak Shaving Application The 2017 demand profiles for Michigan and Oregon used to model the peak shaving application are provided in Appendix A3. Figure 2. 7 shows the LCOE and life-cycle GWP for all scenarios of peak shaving consid ered for both Michigan and Oregon. Due to the differences in the peak demand power and time, the cost varied substantially between the two states. The LCOE obtained for the baseline and ener gy storage scenarios were lower than those reported for a study 27 based in California. 103 The California-based study considered lead-acid battery storage with lower efficiency, and the load profile in California was also not explicitly modeled, which can explain the differences. For the PS0 scenario with a natural gas peaking plant, the LCOE in Oregon was lower than in Michigan. SLBs providing peak ing power (PS1) reduced the LCOE by 39% in Michigan but increased it by 61% in Oregon. New LIBs (PS1n) did not lower the LCOE in both states compared to a natural ga s peaking plant. The consumpti on profile was found to be an important factor in considering the potential economic benefit from SLB use for peak shaving. Figure 2. 7. Levelized cost of elec tricity and global warming potenti al for peak shaving application for scenarios: PS0 (Natural ga s), PS1 (SLB), PS1n (new LIB). The GWP of SLB use for peak shaving depend ed on the baseload used to charge energy storage. In Michigan, there was an increase in GWP with SLBs and new LIBs for peaking power compared with natural gas for two reasons: (1 ) an increased electric ity consumption due to roundtrip efficiency losses, and (2) the use of base load generation in Michigan instead of natural gas generation. Similarly, in Oregon, new LIBs in creased the GWP in comparison with natural gas. However, SLBs reduced it by 27% compared to natural gas. This reduction by SLBs was because Oregon™s baseload generation was less carbon-intensive than natural gas generation. The detailed results of the utility-level pe ak shaving are provided in Appendix A3. 28 2.4 Conclusions Figures 2. 8 (a) and (b) compared the LCOE and life-cycle GWP of conventional, SLB, and new LIB scenarios for various applications at residential and utility levels. Using EV SLBs instead of new batteries had the potential to redu ce the levelized costs and global warming potential for all applications. The maximum reduction was obtained for commercial level applications. Figure 2. 8. Average (a) levelized cost of electric ity and (b) global warming potential for baseline, new LIB scenario, and SLB scenario for various applications at re sidential (without and with EV charging), commercial (fast charging), and utility (peak shaving) levels. Comparing SLBs with the application base line gave insight in to prioritizing SLB applications. SLBs provided benefits at the reside ntial level when compared to rooftop PV alone by reducing the levelized cost by 15-25% and car bon emission by 22-51%, making SLBs attractive to residential consumers as well. For commercial level fast char ging, SLBs reduced the levelized cost of electricity (LCOE) by 12-41% and the global warming potential (GWP) by 7-77% compared to using new batteries. Photovoltaics, along with SLBs, reduc ed the use of grid electricity and provided higher GWP reduction compared to onl y using SLBs. In comparison, the 29 LCOE and GWP reduction for SLB-based peak shaving compared to natural gas plants depended on the electricity consumption profile and generation sources. Important implications were al so identified that could affe ct the cost and carbon emissions for each application. In the case of residential applications, electric ity pricing was the key to LCOE and GWP reduction. The levelized cost of electricity for rooftop solar PV with SLBs was always lower than that for rooftop sola r PV only, irrespective of whether the residential tax credits were applied or not. Adding SLB storag e to rooftop solar PV can provide financial benefits that may otherwise be lost as residential tax credits for PV panels get scaled down to 0% by 2022.91 Similarly, for commercial-level electric vehi cle fast charging applications, the LCOE and GWP benefits depended on both the electricity pricing and the solar irradiance in the location. In cities such as Phoenix with high demand charges, fast charging can be five to six times more expensive than in cities with low or no demand ch arges, such as Portland. The results presented here showed that the addition of SLBs was most favorable for cities with demand costs and high variation in the time-of-use pricing. For example, using SLBs brought the cost of fast charging in Phoenix on par with that of Portla nd. In some cities such as Ne w York City that did not have demand charges, SLBs can still reduce fast chargi ng costs if there is a large difference between high and low peak electricity prices. There is grow ing interest in time-of-use pricing by utilities to decrease peak consumption and reduce the need for peak power plants. 119,120 This suggests that the market for energy storage in fast charging system s could rapidly increase as demand costs or time- of-use pricing are incr easingly established. The analysis of utility-level peak shavi ng revealed that the carbon emission reduction depended on the baseload generation sources. At present, the carbon intensity of electricity generation sources in many locations is higher than that of a natu ral gas power plant. Therefore, 30 global warming potential can increase when SLBs are used with the current electricity grid (as seen in Michigan). However, as more renewabl e energy sources are added at the utility-level, 121 SLBs will eventually deliver lower overall GWP opportunities than natural gas peaking plants (as already seen in Oregon). Identifying potential applications for SLBs is only the first step in the successful deployment of SLBs. Research need s to be carried out to understand if the available SLBs can meet the demand for potential applications. The hi ghly variable in-vehicle life of batteries is another important factor for remanufacturability. Temperature and fast charging can deteriorate the EV battery faster, which, in turn, can reduce the volume of EOL batteries suitable for a second life. To reduce battery degradati on and thereby ensure SLB availabi lity in the future, there is a need to improve EV battery technology, understand battery degradation m echanisms during fast charging, and select suitabl e fast charging techniques. 122Œ124 Apart from changes in SLB supply and demand, other market barriers also need to be considered in future SLB research to support their wide-spread adoption. Some market barriers include remanufacturi ng processes, quality assurance, policy, a nd public opinion. 29Œ31 There is also a need to compare the remanufacturing end-of-life management pathway wi th recycling to consider the cost, carbon, and material trade- offs between them, since using SLBs delays them reaching the recycling and change the raw material requirements over the long-term. In th e next Chapter, these issues are addressed by undertaking the performance testing of end-of-life electric vehicle batteries and also developing a system dynamics model to study the cost, carbon footprint, and material trade-offs with recycling and remanufacturing end-of-life electric vehicle batteries over time in the US. 31 3 End-of-life management of electric vehicle batteries in the US: Weighing the technical feasibility of second-life batteries, and th e cost, carbon, and material trade-offs of remanufacturing and recyclingc Second-life batteries are electr ic vehicle (EV) batteries that have been remanufactured or repurposed for a secondary appl ication at th e end-of-life.113 A fisecond lifefl is possible for end-of- life (EOL) lithium-ion batteries (LIBs) due to th e remaining 75-85% capacity after the in-vehicle use.78 EOL batteries can also be managed by recycling. Howeve r, extending their life (by remanufacturing) provides an oppor tunity to utilize the remaini ng capacity in the EOL batteries before ultimately recycling them . Second-life batteries or SL Bs also reduce raw material extraction, global warming potential, and cumulative energy demand compared to new battery manufacturing.66,125 Electric vehicles are increasing in number in th e US, which will soon lead to the availability of EOL LIBs after their in-vehicle use, signa ling a need for proper end-of-life management. 126 Introducing an additional pathway at end-of-life can change the dyna mics of the market, affecting the demand, supply, cost, and environmental impacts. The EOL batteries are sorted based on their state-of-health and are either remanufactured or r ecycled. The state-of-health is a highly variable parameter that can impact the success of the remanufacturing EOL pathway. The success of remanufacturing as an EOL pathway depends not only on a well-oiled supply chain but also public perception and consumer acceptance that impact the demand for c Parts of this chapter have been reproduced in part with permission from: (1) Kamath, D.; Arsenault, R.; Kim, H. C.; Anctil, A. Economic and Environmental Feasibility of Second-Life Lithium-Ion Batteries as Fast Charging Energy Stor age. Environ. Sci. Technol. 2020, acs.est.9b05883. https://doi.org/10.1021/acs.est.9b05883. Copyright 2020 American Chemical Society. (2) Kamath, D.; Shukla, S.; Arsenault, R.; Kim, H. C.; Anc til, A. Evaluating the Cost and Carbon Footprint of Second- Life Electric Vehicle Batteries in Residential an d Utility-Level Applications. Waste Manag. 2020. https://doi.org/10.1016/j.wasman.2020.05.034. Copyright 2020 Elsevier Ltd. 32 remanufactured products, incl uding remanufactured batteries. 127,128 The supply and demand for remanufactured batteries depend on many factors, such as EV sa les, EOL battery availability, state-of-health, overall storag e demand, and consumer acceptance. The feedbacks between these factors are dynamic in nature and can either constrain or amplify the overall effects. The variation in supply and demand for remanufa ctured batteries can influence their cost and environmental impacts. While remanufacturing can utilize the remaining capacity of EOL batteries, it can also keep the recyclable material in use for longer. The delay in recycling the batteries can lead to changes in battery raw material requirementsŠpossibly, leading to increased mining, cost, and carbon footprint. Decisions on the appropriate EOL pa thway need to take these considerations into account to ensure the long-term sustainability of the chosen pathways. In this chapter, the capacity and performance of end-of-life electric vehicle batteries were studied for three second-life app lications to understand their tec hnical feasibility as second-life batteries. A system dynamics (SD) model was also developed to determine the SLB availability and the influence of the end- of-life management pathway on end-of-life value. The public perception based on focus groups and surveys was included in the model, and the cost, carbon footprint, and material trade-offs with recycl ing and remanufacturing over time in the US were studied. 3.3 Background End-of-life management is a key phase in the life-cycle of a product, and understanding the influence of different pathways on cost, carbon footprint, and material requirements can inform policy decisions. Most studies anal yze the circularity of systems a nd markets using material flow analysis and system dynamics mo deling. Material flow analysis has been used to study the influence of EV diffusion on global lithium demand-supply 129Œ132 and the influence of recycling 33 or recoverability on lithium reserve. 133Œ137 Sverdrup (2016) showed th e availability of large amounts of lithium reserves till 2400, with scarcity starting only around 2110.131 The importance of lithium recycling efficiency was shown in 136,137 , where the recycling process can be successful only if the recycled or secondary lithium replaced primary lithium. The system dynamics (SD) approach has been used to model lithium 131 and uranium 138 markets, which incorporate recycling and reproce ssing, respectively. Studies have also used SD to model the environmental impacts of energy systems 139Œ141 and manufacturing, 142,143 by using emission factors in the model its elf or using the modeled result s for life-cycle assessment or LCA.144 SD has also been used to model the temporal aspects of remanufactured product markets, 145Œ148 and can be used to model the remanufactured battery market. Positive public perception is an important cr iterion for the success of a remanufactured product,127,149 because, without consumer acceptance (i.e ., without remanufactured products being actually bought), environmental im pacts or costs cannot be redu ced. Studies of remanufactured products mostly look at the supply chain and economic benefits while assuming positive public perceptions.150Œ152 Studies on reverse supply chains of end-of-lif e EV batteries also treat public perceptions to be positive when remanufacturing is considered along with recycling. 68,125,126,153,154 Previous work on the cost and environmental impacts of SLBs 17,18,21,68,155 does not analyze the effects of public perception on the remanufacturing facility needed and production rate every year. Little work has been done to integrate actual public pe rceptions into these models. These studies also have assumed second-life duration and pe rformance from modeling studies without any experimental data to support the same.24,66,74 34 SLBs can have high variabili ty in their residual capacity and second life performance depending on their in-vehicle duration and use. 18 The remaining cycles depend on the depth-of- discharge (DOD) and decrease with increasing DOD. Based on battery performance modeling, various studies have assumed different cycle lives for SLBs. For 100% DOD, the remaining number of cycles is ex pected to be 900 to 1600 22, while for 80% DOD, it is expected to be 3,650 to 4,800. 25,156,157 As the DOD decreases to 60-70%, the expe cted number of cy cles remaining is 1,260 to 4,800 number of cycles. 17,26,66 Lifetimes for SLBs have been assumed to vary from 3 to 15 years depending on the second-life applications.17,19,22,24,25,66,156,158 These studies have defined the end-of-life for SLBs as the loss of betw een 20% to 75% of the initial SLB capacity, 22,23,25Œ 27,156,157 which can explain the variation in the lifeti me and the remaining num ber of cycles. This variability needs to be investigated and better understood with the use of experimental data. In addition to capacity fade and cycle life, the roundtrip efficiency is also different for an SLB compared to a new LIB. As st ated earlier, the roundtri p efficiency of a batt ery is the ratio of the energy output to the energy input, and it increases as the energy loss due to storage decreases. SLB roundtrip efficiency is gene rally assumed to be 80-85%. 25,156 However, this might be a pessimistic assumption since round trip efficiency for SLB was 98% for an experimental study 27 and 95% for an SLB demonstration project. 100 One study conducted the e xperimental analysis of EOL batteries in residential and renewables integration applications and found that highly demanding second-life applications are te chnically and economically infeasible. 159 However, the batteries were not obtained after real-life EV use, and therefore, do not incorporat e the real-life variability that would be otherwise present. In this chapter, the aim was to incorporate the influence of public perception of the demand and selling price of second-life batteries and there by model the end-of-life value of electric vehicle 35 batteries from remanufact uring and recycling. Firstly, the cell-level testing of EOL EV batteries was also conducted to study th eir performance in second-life. Secondly, a system dynamics model was built to determine the economic value, carbon footprint, and material requirement changes of two EOL pathways of managing EOL EV batteries. 3.4 Methods 3.4.1 Testing end-of-life elec tric vehicle batteries To determine the second life performance, cell-level te sts were conducted on EV battery cells, which were obtained after their in-vehic le use. To account fo r different second-life applications, three different life-cycle tests were chosen, which included residential storage, peak shaving, and EV fast charging. Thes e life-cycle tests were designe d to characterize the secondary battery lifetime capacity and perfo rmance over time for each of thes e applications. Test 1 for the residential storage application was based on the International Electrotechnical Commission (IEC) standard for secondary cells and batteri es for renewable energy storage, IEC 61427. 160 In contrast, Test 2 for peak shaving was based on the peak shav ing test profile from th e Protocol for Uniformly Measuring and Expressing the Performance of ESS made by Pacific Northwest National Laboratory and Sandia National Laboratories.161 The last application of EV fast charging did not have any standards or protocols th at defined a test; therefore, a life-cycle testing profile for the application was developed to refl ect a high-rate discharging (sta tionary) battery (used for fast charging). This profile was used to design Te st 3 for the EV fast charging application. 36 Figure 3. 1. Results from initial capacity tests on cells showing the mean remaining capacity and frequencies. All the tests were performed on constrained pouc h cells using an eight-channel Arbin MITS Pro 4.23 system at room temperature. The initial EV LIBs capacities were measured using the IEC 61247 method.160 Figure 3. 1 shows the results for 30 cells from an EV battery pack, and the mean value of the remaining capacity was 12.75Ah (+/- 0.2Ah). This valu e was 85% of 15Ah, the initial capacity of a cell from an EV battery pack, implying that the cells had more than the expected 80% capacity remaining. Hence, 12.75Ah was assumed to be the initial capacity for an SLB, and this value was used to calculate the rate of charging and discharging (c-rates) and states of charge (SOCs) for the tests. A minimum of three samples were tested under each test. The EOL condition was defined as a 20% reduction in the capacity of the cell from the start of the test, though some tests were allowed to go longer to understa nd the evolution of performan ce. Reference performance tests (RPTs) were also initiated prior to cycle testing to establish a baseline condition of the battery that can be used to assess any changes in its perfo rmance throughout the te st duration. The RPTs included stored energy, round-trip efficiency, and internal resistance, based on the Protocol for Uniformly Measuring and Expressing the Performance of ESS. 161 In addition to running the RPTs prior to duty cycle testin g, they were also performed at set time intervals through the cycle testing. 37 For Tests 1 and 3, the RPTs we re conducted after every 50 cycl es, whereas for Test 2, they were conducted every 25 cycles. Along with RPTs, the protocol required that specific duty cycle metrics be taken throughout the duration of the cycle testing but did not require the duty cycles to be paused to take the measurements. The duty cycle metrics were the duty cycle throughput and duty cycle efficiency.161 The SOC and c-rate varia tion over the test cycle fo r all three tests are as shown in Figure 3. 2. Figure 3. 2. Test profiles fo r each test showing the SOC and c-rate variation over time. 3.4.2 System dynamics modeling of e nd-of-life management of elect ric vehicle batteries in the US The existing markets of electri c vehicle lithium-ion batteries and stationary storage batteries are expected to change with the intr oduction of remanufactured batteries. To model the remanufactured battery market, a causal loop diagram and, subseque ntly, a system dynamics (SD) model was developed (Figure 3. 3) for the US from 2017 thr ough 2050, with remanufacturing beginning in 2020. In a SD model, the different variables at different points in time are represented by stocks and flows. The main st ocks in this model were the number of EVs, the number of remanufactured batteries, and the remanufacturing facility. Only light-duty EVs were considered 38 in the current study. The sales da ta from the EIA Energy Outlook 2019 121 was used. The end-of- life of an EV battery was assume d to be a truncated distribution, 125 and any EOL battery taken out of the vehicle after ten years is directly recycled. The test results based on the methods given in Section 3.2.1 were not used since they were still ongoing when this model was being built. However, future work can incorporate test result s, provided more replicat es are incorporated due to the variability in the second-life perfor mance. The fraction of batteries reaching a remanufacturing facility that can be remanuf actured in an SLB was assumed to be 95%. The lifetime of SLBs was also considered to be a similar truncated distribution but with half the years as new EV LIBs (see Figure B1 in Appendix B). The selling price of SLBs is the amount that the consumer will pa y for a remanufactured SLB after it is remanufac tured and brought to market. The sell ing price was calculated using a top-down approach based on the methodology in Neubauer and Pesaran (2011) 87 to incorporate public perception in the form of willingness to pay. E q. 3-1 draws the relationship of the selling price of SLBs with the cost of new batteries, willingness to pay for SLBs, and the value of SLBs with that of new batteries. For studying the impact of the h ealth factor, three values were chosen for this variable: 80%, 70%, and 50%. PSLB = Kh × Ku × CN –Eq. 3-1 Where CN = Price of new battery Kh = health factor = (CEOFL-CEOL )/(1-CEOL); CEOFL = capacity at end-of-first-life CEOL = capacity at end-of-life Ku = used product discount factor = 39 The price of new batteries was obtained from the Bloomberg New Energy Finance forecasts 162 till 2030, followed by a 10% annual reduction in price, based on the average annual change from 2015 to 2030. As part of this res earch, a nationwide survey was conducted by our collaborator to understand public opinion and willi ngness to adopt second-life batteries as residential battery storage. 163 There were a total of 1,002 complete responses , which were divided into early adopters (if they owned an EV or a rooftop PV) and the general public. The variable Ku was assumed based on the survey results (speci fically the question of willingness to pay for a second-life battery compar ed to a new battery). 163 End-of-life management of EV LIBs was assumed to start in 2020, with the consumer diffusion moving from early adopters to the general public in 2028. Therefore, the survey results for the appropriate populations were used. Figure 3. 3. System dynamics causal loop diagram for end-of-life management of electric vehicle batteries. 40 Table 3. 1. Assumed inputs for the end-of-life management model. Sl. No. Variable Assumption (unit) Note 1. Electricit y cost 0.0688 ($/kWh) Industrial rate with an increase of 2% a t nominal prices. 121 2. Transportation cost 2.16 ($/mile) With an increase of 3% per year from the ATRI report. 165 The transportation was assumed to be by a 16-ton truck. The distance is assumed to be a minimum of 100 miles and a maximum of 2,500 miles, d ecreasing with increasing remanufacturin g facilities. 3. Remanufacturing labor cost 20.67 The increase was assumed to be 2.8% based on Bureau of Labor Statistics reports, 166 with three shifts, a total of (8 hours of 252 days per year), and ten la borers at the facility at every time. 4. Remanufacturing output 3 (10kWh battery packs per hour) Three shifts of eight hours, each with 252 working days per year, are assumed, similar to the BatPac model. 167 The facility of 115,000 kWh per year capacity 168 (Idjis and Costa) is assumed to have ten laborers per shift. 5. Remanufacturing facility cost 11.35 ($/kWh) Based on 168 (Idjis and Costa) , the cost of a remanufacturing facility was assumed as an investment of $1,124,990 for a capacity of 115,000 kWh/ year, with a fixed cost of $1,192,854 per year. 6. Material cost 4.56 ($/kWh) + 7 ($/kg) for a 23 kWh batter y Material costs were assumed for the battery pack and BMS from Batpac as 4.56 $/kWh + $7/kg for a 23 kWh battery. 167 7. Overhead cost 5 (% of material cost) + 56 (% of labor cost) Total overhead costs were assume d to be 5% of the material cost and 56% of the labor cost. 167 8. Allowable profit for remanufacturin g 10 (%) Assumed. 9. Recycling cost in 2020 4 ($/kg) 169 with a yearly 5% decrease. 10. Increase in recycled material value in 2020 5 (%) Assumed. 11. Inflation rate ( f) 2.5 (%) 102 12. Nominal discount rate (i™) 6.9 (%) 102 13. Start year of EOL managemen t 2020 Assumed. 14. Year of a switch from early adopter (EA) to the general public (GP) 2028 Assumed. 15. Used product discount factor (EA) 61.52 (%) A weighted average of EA survey results for discount needed to purchase SLB. 16. Used product discount factor (GP) 58.53 (%) A weighted average of GP survey results for discount needed to purchase SLB. 17. Demand for SLB versus new LIB (EA) 37 (%) Survey result for direct question (EA). 18. Demand for SLB versus new LIB (GP) 44 (%) Survey result for direct question (GP). 19. Technically reusable fraction 95 (%) Assumed. 20. Collection efficienc y 100 (%) Assumed. 21. EOL battery buying cost for remanufacture r 0 ($/kWh) Assumed. 41 The remanufacturing cost is the amount that the remanufacturer will incur when remanufacturing a second-life ba ttery. The cost was determined using a bottom-up approach, considering the cost of electricity, labor, transp ortation, equipment, and f acility needed for the remanufacturing process, similar to the methodology from previous literature, 126 but modified by adding the demand for SLBs. The demand for energy storage was based on the National Renewable Energy Laboratory™s (NREL™s) standard scenarios Mid-case scenario. 164 The results from the survey (specifically, the question on willingness to choose SLB over a new battery) were used, ignoring the fidon™t knowfl answers to determine the demand for SLBs. 163 No replacement was considered for either EV battery or SLB. All EOL batteries were assumed to be colle cted for EOL management. Of the collected batteries reaching the remanufactu ring facility, 95% were assumed to be technically reusable. Remanufacturing was assumed to be carried out by dismantling to module-level since cell-level dismantling can be difficult and expensive. 170 The initial remanufacturing facility was assumed to have a production capacity of 500 MWh per year. Remanufacturing facilities were added yearly according to the demand. Estimating the costs incu rred in remanufacturing was a challenge, especially given the emerging natu re of the market. Assumptions were made based on previous literature and are prov ided in Table 3. 1. For recycling, manganese, lithium, aluminum, cobalt, copper, nickel, steel, and iron were all recovered at 95% recy cling efficiencies (See Ta ble B1 in Appendix B). 171 The weights of the battery material components were obtained based on the battery chemistries from literature. 125,172 The LIBs were assumed to of LiCoO 2 (LCO), LiMn 2O4 (LMO), LiF 2PO4 (LFP), and LiNiMnCoO 2 (NMC) chemistries. The changing battery chemistries in the future were incorporated in the model based on previous reports and publications on end-of-life electric vehicle batteries 134,173,174 (see 42 Figure 3. 4 (a) for the chemistry fr actions that reach EOL for each year). For the baseline, the NMC chemistry over time was assumed to ch ange from the NMC111 chemistry (LiNi 0.33 Mn0.33 Co0.33 O2) to NMC622 chemistry (LiNi 0.6 Mn0.2 Co0.2 O2), and finally to the NMC811 chemistry (LiNi0.8 Mn0.1 Co0.1 O2), based on the BNEF 2019 report on recycling 169 (see Figure 3. 4 (b)). In this work, three additional cases were considered when the NMC chemistries did not change over time to estimate the impact of battery chemistry on the recycling benefits. Case 1 assumed that all the NMCs to be the NMC111 chemistry (LiNi 0.33 Mn0.33 Co0.33 O2), Case 2 assumed all NMCs to be the NMC622 chemistry (LiNi 0.6 Mn0.2 Co0.2 O2), and lastly, Case 3 assumed all NMCs to be the NMC811 chemistry (LiNi0.8 Mn0.1 Co0.1 O2). Figure 3. 4. Changing chemistries of (a) end-of-life lithium-ion battery chemistries and (b) end- of-life NMC battery chemistries over time. The recycling cost is the cost that the recycler incurs in recycling an end-of-life battery. The material requirement and energy density were assumed to change with the changing chemistries over time based on the values shown in Table B2 in Appendix B. The recycling cost 43 was assumed to be $4/kg in 2020, 169 with a decrease of 5% every y ear. The material prices were obtained from the United States Geological Survey (USGS) datasheets and industry surveys, 175Œ 181 and provided in Table B1 in Appendix B. Figure 3. 5. Detailed calculation of end-of-l ife economic value from remanufacturing and recycling pathways for the EOL manager, i.e ., remanufacturer and recycler, respectively. The net present value (NPV) at EOL was calcula ted, as shown in Figure 3. 5. The end-of- life economic value is the am ount obtained after subtrac ting the cost (recycling and remanufacturing) from the price (SLB selling pric e and secondary material price/value) for the EOL manager (recycler, remanufacturer). The NPV can be obtained using Eq. 3-2, 89,90 where CAnn is the annualized cost, i is the real interest rate, i™ is the nominal discount rate, f is the expected inflation rate, and N is the year considered (Refer Table 3. 1 for the values). NPV = CAnn /(1+i)N –Eq. 3-2 i = (i™-f)/(1+f) –Eq. 3-3 The model equations for the baseline EOL va lue model (considering the NMC chemistries to change over time) are given in Appendix C. The system dynamics model was used to evaluate the changing EOL battery availability and also the im pact of factors, such as battery state-of-health and battery chemistry, on remanuf acturing and recycling value (economic value, carbon footprint, material requirement for nickel and cobalt). 44 To estimate the carbon footprint trade-offs, a life-cycle a ssessment was conducted. Firstly, the number of total EOL batteries, the number of SLBs produced, the total demand for batteries, and the number of recycled batteri es were obtained from the SD model. Using these values, the carbon footprint associated with the production of one kWh of lithium-ion battery for each year from 2017 to 2050 was calculated using the TRACI2.1 method 182 in SimaPro 8.5 software. 108 The system boundary is drawn around the production a nd the end-of-life management phases for the electric vehicle battery, as shown in Figure 3. 6. Figure 3. 6. System boundary c onsidered for this study, including the raw material extraction, electric vehicle batter y production, and the end-of-life management pathwa ys of recycling and remanufacturing. The carbon footprint of EV battery pr oduction was from Chul et al. (2016). 59 The recycling process emissions for the different battery che mistries were obtained fr om Mohr et al. (2020). 183 The recycled materials were assu med to be avoided products and provide carbon bene fits from the recycling process; their inventory was obtained from the Ecoinvent 3.6 database. 184 The remanufacturing carbon f ootprint was obtained from the previous works, 15,113 and the use of an SLB was also assumed to avoid the producti on of a new lithium-ion battery, providing carbon benefits. 45 3.5 Results and Discussion 3.5.1 Testing end-of-life elec tric vehicle batteries Some of the tests were stopped earlier than e xpected due to laborator y closures with the pandemic in 2020. However, the capacity change with cycle number for each sample for each test is presented in Figure 3.7. The cycle number was ba sed on the cycle as defined in the test. Both residential storage and peak shavi ng application test profiles were based on standard tests, and the test duration was converted to the time of real -life application based on the standard. However, since the fast charging test profile had been de veloped, the duration was calculated as follows. The charging demand was assumed to be at a frequency of four charge -discharge cycles per day, i.e., 1,460 cycles per year. It was seen based on the samp les which had undergone complete tests, residential storage can be a promising application for SLBs. Additional samples were added for Te st 1 to confirm this hypothesis. Three of the six completed samples lasted for a time equivalent to more than five years in real-life application. Though samples 7 and 8 showed promise, they were stopped early. For peak shaving, two samples out of three lasted for a time equivalent to more than one year in real- life application. Of the two, sample 1 showed pr omise but has been stopped. The fast charging application was a highly demand ing application. Of the three sa mples, two lasted a duration equivalent to more than one year of real-life applica tion, and one lasted a dur ation equivalent to more than two years of real-life application. The samples also showed capa city readings that had high variability (see Figure 3. 7 (c )), which could be due to the te sting profile and battery health. More testing would be required to arrive at a conclusive result. However, these results can still be used to model second-life applications. 46 Figure 3. 7. Test results for (a) resi dential energy storage, (b) peak shaving, and (c) fast charging applications with the samples given in the legend. 47 3.5.2 System dynamics modeling of e nd-of-life management of elect ric vehicle batteries in the US The system dynamics model was developed ba sed on the causal loop diagram shown in Figure 3. 3. The model was tested for dimensiona l consistency, and an extreme condition test was used to check whether the model solutions were logical under extreme conditions, as recommended in the literature. 185 After the model was tested, the effect of cha nging parameters, such as EOL battery health factor and EOL battery chemistry, on material requirement, economic value, and carbon footprint was studied for recycling and re manufacturing. Figure B2 in Appe ndix B presents the demand for lithium-ion batteries modified due to the availa bility of SLBs. The demand over time for new batteries decreased with the introduction of remanufacturing and SLBs. The health factor (and therefore, SOH) of the EOL battery impacted the price at which a SLB would be sold (based on Eq. 3-1) and the economic value of remanu facturing. The SOH was represented by the health factor , which was assumed to be 80% , 70%, and 50%. Figure 3. 8 (a) shows the change in the SLB price with the health factor in comparison to the new battery price. As the health factor was reduced from 80% to 50%, there was a 37.5% decrease in SLB selling price. Similarly, Figure 3. 8 also shows the value of remanufactur ing. As the SLB selling prices decreased over time (Figure 3. 8 (a)), the value pe r kWh of battery decreased (Figure 3. 8 (c)). The number of available EOL batteries impacted the overall value from remanuf acturing, as seen in Figure 3. 8 (b). Even if the health factor was low, it would be helpful to in crease the perception of SLBs among the early adopters and the general pub lic, not only to increase demand but also to increase the SLB selling prices that are acceptable to consumer s. Currently, the values were obtained from the survey results.163 48 Figure 3. 8. (a) The price of new batteries and second-life batteries with differing health factors (80%, 70%, and 50%). The economic value of remanufacturing for remanufacturer in (a) million $/year and in (b) $/kWh/year when second-life batteries with differing he alth factors (80%, 70%, and 50%). The economic value of recycling for recycl er in (a) million $/year and in (b) $/kWh/year when second-life batteries with differing NMC chemistry changes (Baseline, Case 1 with all NMCs as NMC111, Case 2 with all NMCs as NMC 622, and Case 3 with all NMCs as NMC811). Similarly, the recycling value was calculated fo r the different cases mentioned in the earlier section. Figure B3 in Appendix B shows the ch anging demand for cobalt and nickel due to the availability of recycled cobalt and nickel for th e baseline scenario wher e the NMC chemistry is expected to change over time. With remanufacturing, there was an additional reduction in material demand as the need for producing new batteries was compensated for by SLBs meeting the demand for batteries. Similarly, the avai lability of recycled lithium, ma nganese, aluminum, steel, iron, and copper, with respect to the EV ba ttery life-cycle, was also obtained. The initial decline in the cobalt demand was due to the NMC chemistry changi ng over time. As NMC111 batteries will get recycled, more secondary cobalt will be obtained. But as the chemistry will progress towards NMC811, the cobalt obtained from re cycling will decrease, as exp ected. The net present economic 49 value of recycling is also pres ented in Figure 3. 8. Over time , the economic value of recycling increased in all cases, as shown in Figure 3. 8 (d ) and (e). The initial ec onomic value was negative but increased over time. Figure 3. 9. Economic value of recycling and re manufacturing in (a) mi llion $/year and (b) $/kWh/year for recycling only vs. recycling with remanufacturing pathways for the baseline. The total end-of-life economic value of both recy cling only and recycling with remanufacturing pathways is shown in (c) million $/year and (d) $/kWh/year for the baseline. Figure 3. 9 presents the value of the two end-o f-life pathways separately in (a) and (b), and in total in (c) and (d). As seen in Figure 3. 9 (a), the value from the remanufacturing process was less than 100 million $ at any given time compared to the recycling value, which could be higher than 1,200 million $ by 2050. However, when the total end-of-life value was considered (Figure 3. 9 (c)), the value from remanufacturing reduced th e losses from the recycling in the initial years (2020-2025). Considering the value in $ per kWh gives a better idea in this regard (Figure 3. 9 (b) and (d)). The time when the recy cling value was lowest corresponds to when the remanufacturing value was the highest, allowing them to compensate for each other. However, as the new lithium- 50 ion battery's cost decreased, the remanufacturing value decreased, re ducing the overall end-of-life value of the recycling with the remanufacturing pathway. Figure 3. 10. (a) Total battery capacity needed to be produced equivalent to one kWh capacity demand. The carbon footprint in kgCO 2 eq. per kWh of battery capa city, given that (b) only recycling is considered and (b) recycling with remanufacturing is considered. The carbon footprint of the two pathways was determined per kWh of battery demand each year. Figure 3. 10 (a) presents the total battery capacity that needs to be produced to meet one kWh demand. When remanufacturing was added to the end-of-life mana gement pathway, SLBs avoided the production of new batteries, reducing the finet production.fl Figure 3. 10 (b) and (c) show the carbon footprint associated with the recycling and remanufactu ring pathways, respectively. The 51 figures show the carbon footprint a ssociated with battery production in gray, the recycling process in light red, and the remanufacturing process in dark red. The avoided carbon footprint due to recycling (recycled materials) is shown in light green, and that due to remanufacturing (SLB use) is shown in dark green. Compared to the recycling only pathway, the recycling with remanufacturing pathway had a net reduction in the carbon footprint per kWh. Overall, the results showed the complementar y nature of the two pa thways. In terms of carbon footprint, there was a 2-16% reduction when remanufacturing is adde d to the EOL pathway. This was also seen in terms of materials, especia lly cobalt and nickel. With the reduction in overall battery production requirement, coup led with recycled materials, there was a reduction in materials when remanufacturing was added to the recycling pathway. However, in terms of the cost trade- offs, the economic value from remanufacturing is lower than that from recycling, especially beyond 2035, due to the lower price of new batteries, lower demand for SLBs (compared to total demand), and the impact of public perception. Re manufacturing could still compensate for the losses otherwise incurred by recycling in the initi al years. The value from remanufacturing can be increased by increasing the percei ved value of SLBs in the eyes of the consumers by providing proper warranty and certification. Policy can al so support remanufacturing by providing subsidies and incentives, such as the tax credits for photovoltaics and electric vehicles. 3.6 Conclusions In this chapter, we conducted cell-level te sting of EOL EV batteries to ascertain their performance in three second-life applications. Due to the pandemic in 2020, some of the tests were stopped earlier due to laboratory closures, and additio nal replicates could not be added to test more variability. However, the results from a minimum of three samples were obtained for each test. Based on the completed samples, the residential energy storage application was identified as a 52 promising second-life ap plication for EOL EV batteries. Thou gh incomplete, the results presented in this chapter can be used for mo deling the second-life applications. A system dynamics model was also develope d to evaluate the changing EOL battery availability and the impact of factors, such as battery state-of-health a nd battery chemistry, on the end-of-life value. It was found that as EOL battery state-of-health decrease d (in terms of health factor), the selling price, as well as the remanu facturing value decreased. As the health factor reduced from 80% to 50%, there was a 37.5% decrea se in SLB selling price. If subsidies or other incentives could reduce remanufacturing costs, batte ries with lower health factors could still be remanufactured and sold with appropriate warranties. Currently, public perception was incorporated by using the results fro m the surveys in a companion paper. 163 However, these scores can be increased by appropriate warranty and prici ng, as noted in the said paper. These can impact the results presented here. The impact of battery ch emistry was seen to be key for recycling value, as chemistries with lower cobalt content will provide lower value. Changes in recycling efficiencies and recy cling technologies, such as direct recycling, can incr ease the value despite the lower cobalt content. Remanufacturing of EOL electric vehicle batter ies can, therefore, support reliable and stable energy, which is also economically and environmentally sustainable. Considering the different aspects of remanufacturing can further stre ngthen the end-of-life ma nagement of electric vehicle batteries. Remanufacturing can not only make the energy sector sustainable, but it can also support the electric vehicle industry in reducing battery wast e and also providing charging infrastructure. To enable rema nufacturing, however, there needs to be appropriate policy support that can incentivize SLB use by providing tax credits or subsidie s, as was the case with PVs. 91 53 Mandatory EOL management of EV batteries can also support both remanufacturing and recycling, encouraging original equipment manufacturers to innovate in these areas. Energy storage, however, is just one piece in the larger jigsaw puzzle of providing reliable and stable energy for the world. We will also need to consider energy generation to fully access sustainable energy. While we have discussed the integration of batteries with photovoltaics in Chapter 2, large-scale deployment of photovoltaics, and renewable energy sources in general, can still lead to reliability issues, which are usually countered with energy storage and fossil fuel generation. A low-carbon alternative to fossil fuel generation is nuclear energy. However, changes in technologies in different stages of the nuclear fuel cycle have risen w ith changes in resource availability, such as in the uranium extractio n stage. Proper modeling of these changes is imperative to ensure the proper modeling of the e nvironmental impacts of the nuclear fuel cycle, which we will discuss more in detail in Chapter 4. 54 4 A comprehensive life-cycle assessment of uranium extraction by alkaline in-situ leaching in the US In this chapter, we focus on uranium fuel for nuclear energy, especially the uranium extraction method of alkaline in -situ leaching, which is the most prominent method in the US. Since the late 1950s, nuclear energy has become a major source of electricity generation around the world, meeting 11% of the wo rld™s electricity demand in 2018. 186 Future electricity generation sources are expected to have lower associated carbon footprint or global warming potential due to a shift in energy and environm ental policies across the globe. 2 In a carbon-cons trained world, the combination of low cost and low carbon emissi ons, along with the maturity in technology and flexibility with newer technology, has made nuclear energy a desirable energy source. 187 At the same time, there is uncertainty in the future world nuclear energy cap acity due to the impact of the Fukushima Daiichi accident and renewable energy prices on national nuclear policies of different countries. Therefore, the International Atomic Energy Agency (IAEA) has created various growth scenarios. The best-case scenario considers dela yed retirements of exis ting plants and faster capacity additions, projecting the 2018 world nu clear power plant capac ity to almost double by 2050. Conversely, the worst-case scenario assu mes faster retirements and limited capacity additions, leading to the nuclear power plant capacity reducing to 1990 levels by 2040 and then increasing again (see Figure 4. 1).188 Nuclear energy has a comparable carbon fo otprint with renewable energy sources.189 The global warming potential of nuclear energy base d on life-cycle assessment (LCA) studies ranges from 3.2 to 172.5 gCO 2 eq./ kWh. 33,34,43Œ51,35Œ42 This wide variation can be attributed to the differences in the system boundaries or the assumptions in each lif e-cycle phase. For instance, the fuel requirement and en richment depend on the type and mode l of nuclear power plants, thereby 55 changing the GWP of nuclear energy. For exampl e, a Canada Deuterium Uranium reactor has a GWP of 3.2 gCO2 eq./ kWh,33 while a pressurized light wate r reactor has a GWP of 34 gCO2 eq./ kWh.40 Figure 4. 1. World nuclea r capacity from 1970 to 2017 with projections of nuc lear capacity until 2050 as per International Atomic Energy Agency (IAEA), 188 along with uranium production and demand190 between 1970-2018. The uranium extraction phase contribution is 3 to 47% of the total nuclear fuel cycle GWP, 33,34,52Œ54,35Œ42 as shown in Figure 4. 2, with an average of 38%. 45 As ore grades decrease, the uranium extraction process efficiency can decr ease unless compensated for by the technology advancements in the future. The use of an ore grade with 0.01 % U3O8 instead of 0.15% increases the nuclear fuel cycle GWP from 34 gCO 2 eq./ kWh to 60 gCO 2 eq./ kWh (77% increase). 40 A separate study echoed these findings showing a 188% increase in th e nuclear fuel cycle GWP with the reduction of ore grades from 10% to 0.013%.37 56 Figure 4. 2. Global warming potential (gCO 2 eq./ kWh) of the nuclear fu el cycle from various life- cycle assessment studies. 4.1 Long-term environmental sustainability of uranium extraction Open-pit mining, underground mining, and in-situ leaching are the three main methods of uranium extraction. 55 Open-pit and underground mining met hods are conventional methods that include excavation, blasting, and milli ng of the ore. For in-situ leaching (ISL), a leaching solution (also known as lixiviant) is pumped into the uraniu m ores through vertical bore wells, as shown in Figure 4. 3 (a). The uranium ions leach directly from the ore into the lixiviant, which gets pumped back to the surface. 191 The uranium is subsequently separa ted from the lixiviant, generally using an ion-exchange process.191 Open-pit and undergrou nd mining methods are suitable fo r high-grade uranium ores. In contrast, in-situ leaching is more economical for lower-grade ores. As uranium demand and production increased (see Figure 4. 1), 188 higher-grade resources were mined, leading to a global decrease in uranium ore grades. 40 In-situ leaching has lower capital costs than conventional mining methods, especially for lower grade ores. 192 While open-pit and underground mining methods were most prominent in the early 1950s, now, in-situ l eaching is emerging as the most suitable method 57 to meet the uranium demand. Figur e 4. 3 (b) shows the trend of th e three production methods from 1996 to 2018. The in-situ leaching process contributed to 13% of the world™s uranium production in 1996, increasing to 44% in 2011 and 55% in 2018, showing an increasing contribution over time.55 Figure 4. 3. (a) Schematic of in-situ leaching with a hexagonal well st ructure with a central recovery well passing through an aquifer and an ore body; (b) fract ion of world uranium production by extraction method. Most life-cycle assessment (LCA) studies of uranium extraction have focused on the GWP of open pit and underground mining, 33,34,50,193,194,35Œ39,41,42,46 despite in-situ leaching being the current common method of uranium extractio n. The average uranium extraction GWP is 2.66 gCO2 eq./kWh. 33,34,50,193Œ197,35Œ38,40Œ42,46 A recent estimate of the GWP of Canadian mining with relatively high ore grades of 3.81% U 3O8 was 0.53-1.1 gCO 2 eq./kWh. 194 The few LCA studies on ISL have focused on the ac idic process, typically in Australia, with varying results. While one study reported the GWP of extracting uranium in the Beverley mine, Australia as 38.1 kgCO 2 eq./ kg U3O8,195 another study estimated the GWP to be twice as much. 197 However, there were differences in assumed material inputs, such as choice of ion-exchange re sin. The latter study also reported in-situ leaching to have higher energy and material consumption than open-pit and underground mining methods in Australia for similar ore grades. 58 At present, the environmental impact of ur anium extraction through alkaline ISL is not well-represented. Acidic ISL has higher uranium recovery than alkaline ISL, making it relatively common and, therefore, we ll-represented in literature. However, when the ore body consists of significant amounts of acid-consumin g minerals, such as gypsum or limestone, alkaline ISL should be used. 192 Acidic ISL has also come under criticism due to concerns associated with site remediation and groun dwater restoration. 198,199 The lixiviant is usuall y pumped back into the aquifer, where natural attenuation is expected to reduce the cont aminant concentration over time. However, this process can take decades, depending on the local geology, and th ere is a lack of data on the mechanisms and the required time for natural restoration. 198 When considering the restoration cost, acidic ISL was not necessarily cost-effective compared to alkaline ISL. Acidic ISL can also cause mineral alterations and por e-plugging through chemical processes, which are challenging to address.200 Alkaline ISL can play a more prominent role in uranium extraction as the uranium ore grades decrease further, given it s relatively lower impacts and cost s. The US relies on alkaline ISL for domestic uranium extraction. With recent repo rts calling for increasing uranium self-reliance in the US, 201 using alkaline ISL for uran ium production is likely to increase, which needs to be included in the US nuclear fuel cycle. To en sure that nuclear energy remains a low-carbon energy source, changes in the processes, especially uran ium extraction, need to be appropriately modeled and assessed. The main goals of this chapter were to (1) evaluate the life-cycle environmental impacts of alkaline in-situ leaching for ur anium production, (2) assess the eff ect of ore grade variation on the environmental impacts of alkaline in-situ leach ing, and (3) compare our results with those in the existing literature based on ore grade and extraction method. To accomplish this, the focus was 59 on the alkaline in-situ leaching pr ocess for uranium extraction in US-based mines since the process is most common in the US. 4.2 Methods The goal of this work was to evaluate the li fe-cycle environmental impact of alkaline ISL for uranium extraction. A cradle-t o-gate life-cycle assessment of the US-based alkaline ISL process was conducted. The cradle-to-gate global warming pote ntial, acidification potential, and land use of uranium extraction from ISL, includin g the ion exchange process, were evaluated using the ReCiPe Midpoint (H) met hod in SimaPro v8.5 software. 108 Figure 4. 4 shows the cradle-to-gate life-cycle stages of uranium ore extraction using in-situ leaching in the US. The system boundary included the uranium extraction proce ss and the transportation of uranium ore concentrates to the refining facility. The downstream pro cesses of uraniu m conversion, enrichment, fuel rod fabricati on, nuclear energy gene ration (reactor construction, operation, and decommissioning), and spent fuel management were excluded. All processes of mining and chemical production were assumed to occur in the US. The functiona l unit was one kgU of uranium extracted and transported to a uranium refining facility. Mining was assumed to last for two years per mine, as typically se en in many mine sides. 196 The full nuclear fuel cycle was not explicitly modeled in this work. In the US, currently, seven out of the eight uranium mines use alkaline in-situ leaching. 55 A typical hexagonal well-field pattern with 20 m si des, a central recovery well, and a well depth equaling the ore depth196 was considered as shown in Figur e 4. 4 (a). High-alloyed steel was assumed for the steel used in the wells. 196 Uranium ore grades typically range from 0.036% to 0.26% in the US. 202Œ205 Uranium resources with hi gher ore grades of 0.4% U 3O8 have also been 60 found in Roca Honda, New Mexico. 205 For this study, the uranium ore grade range was 0.036% to 0.4% U 3O8. The ore depth was 120m, which is typical for the US. 196,206 The choice of the lixiviant depends on the composition of the host rock and ore, the cost of reagents, recovery rate, and environmental considerations.199 Since most uranium ores in th e US are carbonate -based, alkaline lixiviants are common in the US. Generally, in th e US, the ISL method uses an alkaline lixiviant (sodium carbonate) in water 207 and air208 to also prevent contaminati on of drinking water aquifers that may exist near the ore. In this analysis , the carbonate concentration in the lixiviant was assumed to be 0.3-1.5 g per liter for ore grades ranging from 0.03 to 0.05% U3O8.199 Figure 4. 4. System boundary for this study, shown as a dotted line, includes the uranium extraction process and ion exchange process. The recovery rate is a function of the ore gr ade, as derived by prev ious studies on uranium extraction 37,40,209,210 and shown in Eq. 4-1 and Eq. 4-2. Eq. 4-1 is based on the mining and milling yield from existing mine (open-pit, underground, and ISL) and labor atory-scale data. In contrast, Eq. 4-2 is based on the mining yield from exis ting mine (open-pit, underground, and ISL) data. 53 Eq. 4-2 has slightly lower extraction recoveries compared to Eq. 4-1, leading to lower carbon intensities. 40,53 The average recovery rates were calcul ated using the two e quations. Alkaline ISL has lower yields compared to acid ic ISL. The overall rate was assu med to be 90% of that obtained from Eq. 4-1 and 4-2 to account for the lo wer recovery rate of alkaline ISL. 192,199 Section D1 in 61 Appendix D provides the recovery rates obtai ned from Eq. 4-1 and 4-2 for each ore grade considered. Recovery rate %98.07.23 log%UO –Eq. 4-1 ecovery rate %94.702.97 log %UO –Eq. 4-1 The average uranium concentration in the recove ry liquid (lixiviant with leached uranium) was 120-150 mg/l.211 Typically, 1-3% of the lixiv iant is lost due to production bleed, which is the amount of lixiviant discarded to ensure a negati ve water balance that min imizes the spread of contaminants in the groundwater. 211 To include loss due to producti on bleed as well as evaporation, a conservative estimate of 5% water loss le d to 7,018-8,772 liters of water to extract one kgU. 212 A weak acidic cationic exchange resin (IRC-84) was assumed to separate uranium from the alkaline lixiviant. 213 Due to a lack of information on alkaline ISL methods, the amount of resin used was assumed to be s imilar to acidic ISL methods 195 after accounting for differences in ore grade. Similarly, the emissions to air, water, and soil during the leaching process were based on results from previous work on acidic ISL.196 In a typical ion-exchange process, the le achate first flows through the resin, and the uranium ions are adsorbed. The resin is subseque ntly flushed with a concentrated brine solution, which displaces the uranium ions in to the solution. In this analysis , the so-produced eluent solution was assumed to have a uranium concentration of 8-20 g/l. 211 Hydrogen peroxide was assumed to be used for precipitating uranyl peroxide, which was then settled, washed, filtered, dewatered, and finally transported to the refining facili ty, as is typical of these operations. 211 The freight transport to the refining facility was assumed to be 114 ton- miles by road, which is t ypical of transportation of mining goods (non-oil and petroleum) fo r the year 2017. 214 The uranium was transported to the 62 refining facility in a steel drum weighing 345 kg. The steel drum can carry 360 kg of uranium ore concentrates or 305 kg of uranium equivalent. 215 The processes and flows of electricity and chemical production were from th e Ecoinvent database v. 3.5 216 and DATASMART LCI databases,110 which corresponded to the US conditions when necessary. Tables D2 and D3 in Appendix D show the input flows for the uranium extraction process, including the uranium ore, water, energy, chemicals, other resources , and transportation for the two cases. These results were also compar ed with previous life-cycle assessment studies of uranium extraction, using the most common metricŠgloba l warming potential (GWP). However, it is worth noting that the GWP might not be a key me tric for the environmental impact of uranium extraction. LCA literature since 1989 was system atically collected a nd reviewed using the methodology in Warner et al. (2012). 217 Articles were collected that either modeled the global warming potential of uranium extraction or that of the entire nuclear cycle (irrespective of reactor technology) and provided the contribution of uranium extraction. The functional unit was converted to kWh to match all studies. Currently, most of the installed nuclear reactors are pressu rized light water reactors (PWRs). 218 An average PWR uses ~3kg of enriched uranium per GW h electricity produced, which translates to a requirement of 27 kg of natural uranium in a typi cal nuclear reactor, or 20 kg in a newer reactor for every GWh electricity produced.219 Therefore, a factor of 2.7×10-5 kgU/kWh was used to convert kgCO2 eq./ kgU from this study (and any other study coll ected in the literatu re review) to gCO 2 eq./kWh, based on the methodology developed by Parker et al. (2016).194 63 4.3 Results and Discussion The associated GWP, terrestrial acidificati on, and land use for ex tracting one kgU were calculated for select ore grad es. Figure 4. 5 presents the gl obal warming potential (GWP), terrestrial acidification, and water consumption for uranium extraction using alkaline ISL in the US for the lowest ore gr ade considered (0.036% U 3O8) to err on the conservative side. Figure 4.5 shows the total value as well as the percentage contribution from the indi vidual components (such as steel, water, chemicals). The total GWP for the process was 130.37 kgCO2 eq./ kgU. Previous studies on acidic ISL (o re grades of 0.2-0.24% U 3O4) did not consider transportation and steel in the wells,195,197 leading to lower GWP values co mpared to our results (38, 73.895 kgCO 2 eq./ kgU, respectively). Figure 4. 5. Global warming potential (GWP), terrestrial acidificat ion (Ter. Acid.), and land use of uranium extraction using alkaline ISL in the US for ores of grade 0.036% U3O8. The use of diesel was the highest contributo r to the GWP and terrestrial acidification (60%, 74%, respectively) associated with ur anium extraction. Diesel provides energy for the machinery in the uranium mines, which are mostly in remote locations. However, using renewable energy sources, along with energy storage, coul d bring down the impacts due to energy use for 64 uranium extraction. A previous study on acidic ISL al so found fossil fuel use to contribute majorly to the GWP of uranium extraction. 197 Another study on acidic ISL assumed electricity from the grid, leading to even lower GWP (38.1 kgCO 2 eq./ kgU). 195 The assumptions in the studies, such as energy source, can lead to underestimating the global warming potential of uran ium extraction, especially given the remote locations of most mine s. These studies also did not consider the steel used for wells and transportati on. Excluding the steel used in th e wells and transportation, our GWP value was 100.1 kgCO2 eq./ kgU for alkaline ISL of 0.036% U 3O8 ore grade, while the reported GWP for acidic ISL of 0.2% U 3O8 ore grade was 73.895 kgCO 2 eq./ kgU. 197 Chemical production contributed the most to land use, accounting for 52% of the total, due to the land requirements of limestone mining. The second-largest contribu tor to land use was the land occupation and land use associated with the wells themselves, ac counting for 20% of the total. Reducing the requirement for chemicals can be done in various ways . Recent studies have explored the combination of in-situ leaching and bioleac hing, where microorganisms can regenerate the lixiviant, thereby reducing the requirement of chemicals. 220,221 65 Figure 4. 6. (a) GWP (kgCO2 eq./ kgU), (b) terrestrial acidification (kg SO 2 eq./ kgU), and (c) land use (m2a /kgU) of uranium extraction using ISL in the US as a function of ore grade. The recovery rate decreases with decreasing ore grade, leading to higher environmental impacts. Figure 4. 6 emphasizes this point by s howing the increase in global warming potential, terrestrial acidification, and land use with decreasing ore grades (from 0.4% to 0.036% U 3O8). As the ore grades fell from 0.4% to 0.036%, the GWP and terrestrial acidification increased by 12.9%, while land use increased by 42.1%, on average. The d ecrease in ore grades led to a considerable increase in land transformation and land occupation during the ISL process, which contributed considerably to the land use but not to GWP and terrestrial acidification (s een as blue in Figure 4. 6). Detailed results are provided in Section D3 of Appendix D. 4.3.1 Comparison with existing literature To understand the effect of cha nging extraction methods, these results were also compared with previous lite rature. The studies included in this compar ison are organized in chronological order in Table 4. 1, al ong with the met hodological details and assumptions. Also, the GWP of 66 uranium extraction and the full nuclear cycle in gCO 2 eq./kWh are presented in Table 4. 1. When the functional unit differed, the results were converted into the units of gCO 2 eq./kWh using a factor of 2.7×10 -5 kgU/kWh. 194 It should be noted that until 2011, in-situ leaching had not been considered in LCA studies or was approximated as open-pit or underground mining.37,222 Figure 4. 7 presents the GWP from this work and literature in terms of gCO 2 eq./ kWh. The results from studies that did not report the ore grades are given in the shaded area in gray. When considering the studies that report the ore grade, the GWP associated with uranium extraction was seen to increase as the ore grade decreases. The studies on ISL are shown as half- filled dots, and the current study is shown in ye llow. The median GWP of uranium extraction from all the reported estimates from literature was 1.56 gCO 2 eq./ kWh of electricity produced. The maximum GWP value was 67 gCO 2 eq./kWh 37 for the open-pit and underground mining of ore grades 0.06% U3O8. The minimum reported GWP for uranium extraction was for high ore grades of 12.7% U 3O8,38 typical of Canadian ores. The lower GWP values were also supported by other studies,33,42,194 which can be attributed to the high ore grades in Canada. The GWP increased with a decrease in ore grade, as found in this st udy as well as in literature (Figure 4. 7). The GWP of the entire nuclear fuel cycle could increase by 75% when ore grades fall from 0.15% to 0.01% U 3O840, reiterating the importance of ore grades to the environmental impact of the full nuclear fuel cycle. The study was motivated by the decrease in Australian ore grades from ~20% in the early 1900s to <1% in th e early 2000s. However, in-situ leaching was assumed to contribute only 28% of th e total uranium production, whereas this number is currently closer to 55% in the world. 55 Three LCA studies have explicitly modeled acidic ISL for Australian conditions.195Œ197 67 Table 4. 1. Review of nuclear fuel cycle life-cycle assessment studies , showing the global warming potential of mining and mill ing stage, as well as the total for 1 kWh of electricity, along with methodological details and assumptions. Authors (Year of publication) (Ref) GWP of fuel extraction stage (gCO 2 eq./kWh) Total GWP (gCO 2 eq./kWh) Fuel extraction contribution (%) Functional unit in the study Uranium mining method # Uranium ore grade (%) Mining location Full study location LCA method Data type* Reactor technology Life cycle stages considered San Martin (1989) 52 1.5 7.8 19.2 kWh Not specified USA PCA T BWR Full, without waste disposal, dismantlin g Andseta et al. (1998) 33 0.22 3.2 6.9 - 8.8% TWh OP 75%/ UG 25% 1.5 Canada Canada PCA E CANDU Full White and Kulcinski (2000) 34 0.4 15 2.7 GWh Not specified USA Hybrid T PWR Full Dones et al. (2005) 35 1.5 5-12 12.5 - 30 kWh OP /UG Not specified USA Europe PCA E PWR, BWR Full Hondo (2005) 36 0.9 22.2 4.1 kWh Not specified Australia, Canada Japan Hybrid E BWR Full Van Leeuwen and Smith (2005) 37,222 2.95-67 108.19- 172.5 2.7 Œ 38.8 kWh OP 60%/ UG 40% 0.15- 0.01 World World Hybrid T/E PWR Only fuel extraction Lenzen (2006) values from Beerten et al.(2009) 53 3.77 57.69 6.5 kWh Not specified Australia Hybrid T PWR Full Fthenakis and Kim (2007) 38 0.1, 1.7, 5.5 16, 24, 55 0.6, 7, 10 kWh Not specified 12.7, 0.2, 0.05 Australia, US, Canada US Hybrid T/E LWR Full Beerten et al. (2009) 53 0.83 7.72 10.8 kWh OP 62%/ UG 38% 0.2 Belgium Belgium Hybrid T/E PWR Full Doka (2011) Data from Ecoinvent 196 ~2.22 g per kWh (81.48 kg per kg uranium) NA kg ore ISL 0.168 Australia Australia PCA E NA Only fuel extraction 68 Table 4. 1 (cont™d). Authors (Year of publication) (Ref) GWP of fuel extraction stage (gCO 2 eq./kWh) Total GWP (gCO 2 eq./kWh) Fuel extraction contribution (%) Functional unit in the study Uranium mining method # Uranium ore grade (%) Mining location Full study location LCA method Data type* Reactor technology Life cycle stages considered Haque and Norgate (2014) 195 ~1.02 (38 kg per kg uranium) NA kg ore ISL 0.24 Australia Australi a PCA E NA Only fuel extraction Mudd (2014) 193 ~0.8856 (32. 8 kg per kg U3O8) NA kg ore OP/ UG/ ISL Not specified World NA PCA E NA Only fuel extraction Norgate et al. (2014) 40 1.3-27.4 33.9- 60 3.8 Œ 45.7 MWh OP 27%/ UG 45%/ ISL 28% 0.15 - 0.01 Australia Australi a PCA T/E PWR Full Poinssot et al. (2014) 39 1.704 5.29 32.2 kWh OP+UG 59%, ISL 41% Not specified Canada, Niger, Kazakhstan France PCA E PWR Full Zafrilla et al. (2014) 50 3.79 21.3 17.81 kWh OP/ UG Not specified World Spain Hybrid E BWR Full Ashley et al. (2015) 41 2 6.6 30.3 kWh OP-25%/ UG 25%/ ISL 50% 0.134 World World PCA E PWR Full Parker et al. (2016) 194 2.187, 1.728, 0.918 NA kWh, kg ore OP, UG, RB 0.74, 1.54, 4.53 Canada NA PCA E PWR Only fuel extraction Siddique and Dincer (2017) 42 1.61 3.41 47.2 kWh OP, UG 0.59 - 18.84 Canada Canada PCA E CANDU Full Farjana et al. (2018) 197 ~0.169, 0.72, 1.995 (6.273, 26.67, 73.895 kg per kg uranium) NA kg ore OP, UG, ISL 0.2 Australia NA PCA E NA Only fuel extraction # OP: open-pit mining, UG: underg round mining, ISL: in-situ leaching, RB: raise-bor e mining; * Data type is denoted by E when e mpirical or T when theoretical or T/E when it is a combination of empirical and th eoretical; ~converted to kWh b y assumin g a factor of 2.7×10-5 k g/kWh for a PWR 69 The GWP of uranium extraction depends on th e extraction method, as shown in previous literature.194,197 The Australian study found ISL to be the most GWP-intensive of the three methods for the same ore grade. 197 Only a few studies have analyzed the GWP of in-situ leaching. The literature has focused on Australian conditions with the acidic process, 42,193,195 which includes an estimate from the widely used life-cyc le inventory databa se, Ecoinvent v. 3.5. 196 The average GWP of the acidic ISL process in previous literature was 1.745 gCO 2 eq./kWh, 40,195Œ197 which is higher than the median reported for the open-pit and underground mining processes (1.56 gCO2 eq./kWh). These results are in agre ement with the earlier mentioned stu dy. The GWP of ISL has been studied for ore grades ranging from 0.15% to 0.24%. In reality, th e grades for in-situ leaching can be much lower than those studied. For example, the ores in the US can go as low as 0.036% U 3O8.199 Decreasing ore grades over time can increase en ergy requirements and the associated GWP with in-situ leaching, similar to open pit and unde rground mining methods. The GWP of uranium extraction from the analysis presented in this chapter was 3.03 to 3.52 gCO 2 eq./kWh, averaging at 3.21 gCO 2 eq./kWh, almost twice the value of the av erage GWP of acidic ISL in the literature (1.74 gCO2 eq./kWh). It is also observed that th e GWP of the alkaline ISL proce ss was inversely proportional to the ore grade, as was seen for open-pit and underground mining.37,40 However, the GWP of ISL at 0.036% U 3O8 was 3.46-3.52 gCO 2 eq./kWh, which was lower than the open-pit and underground mining GWP at comparable ore grades of ~0.05-0.06% U 3O8 (5.5-67 gCO 2 eq./kWh). 37,38 The variation can be attributed to the differences in the processes (such as bl asting and chemical use) between ISL and conventional methods. 70 Figure 4. 7. GWP (gCO2 eq./ kWh) of uranium extr action using alkaline ISL in the US in this study in yellow (Cases 1 and 2, fu ll system boundary) compared with previous literature (in gray). The half-filled circles denote ISL. The studies that did not specify/ report the ore grades are shown separately in the shaded area in gray. It should be noted that the analysis in this chapter considered the transportation phase and the steel used in the wells, and the results corr esponded to the system boundary shown in Figure 4. 7. When disregarding the transportation phase and the steel used in the wells, the GWP for 0.036% U 3O8 in Case 1 was 2.7 gCO 2 eq. /kWh. This value was higher than the median GWP from all estimates for uranium extraction (1.56 gCO2 eq./ kWh) and also higher than that for acidic ISL. But when considering the effect of ore grade, these results were lower for comparab le ore grades, as previously stated. It is also important to model transportation an d steel. Together, they contribute to 23% of the uraniu m extraction GWP based on the system boundary in Figure 4. 7. 71 Most studies underestimate the GWP of the full nuc lear fuel-cycle by considering the average uranium extraction method in their analysis, without transportation or steel. This study has added to the life-cycle inventory of alkaline ISL while estimating the contributi on of transportation and steel on the environmental impacts of uranium extraction. 4.3.2 Implications of the changes in uranium demand, production, and ownership The demand and production for uranium have not always coincided geographically. Over time, there have been changes due to the public perception of nuclear energy and the policies at the governmental level. Another as pect related to uranium use is the ownership of the mines and the uranium produced. In c ountries such as China, Russia, a nd India, the state governments fully control uranium mining.55,188,223Œ225 With time, some countries have opened their doors for private and overseas mining companies to explore and mine. Some of the bigger companies with many mines across the world are the Russian Rosatom, the French Orano, and the Chinese China National Nuclear Corporation.55,188,223Œ225 Figure 4. 8 shows the demand, production, and ownership of uranium in 1988, 2017, and the projection for 2025. This data was obtained fro m the mine data from the Uranium Red Books of years 1988, 2007, 2011, and 2018 as well as the compilation of 40 years Red Books. 55,188,223Œ225 The total production values for the countries were obtained from the World Mining Data website.226 The nominal production from each mine was pr oportionally increased or decreased to match the year™s total production. The ownershi p of the produced uran ium (country, foreign, domestic, govt., private) was determined by multiplying the corresponding share of the owning entity with the mine™s production in the considered year. But for the future production, either the nominal production or the 2017 data was used, depending on the availability. 72 Figure 4. 8. The demand, production, and ownership of uranium produced in 1988, 2017, and 2025. All units are in tU. 73 In 1988, the demand, production, and ownership we re all concentrated in the then USSR and the US. In 2017, most of the production and ownership moved to Kazakhstan, Australia, and Canada, whereas most of the de mand was still concentrated in the US, along with China and Russia. The projected production and ownership in 2025 are expected to remain similar, with most countries in Europe and Asia playin g a larger role. Another important aspect to be noted is that the main producers of uranium, such as Australi a and Kazakhstan, did not have any demand for the product in 2017 and are projected to not have a dema nd in the future (2025) as well. In the future, the demand is expected to get concentrated in the Americas and Asia, with many countries planning on reducing their nuclear power. This detail can be expected to play an important role in the type of extraction method as well as trans portation. As the distance between the centers of production, ownership, and demand for uranium change or even increase, the associated transportation-related carbon footprint can have an increased impact. Modeling these relationships can also improve the life-cycle assessment of th e full nuclear fuel cycle and provide further implications on the supply of uranium. 4.4 Conclusions This chapter addressed the lack of a compre hensive life-cycle assessment of uranium extraction using the alkaline in-situ leaching pro cess. These results were modeled for the US scenario but can be applied elsewhere for alkaline in-situ leaching if the chemical use and energy sources are similar. It was f ound that the overall globa l warming potential of the alkaline ISL method in the US was higher than those reported for ISL in the literature, which can be attributed to the differences in assumed ore grades and th e process itself. However, for comparable ore grades, alkaline ISL had lower global warming potential than conventi onal uranium extraction methodsŠopen-pit and underground mining. He nce, the commonly held assumption of 74 approximating in-situ leach ing as conventional met hods can lead to the wr ongful estimation of the GWP of the nuclear fuel cycle. On a similar not e, when considering in-situ leaching, acidic and alkaline processes also differ in their environmental impacts and n eed to be modeled separately. With the addition of intermittent renewable en ergy sources to the grid, nuclear energy can provide the baseload and also al low for faster ramp rates to ensure stable energy generation. 32 While nuclear energy can generate low-carbon electri city, the different life-cycle phases and the changing technologies, especially in the uraniu m extraction phase, have an impact on the full nuclear-fuel cycle. As the carbon footprint of alka line ISL is high, future nuclear energy can have a higher carbon footprint and needs to be assessed by considering the changes in ore grades and processes. These need to be appropriately mode led to ensure that nuclear energy remains a low- carbon option that aids in decarbon izing the electricity grid. 75 5 Conclusions and major contributionsd This dissertation focused on the prospective life -cycle assessment of second-life electric vehicle batteries and uranium extraction in th e US. The two options are complementary and necessary for the decarbonization of the energy s ector. While the methods used to assess these options differ in certain respects, key implications of the obtained results we re discussed that are relevant at both policy and the broader research levels. The following two sections separate these implications based on the two energy options. 5.1 Sustainability implications of second-life electric vehicle batteries Firstly, second-life batteries were shown to ha ve the potential to lower the cost and carbon footprint for residential, commercial, and utility-l evel peak shaving applications. Compared to new LIBs, SLBs lowered the levelized cost of electricity and global warming potential in all applications and locations consid ered (Chapter 2). However, en ergy storage applications will require SLBs in large volumes and capacities. Th e uncertainty in SLB quality and availability may prove to be a limitation. As lithium-ion battery chemistries change ove r time, the corresponding battery degradation and second-life performance can change. 23,24,27 To ensure safety, economic viability, and avoid premature failure, SLBs need th orough testing and characterization of second-life performance and lifetime, 18 which has resulted in testing protocols227 and an Underwriters Laboratory standard d Parts of this chapter have been reproduced in part with permission from: (1) Kamath, D.; Arsenault, R.; Kim, H. C.; Anctil, A. Economic and Environmental Feasibility of Second-Life Lithium-Ion Batteries as Fast Charging Energy Stor age. Environ. Sci. Technol. 2020, acs.est.9b05883. https://doi.org/10.1021/acs.est.9b05883. Copyright 2020 American Chemical Society. (2) Kamath, D.; Shukla, S.; Arsenault, R.; Kim, H. C.; Anc til, A. Evaluating the Cost and Carbon Footprint of Second- Life Electric Vehicle Batteries in Residential an d Utility-Level Applications. Waste Manag. 2020. https://doi.org/10.1016/j.wasman.2020.05.034. Copyright 2020 Elsevier Ltd. 76 UL1974.228 Moderate cycling and appropriate choice of second-life applications can also ensure that SLBs do not undergo increased degradation, as was seen in Chapter 3 with the residential application compared to the more aggressive fast charging application. With changes in battery chemistry, the future EOL battery stream is expected to have multiple battery chemistries, impacting the recycling process, as seen in Chapter 3. Suitable efforts, such as labeling and pre-sorting, can improve the purity of the recycling output stream even for multiple battery chemistries. 229 Solutions applicable for EOL vehicles can also be applied to EOL management of SLBs, such as worldwide regulati ons on reverse logistics and the use of renewable energy for various processes. 230 Technologies, such as internet-of- things, can be adopted to obtain relevant battery information, e.g., the disassemb ly method, which can streamline the recycling process and the repurposing process.231 Remanufacturing can divert the flow of end-of-life batteries away from recycling. On the other ha nd, using SLBs in applications, which would otherwise have required new LIBs, can reduce new mining. We see in Chapter 3 that this was indeed the case, reducing material requi rement and carbon footprint by introducing remanufacturing along with recycling as an EOL management pathway. Stakeholders such as the automotive indus try and battery manufacturers are already involved in SLB dem onstration projects,5,19,28 showing interest in remanufacturing from different sectors. However, the SLB market is still in a nascent stage. The successful deployment of SLBs for various applications presented in this disser tation and the broader literature depends on market variables, such as end-of-life battery availabil ity, testing capability, repurposing capability, quality assurance, transportation, a nd other supply chain logistics.29,30 Analysis of end-of-life battery availability125,232 and the effect of market va riables on the SLB supply chain 126 can provide us with important lessons to ensure reliable SLB availability. Strategies for successful 77 remanufacturing have also been identified in the broader literature, such as design for disassembly, policy support, and standardization of product and component designs. 126 Consumer perception of SLBs is another important f actor for large-scale adoption, 31 addressed in this dissertation as part of the system dynamics model developed in Chapter 4. Extending the life of end-of-life electric ve hicle batteries through the remanufacturing process provides an opportunity to utilize a pr oduct for energy storage applications that would otherwise be considered waste. C oupling the results in Chapters 3 a nd 4, i.e., integrating technical, economic, and environmental feasibility of SLBs with market variables, end-of-life battery availability, and consumer perception will also allow for further research that can ensure the successful deployment of SLBs. 5.2 Sustainability implications of uranium fuel for nuclear energy Nuclear energy has been touted as a low cost, low carbon ener gy source that can provide both base load and provide flexibility to th e electric grid by co untering the renewable intermittency. 32 However, with changes in the uranium fuel sources, as shown in Figure 4.8, and the technology of extracting uraniu m, as shown in Figure 4. 3, th e nuclear fuel cycle processes need to be updated to represent the real-life scenarios. To this end, a life-cycle assessment of the alkaline in-situ leaching process of uraniu m extraction was carried out in Chapter 4. The use of in-situ leaching is likely to incr ease with the combined effect of growing uranium demand and decreasing uranium ore grad es. Alkaline in-situ leaching can also gain momentum as the call for uranium self-reliance in the US intensifies. Other countries, such as China, have also prioritized in-situ leaching and have found deposits where al kaline or weak acidic ISL would be more feasible than acidic ISL. 233 78 Coupled with the acidic process, alkaline in -situ leaching can incr easingly contribute to uranium production in the US and the world. As the carbon footprin t associated with alkaline in- situ leaching is almost twice that of the acidic process, there is a risk that future nuclear energy can have a higher carbon footprint. Thus, this life-cycle assessment of alkaline in-situ leaching can be useful for estimating the environmental imp acts of the nuclear fuel cycle. With changes in the demand, production, and ownershi p of uranium, the tr ansportation distances can also increase, increasing the carbon footprint of uranium extraction. These cha nges can also impact uranium supply and need to be included in the sustainabili ty analysis of uranium fuel and the full nuclear fuel cycle. It is especially important in the c ontext of decarbonization of electricity generation to ensure that there is no under- or over-estimation of GWP of elec tricity generation options, which can lead to decisions that can have long-term implications. 5.3 Conclusions Based on the research presented here, the resu lts for SLB-based appl ications show lower cost and carbon emissions compared to new batter ies. Similarly, the results obtained by combining the effect of changes in ore grades and tec hnology can be included in the nuclear fuel cycle modeling to better estimate th e environmental impacts of fu ture nuclear en ergy generation. Therefore, it is important to id entify anticipated fu ture changes in reso urces or technology and include them in the present-day su stainability analysis of electrici ty generation. As shown in this dissertation, prospective life-cycle assessment is a key analysis t ool to analyze various low-carbon energy options to ensure the low cost and envi ronmental impacts of electricity generation today and in the future. 79 APPENDICES 80 APPENDIX A Supplementary Information for Chapter 2 81 A1. Life-Cycle Inventory Details Table A 1. Life-cycle inventory for various components considered in the analysis. Details Reference Electricity Greenhouse gas emissions of electri city production according to the North American Electric Reliabilit y Corporation (NERC) in 2016 234 Photovoltaic system Installation of 1kWp multi-Si photov oltaic system without inverter for US 235 Second life battery Data for GWP and CED for the enclosure (30%+/- 20%) and pack manufacturing (100%) of a new LIB from Kim et al. (2016) used to obtain second life batter y inventor y 25,59 New lithium- ion batter y Inventory of first life EV battery based on GWP from Kim et al. (2016) 59 Inverte r Production of 500kW inverter in the US 235 Battery transportation Average transportation distances for electronics in the US: 1,996 tkm b y lorr y 16-32 metric ton 236 A2. Residential Energy Storage Application The results from two companion pape rs were used for this analysis. 114,115 The cumulative annual modeled values of the residential household applia nce demand for each location are listed in Table A2. Table A 2. Yearly residential electri city demand for each location. City Residential household appliance demand (MWh/year) Detroit (DET) 7.12 Los Angeles (LA) 6.50 New York City (NYC) 6.81 Phoenix (PHX) 12.85 Portland (PTD) 6.39 The results of the residential application in term s of the component design (the capacities of PV, energy storage, inverter, and grid purchases), the economic analysis (L COE and total annualized cost), and the life-cycle assess ment (GWP) for each c onfiguration in each location are given in Tables A3 (Residential household appliance (R) application) and A4 (Residential household appliance with EV chargi ng (REV) application). 82 Table A 3. Results of component design, economic analysis and life-cycle assessment for residential household appliance energy storage application (R). # Location Configuration Component Design Economic Analysis Life-Cycle Assessment PV (kW) Battery storage (kWh) Inverter (kW) Grid Purchase (kWh) LCOE (¢/kWh) Annualized Cost ($/ year) LCOE (¢/kWh) with 30% rebate Annualized Cost ($/ year) with 30% rebate GWP (kg CO 2 eq. per household) 1 DET R0 0 0 5 7,122 14.34 1,021 14.34 1,021 5,107 R1 5 0 5 4,599 13.53 963 12.28 874 3,298 R2 5 15 5 2,045 10.57 753 8.85 630 1,825 R2n 5 5 5 3,371 13.65 972 11.81 841 2,829 2 LA R0 0 0 5 6,503 18.44 1,199 18.44 1,199 2,814 R1 5 0 5 3,971 16.93 1,101 15.57 1,012 2,058 R2 5 20 5 475 10.71 696 8.63 561 571 R2n 5 10 5 1,232 14.47 941 11.80 767 1,017 3 NYC R0 0 0 5 6,880 15.6 1,074 15.6 1,074 3,437 R1 5 0 5 4,402 15.02 1,033 13.73 944 2,539 R2 5 15 5 1,880 11.4 784 9.62 662 1,298 R2n 5 10 5 2,104 14.31 985 11.78 811 1,535 4 PHX R0 0 0 5 12,853 15.29 1,965 15.29 1,965 7,135 R1 5 0 5 8,030 15.02 1,930 14.33 1,842 4,797 R2 5 20 5 3,991 12.5 1,606 11.39 1,465 2,585 R2n 5 15 5 4,218 14.90 1,915 13.21 1,698 2,897 5 PTD R0 0 0 5 6,387 14.94 954 14.94 954 2,438 R1 5 0 5 4,216 15.39 982 14.0 849 1,949 R2 5 15 5 1,804 12.14 775 10.22 653 1,047 R2n 5 5 5 2,999 14.66 937 12.61 805 1,557 83 Table A 4. Results of component design, economic analysis, and life-cycle assessment for residential energy storage application with EV charging (REV). # Location Configuration Component Design Economic Analysis Life-Cycle Assessment PV (kW) Battery storage (kWh) Inverter (kW) Grid Purchase (kWh) LCOE (¢/kWh) Annualized Cost ($/ year) LCOE (¢/kWh) with 30% rebate Annualized Cost ($/ year) with 30% rebate GWP (kg CO 2 eq. per household) 1 DET REV0 0 0 5 16,758 12.14 2035 12.14 2,035 12,018 REV1 5 0 5 14,235 11.80 1,977 11.27 1,888 10,548 REV2 5 20 5 10,960 10.31 1,728 9.51 1,594 8,225 REV2n 5 5 5 13,007 11.55 1,936 10.77 1,805 9,739 2 LA REV0 0 0 5 16,139 16.7 2,694 16.7 2,694 6,983 REV1 5 0 5 13,607 16.09 2,596 15.54 2,508 6,227 REV2 5 20 5 9,302 13.02 2,102 12.13 1,957 4,397 REV2n 5 20 5 8,628 14.70 2,373 13.10 2,114 4,098 3 NYC REV0 0 0 5 16,516 7.51 1,240 7.51 1,240 8,252 REV1 5 0 5 14,038 7.26 1,199 6.72 1,199 7,353 REV2 5 15 5 11,516 5.75 950 5.01 828 6,112 REV2n 5 10 5 11,740 6.97 1,151 5.91 977 6,349 4 PHX REV0 0 0 5 22,489 10.99 2,472 10.99 2,472 12,484 REV1 5 0 5 17,666 10.84 2,437 10.45 2,349 10,147 REV2 5 20 5 13,617 9.43 2,117 8.8 1,976 7,929 REV2n 5 15 5 13,853 10.77 2,422 9.81 2,205 8,048 5 PTD REV0 0 0 5 16,023 11.6 1,859 11.6 1,859 6,117 REV1 5 0 5 13,852 11.78 1,887 11.23 1,799 5,628 REV2 5 20 5 10,581 10.23 1,639 9.34 1,505 4,404 REV2n 5 5 5 12,635 11.49 1,842 10.67 1,710 5,235 A3. Commercial-level Fast Charging Application The results of the fast charging ap plication in terms of the component design (t he capacities of PV, energy storage, inverter, and grid purchases), the economic analysis (L COE and total annualized cost), and the life-cycle assess ment (GWP) for each c onfiguration in each location are given in Tables A5 (new LIB), and A6 (SLB). 84 Table A 5. Results of the fast charging application in terms of component design, economic analysis and life-cycle assessment when using new LIB. # Location Configuration Component Design Economic Analysis Life-Cycle Assessment PV (kW) SLB (kWh) Inverter (kW) Grid Purchase (kWh) LCOE (¢/kWh) Net Present Cost ($) Total Annualized Cost ($/ year) GWP (tonnes CO 2 eq. per 10,000 charges) 1 Detroit FS0 0 0 111 39,014 0.62 189,502 23,706 193.11 FS1n 0 108 8 42,719 0.24 73,439 9,187 230.04 FS2n 2 108 8 40,774 0.24 73,249 9,163 221.34 FS3n 50 907 0 0 0.48 145,508 18,202 150.86 2 Los Angeles FS0 0 0 111 39,014 0.26 80,469 10,066 117.09 FS1n 0 108 8 42,719 0.13 39,516 4,943 146.80 FS2n 2 108 8 40,361 0.13 39,002 4,879 140.67 FS3n 50 270 0 0 0.17 52,288 6,541 61.68 3 New York City FS0 0 0 111 39,014 0.17 52,264 6,538 134.99 FS1n 0 108 8 42,719 0.21 63,756 7,975 166.40 FS2n 10 108 8 35,687 0.18 54,589 6,829 146.63 FS3n 50 594 0 0 0.33 99,688 12,470 107.04 4 Phoenix FS0 0 0 111 39,014 0.95 290,531 36,344 149.82 FS1n 0 108 8 42,719 0.21 62,792 7,855 182.64 FS2n 3 108 8 38,871 0.20 61,630 7,710 169.25 FS3n 50 227 0 0 0.15 45,969 5,750 55.66 5 Portland FS0 0 0 111 39,014 0.17 52,371 6,551 103.50 FS1n 0 108 8 42,719 0.23 70,158 8,776 131.92 FS2n 9 108 8 36,614 0.21 65,022 8,134 119.83 FS3n 50 1,372 0 0 0.70 213,448 26,701 215.96 85 Table A 6. Results of the fast charging application in terms of component design, economic analysis and life-cycle assessment when using SLBs. # Location Configuration Component Design Economic Analysis Life-Cycle Assessment PV (kW) SLB (kWh) Inverter (kW) Grid Purchase (kWh) LCOE ($/kWh) Net Present Cost ($) Total Annualized Cost ($/ year) GWP (tonnes CO 2 eq. per 10,000 charges) 1 Detroit FS0 0 0 111 39,014 0.62 189,502 23,706 193.11 FS1 0 111 8 42,718 0.21 63,007 7,882 212.97 FS2 2 111 8 40,775 0.21 62,951 7,875 204.37 FS3 50 903 0 0 0.29 87,306 10,921 48.65 2 Los Angeles FS0 0 0 111 39,014 0.26 80,469 10,066 117.09 FS1 0 111 8 42,719 0.10 31,072 3,887 129.73 FS2 2 111 8 40,362 0.10 30,741 3,846 123.70 FS3 50 258 0 0 0.11 34,079 4,263 30.96 3 New York City FS0 0 0 111 39,014 0.17 52,264 6,538 134.99 FS1 0 111 8 42,719 0.18 54,026 6,758 149.32 FS2 10 101 8 35,684 0.15 44,550 5,573 129.77 FS3 50 599 0 0 0.20 62,213 7,783 40.31 4 Phoenix FS0 0 0 111 39,014 0.95 290,531 36,344 149.82 FS1 0 111 8 42,719 0.17 52,558 6,575 165.56 FS2 3 101 8 38,870 0.16 49,575 6,202 152.11 FS3 50 240 0 0 0.10 31,037 3,883 29.95 5 Portland FS0 0 0 111 39,014 0.17 52,371 6,551 103.5 FS1 0 111 8 42,719 0.20 61,021 7,633 114.84 FS2 9 101 8 36,610 0.18 55,724 6,871 102.89 FS3 50 1,372 0 0 0.41 126,086 15,773 61.54 86 A4. Utility-level Peak Shaving Application The 2017 hourly net generation data for the Midwest and Northwest regions were obtained from the EIA™s U.S. Electric System Operating dataset 118 and are shown in Figure S2. The generation data were reduced in proportion to the 2017 annual demand for Michigan and Oregon 118, and used as a proxy for the 2017 demand for both states (F igures A1 and A2). The maximum demand was 20,810 MWh for Michigan and 10,455 MWh for Oregon. Figure A 1. Electricity generation data for the Midwest region in grey, 2017 Michigan electricity demand in red, and the net peak demand target in blue. Figure A 2. Electricity generation data for the Northwest region in grey, 2017 Oregon electricity demand in red, and the net peak demand target in blue. The results of the peak shaving application in terms of the component design (the capacities of PV, energy storage, inverter, and grid purchases), the economic analysis (LCOE and total 0206080100 120 140 45Net Peak Demand Target Electricity power (GW) Northwest Oregon Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 87 annualized cost), and the life-cycle assessment (GWP) for each conf iguration in each location are given in Tables A7. Table A 7. Results of component design, economic analysis, and life-cycle assessment for utility- level Peak Shaving (PS). # Location Configuration Component Design Economic Analysis Life-Cycle Assessment Nautral gas (MW) Battery storage (MWh) Inverter (MW) Grid Purchase (MWh/year) LCOE ($/kWh) Annualized Cost ($/ year) GWP (kg CO 2 eq. per kWh) 1 Michigan PS0 4,162 0 0 0 0.6601 403,716,715 0.74 PS1 0 30,000 5,500 675,494 0.4017 245,687,253 0.87 PS1n 0 25,000 5,500 649,821 1.1376 696,301,512 1.37 2 Oregon PS0 2,091 0 0 0 0.4612 213,265,921 0.74 PS1 0 30,000 7,000 473,367 0.7439 317,671,525 0.54 PS1n 0 25,000 7,000 472,689 1.5408 657,219,231 1.30 88 APPENDIX B Supplementary Information for Chapter 3 89 B1 Assumptions in the System Dynamics Model Some of the assumptions used in the system dynamics model incl ude the chosen end-of-life in years for the batteries (see Figure B1), the recy cling efficiencies and th e place of materials (see Table B1), and the battery components and the average energy density based on the battery chemistry (see Table B2). Figure B 1. Percentage of batteries reaching end-of -life at different years for EV batteries and second-life batteries. Table B 1. Recycling efficiencies (REs) for different battery components as the baseline assumptions,171 along with their prices175Œ181 in $/kg. Material Lithium Cobalt Manganese Nickel Iron Steel Aluminum Copper RE (%) 95 95 95 95 95 95 95 95 Price ($/k g) 6.80 36.04 0.01 14.00 0.31 0.25 2.20 6.17 Table B 2. Battery components for different battery chemistries125,172 and the average energy density237 om kWh/kg. Battery chemistry Material (kg/kWh) Energy density (kWh/kg) Lithium Cobalt Manganese Nickel Iron Steel Aluminum Copper LiCoO2 0.119 1.01 0 0.071 0 0.963 0.304 0.426 0.175 LiMn2O4 0.104 0 1.37 0 0 1.105 0.075 0.075 0.125 LiF2PO4 0.084 0 0 0 0.68 2.348 0.457 0.571 0.105 NMC111 0.139 0.394 0.392 0.367 0 0.866 0.263 0.390 0.185 NMC622 0.126 0.214 0.641 0.2 0 0.866 0.263 0.390 NMC811 0.111 0.094 0.75 0.088 0 0.866 0.263 0.390 2345678910111213 0204060 SLB EV battery Years Percentage of batteries reaching end-of-life (%) 90 B2. Intermediate Results Intermediate results from the system dynamics model include the change in lithium-ion battery demand due to the availability of SLBs (see Fi gure B2) and the change in cobalt and nickel demand with recycling and remanufacturing (see Figure B3). Figure B 2. Change in lithium-ion battery demand due to the availability of SLBs. Figure B 3. Change in (a) cobalt and (b) nickel demand with recycli ng and remanufacturing. 2020202520302035204020452050 2017 050100 150200250 300Total demand (EV demand included) Recycling only With remanufacturing Storage demand (GWh/year) 91 APPENDIX C System Dynamics Model Equations for Chapter 3 92 C1. Initialization Equations : s BEV_stock = 0.002986 : s hyb_stock = 3.57392 : s PHEV10_stock = 0.147764 : s PHEV40_stock = 0.127634 : s SLB_stock = 0 : c Start_year_for_reman = 2020 : c demand_for_energy_storage = TIME : c year_of_switch_from_EA_to_GP = 2028 : c PP_GP = 0.44 : c PP_EA = 0.37 : c PP_total = IF TIME> (year_of_switch_from_EA_to_GP-1) THEN PP_GP ELSE PP_EA : c SLBdemand = IF TIME> (Start_year_for_reman-1) THEN demand_for_energy_storage*PP_total ELSE 0 : S reman_capacity = IF SLBdemand>0 THEN 500000 ELSE 0 : c EV100in_gf = TIME : f EV100_in = EV100in_gf : c EV200in_gf = TIME : f EV200_in = EV200in_gf : c EV300in_gf = TIME : f EV300_in = EV300in_gf : f BEV_out_a = 0.1*BEV_stock/6 : f BEV_out_b = 0.4*BEV_stock/8 : f BEV_out_c = 0.4*BEV_stock/10 : f BEV_out_d = 0.1*BEV_stock/12 : c collectionefficiency = 1 : c hyb_gf = TIME : f hyb_in = hyb_gf : f hyb_out_a = 0.1*hyb_stock/6 : f hyb_out_b = 0.4*hyb_stock/8 : c hyb_battery_energy = 5.3 : c hyb_battery_energy_out_SLB = (hyb_out_a+hyb_out_b)*hyb_battery_energy : c PHEV10_gf = TIME : f PHEV10_in = PHEV10_gf : f PHEV10_out_a = 0.1*PHEV10_stock/6 : f PHEV10_out_b = 0.4*PHEV10_stock/8 : c PHEV10_battery_energy = 4.4 : c PHEV10_battery_energy_out_SLB = (PHEV10_out_a+PHEV10_out_b)*PHEV10_battery_energy : c PHEV40_gf = TIME : f PHEV40_in = PHEV40_gf : f PHEV40_out_a = 0.1*PHEV40_stock/6 : f PHEV40_out_b = 0.4*PHEV40_stock/8 : c PHEV40_battery_energy = 18 : c PHEV40_battery_energy_out_SLB = (PHEV40_out_a+PHEV40_out_b)*PHEV40_battery_energy : c BEV_battery_energy = 39 : c BEV_battery_energy_out_SLB = (BEV_out_a+BEV_out_b)*BEV_battery_energy : c remaining_capacity = 0.95 : c sum_total_cap_eol_bat_SLB = (hyb_battery_energy_out_SLB+PHEV10_battery_energy_out_SLB+PHEV40_battery_energy_out_SLB+BEV_batt ery_energy_out_SLB)*remaining_capacity : c collectedEOLB_SLB = collectionefficiency*sum_total_cap_eol_bat_SLB : c technically_feasible_batteries_fraction = 0.95 : c technically_feasible_collectedEOLB = collect edEOLB_SLB*technically_feas ible_batteries_fraction : c reman_SLB_possible = IF SLBdemand>0 THEN (IF technically_feasible_collect edEOLB>SLBdemand THEN SLBdemand ELSE technically_feasible_collectedEOLB) ELSE 0 93 : c addition_over_n_years = IF reman_SLB_possi ble>reman_capacity THEN (reman_SLB_possible- reman_capacity) ELSE 0 : f capacity_addition = IF TIME > (Start_year_for_reman-2) THEN addition_over_n_years ELSE 0 : f hyb_out_c = 0.4*hyb_stock/10 : f hyb_out_d = 0.1*hyb_stock/12 : f PHEV10_out_c = 0.4*PHEV10_stock/10 : f PHEV10_out_d = 0.1*PHEV10_stock/12 : f PHEV40_out_c = 0.4*PHEV40_stock/10 : f PHEV40_out_d = 0.1*PHEV40_stock/12 : c reman_SLB = IF SLBdemand>0 THEN reman_capac ity*technically_feasible_batteries_fraction ELSE 0 : f SLB_in = reman_SLB : f SLB_out_a = 0.1*SLB_stock/3 : f SLB_out_b = 0.4*SLB_stock/4 : f SLB_out_c = 0.4*SLB_stock/5 : f SLB_out_d = 0.1*SLB_stock/6 : c cost_of_labor_per_hour = 20.67 : c number_of_hours_per_year_per_laborer = 8*252 : c output_per_hour = 3 : c output_per_hour_kWh = 10*output_per_hour : c number_of_shifts = 3 : c number_of_laborers_needed = reman_SLB/(output_per_hour_kWh*number_of_hours_per_year_per_laborer*number_of_shifts) : c labor_in_facility = 10 : c min_labor_needed = MIN(number_of_laborers _needed, (labor_in_facility*reman_capacity/115000)) : c cost_of_labor_in_facility = cost_of_labor_per_hour *number_of_hours_per_year_per _laborer*min_labor_needed : c Cost_of_labor_per_SLB_kWh_without_increase = IF reman_SLB>0 THEN cost_of_labor_in_facility/reman_SLB ELSE 0 : c percentage_increase_in_labor_cost = 0.028 : c increase_in_labor_cost = Cost_of_labor_per_SLB_kWh_without_increase*(TIME- 2019)*percentage_increase_in_labor_cost : c Cost_of_labor_per_SLB_kWh = Cost_of_labor_per_SLB_kWh_without_increase+increase_in_labor_cost : c cost_of_reman_capacity = ((1124990/10)+1192854)/115000 : c cost_of_reman_capacity_total = re man_capacity*cost_of_reman_capacity : c cost_of_reman_capacity_per_SLB_k Wh = IF reman_SLB>0 THEN cost_of_reman_capacity_total/reman_SLB ELSE 0 : c electricity_per_kWh = 7.5 : c cost_of_electricity = 0.0688 : c percentage_increase_in_cost_of_electricity = 0.02 : c increase_in_cost_of_electricity_ove r_year = cost_of_electricity*(TIME- 2019)*percentage_increase_in_cost_of_electricity/(32) : c cost_of_electricity_with_increase = cost_of_electricity+increase_in_cost_of_electricity_over_year : c cost_of_electricity_per_SLB_kWh = electricity_per_kWh*cost_of_electricity_with_increase/(1+technically_feasible_batteries_fraction) : c cost_of_materials_constant_per_kWh = (15+20+30)/23 : c cost_of_materials_others_per_kWh = (30+10)/23 : c "kWh/kg" = TIME : c cost_of_materials_constant_based_on_mass_per_kWh = (7/"kWh/kg")/23 : c cost_of_materials_per_SLB_kWh = cost_of_materials_constant_per_kWh+cost_of_materials_others_per_kWh+cost_of_materials_constant_based_on_ mass_per_kWh : c transporation_cost_per_mile = 2.16 : c percentage_increase_in_transportation_cost = 0.03 : c transportation_costs_per_mile_for_the_year = transporation_cost_per_mile*((1+percentage_increase_in_transportation_cost)^(TIME-2020)) : c transport_truck_tonnage = 16 : c transportation_costs_per_tonmile = transportation_ costs_per_mile_for_the_year/transport_truck_tonnage 94 : c transport_distance_min_input = 100 : c transport_distance_from _reman_capacity = 2524.2424-(reman_capacity/20625) : c transport_distance_possible = MAX( transport_distance_min_input, tran sport_distance_from_reman_capacity) : c tonne_per_SLB_kWh = 0.00110231/"kWh/kg" : c transport_cost_per_kWh = transportation_costs_per_tonmile*transport_distance_possible*tonne_per_SLB_kWh : c transportation_costs_per_SLB_kWh = transport_cost_per_kWh/(1+technically_feasible_batteries_fraction) : c total_cost_per_SLB_kWh = Cost_of_labor_per_SLB_kWh+cost_o f_reman_capacity_per_SLB_kWh+cost_o f_electricity_per_SLB_kWh+cost_ of_materials_per_SLB_kWh+transportation_costs_per_SLB_kWh : c overhead_and_launch_costs = (0.05*cost_of_materials_per_SLB_kWh)+(0.65*Cost_of_labor_per_SLB_kWh) : c EOL_battery_buying_cost = 0 : c allowable_max_cost_with_overheads = total_cost_per_SLB_kWh+overhead_and_launch_costs+EOL_battery_buying_cost : c allowable_profit_% = 0.1 : c allowable_max_cost = allowable_max_ cost_with_overheads*(1+allowable_profit_%) : c Aluminum_RE = 0.95 : c LiCoO2_fraction = TIME : c LiMn2O4_fraction = TIME : c LiFePO4_fraction = TIME : c NMC_fraction = TIME : c aluminum_fraction = (LiCoO2_fraction*0.304)+(LiMn2O4_fraction*0.075)+(LiFePO4_fraction*0.457)+(NMC_fraction*0.263) : c total_BEV_out = BEV_out_a+BEV_out_b+BEV_out_c+BEV_out_d : c BEV_battery_energy_out_per_year = BEV_battery_energy*total_BEV_out : c total_hyb_out = hyb_out_a+hyb_out_b+hyb_out_c+hyb_out_d : c hyb_battery_energy_out_per_year = hyb_battery_energy*total_hyb_out : c total_PHEV10_out = PHEV10_out_a+PHEV10_out_b+PHEV10_out_c+PHEV10_out_d : c PHEV10_battery_energy_out_per_year = PHEV10_battery_energy*total_PHEV10_out : c total_PHEV40_out = PHEV40_out_a+PHEV40_out_b+PHEV40_out_c+PHEV40_out_d : c PHEV40_battery_energy_out_per_year = PHEV40_battery_energy*total_PHEV40_out : c total_EV_battery_energy_per_year = BEV_battery_energy_out_per_year+hyb_battery_energy_out_per_year+PHEV10_battery_energy_out_per_year+P HEV40_battery_energy_out_per_year : c sum_total_cap_eol_bat = total_EV_battery_energy_per_year*remaining_capacity : c collectedEOLB = collectioneffi ciency*sum_total_cap_eol_bat : c total_SLB_out = SLB_out_a+SLB_out_b+SLB_out_c+SLB_out_d : c recycledEOLB = IF TIME> Start_y ear_for_reman THEN (collectedEOLB- reman_SLB+total_SLB_out)/remaining_capacity ELSE 0 : c Aluminum_after_recycl = Aluminum_RE*aluminum_fraction*recycledEOLB : c aluminum_cost = 1/0.453592 : c PHEV40_battery_energy_in_per_year = PHEV40_battery_energy*PHEV40_in : c hyb_battery_energy_in_per_year = hyb_battery_energy*hyb_in : c PHEV10_battery_energy_in_per_year = PHEV10_battery_energy*PHEV10_in : c BEV_battery_energy_in_per_year = BEV_battery_energy*(EV100_in+EV200_in+EV300_in) : c sum_total_cap_in_bat = PHEV40_battery_energy_in_per_year+hyb_battery_energy_in_per_year+PHEV10_battery_energy_in_per_year+B EV_battery_energy_in_per_year : c total_LIB_demand = demand_for_energy_storage+sum_total_cap_in_bat : c Aluminum_in = aluminum_fraction*total_LIB_demand : c Aluminum_demand_modified = Aluminum_in-Aluminum_after_recycl : c aluminum_revenue = Alumin um_after_recycl*aluminum_cost : c nominal_discount_rate = 0.069 : c expected_inflation_rate = 0.025 : c real_discount_rate = (nominal _discount_rate-expected_inflation_rate)/(1+expected_inflation_rate) : c buying_cost_npv = IF TIME>Start_year_for_reman THEN (EOL_battery_buying_cost)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 95 : c Cobalt_RE = 0.95 : c NMC_avg_Co = TIME : c Cobalt_fraction = (LiCoO2_fraction*1.01)+(LiMn2O4_fraction*0)+(LiFePO4_fraction*0)+(NMC_fraction*NMC_avg_Co) : c Cobalt_after_recycl = Cobalt_ RE*Cobalt_fractio n*recycledEOLB : c cobalt_cost = 36.04 : c Cobalt_in = Cobalt_fraction*total_LIB_demand : c Cobalt_demand_modified = Cobalt_in-Cobalt_after_recycl : c cobalt_revenue = Cobalt_after_recycl*cobalt_cost : c Copper_RE = 0.95 : c Copper_fraction = (LiCoO2_fraction*0.426)+(LiMn2O4_fraction*0.075)+(LiFePO4_fraction*0.571)+(NMC_fraction*0.390) : c Copper_after_recycl = Copper_ RE*Copper_fraction*recycledEOLB : c copper_cost = 6.17 : c Copper_in = Copper_fraction*total_LIB_demand : c Copper_demand_modified = Copper_in-Copper_after_recycl : c copper_revenue = Copper_after_recycl*copper_cost : c cost_of_new_battery_BNEF = TIME : c Electricity_cost_npv = IF TIME>Start_year_for_reman THEN (cost_of_electricity_per_SLB_kWh)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c EOL_after_SLB = 0.5 : c lithium_cost = 6.8 : c Lithium_RE = 0.95 : c NMC_avg_Li = TIME : c Lithium_fraction = (LiCoO2_fraction*0.119)+(LiMn2O4_fraction*0.104)+(LiFePO4_fraction*0.084)+(NMC_fraction*NMC_avg_Li) : c Lithium_after_recycl = Lithiu m_RE*Lithium_fractio n*recycledEOLB : c lithium_revenue = lithium_cost*Lithium_after_recycl : c manganese_cost = 6.6/1000 : c Manganese_RE = 0.95 : c NMC_avg_Mn = TIME : c Manganese_fraction = (LiCoO2_fraction*0)+(LiMn2O4_fraction*1.37)+(LiFePO4_fraction*0)+(NMC_fraction*NMC_avg_Mn) : c Manganese_after_recycl = Manganes e_RE*Manganese_fraction*recycledEOLB : c manganese_revenue = manganese_cost*Manganese_after_recycl : c iron_cost = 312.29/1000 : c Iron_RE = 0.95 : c Iron_fraction = (LiCoO2_fraction*0)+(LiMn2O4_fract ion*0)+(LiFePO4_fraction*0.68)+(NMC_fraction*0) : c Iron_after_recycl = Iron_RE*Iron_fraction*recycledEOLB : c iron_revenue = iron_cost*Iron_after_recycl : c Steel_RE = 0.95 : c Steel_fraction = (LiCoO2_fraction*0.963)+(LiMn2O4_fraction*1.105)+(LiFePO4_fraction*2.348)+(NMC_fraction*0.866) : c Steel_after_recycl = Steel_R E*Steel_fraction*recycledEOLB : c steel_cost = 249.22/1000 : c steel_revenue = Steel_after_recycl*steel_cost : c Nickel_RE = 0.95 : c NMC_avg_Ni = TIME : c Nickel_fraction = (LiCoO2_fraction*0.071)+(LiMn2O4_fraction*0)+(LiFePO4_fraction*0)+(NMC_fraction*NMC_avg_Ni) : c Nickel_after_recycl = Nickel_R E*Nickel_fraction*recycledEOLB : c nickel_cost = 14 : c nickel_revenue = Nickel_after_recycl*nickel_cost : c revenue_change_yearly = 0.05 96 : c total_revenue_recycling = (lithium_revenue+manganese_revenue+i ron_revenue+steel_revenue+nickel_ revenue+copper_revenue+aluminum_r evenue+cobalt_revenue)*((1+revenue_change_yearly)^(TIME-2020)) : c recycling_cost_in_2020_per_kg = 4 : c recycling_cost_change_yearly = -0.05 : c recycling_cost_per_kg_per_year = recycling_cost_in_2020_per_kg*((1+recycling_cost_change_yearly)^(TIME- 2020)) : c recycling_cost = recycling_cost_per_kg_per_year/"kWh/kg" : c total_cost_recycling = recycling_cost*recycledEOLB : c EOL_recycling_benefit = total_re venue_recycling-tota l_cost_recycling : c EOL_Recycling_value = EOL_recycling_be nefit/(1000000*((1+real_di scount_rate)^(TIME- Start_year_for_reman))) : c EOL_recycling_value_per_kWh = IF recycledEOLB >0 THEN EOL_Recycling_val ue*1000000/recycledEOLB ELSE 0 : c health_factor = (remaining_capac ity-EOL_after_SLB)/(1-EOL_after_SLB) : c value_of_throughput_new_battery = 1 : c value_parameter = health_factor/value_of_throughput_new_battery : c UPDF_GP = 0.5853 : c UPDF_EA = 0.6152 : c UPDF = IF TIME> (year_of_switch_from_EA_to_GP-1) THEN UPDF_GP ELSE UPDF_EA : c used_product_discount_factor = IF TIMEStart_year_for_r eman THEN (selling_price_of_SLB- allowable_max_cost)*reman_SLB/(1000000*((1+real_discount_rate)^(TIME-Start_year_for_reman))) ELSE 0 : c EOL_remanufacturing_value_per_kWh = IF reman_SLB>0 THEN EOL_value_from_remanufacturing*1000000/reman_SLB ELSE 0 : c total_EOL_value = IF TIME>Start_year_for_reman THEN EOL_Recycling_value+EOL_value_ from_remanufacturing ELSE 0 : c EOL_total_value_per_kWh = IF TIME>Start_year_for_reman THEN (total_EOL_value*1000000/(reman_ SLB+recycledEOLB)) ELSE 0 : c Equipment_and_facility_cost_npv = IF TIME>Start_year_for_reman THEN (cost_of_reman_capacity_per_SLB_kWh)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c Iron_in = Iron_fraction*total_LIB_demand : c Iron_demand_modified = Iron_in-Iron_after_recycl : c labor_cost_npv = IF TIME >Start_year_for_reman THEN (Cost_of_labor_per_SLB_kWh)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c Lithium_in = Lithium_fraction*total_LIB_demand : c Lithium_demand_modified = L ithium_in-Lithium_after_recycl : c Manganese_in = Manganese_fraction*total_LIB_demand : c Manganese_demand_modified = Manganese_in-Manganese_after_recycl : c Material_cost_npv = IF TIME>Start_year_for_reman THEN (cost_of_materials_per_SLB_kWh)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c modified_LIB_demand = IF TIME>Start_year_for_reman THEN total_LIB_demand-reman_SLB ELSE total_LIB_demand : c Nickel_in = Nickel_fraction*total_LIB_demand : c Nickel_demand_modified = Nickel_in-Nickel_after_recycl : c overhead_and_profit_per_kwh = allowable_max_cost-total_cost_per_SLB_kWh-EOL_battery_buying_cost : c Overhead_and_profit_npv = IF TIME>Start_year_for_reman THEN (overhead_and_profit_per_kwh)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c recycling_cost_in_2020 = 46.2 : c SLB_tonnes = 1.1/("kWh/kg"*1000) : c Steel_in = Steel_fraction*total_LIB_demand : c Steel_demand_modified = Steel_in-Steel_after_recycl : c sum_total_cap_eol_bat_recycle = sum_total_cap_eol_bat-sum_total_cap_eol_bat_SLB : c total_revenue_recycling_per_kwh = IF recycledEOLB>0 THEN total_re venue_recycling/recycledEOLB ELSE 0 97 : c Transport_cost_npv = IF TIME>Start_year_for_reman THEN (transportation_costs_per_SLB_kWh)/((1+real_disc ount_rate)^(TIME-Start _year_for_reman)) ELSE 0 C2. Runtime Equations : s BEV_stock(t) = BEV_stock(t - dt) + (EV100_in + EV200_in + EV300_in - BEV_out_a - BEV_out_b - BEV_out_c - BEV_out_d) * dt : s hyb_stock(t) = hyb_stock(t - dt) + (hyb_in - hyb_out_a - hyb_out_b - hyb_out_c - hyb_out_d) * dt : s PHEV10_stock(t) = PHEV10_stock(t - dt) + (PHEV10_in - PHEV10_out_a - PHEV10_out_b - PHEV10_out_c - PHEV10_out_d) * dt : s PHEV40_stock(t) = PHEV40_stock(t - dt) + (PHEV40_in - PHEV40_out_a - PHEV40_out_b - PHEV40_out_c - PHEV40_out_d) * dt : s SLB_stock(t) = SLB_stock(t - dt) + (SLB_in - SLB_out_a - SLB_out_b - SLB_out_c - SLB_out_d) * dt : S reman_capacity(t) = reman_capacity(t - dt) + (capacity_addition) * dt : c demand_for_energy_storage = GRAPH(TIME) (2020.00, 2666670), (2022.00, 4e+06), (2024.00, 5333330 ), (2026.00, 5333330), (2028.00, 4e+06), (2030.00, 2666670), (2032.00, 2666670), (2034.00, 4e+06), (2036 .00, 6666670), (2038.00, 9333330), (2040.00, 18666670), (2042.00, 25333330), (2044.00, 33333330), (2046.00, 50666670), (2048.00, 1.04e+08), (2050.00, 165333330) : c PP_total = IF TIME> (year_of_switch_from_EA_to_GP-1) THEN PP_GP ELSE PP_EA : c SLBdemand = IF TIME> (Start_year_for_reman-1) THEN demand_for_energy_storage*PP_total ELSE 0 : c EV100in_gf = GRAPH(TIME) (2017.00, 36860.0), (2018.00, 19825.0), (2019.00, 75341.0 ), (2020.00, 86740.0), (2021.00, 105663.0), (2022.00, 129707.0), (2023.00, 136803.0), (2024.00, 153022.0), (2025.00, 163241.0), (2026.00, 159558.0), (2027.00, 160771.0), (2028.00, 154415.0), (2029.00, 156671.0), (2030.00, 157530.0), (2031.00, 161096.0), (2032.00, 155506.0), (2033.00, 158267.0), (2034.00, 163282.0), (2035.00, 168110.0), (2036.00, 173956.0), (2037.00, 176991.0), (2038.00, 183759.0), (2039.00, 191141.0), (2040.00, 199681.0), (2041.00, 206635.0), (2042.00, 214662.0), (2043.00, 223435.0), (2044.00, 233428.0), (2045.00, 243772.0), (2046.00, 255642.0), (2047.00, 267801.0), (2048.00, 280834.0), (2049.00, 295111.0), (2050.00, 311008.0) : f EV100_in = EV100in_gf : c EV200in_gf = GRAPH(TIME) (2017.00, 39031), (2018.00, 48594), (2019.00, 173583), (2020.00, 221596), (2021.00, 241157), (2022.00, 256255), (2023.00, 279229), (2024.00, 313864), (2025.00, 3438 88), (2026.00, 343982), (2027.00, 348808), (2028.00, 355275), (2029.00, 373298), (2030.00, 381611), (2031.0 0, 408645), (2032.00, 433491), (2033.00, 464839), (2034.00, 500733), (2035.00, 535643), (2036.00, 5713 34), (2037.00, 604425), (2038.00, 642147), (2039.00, 679570), (2040.00, 720112), (2041.00, 743004), (2042.0 0, 763037), (2043.00, 780672), (2044.00, 799028), (2045.00, 816715), (2046.00, 835056), (2047.00, 8499 53), (2048.00, 865570), (2049.00, 880396), (2050.00, 893826) : f EV200_in = EV200in_gf : c EV300in_gf = GRAPH(TIME) (2017.00, 32202), (2018.00, 137843), (2019.00, 221603), (2020.00, 290816), (2021.00, 372729), (2022.00, 399582), (2023.00, 425971), (2024.00, 442494), (2025.00, 4549 65), (2026.00, 459696), (2027.00, 473874), (2028.00, 493811), (2029.00, 519854), (2030.00, 562392), (2031.0 0, 608889), (2032.00, 650409), (2033.00, 694481), (2034.00, 738001), (2035.00, 777088), (2036.00, 8111 03), (2037.00, 840807), (2038.00, 866144), (2039.00, 885781), (2040.00, 902296), (2041.00, 921157), (2042.0 0, 939614), (2043.00, 960781), (2044.00, 982347), (2045.00, 1001593), (2046.00, 1021100), (2047.00, 1037172) , (2048.00, 1051891), (2049.00, 1069414), (2050.00, 1083464) : f EV300_in = EV300in_gf : f BEV_out_a = 0.1*BEV_stock/6 : f BEV_out_b = 0.4*BEV_stock/8 : f BEV_out_c = 0.4*BEV_stock/10 : f BEV_out_d = 0.1*BEV_stock/12 : c hyb_gf = GRAPH(TIME) (2017.00, 418124.0), (2018.00, 497531.0), (2019.00, 497566.0), (2020.00, 497038.0), (2021.00, 493931.0), (2022.00, 493773.0), (2023.00, 516047.0), (2024.00, 542335.0), (2025.00, 553572.0), (2026.00, 578453.0), (2027.00, 608728.0), (2028.00, 637378.0), (2029.00, 666725.0), (2030.00, 694289.0), (2031.00, 725454.0), (2032.00, 744822.0), (2033.00, 762482.0), (2034.00, 779577.0), (2035.00, 791153.0), (2036.00, 800683.0), 98 (2037.00, 809566.0), (2038.00, 819083.0), (2039.00, 825239.0), (2040.00, 832722.0), (2041.00, 840717.0), (2042.00, 843018.0), (2043.00, 844323.0), (2044.00, 848329.0), (2045.00, 851079.0), (2046.00, 852109.0), (2047.00, 848900.0), (2048.00, 844630.0), (2049.00, 841638.0), (2050.00, 836532.0) : f hyb_in = hyb_gf : f hyb_out_a = 0.1*hyb_stock/6 : f hyb_out_b = 0.4*hyb_stock/8 : c hyb_battery_energy_out_SLB = (hyb_out_a+hyb_out_b)*hyb_battery_energy : c PHEV10_gf = GRAPH(TIME) (2017.00, 22827.0), (2018.00, 29022.0), (2019.00, 32499.0 ), (2020.00, 37421.0), (2021.00, 53468.0), (2022.00, 62236.0), (2023.00, 67023.0), (2024.00, 71854.0), (2025.00, 78188.0), (2026.00, 79583.0), (2027.00, 81446.0), (2028.00, 82679.0), (2029.00, 83616.0), (2030.00, 84385.0 ), (2031.00, 85340.0), (2032.00, 86039.0), (2033.00, 86582.0), (2034.00, 87349.0), (2035.00, 87777.0), (2036.00, 88601.0), (2037.00, 89745.0), (2038.00, 88789.0), (2039.00, 89421.0), (2040.00, 90217.0), (2041.00, 90522.0 ), (2042.00, 90703.0), (2043.00, 91057.0), (2044.00, 91530.0), (2045.00, 91753.0), (2046.00, 92020.0), (2047.00, 92236.0), (2048.00, 92220.0), (2049.00, 92279.0), (2050.00, 92619.0) : f PHEV10_in = PHEV10_gf : f PHEV10_out_a = 0.1*PHEV10_stock/6 : f PHEV10_out_b = 0.4*PHEV10_stock/8 : c PHEV10_battery_energy_out_SLB = (PHEV10_out_a+PHEV10_out_b)*PHEV10_battery_energy : c PHEV40_gf = GRAPH(TIME) (2017.00, 47444.0), (2018.00, 107649.0), (2019.00, 113503.0 ), (2020.00, 118546.0), (2021.00, 110668.0), (2022.00, 101893.0), (2023.00, 98666.0), (2024.00, 99143.0), (202 5.00, 96735.0), (2026.00, 98697.0), (2027.00, 103601.0), (2028.00, 109448.0), (2029.00, 121749.0), (2030.00, 128817.0), (2031.00, 135848.0), (2032.00, 142723.0), (2033.00, 149379.0), (2034.00, 156815.0), (2035.00, 163128.0), (2036.00, 169391.0), (2037.00, 174708.0), (2038.00, 179846.0), (2039.00, 185177.0), (2040.00, 190777.0), (2041.00, 193623.0), (2042.00, 196556.0), (2043.00, 198553.0), (2044.00, 200789.0), (2045.00, 202810.0), (2046.00, 204662.0), (2047.00, 205672.0), (2048.00, 206798.0), (2049.00, 207491.0), (2050.00, 208189.0) : f PHEV40_in = PHEV40_gf : f PHEV40_out_a = 0.1*PHEV40_stock/6 : f PHEV40_out_b = 0.4*PHEV40_stock/8 : c PHEV40_battery_energy_out_SLB = (PHEV40_out_a+PHEV40_out_b)*PHEV40_battery_energy : c BEV_battery_energy_out_SLB = (BEV_out_a+BEV_out_b)*BEV_battery_energy : c sum_total_cap_eol_bat_SLB = (hyb_battery_energy_out_SLB+PHEV10_battery_energy_out_SLB+PHEV40_battery_energy_out_SLB+BEV_batt ery_energy_out_SLB)*remaining_capacity : c collectedEOLB_SLB = collectionefficiency*sum_total_cap_eol_bat_SLB : c technically_feasible_collectedEOLB = collect edEOLB_SLB*technically_feas ible_batteries_fraction : c reman_SLB_possible = IF SLBdemand>0 THEN (IF technically_feasible_collect edEOLB>SLBdemand THEN SLBdemand ELSE technically_feasible_collectedEOLB) ELSE 0 : c addition_over_n_years = IF reman_SLB_possi ble>reman_capacity THEN (reman_SLB_possible- reman_capacity) ELSE 0 : f capacity_addition = IF TIME > (Start_year_for_reman-2) THEN addition_over_n_years ELSE 0 : f hyb_out_c = 0.4*hyb_stock/10 : f hyb_out_d = 0.1*hyb_stock/12 : f PHEV10_out_c = 0.4*PHEV10_stock/10 : f PHEV10_out_d = 0.1*PHEV10_stock/12 : f PHEV40_out_c = 0.4*PHEV40_stock/10 : f PHEV40_out_d = 0.1*PHEV40_stock/12 : c reman_SLB = IF SLBdemand>0 THEN reman_capac ity*technically_feasible_batteries_fraction ELSE 0 : f SLB_in = reman_SLB : f SLB_out_a = 0.1*SLB_stock/3 : f SLB_out_b = 0.4*SLB_stock/4 : f SLB_out_c = 0.4*SLB_stock/5 : f SLB_out_d = 0.1*SLB_stock/6 : c number_of_hours_per_year_per_laborer = 8*252 : c output_per_hour_kWh = 10*output_per_hour 99 : c number_of_laborers_needed = reman_SLB/(output_per_hour_kWh*number_of_hours_per_year_per_laborer*number_of_shifts) : c min_labor_needed = MIN(number_of_laborers _needed, (labor_in_facility*reman_capacity/115000)) : c cost_of_labor_in_facility = cost_of_labor_per_hour *number_of_hours_per_year_per _laborer*min_labor_needed : c Cost_of_labor_per_SLB_kWh_without_increase = IF reman_SLB>0 THEN cost_of_labor_in_facility/reman_SLB ELSE 0 : c increase_in_labor_cost = Cost_of_labor_per_SLB_kWh_without_increase*(TIME- 2019)*percentage_increase_in_labor_cost : c Cost_of_labor_per_SLB_kWh = Cost_of_labor_per_SLB_kWh_without_increase+increase_in_labor_cost : c cost_of_reman_capacity = ((1124990/10)+1192854)/115000 : c cost_of_reman_capacity_total = re man_capacity*cost_of_reman_capacity : c cost_of_reman_capacity_per_SLB_k Wh = IF reman_SLB>0 THEN cost_of_reman_capacity_total/reman_SLB ELSE 0 : c increase_in_cost_of_electricity_ove r_year = cost_of_electricity*(TIME- 2019)*percentage_increase_in_cost_of_electricity/(32) : c cost_of_electricity_with_increase = cost_of_electricity+increase_in_cost_of_electricity_over_year : c cost_of_electricity_per_SLB_kWh = electricity_per_kWh*cost_of_electricity_with_increase/(1+technically_feasible_batteries_fraction) : c cost_of_materials_constant_per_kWh = (15+20+30)/23 : c cost_of_materials_others_per_kWh = (30+10)/23 : c "kWh/kg" = GRAPH(TIME) (2017.00, 0.142), (2018.00, 0.142), (2019.00, 0.142), (2 020.00, 0.142), (2021.00, 0.1432), (2022.00, 0.1444), (2023.00, 0.1456), (2024.00, 0.1468), (2025.00, 0.148), (2026.00, 0.1488), (2027.00, 0.1496), (2028.00, 0.1504), (2029.00, 0.1512), (2030.00, 0.152), (2031.00, 0.15348) , (2032.00, 0.15496), (2033.00, 0.15644), (2034.00, 0.15792), (2035.00, 0.1594), (2036.00 , 0.1602), (2037.00, 0.161), (2038.00, 0.1618), (2039.00, 0.1626), (2040.00, 0.1634), (2041.00, 0.16354), (2042.00, 0.16368), (204 3.00, 0.16382), (2044.00, 0. 16396), (2045.00, 0.1641), (2046.00, 0.16426), (2047.00, 0.164 42), (2048.00, 0.16458), (2049.00, 0.16474), (2050.00, 0.1649) : c cost_of_materials_constant_based_on_mass_per_kWh = (7/"kWh/kg")/23 : c cost_of_materials_per_SLB_kWh = cost_of_materials_constant_per_kWh+cost_of_materials_others_per_kWh+cost_of_materials_constant_based_on_ mass_per_kWh : c transportation_costs_per_mile_for_the_year = transporation_cost_per_mile*((1+percentage_increase_in_transportation_cost)^(TIME-2020)) : c transportation_costs_per_tonmile = transportation_ costs_per_mile_for_the_year/transport_truck_tonnage : c transport_distance_from _reman_capacity = 2524.2424-(reman_capacity/20625) : c transport_distance_possible = MAX( transport_distance_min_input, tran sport_distance_from_reman_capacity) : c tonne_per_SLB_kWh = 0.00110231/"kWh/kg" : c transport_cost_per_kWh = transportation_costs_per_tonmile*transport_distance_possible*tonne_per_SLB_kWh : c transportation_costs_per_SLB_kWh = transport_cost_per_kWh/(1+technically_feasible_batteries_fraction) : c total_cost_per_SLB_kWh = Cost_of_labor_per_SLB_kWh+cost_o f_reman_capacity_per_SLB_kWh+cost_o f_electricity_per_SLB_kWh+cost_ of_materials_per_SLB_kWh+transportation_costs_per_SLB_kWh : c overhead_and_launch_costs = (0.05*cost_of_materials_per_SLB_kWh)+(0.65*Cost_of_labor_per_SLB_kWh) : c allowable_max_cost_with_overheads = total_cost_per_SLB_kWh+overhead_and_launch_costs+EOL_battery_buying_cost : c allowable_max_cost = allowable_max_ cost_with_overheads*(1+allowable_profit_%) : c LiCoO2_fraction = GRAPH(TIME) (2017.00, 0.1), (2018.00, 0.1), (2019.00, 0.1), (2020.00 , 0.1), (2021.00, 0.1), (2022.00, 0.1), (2023.00, 0.1), (2024.00, 0.1), (2025.00, 0.1), (2026.00, 0.1), (2027.00 , 0.1), (2028.00, 0.1), (2029.00, 0.1), (2030.00, 0.1), (2031.00, 0.092), (2032.00, 0.084), (2033.00, 0.076), (2034.0 0, 0.068), (2035.00, 0.06), (2036.00, 0.06), (2037.00, 0.06), (2038.00, 0.06), (2039.00, 0.06), (2040.00, 0.06) , (2041.00, 0.058), (2042.00 , 0.056), (2043.00, 0.054), (2044.00, 0.052), (2045.00, 0.05), (2046. 00, 0.05), (2047.00, 0.05), (2048.00, 0.05), (2049.00, 0.05), (2050.00, 0.05) : c LiMn2O4_fraction = GRAPH(TIME) (2017.00, 0.3), (2018.00, 0.3), (2019.00, 0.3), (2020.00, 0. 3), (2021.00, 0.28), (2022.00, 0.26), (2023.00, 0.24), (2024.00, 0.22), (2025.00, 0.2), (2026.00, 0.2), (2027.00 , 0.2), (2028.00, 0.2), (2029.00, 0.2), (2030.00, 0.2), (2031.00, 0.19), (2032.00, 0.18), (2033.00, 0.17), (2034.00, 0.16), (2035.00, 0.15), (203 6.00, 0.15), (2037.00, 0.15), 100 (2038.00, 0.15), (2039.00, 0.15), (2040.00, 0.15), (2041.0 0, 0.148), (2042.00, 0.146), (2043.00, 0.144), (2044.00, 0.142), (2045.00, 0.14), (2046.00, 0.14), (2047.00, 0. 14), (2048.00, 0.14), (2049. 00, 0.14), (2050.00, 0.14) : c LiFePO4_fraction = GRAPH(TIME) (2017.00, 0.3), (2018.00, 0.3), (2019.00, 0.3), (2020.00 , 0.3), (2021.00, 0.3), (2022.00, 0.3), (2023.00, 0.3), (2024.00, 0.3), (2025.00, 0.3), (2026.00, 0.29), (2027.00, 0. 28), (2028.00, 0.27), (2029.00, 0.26), (2030.00, 0.25), (2031.00, 0.24), (2032.00, 0.23), (2033.00, 0.22), (2034.00, 0.21), (2035.00, 0.2), (2036.00, 0.19), (2037.00, 0.18), (2038.00, 0.17), (2039.00, 0.16), (2040.00, 0.15), (2041.00, 0.15), (2042.00, 0.15), (204 3.00, 0.15), (2044.00, 0.15), (2045.00, 0.15), (2046.00, 0.148), ( 2047.00, 0.146), (2048.00, 0.144), (2049.00, 0.142), (2050.00, 0.14) : c NMC_fraction = GRAPH(TIME) (2017.00, 0.3), (2018.00, 0.3), (2019.00, 0.3), (2020.00, 0. 3), (2021.00, 0.32), (2022.00, 0.34), (2023.00, 0.36), (2024.00, 0.38), (2025.00, 0.4), (2026.00, 0.41), (2027.00, 0.42), (2028.00, 0.43), (2029.00, 0.44), (2030.00, 0.45), (2031.00, 0.478), (2032.00, 0.506), (2033.00, 0.534), (2034 .00, 0.562), (2035.00, 0.59), (2036.00, 0.6), (2037.00, 0.61), (2038.00, 0.62), (2039.00, 0.63), (2040.00, 0.64) , (2041.00, 0.644), (2042.00 , 0.648), (2043.00, 0.652), (2044.00, 0.656), (2045.00, 0.66), (2046.00, 0.662), (2047.0 0, 0.664), (2048.00, 0.666), (2049.00, 0.668), (2050.00, 0.67) : c aluminum_fraction = (LiCoO2_fraction*0.304)+(LiMn2O4_fraction*0.075)+(LiFePO4_fraction*0.457)+(NMC_fraction*0.263) : c total_BEV_out = BEV_out_a+BEV_out_b+BEV_out_c+BEV_out_d : c BEV_battery_energy_out_per_year = BEV_battery_energy*total_BEV_out : c total_hyb_out = hyb_out_a+hyb_out_b+hyb_out_c+hyb_out_d : c hyb_battery_energy_out_per_year = hyb_battery_energy*total_hyb_out : c total_PHEV10_out = PHEV10_out_a+PHEV10_out_b+PHEV10_out_c+PHEV10_out_d : c PHEV10_battery_energy_out_per_year = PHEV10_battery_energy*total_PHEV10_out : c total_PHEV40_out = PHEV40_out_a+PHEV40_out_b+PHEV40_out_c+PHEV40_out_d : c PHEV40_battery_energy_out_per_year = PHEV40_battery_energy*total_PHEV40_out : c total_EV_battery_energy_per_year = BEV_battery_energy_out_per_year+hyb_battery_energy_out_per_year+PHEV10_battery_energy_out_per_year+P HEV40_battery_energy_out_per_year : c sum_total_cap_eol_bat = total_EV_battery_energy_per_year*remaining_capacity : c collectedEOLB = collectioneffi ciency*sum_total_cap_eol_bat : c total_SLB_out = SLB_out_a+SLB_out_b+SLB_out_c+SLB_out_d : c recycledEOLB = IF TIME> Start_y ear_for_reman THEN (collectedEOLB- reman_SLB+total_SLB_out)/remaining_capacity ELSE 0 : c Aluminum_after_recycl = Aluminum_RE*aluminum_fraction*recycledEOLB : c aluminum_cost = 1/0.453592 : c PHEV40_battery_energy_in_per_year = PHEV40_battery_energy*PHEV40_in : c hyb_battery_energy_in_per_year = hyb_battery_energy*hyb_in : c PHEV10_battery_energy_in_per_year = PHEV10_battery_energy*PHEV10_in : c BEV_battery_energy_in_per_year = BEV_battery_energy*(EV100_in+EV200_in+EV300_in) : c sum_total_cap_in_bat = PHEV40_battery_energy_in_per_year+hyb_battery_energy_in_per_year+PHEV10_battery_energy_in_per_year+B EV_battery_energy_in_per_year : c total_LIB_demand = demand_for_energy_storage+sum_total_cap_in_bat : c Aluminum_in = aluminum_fraction*total_LIB_demand : c Aluminum_demand_modified = Aluminum_in-Aluminum_after_recycl : c aluminum_revenue = Alumin um_after_recycl*aluminum_cost : c real_discount_rate = (nominal _discount_rate-expected_inflation_rate)/(1+expected_inflation_rate) : c buying_cost_npv = IF TIME>Start_year_for_reman THEN (EOL_battery_buying_cost)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c NMC_avg_Co = GRAPH(TIME) (2017.00, 0.394), (2018.00, 0.394), (2019.00, 0.394), (2020.00, 0.394), (2021.00, 0.394), (2022.00, 0.394), (2023.00, 0.394), (2024.00, 0.394), (2025.00, 0.394), (2 026.00, 0.3922), (2027.00, 0.3904), (2028.00, 0.3886), (2029.00, 0.3868), (2030.00, 0.385), (2031.00, 0.3688), (2032.00, 0.3526), (2033.00, 0.3364), (2034.00, 0.3202), (2035.00, 0.304), (2036.00, 0.27925), (2037.00, 0.2545), (2038.00, 0.22975), (2039.00, 0.205), (2040.00, 0.18025), (2041.00, 0.171545455), (2042.00, 0. 162840909), (2043.00, 0.154136364), (2044.00, 0.145431818), (2045.00, 101 0.136727273), (2046.00, 0.132122117), (2047.00, 0.127516 961), (2048.00, 0.122911805), (2049.00, 0.118306649), (2050.00, 0.113701493) : c Cobalt_fraction = (LiCoO2_fraction*1.01)+(LiMn2O4_fraction*0)+(LiFePO4_fraction*0)+(NMC_fraction*NMC_avg_Co) : c Cobalt_after_recycl = Cobalt_ RE*Cobalt_fractio n*recycledEOLB : c Cobalt_in = Cobalt_fraction*total_LIB_demand : c Cobalt_demand_modified = Cobalt_in-Cobalt_after_recycl : c cobalt_revenue = Cobalt_after_recycl*cobalt_cost : c Copper_fraction = (LiCoO2_fraction*0.426)+(LiMn2O4_fraction*0.075)+(LiFePO4_fraction*0.571)+(NMC_fraction*0.390) : c Copper_after_recycl = Copper_ RE*Copper_fraction*recycledEOLB : c Copper_in = Copper_fraction*total_LIB_demand : c Copper_demand_modified = Copper_in-Copper_after_recycl : c copper_revenue = Copper_after_recycl*copper_cost : c cost_of_new_battery_BNEF = GRAPH(TIME) (2010.00, 1160), (2011.00, 899), (2012.00, 707), (2013.00, 650), (2014.00, 577), (2015.00, 373), (2016.00, 288), (2017.00, 214), (2018.00, 176), (2019.00, 162.3333333), (2020.00, 148.6666667), (2021.00, 135), (2022.00, 121.3333333), (2023.00, 107.6666667), (2024.00, 94), (202 5.00, 88.66666667), (2026.00, 83.33333333), (2027.00, 78), (2028.00, 72.66666667), (2029.00, 67.33333333), (2030.00, 62), (2031.00, 61.38), (2032.00, 60.7662), (2033.00, 60.158538), (2034.00, 59.55695262), (2035.0 0, 58.96138309), (2036.00, 58.37176926), (2037.00, 57.78805157), (2038.00, 57.21017105), (2039.00, 56.63 806934), (2040.00, 56.07168865), (2041.00, 55.51097176), (2042.00, 54.95586205), (2043.00, 54.40630343), (204 4.00, 53.86224039), (2045.00, 53.32361799), (2046.00, 52.79038181), (2047.00, 52.26247799), (2048.00, 51.73 985321), (2049.00, 51.22245468), (2050.00, 50.71023013) : c Electricity_cost_npv = IF TIME>Start_year_for_reman THEN (cost_of_electricity_per_SLB_kWh)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c NMC_avg_Li = GRAPH(TIME) (2017.00, 0.139), (2018.00, 0.139), (2019.00, 0.139), (2020.00, 0.139), (2021.00, 0.139), (2022.00, 0.139), (2023.00, 0.139), (2024.00, 0.139), (2025.00, 0.139), (2 026.00, 0.13887), (2027.00, 0.13874), (2028.00, 0.13861), (2029.00, 0.13848), (2030.00, 0.13835), (2031.00, 0.136914), (2032.00, 0.13547 8), (2033.00, 0.134042), (2034.00, 0.132606), (2035.00, 0.13117), (2036.00, 0.128936), ( 2037.00, 0.126702), (2038.00, 0.124468), (2039.00, 0.122234), (2040.00, 0.12), (2041.00, 0.119181818), (2042 .00, 0.118363636), (2043.00, 0.117545455), (2044.00, 0.116727273), (2045.00, 0.115909091), (2046.00, 0.115419 81), (2047.00, 0.114930529), (2048.00, 0.114441248), (2049.00, 0.113951967), (2050.00, 0.113462687) : c Lithium_fraction = (LiCoO2_fraction*0.119)+(LiMn2O4_fraction*0.104)+(LiFePO4_fraction*0.084)+(NMC_fraction*NMC_avg_Li) : c Lithium_after_recycl = Lithiu m_RE*Lithium_fractio n*recycledEOLB : c lithium_revenue = lithium_cost*Lithium_after_recycl : c manganese_cost = 6.6/1000 : c NMC_avg_Mn = GRAPH(TIME) (2017.00, 0.367), (2018.00, 0.367), (2019.00, 0.367), (2020.00, 0.367), (2021.00, 0.367), (2022.00, 0.367), (2023.00, 0.367), (2024.00, 0.367), (2025.00, 0.367), (2 026.00, 0.36533), (2027.00, 0.36366), (2028.00, 0.36199), (2029.00, 0.36032), (2030.00, 0.35865), (2031.00, 0.343592), (2032.00, 0.32853 4), (2033.00, 0.313476), (2034.00, 0.298418), (2035.00, 0.28336), (2036.00, 0.2603505), (2037.00, 0.237341), (2038.00, 0.2143315), (2039.00, 0.191322), (2040.00, 0.1683125), (2041.00, 0.160216667), (2042.00, 0.152120833), (204 3.00, 0.144025), (2044.00, 0.135929167), (2045.00, 0.127833333), (2046.00, 0.123544 279), (2047.00, 0.119255224), (2048.00, 0.114966169), (2049.00, 0.110677114), (2050.00, 0.10638806) : c Manganese_fraction = (LiCoO2_fraction*0)+(LiMn2O4_fraction*1.37)+(LiFePO4_fraction*0)+(NMC_fraction*NMC_avg_Mn) : c Manganese_after_recycl = Manganes e_RE*Manganese_fraction*recycledEOLB : c manganese_revenue = manganese_cost*Manganese_after_recycl : c iron_cost = 312.29/1000 : c Iron_fraction = (LiCoO2_fraction*0)+(LiMn2O4_fract ion*0)+(LiFePO4_fraction*0.68)+(NMC_fraction*0) : c Iron_after_recycl = Iron_RE*Iron_fraction*recycledEOLB : c iron_revenue = iron_cost*Iron_after_recycl : c Steel_fraction = (LiCoO2_fraction*0.963)+(LiMn2O4_fraction*1.105)+(LiFePO4_fraction*2.348)+(NMC_fraction*0.866) 102 : c Steel_after_recycl = Steel_R E*Steel_fraction*recycledEOLB : c steel_cost = 249.22/1000 : c steel_revenue = Steel_after_recycl*steel_cost : c NMC_avg_Ni = GRAPH(TIME) (2017.00, 0.392), (2018.00, 0.392), (2019.00, 0.392), (2020.00, 0.392), (2021.00, 0.392), (2022.00, 0.392), (2023.00, 0.392), (2024.00, 0.392), (2025.00, 0.392), (2 026.00, 0.39449), (2027.00, 0.39698), (2028.00, 0.39947), (2029.00, 0.40196), (2030.00, 0.40445), (2031.00, 0.424466), (2032.00, 0.44448 2), (2033.00, 0.464498), (2034.00, 0.484514), (2035.00, 0.50453), (2036.00, 0.534749), ( 2037.00, 0.564968), (2038.00, 0.595187), (2039.00, 0.625406), (2040.00, 0.655625), (2041.00, 0.66596060 6), (2042.00, 0.676296212), (2043.00, 0.686631818), (2044.00, 0.696967424), (2045.00, 0.70730303), (2046. 00, 0.71226332), (2047.00, 0.717223609), (2048.00, 0.722183899), (2049.00, 0.727144188), (2050.00, 0.732104478) : c Nickel_fraction = (LiCoO2_fraction*0.071)+(LiMn2O4_fraction*0)+(LiFePO4_fraction*0)+(NMC_fraction*NMC_avg_Ni) : c Nickel_after_recycl = Nickel_R E*Nickel_fraction*recycledEOLB : c nickel_revenue = Nickel_after_recycl*nickel_cost : c total_revenue_recycling = (lithium_revenue+manganese_revenue+i ron_revenue+steel_revenue+nickel_ revenue+copper_revenue+aluminum_r evenue+cobalt_revenue)*((1+revenue_change_yearly)^(TIME-2020)) : c recycling_cost_per_kg_per_year = recycling_cost_in_2020_per_kg*((1+recycling_cost_change_yearly)^(TIME- 2020)) : c recycling_cost = recycling_cost_per_kg_per_year/"kWh/kg" : c total_cost_recycling = recycling_cost*recycledEOLB : c EOL_recycling_benefit = total_re venue_recycling-tota l_cost_recycling : c EOL_Recycling_value = EOL_recycling_be nefit/(1000000*((1+real_di scount_rate)^(TIME- Start_year_for_reman))) : c EOL_recycling_value_per_kWh = IF recycledEOLB >0 THEN EOL_Recycling_val ue*1000000/recycledEOLB ELSE 0 : c health_factor = (remaining_capac ity-EOL_after_SLB)/(1-EOL_after_SLB) : c value_parameter = health_factor/value_of_throughput_new_battery : c UPDF = IF TIME> (year_of_switch_from_EA_to_GP-1) THEN UPDF_GP ELSE UPDF_EA : c used_product_discount_factor = IF TIMEStart_year_for_r eman THEN (selling_price_of_SLB- allowable_max_cost)*reman_SLB/(1000000*((1+real_discount_rate)^(TIME-Start_year_for_reman))) ELSE 0 : c EOL_remanufacturing_value_per_kWh = IF reman_SLB>0 THEN EOL_value_from_remanufacturing*1000000/reman_SLB ELSE 0 : c total_EOL_value = IF TIME>Start_year_for_reman THEN EOL_Recycling_value+EOL_value_ from_remanufacturing ELSE 0 : c EOL_total_value_per_kWh = IF TIME>Start_year_for_reman THEN (total_EOL_value*1000000/(reman_ SLB+recycledEOLB)) ELSE 0 : c Equipment_and_facility_cost_npv = IF TIME>Start_year_for_reman THEN (cost_of_reman_capacity_per_SLB_kWh)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c Iron_in = Iron_fraction*total_LIB_demand : c Iron_demand_modified = Iron_in-Iron_after_recycl : c labor_cost_npv = IF TIME >Start_year_for_reman THEN (Cost_of_labor_per_SLB_kWh)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c Lithium_in = Lithium_fraction*total_LIB_demand : c Lithium_demand_modified = L ithium_in-Lithium_after_recycl : c Manganese_in = Manganese_fraction*total_LIB_demand : c Manganese_demand_modified = Manganese_in-Manganese_after_recycl : c Material_cost_npv = IF TIME>Start_year_for_reman THEN (cost_of_materials_per_SLB_kWh)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c modified_LIB_demand = IF TIME>Start_year_for_reman THEN total_LIB_demand-reman_SLB ELSE total_LIB_demand : c Nickel_in = Nickel_fraction*total_LIB_demand : c Nickel_demand_modified = Nickel_in-Nickel_after_recycl 103 : c overhead_and_profit_per_kwh = allowable_max_cost-total_cost_per_SLB_kWh-EOL_battery_buying_cost : c Overhead_and_profit_npv = IF TIME>Start_year_for_reman THEN (overhead_and_profit_per_kwh)/((1+real_discount_rate)^(TIME-Start_year_for_reman)) ELSE 0 : c SLB_tonnes = 1.1/("kWh/kg"*1000) : c Steel_in = Steel_fraction*total_LIB_demand : c Steel_demand_modified = Steel_in-Steel_after_recycl : c sum_total_cap_eol_bat_recycle = sum_total_cap_eol_bat-sum_total_cap_eol_bat_SLB : c total_revenue_recycling_per_kwh = IF recycledEOLB>0 THEN total_re venue_recycling/recycledEOLB ELSE 0 : c Transport_cost_npv = IF TIME>Start_year_for_reman THEN (transportation_costs_per_SLB_kWh)/((1+real_disc ount_rate)^(TIME-Start _year_for_reman)) ELSE 0 C3. Time Specs STARTTIME=2017 STOPTIME=2050 DT=1 INTEGRATION=EULER RUNMODE=NORMAL PAUSEINTERVAL=0 104 APPENDIX D Supplementary Information for Chapter 4 105 D1. Uranium Recovery Rates The overall recovery rate was assumed to be 90% of the recovery rate based on prev ious literature. Similar to previous work 40, Case 1 was assumed to have the re covery rate dependent on ore grade based on Eq. 1,37,40 and Case 2 was based on Eq. 2 from 209 . While the correlation coefficient was low for the regression equation Eq. 2 reported by 209, it was based on existi ng uranium mines data. Eq. 1, however, also included laboratory-scale results and undisclosed assumptions 210 . %98.07.23 log % __ Eq. 1 %94.702.97 log % __ Eq. 2 Table D 1. Recovery rate s from solution and Recovery rates total in percentage for Cases 1 and 2. Uranium ore grade (%U3O8) Recovery rate from solution (Case 1) (%) Recovery rate total (Case 1) (%) Recovery rate from solution (Case 2) (%) Recovery rate total (Case 2) (%) 0.036 82.93 74.64 84.84 76.35 0.08 89.30 80.37 87.21 78.49 0.13 92.32 83.09 88.65 79.78 0.18 93.99 84.59 89.61 80.65 0.23 95.06 85.55 90.34 81.31 0.28 95.79 86.21 90.92 81.83 0.33 96.32 86.69 91.41 82.27 0.40 96.86 87.17 91.98 82.78 D2. ISL Inventory Tables Based on the total recovery rate s obtained for Cases 1 and 2, as shown in Table D1, the amounts of chemicals, water, resin, steel, and land were calculated as given in Tables D2 and D3. Due to limited data, various assumptions are made to obtain the different values of the processes and flows based on available data, as follows. The total steel was assumed to be 7,200 kg over a 20m side hexagonal we ll-field and depth 120m. The steel per area was 6.93 kg/m2. The uranium in the ground wa s obtained by dividing the obtained uranium (1kgU) by the recovery rate. The total kgU obtained per m 2 based on 196 was 32 kgU for an ore grade of 0.168% U3O8 and 75% recovery. This implied a total uranium ore weight of 29,968 kg/m 2. The kgU obtained per m2 was calculated from the recovery rate and ore weight per area, and the associated area per kgU was also calculated. The steel per kgU was then calculated using the area per kgU. 106 The average uranium concentration in the production fluid was 120-150mg/l. 211 With a 95% recovery during the ion-exchange process, th e amount of water required was 7,018-8,772 liters to extract one kgU. The water used for the lixiviant was assumed to be 7,895 liters per kgU for an ore grade of 0.058% U3O8. The carbonate concentration in the lixiviant is 0.3-1.5 g/l for ore grades of 0.03-0.05% U 3O8. In this study, the concentration was assumed to be 0.9 g/l for an ore grade of 0.4% U3O8, which was multiplied by the water used to obtain carbonate use for that grade. The diesel used was 886.6 MJ per kgU for 75% recovery. For the ion exchange circuit, a 44 m 3 system 211 of weak acidic cationic resin 213 was assumed. The production was 11.8 mil. lbs U3O8 from 2000 to 2018 for a uranium grade of 0.08% 238 . The density was assumed to be 760 kg/m 3.239 The brine needed to wash the resin was determined from the uranium concentration in the eluant. The uranium concentration after wash was 8 to 20 g/l for ore rages of 0.05 to 0.25% U 3O8. The uranium concentration was assumed to be 14gU/l for an ore grade of 0.15% U3O8, and therefore the concentrated brine solution amount was 71.43 l/kgU. The hydrogen peroxide to precipitat e the uranium from the solution was 0.61 kg/kgU at 75% recovery rate for acidic ISL process. 195 Due to the lack of data for al kaline ISL, the same assumption had been used for the m odeled processes. An inversely proportional relations hip with recovery rates was a ssumed to obtain the water, carbonate, diesel, ion exchange re sin, concentrated brine soluti on, and hydrogen peroxide amounts for different ore grades considered. 107 Table D 2. Inventory for Case 1 (from Eq. 1). Uranium ore grade (%U 3O8) 0.036 0.08 0.13 0.18 0.23 0.28 0.33 0.4 Recovery rate (%) 74.64 80.37 83 .09 84.59 85.55 86 .21 86.69 87.17 Process/ detail Process used Water use (liters/kgU) Water, decarbonised, at user {GLO}| market for | Alloc Def, U 8277.1 1 7686.6 9 7435.00 7303.18 7221.4 0 7165.9 4 7126.24 7087.16 Carbonate use (kg/kgU) Sodium carbonate from ammonium chloride production, at plant/GLO US-EI U 6.45 5.99 5.79 5.69 5.63 5.58 5.55 5.52 Diesel use (MJ/kgU) Diesel, burned in diesel-electric generating set {GLO}| market for | Alloc Def, U 890.90 827.35 800.26 786.08 777.27 771.30 767.03 762.82 Steel (kg/kgU) Steel, chromi um steel 18/8 {GLO}| market for | Alloc Def, U 0.86 0.36 0.21 0.15 0.12 0.10 0.08 0.07 Area used (m 2/kgU) Transformation, to mineral extraction site 0.12 0.05 0.03 0.02 0.02 0.01 0.01 0.01 Area used (m 2/kgU) Transformation, to unknown 0.12 0.05 0.03 0.02 0.02 0.01 0.01 0.01 Area used (m 2/kgU) Transformation, from mineral extraction site 0.12 0.05 0.03 0.02 0.02 0.01 0.01 0.01 Area used (m 2/kgU) Transformation, from unknown 0.12 0.05 0.03 0.02 0.02 0.01 0.01 0.01 Occupation (7 years) (m2a) Occupation, mineral extraction site 0.87 0.36 0.22 0.15 0.12 0.10 0.08 0.07 Uranium in ground (kg/kgU) Uranium 1.34 1.24 1.20 1.18 1.17 1.16 1.15 1.15 Cationic resin (kg/kgU) Cationic resin, at plant/US* US-EI U 0.0079 0.0074 0.0071 0.0070 0.0069 0.0069 0.0068 0.0068 Hydrogen peroxide (kg/kgU) Hydrogen peroxide, 50% in H2O, at plant/US- US-EI U 0.61 0.57 0.55 0.54 0.53 0.53 0.53 0.52 Brine needed (l/kgU) Sodium chloride, brine solution, at plant/US- US-EI U 80.18 74.46 72.02 70.75 69.95 69.42 69.03 68.65 Transportation (tkm) Transport, lorry 3.5-20t, fleet average/US* US-EI U 68.79 68.79 68.79 68.79 68.79 68.79 68.79 68.79 Uranium transport drum (k g) Steel, chromium steel 18/8 {GLO}| market for | Alloc Def, U 1.13 1.13 1.13 1.13 1.13 1.13 1.13 1.13 108 Table D 3. Inventory for Case 2 (from Eq. 2). Uranium ore grade (%U 3O8) 0.036 0.08 0.13 0.18 0.23 0.28 0.33 0.4 Recovery rate (%) 76.35 78.49 79 .78 80.65 81.31 81 .83 82.27 82.78 Process/ detail Process used Water use (liters/kgU) Water, decarbonised, at user {GLO}| market for | Alloc Def, U 8026.6 5 7808.6 4 7681.78 7599.03 7537.87 7489.4 9 7449.56 7403.34 5 Carbonate use (kg/kgU) Sodium carbonate from ammonium chloride production, at plant/GLO US-EI U 6.69 6.51 6.40 6.33 6.28 6.24 6.21 6.17 Diesel use (MJ/kgU) Diesel, burned in diesel-electric generating set {GLO}| market for | Alloc Def, U 870.88 847.23 833.46 824.49 817.85 812.60 808.27 803.25 Steel (kg/kgU) Steel, chromi um steel 18/8 {GLO}| market for | Alloc Def, U 0.84 0.37 0.22 0.16 0.12 0.10 0.09 0.07 Area used (m 2/kgU) Transformation, to mineral extraction site 0.12 0.05 0.03 0.02 0.02 0.01 0.01 0.01 Area used (m 2/kgU) Transformation, to unknown 0.12 0.05 0.03 0.02 0.02 0.01 0.01 0.01 Area used (m 2/kgU) Transformation, from mineral extraction site 0.12 0.05 0.03 0.02 0.02 0.01 0.01 0.01 Area used (m 2/kgU) Transformation, from unknown 0.12 0.05 0.03 0.02 0.02 0.01 0.01 0.01 Occupation (7 years) (m2a) Occupation, mineral extraction site 0.85 0.37 0.23 0.16 0.12 0.10 0.09 0.07 Uranium in ground (kg/kgU) Uranium 1.30 1.27 1.25 1.24 1.2 1.22 1.22 1.21 Cationic resin (kg/kgU) Cationic resin, at plant/US* US-EI U 0.0076 0.0074 0.0073 0.0072 0.0071 0.0071 0.0070 0.0070 Hydrogen peroxide (kg/kgU) Hydrogen peroxide, 50% in H2O, at plant/US- US-EI U 0.60 0.58 0.57 0.57 0.56 0.56 0.56 0.55 Brine needed (l/kgU) Sodium chloride, brine solution, at plant/US- US-EI U 74.99 72.96 71.77 71.00 70.43 69.97 69.60 69.17 Transportation (tkm) Transport, lorry 3.5-20t, fleet average/US* US-EI U 68.79 68.79 68.79 68.79 68.79 68.79 68.79 68.79 Uranium transport drum (kg) Steel, chromium steel 18/8 {GLO}| market for | Alloc Def, U 1.13 1.13 1.13 1.13 1.13 1.13 1.13 1.13 109 D3. LCA Results Based on life-cycle inventories for Cases 1 and 2, the LCA results are given in Tables D4, D5, and D6. The cradle-to-gate GWP, acidification potential, and water consumption of uranium extraction from ISL, incl uding the ion-exchange process, were evaluat ed by using the ReCiPe Midpoint (H) me thod in SimaPro v8.5 software. 108 The functional unit wa s one kgU of uranium yellowcake transported to a uranium refining facility after extraction using the in-sit u leaching process. Henceforth, the functional unit would be re ferred to as kgU. Table D 4. LCA results for 0.036% ore grade for Case 1 (Eq. 1), w ith contributions from each process (Functional unit: 1 kgU). Impact category Unit Total Uranium, in yellowcake {GLO}| uranium production, in yellowcake, in-situ leaching | Alloc Def, U_Alkaline_depth120_US Steel, chromium steel 18/8 {GLO}| market for | Alloc Def, U Sodium carbonate from ammonium chloride production, at plant/GLO US-EI U Water, decarbonised, at user {GLO}| market for | Alloc Def, U Cationic resin, at plant/US* US-EI U Hydrogen peroxide, 50% in H2O, at plant/US- US-EI U Sodium chloride, brine solution, at plant/US- US-EI U Uranium, in yellowcake {GLO}| uranium production, in yellowcake, in-situ leaching | Alloc Rec, U_emissions Uranium transportation Diesel, burned in diesel-electric generating set {GLO}| market for | Alloc Def, U Global warming kgCO 2 eq 130.37 0.00 9.30 8.71 0.06 0.01 0.87 11.57 0.00 20.97 78.87 Stratospheric ozone depletion kgCFC 11 eq 9.4E-05 0.0E+ 00 2.7E- 06 2.0E- 06 1.9E- 08 1.1E- 09 1.9E- 07 3.7E-06 0.0E +00 7.5E- 06 7.8E-05 Ionizing radiation kBqCo- 60 e q 37.88 0.00 0.41 1.30 0.00 0.00 0.13 2.74 31.40 0.98 0.92 Ozone formation, Human health kgNO x eq 1.59 0.00 0.02 0.04 0.00 0.00 0.00 0.05 0.00 0.18 1.29 110 Table D 4 (cont™d). Impact category Unit Total Uranium, in yellowcake {GLO}| uranium production, in yellowcake, in-situ leaching | Alloc Def, U_Alkaline_depth120_US Steel, chromium steel 18/8 {GLO}| market for | Alloc Def, U Sodium carbonate from ammonium chloride production, at plant/GLO US-EI U Water, decarbonised, at user {GLO}| market for | Alloc Def, U Cationic resin, at plant/US* US-EI U Hydrogen peroxide, 50% in H2O, at plant/US- US-EI U Sodium chloride, brine solution, at plant/US- US-EI U Uranium, in yellowcake {GLO}| uranium production, in yellowcake, in-situ leaching | Alloc Rec, U_emissions Uranium transportation Diesel, burned in diesel-electric generating set {GLO}| market for | Alloc Def, U Fine particulate matter formation kgPM 2.5 eq 0.27 0.00 0.03 0.01 0.00 0.00 0.00 0.02 0.00 0.01 0.20 Ozone formation, Terrestrial ecos ystems kgNO x eq 1.64 0.00 0.03 0.06 0.00 0.00 0.01 0.06 0.00 0.19 1.30 Terrestrial acidification kgSO 2 eq 0.81 0.00 0.04 0.03 0.00 0.00 0.00 0.05 0.00 0.09 0.60 Freshwater eutrophication kgP eq 0.13 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.11 0.00 0.00 Terrestrial ecotoxicity kg1,4- DCB e 0.11 0.00 0.04 0.01 0.00 0.00 0.00 0.02 0.00 0.01 0.02 Freshwater ecotoxicity kg1,4- DCB e 4.18 0.00 0.52 0.18 0.01 0.00 0.03 0.89 2.18 0.23 0.15 Marine ecotoxicity kg1,4- DBC e 6.20 0.00 0.76 0.33 0.01 0.00 0.04 1.35 3.05 0.37 0.28 Human carcinogenic toxicit y kg1,4- DBC e 6.67 0.00 3.39 0.27 0.01 0.00 0.10 0.95 0.96 0.70 0.29 Human non-carcinogenic toxicit y kg1,4- DBC e 2368.25 0.00 282.60 211.28 2.37 0.17 18.36 1061.28 331.49 288.39 172.32 Land use m 2a crop eq 3.03 0.63 0.31 0.65 0.00 0.00 0.06 0.87 0.00 0.34 0.16 111 Table D 4. (cont™d). Impact category Unit Total Uranium, in yellowcake {GLO}| uranium production, in yellowcake, in-situ leaching | Alloc Def, U_Alkaline_depth120_US Steel, chromium steel 18 /8 {GLO}| market for | Alloc Def, U Sodium carbonate from ammonium chloride production, at plant/GLO US-EI U Water, decarbonised, at user {GLO}| market for | Alloc De f, U Cationic resin, at plant/US* US-EI U Hydrogen peroxide, 50% in H2O, at plant/US- US-EI U Sodium chloride, brine solution, at plant/US- US-EI U Uranium, in yellowcake {GLO}| uranium production, in yellowcake, in-situ leaching | Alloc Rec, U_emissions Uranium transportation Diesel, burned in diesel-electric generating set {GLO}| market for | Alloc Def, U Mineral resource scarcity kgCu eq 37.55 33.76 3.33 0.06 0.00 0.00 0.00 0.30 0.00 0.05 0.04 Fossil resource scarcit y kg oil e q 40.48 0.00 2.02 2.36 0.01 0.01 0.27 2.98 0.00 6.81 26.02 Water consumption m 3 97.24 0.00 0.09 18.61 8.13 0.01 1.73 54.76 0.00 13.76 0.16 112 Table D 5. LCA results for all uranium ore grades considered for Case 1 (Eq. 1) (Functional unit: 1 kgU). Uranium grade (% U 3O8) 0.036 0.08 0.13 0.18 0.23 0.28 0.33 0.4 Impact category Unit Global warming kgCO 2 eq 130.37 120.89 117.17 115.28 114.13 113.36 112.81 112.27 Stratospheric ozone depletion kgCFC 11 eq 9.4E-05 8.7E-05 8.5E-05 8.3E-05 8.2E-05 8.2E-05 8.1E-05 8.1E-05 Ionizing radiation kBqCo-60 eq 37.88 35 .17 34.03 33.44 33. 07 32.82 32.64 32.47 Ozone formation, Human health kgNO x eq 1.59 1.48 1.44 1.41 1.40 1.39 1.38 1.38 Fine particulate matter formation kgPM 2.5 eq 0.27 0.25 0.24 0.23 0.23 0.23 0.23 0.23 Ozone formation, Terrestrial ecosystems kgNO x eq 1.64 1.53 1.49 1.46 1.45 1.44 1.43 1.43 Terrestrial acidification kgSO 2 eq 0.81 0.75 0.73 0.72 0.71 0.70 0.70 0.70 Freshwater eutrophication kgP eq 0.13 0.12 0.11 0.11 0.11 0.11 0.11 0.11 Terrestrial ecotoxicity kg1,4-DCB e 0.11 0.10 0.09 0.09 0.09 0.09 0.09 0.08 Freshwater ecotoxicity kg1,4-DCB e 4.18 3.81 3.67 3.60 3.55 3.52 3.50 3.48 Marine ecotoxicity kg1,4-DBC e 6.20 5.64 5.43 5.33 5.27 5.22 5.19 5.17 Human carcinogenic toxicity kg1,4-DBC e 6.67 5.64 5.31 5.16 5.08 5.02 4.99 4.95 Human non-carcinogenic toxicity kg1,4-DBC e 2368.25 2168.97 2093.68 2056.21 2033.58 2018.46 2007.71 1997.17 Land use m 2a crop eq 3.03 2.46 2.28 2.20 2.15 2.12 2.09 2.07 Mineral resource scarcity kgCu eq 37.55 34.27 32.99 32.34 31.95 31.68 31.49 31.31 Fossil resource scarcity kg oil eq 40.48 37.71 36.60 36.03 35. 69 35.45 35.29 35.12 Water consumption m 3 97.24 91.27 88.73 87.40 86.57 86.01 85.61 85.22 113 Table D 6. LCA results for all uranium ore grades considered for Case 2 (Eq. 2) (Functional unit: 1 kgU). Uranium grade (% U 3O8) 0.036 0.08 0.13 0.18 0.23 0.28 0.33 0.4 Impact category Unit Global warming kgCO 2 eq 128.06 123.19 120.96 119.66 118.75 118.05 117.49 116.85 Stratospheric ozone depletion kgCFC 11 eq 9.2E-05 8.9E-05 8.8E-05 8.7E-05 8.6E-05 8.6E-05 8.5E-05 8.5E-05 Ionizing radiation kBqCo-60 eq 37.01 35 .95 35.36 34.98 34. 70 34.48 34.30 34.09 Ozone formation, Human health kgNO x eq 1.55 1.51 1.49 1.47 1.46 1.46 1.45 1.44 Fine particulate matter formation kgPM 2.5 eq 0.27 0.25 0.25 0.24 0.24 0.24 0.24 0.24 Ozone formation, Terrestrial ecosystems kgNO x eq 1.61 1.57 1.54 1.53 1.52 1.51 1.50 1.49 Terrestrial acidification kgSO 2 eq 0.80 0.77 0.75 0.75 0.74 0.74 0.73 0.73 Freshwater eutrophication kgP eq 0.12 0.12 0.12 0.12 0.12 0.11 0.11 0.11 Terrestrial ecotoxicity kg1,4-DCB e 0.11 0.10 0.09 0.09 0.09 0.09 0.09 0.09 Freshwater ecotoxicity kg1,4-DCB e 4.08 3.86 3.77 3.72 3.69 3.66 3.64 3.62 Marine ecotoxicity kg1,4-DBC e 6.04 5.72 5.59 5.52 5.47 5.43 5.40 5.36 Human carcinogenic toxicity kg1,4-DBC e 6.56 5.68 5.40 5.26 5.18 5.13 5.09 5.05 Human non-carcinogenic toxicity kg1,4-DBC e 2292.78 2178.83 2130.95 2104.14 2085.94 2072.32 2061.51 2049.40 Land use m 2a crop eq 2.98 2.51 2.35 2.28 2.23 2.20 2.18 2.16 Mineral resource scarcity kgCu eq 36.74 35.04 34.27 33.82 33.51 33.27 33.07 32.86 Fossil resource scarcity kgoil eq 39.76 38.44 37.80 37.42 37. 15 36.94 36.77 36.57 Water consumption m 3 94.09 91.89 90.62 89.79 89.17 88.69 88.29 87.82 114 REFERENCES 115 REFERENCES (1) Sioshansi, F. Generating Electricity in a Carbon-Constrained World; 2010. https://doi.org/10.1016/C2009-0-20022-5. (2) International Atomic Energy Agency. Transitions to Low Carbon El ectricity Systems: Key Economic and Investment Trends; 2019. (3) Muradov, N. Z.; Vezirouglu, T. N. Carbon-Neutral Fuels and Energy Carriers ; Green Chemistry and Chemical Engineering; Taylor & Francis, 2011. (4) Wang, M. Fuel-Cycle Greenhouse Gas Emissi ons Impacts of Altern ative Transportation Fuels and Advanced Vehicle Technologies. Transp. Res. Rec. J. Transp. Res. Board 1999, 1664 (1), 9Œ17. https://doi.org/10.3141/1664-02. (5) Stringer, D.; Ma, J. Where 3 Million Electri c Vehicle Batteries Will Go When They Retire https://www.bloomberg.com/news/features/2018-06-27/where-3-million-electric-vehicle-batteries-will-go-when-they-retire (accessed Oct 22, 2018). (6) Abujarad, S. Y.; Mustafa, M. W.; Jamian, J. J. Recent Approaches of Unit Commitment in the Presence of Intermittent Rene wable Energy Resources: A Review. Renewable and Sustainable Energy Reviews. 2017. https://doi.org/10.1016/j.rser.2016.11.246. (7) Das, P.; Mathur, J.; Bhakar, R.; Kanudia, A. Implications of Short-Term Renewable Energy Resource Intermittency in L ong-Term Power System Planning. Energy Strateg. Rev. 2018. https://doi.org/10.1016/j.esr.2018.06.005. (8) Jones-Albertus, B. Confronting the Duck Cu rve: How to Address Ov er-Generation of Solar Energy https://www.energy.gov/eere/articles/confronting-duck-curve-how-address-over-generation-solar-energy (accessed Dec 12, 2020). (9) United Nations Economic Commission for Europe. How Natural Gas Can Support the Uptake of Renewable Energy ; 2019. (10) Babatunde, O. M.; Munda, J. L.; Hamam, Y. A Comprehensive State- of-the-Art Survey on Power Generation Expansion Planning with In termittent Renewable Energy Source and Energy Storage. International Journal of Energy Research . 2019. https://doi.org/10.1002/er.4388. (11) Global Battery Energy Storage Market Could Grow to $546B, Says Analyst. Renewable Energy World. February 2020. (12) U.S. Energy Information Administration. Utility-scale battery storage costs decreased nearly 70% between 2015 and 2018 https://www.eia.gov/todayinenergy/detail.php?id=45596#:~:text=The average energy capacity cost,to %24625%2FkWh in 2018. (accessed Nov 11, 2020). 116 (13) Johann, A.; Madlener, R. Profitability of Energy Storage for Raising Self-Consumption of Solar Power: Analysis of Different Household Types in Germany. In Energy Procedia ; Elsevier B.V., 2014; Vol. 61, pp 2206Œ2210. https://doi.org/10.1016/j.egypro.2014.12.110. (14) Uddin, K.; Gough, R.; Radclif fe, J.; Marco, J.; Jennings, P. Techno-Economic Analysis of the Viability of Residential Photovoltaic Syst ems Using Lithium-Ion Batteries for Energy Storage in the United Kingdom. Appl. Energy 2017, 206 (July), 12Œ21. https://doi.org/10.1016/j.apenergy.2017.08.170. (15) Kamath, D.; Arsenault, R.; Kim, H. C.; Anctil, A. Economic and Environmental Feasibility of Second-Life Lithium -Ion Batteries as Fast-C harging Energy Storage. Environ. Sci. Technol. 2020, acs.est.9b05883. https://doi.org/10.1021/acs.est.9b05883. (16) Chediak, M. The Latest Bull Case fo r Electric Cars: the Cheapest Batteries Ever https://www.bloomberg.com/news/articles/2017-12-05/latest-bull-case-for-electric-cars-the-cheapest-batteries-ev er (accessed Nov 22, 2018). (17) Neubauer, J. S.; Pesaran, A.; Willia ms, B.; Ferry, M.; Eyer, J. A Techno-Economic Analysis of PEV Battery Second Use: Repur posed-Battery Selling Price and Commercial and Industrial End-User Value. In SAE Technical Papers; 2012. https://doi.org/10.4271/2012-01-0349. (18) Cready, E.; Lippert, J.; Pihl , J.; Weinstock, I.; Symons, P. Technical and Economic Feasibility of Applying Used EV Ba tteries in Stationary Applications ; 2003. https://doi.org/SAND2002-4084. (19) Canals Casals, L.; Barbero, M.; Corchero, C. Reused Second Life Ba tteries for Aggregated Demand Response Services. J. Clean. Prod. 2019, 212, 99Œ108. https://doi.org/10.1016/j.jclepro.2018.12.005. (20) Desarnaud, G. The second life battery cycle: after about 10 years in vehicle, lithium-ion batteries can be reused for another purpose and thereby begin a fisecond lifefl https://www.capgemini.com/us-en/2019/04/second-life-batteries-a-sustainable-business-opportunity-not-a-conundrum/ (accessed Feb 23, 2020). (21) Sun, S. I.; Chipperfield, A. J.; Kiaee, M.; Wills, R. G. A. Effects of Market Dynamics on the Time-Evolving Price of Second-Li fe Electric Vehicle Batteries. J. Energy Storage 2018, 19 (6), 41Œ51. https://doi.org/10.1016/j.est.2018.06.012. (22) Strickland, D.; Chittock, L.; Stone, D. A.; Foster, M. P.; Price, B. Estimation of Transportation Battery Second Life for Use in Electricity Grid Systems. IEEE Trans. Sustain. Energy 2014, 5 (3), 795Œ803. https://doi.org/10.1109/TSTE.2014.2303572. (23) Casals, L. C.; García, B. A.; Aguesse, F.; Iturrondobeitia, A. S econd Life of Electric Vehicle Batteries: Relation between Material s Degradation and Environmental Impact. Int. J. Life Cycle Assess. 2017, 22 (1), 82Œ93. https://doi.org/10.1007/s11367-015-0918-3. 117 (24) Richa, K.; Babbitt, C. W.; Nenadic, N. G.; Gaustad, G. Environmental Trade-Offs across Cascading Lithium-Ion Battery Life Cycles. Int. J. Life Cycle Assess. 2017, 22 (1), 66Œ81. https://doi.org/10.1007/s11367-015-0942-3. (25) Ahmadi, L.; Young, S. B.; Fowler, M.; Fraser , R. A.; Achachlouei, M. A. A Cascaded Life Cycle: Reuse of Electric Vehicle Lithium-I on Battery Packs in Energy Storage Systems. Int. J. Life Cycle Assess. 2017, 22 (1), 111Œ124. https://doi .org/10.1007/s11367-015-0959-7. (26) Assunção, A.; Moura, P. S.; de Almeida, A. T. Technical and Economic Assessment of the Secondary Use of Repurposed Electric Vehicl e Batteries in the Re sidential Sector to Support Solar Energy. Appl. Energy 2016, 181, 120Œ131. https://doi.org/10.1016/j.apenergy.2016.08.056. (27) Bobba, S.; Podias, A.; Di Persio, F.; Messagie, M.; Tecchio, P.; Cusenza, M. A.; Eynard, U.; Mathieux, F.; Pfrang, A. Sustainability Assessment of Second Life Application of Automotive Batteries (SASLAB): JRC Explor atory Research (2016-2017): Final Technical Report: August 2018; EUR 29321 EN, Public ations Office of the European Union, Luxembourg, 2018, ISBN 978-92-79-928; 2018. https://doi.org/10.2760/53624. (28) Martinez-Laserna, E.; Gandiaga, I.; Sa rasketa-Zabala, E.; Badeda, J.; Stroe, D.-I.; Swierczynski, M.; Goikoetxea, A. Battery Second Life: Hype, Hope or Reality? A Critical Review of the State of the Art. Renew. Sustain. Energy Rev. 2018, 93, 701Œ718. https://doi.org/10.1016/j.rser.2018.04.035. (29) Canals Casals, L.; Amante García, B. A Review of the Complexities of Applying Second Life Electric Car Batteries on Energy Businesses. In Energy Systems Conference ; 2014. https://doi.org/10.13140/2.1.2571.1362. (30) Jiao, N.; Evans, S. Secondary Use of El ectric Vehicle Batteries and Potential Impacts on Business Models. J. Ind. Prod. Eng. 2016, 33 (5), 348Œ354. https://doi.org/10.1080/21681015.2016.1172125. (31) Brauer, S.; Monhof, M.; Kl or, B.; Plenter, F.; Beverungen, D.; Siemen, C. Residential Energy Storage from Repurposed Electric Vehi cle Batteries: Market Overview and Development of a Service-Centered Business Model. In 2016 IEEE 18th Conference on Business Informatics (CBI) ; IEEE, 2016; pp 143Œ152. https://doi.org/10.1109/CBI.2016.24. (32) Jenkins, J. D.; Zhou, Z.; Ponciroli, R.; Vilim, R. B.; Ganda, F.; de Sisternes, F.; Botterud, A. The Benefits of Nuclear Flexibility in Power System Operations with Renewable Energy. Appl. Energy 2018, 222 (March), 872Œ884. https://doi.org/10.1016/j.apenergy.2018.03.002. (33) Andseta, S.; Thompson, M. J.; Jarrell, J.; Pendergast, D. CANDU Reactors and Greenhouse Gas Emissions; 1998. (34) White, S. W.; Kulcinski, G. L. Birth to Death Analysis of the Energy Payback Ratio and 118 CO2 Gas Emission Rates from Coal, Fission, Wind, and DT-Fusion Electrical Power Plants. Fusion Eng. Des. 2000, 48 (3Œ4), 473Œ481. https://doi.org/10.1016/S0920- 3796(00)00158-7. (35) Dones, R.; Heck, T.; Emmenegger, M. F. ; Jungbluth, N. Life Cycle Inventories for the Nuclear and Natural Gas Energy Systems, and Examples of Uncertainty Analysis. Int. J. Life Cycle Assess. 2005. https://doi.org/10.1065/lca2004.12.181.2. (36) Hondo, H. Life Cycle GHG Emission Analys is of Power Generation Systems: Japanese Case. Energy 2005, 30 (11-12 SPEC. ISS.), 2042Œ2056. https://doi.org/10.1016/j.energy.2004.07.020. (37) van Leeuwen, J. W. S.; Smith, P. From Ore to Electricity Energy Production and Uranium Resources Sixth Revision. In Nuclear Power : the Energy Balance; 2005; pp 1Œ24. (38) Fthenakis, V. M.; Kim, H. C. Greenhous e-Gas Emissions from Solar Electric- and Nuclear Power: A Life-Cycle Study. Energy Policy 2007. https://doi.org/10.1016/j.enpol.2006.06.022. (39) Poinssot, C.; Bourg, S.; Ouvrier, N.; Combernoux, N.; Rostaing, C.; Vargas-Gonzalez, M.; Bruno, J. Assessment of the Environmental Footprint of Nuclear Energy Systems. Comparison between Closed and Open Fuel Cycles. Energy 2014, 69, 199Œ211. https://doi.org/10.1016/j.energy.2014.02.069. (40) Norgate, T.; Haque, N.; Koltun, P. The Impact of Uranium Ore Grade on the Greenhouse Gas Footprint of Nuclear Power. J. Clean. Prod. 2014, 84 (1), 360Œ367. https://doi.org/10.1016/j.jclepro.2013.11.034. (41) Ashley, S. F.; Fenner, R. A.; Nuttall, W. J.; Parks, G. T. Life-Cycle Impacts from Novel Thorium-Uranium-Fuelled Nuclear Energy Systems. Energy Convers. Manag. 2015, 101, 136Œ150. https://doi.org/10.1016/j.enconman.2015.04.041. (42) Siddiqui, O.; Dincer, I. Comparative Asse ssment of the Environmen tal Impacts of Nuclear, Wind and Hydro-Electric Power Plants in Ontario: A Life Cycle Assessment. J. Clean. Prod. 2017, 164, 848Œ860. https://doi.org/10.1016/j.jclepro.2017.06.237. (43) Solberg-Johansen, B. Environmental Life Cycle Assessment of the Nuclear Fuel Cycle: Volume 1, University of Surrey, 1998, Vol. 1. (44) Lee, Y. E.; Lee, K. J.; Lee, B. W. En vironmental Assessment of Nuclear Power Generation in Korea. Prog. Nucl. Energy 2000, 37 (1Œ4), 113Œ118. https://doi.org/10.1016/S0149- 1970(00)00034-2. (45) Sovacool, B. K. Valuing the Greenhouse Gas Emissions from Nucl ear Power: A Critical Survey. Energy Policy 2008, 36 (8), 2950Œ2963. https://doi.org/10.1016/j.enpol.2008.04.017. (46) Lenzen, M. Life Cycl e Energy and Greenhouse Gas Emi ssions of Nuclear Energy: A 119 Review. Energy Convers. Manag. 2008. https://doi.org/10.1016/j.enconman.2008.01.033. (47) Stamford, L.; Azapagic, A. Sustainability Indicators for th e Assessment of Nuclear Power. Energy 2011, 36 (10), 6037Œ6057. https://doi.org/10.1016/j.energy.2011.08.011. (48) van der Zwaan, B. The Role of Nuclear Power in Mitigating Emissions from Electricity Generation. Energy Strateg. Rev. 2013, 1 (4), 296Œ301. https://doi.org/10.1016/j.esr.2012.12.008. (49) Nian, V.; Chou, S. K.; Su, B.; Bauly, J. Life Cycle Analysis on Carbon Emissions from Power Generation Œ The Nuclear Energy Example. Appl. Energy 2014, 118, 68Œ82. https://doi.org/10.1016/j.apenergy.2013.12.015. (50) Zafrilla, J. E.; Cadarso, M.-Á.; Monsal ve, F.; de la Rúa, C. How Carbon-Friendly Is Nuclear Energy? A Hybrid MRIO-LCA Model of a Spanish Facility. Environ. Sci. Technol. 2014, 48 (24), 14103Œ14111. https://doi.org/10.1021/es503352s. (51) Gibon, T.; Arvesen, A.; Hertwich, E. G. Life Cycle Assessment Demonstrates Environmental Co-Benefits and Trade-Offs of Low-Carbon Electricity Supply Options. Renew. Sustain. Energy Rev. 2017, 76 (March), 1283Œ1290. https://doi.org/10.1016/j.rser.2017.03.078. (52) San Martin, R. L. Environmental Emissions from Ener gy Technology Systems: The Total Fuel Cycle; 1989. https://doi.org/10.2172/860715. (53) Beerten, J.; Laes, E.; Meskens, G.; D™ haeseleer, W. Greenhouse Gas Emissions in the Nuclear Life Cycle: A Balanced Appraisal. Energy Policy 2009, 37 (12), 5056Œ5068. https://doi.org/10.1016/j.enpol.2009.06.073. (54) Lenzen, M. Life-Cycle Energy Balance and Greenhous e Gas Emissions of Nuclear Energy in Australia; 2006. (55) Nuclear Energy Agency; International At omic Energy Agency. Uran ium 2018: Resources, Production and Demand. Nuclear Energy Agency and International Atomic Energy Agency 2018. (56) Ai, N.; Zheng, J.; Chen, W. Q. U.S. E nd-of-Life Electric Vehicl e Batteries: Dynamic Inventory Modeling and Spatial Analysis for Regional Solutions. Resour. Conserv. Recycl. 2019. https://doi.org/10.1016/j.resconrec.2019.01.021. (57) Tong, S.; Fung, T.; Klein, M. P.; Weisbach, D. A.; Park, J. W. De monstration of Reusing Electric Vehicle Battery fo r Solar Energy Storage and Demand Side Management. J. Energy Storage 2017, 11, 200Œ210. https://doi.org/10.1016/j.est.2017.03.003. (58) Lutsey, N.; Nicholas, M. Update on Electric Vehicle Costs in the United States through 2030; 2019. (59) Kim, H. C.; Wallington, T. J.; Arsenault, R.; Bae, C.; Ahn, S.; Lee, J. Cradle-to-Gate 120 Emissions from a Commercial Electric Vehicle Li-Ion Battery: A Comparative Analysis. Environ. Sci. Technol. 2016, 50 (14), 7715Œ7722. https://doi.org/10.1021/acs.est.6b00830. (60) Qiao, Q.; Zhao, F.; Liu, Z.; Hao, H. Elect ric Vehicle Recycling in China: Economic and Environmental Benefits. Resour. Conserv. Recycl. 2019, 140 (July 2018), 45Œ53. https://doi.org/10.1016/j.resconrec.2018.09.003. (61) Xiong, S.; Ji, J.; Ma, X. Environmenta l and Economic Evaluation of Remanufacturing Lithium-Ion Batteries from Electric Vehicles. Waste Manag. 2020, 102, 579Œ586. https://doi.org/10.1016/j.wasman.2019.11.013. (62) Mayyas, A.; Steward, D.; Mann, M. The Case for Recycling: Overview and Challenges in the Material Supply Chain for Automotive Li-Ion Batteries. Sustain. Mater. Technol. 2019, 19, e00087. https://doi.org/10.1016/j.susmat.2018.e00087. (63) Wuschke, L.; Jäckel, H.-G.; Leißner, T.; Peuker, U. A. Crushing of Large Li-Ion Battery Cells. Waste Manag. 2019, 85, 317Œ326. https://doi.org/10.1016/j.wasman.2018.12.042. (64) D™Adamo, I.; Rosa, P. A Structured Lite rature Review on Obsolete Electric Vehicles Management Practices. Sustainability 2019, 11 (23), 6876. https://doi.org/10.3390/su11236876. (65) Choi, Y.; Rhee, S.-W. Curre nt Status and Perspectives on R ecycling of End-of-Life Battery of Electric Vehicle in Korea (Republic Of). Waste Manag. 2020, 106, 261Œ270. https://doi.org/10.1016/j.wasman.2020.03.015. (66) Ahmadi, L.; Yip, A.; Fowler, M.; Young, S. B. ; Fraser, R. A. Environmental Feasibility of Re-Use of Electric Vehicle Batteries. Sustain. Energy Technol. Assessments 2014, 6, 64Œ74. https://doi.org/10.1016/j.seta.2014.01.006. (67) Cicconi, P.; Landi, D.; Morb idoni, A.; Germani, M. Feasibility Analysis of Second Life Applications for Li-Ion Cells Us ed in Electric Powertrain Us ing Environmental Indicators. In 2012 IEEE International Energy Conf erence and Exhibition (ENERGYCON) ; IEEE, 2012; pp 985Œ990. https://doi.org/10.1109/EnergyCon.2012.6348293. (68) Richa, K.; Babbitt, C. W.; Gaustad, G. Ec o-Efficiency Analysis of a Lithium-Ion Battery Waste Hierarchy Inspired by Circular Economy. J. Ind. Ecol. 2017, 21 (3), 715Œ730. https://doi.org/10.1111/jiec.12607. (69) Bhatnagar, D.; Currier, A.; Hernandez, J.; Ma, O.; Kirby, B. Market and Policy Barriers to Energy Storage Deployment. Sandia Natl. Lab. 2013, No. September, 58. (70) Tervo, E.; Agbim, K.; DeAngelis, F.; He rnandez, J.; Kim, H. K.; Odukomaiya, A. An Economic Analysis of Residential Photovoltaic Systems with Lithiu m Ion Battery Storage in the United States. Renew. Sustain. Energy Rev. 2018, 94 (May), 1057Œ1066. https://doi.org/10.1016/j.rser.2018.06.055. (71) Assunção, A.; Moura, P. S.; de Almeida, A. T. Technical and Economic Assessment of the 121 Secondary Use of Repurposed Electric Vehicl e Batteries in the Re sidential Sector to Support Solar Energy. Appl. Energy 2016, 181, 120Œ131. https://doi.org/10.1016/j.apenergy.2016.08.056. (72) Gladwin, D. T.; Gould, C. R.; Stone, D. A. ; Foster, M. P. Viability of fisecond-Lifefl Use of Electric and Hybridelect ric Vehicle Battery Packs. IECON Proc. (Industrial Electron. Conf. 2013, 1922Œ1927. https://doi.org/10.1109/IECON.2013.6699425. (73) Madlener, R.; Kirmas, A. Economic Viability of Second Us e Electric Vehicle Batteries for Energy Storage in Residential Applications. In Energy Procedia ; 2017; Vol. 105, pp 3806Œ 3815. https://doi.org/10.1016/j.egypro.2017.03.890. (74) Heymans, C.; Walker, S. B.; Young, S. B. ; Fowler, M. Economic Analysis of Second Use Electric Vehicle Batteries for Residential Energy Storage and Load-Levelling. Energy Policy 2014, 71, 22Œ30. https://doi.org/10.1016/j.enpol.2014.04.016. (75) Mirzaei Omrani, M.; Jannesari, H. Econo mic and Environmental Assessment of Reusing Electric Vehicle Lithium-Ion Batteries for Load Leveling in th e Residential, Industrial and Photovoltaic Power Plants Sectors. Renew. Sustain. Energy Rev. 2019, 116 (August), 109413. https://doi.org/10.1016/j.rser.2019.109413. (76) Cusenza, M. A.; Guarino, F.; Longo, S.; Mistretta, M.; Cellura, M. Reuse of Electric Vehicle Batteries in Buildings: An Integrat ed Load Match Analysis and Life Cycle Assessment Approach. Energy Build. 2019, 186, 339Œ354. https://doi.org/10.1016/j.enbuild.2019.01.032. (77) Faria, R.; Marques, P.; Garcia, R.; Moura, P.; Freire, F.; Delgado, J.; De Almeida, A. T. Primary and Secondary Use of Electric Mobility Batteries from a Life Cycle Perspective. J. Power Sources 2014, 262, 169Œ177. https://doi.org/10.1016/j.jpowsour.2014.03.092. (78) Kamath, D.; Arsenault, R.; Kim, H. C.; Anctil, A. Economic and Environmental Feasibility of Second-Life Lithium-Ion Batterie s as Fast Charging Energy Storage. Environ. Sci. Technol. 2020, acs.est.9b05883. https://doi.org/10.1021/acs.est.9b05883. (79) Arizona Public Service Electric Comp any. Rate Schedule E-32T OU S: Small General Service (21 KW - 100 KW) Time of Use. Phoenix, Arizona 2010, pp 1Œ5. (80) Consolidated Edison Company of New Yo rk Inc. Schedule for Electricity Service. PSC No: 10- Electricity . New York City, New York 2012. (81) Silva, C.; Sousa, D. M.; Roque, A. Charging Electric Vehicles from Photovoltaic Generation with Intermedia te Energy Storage. In 2017 6th International Conference on Clean Electrical Power (ICCEP); IEEE, 2017; pp 596Œ601. https://doi.org/10.1109/ICCEP.2017.8004749. (82) Pires, V. F.; Roque, A.; Sousa, D. M.; Margato, E. Management of an Electric Vehicle Charging System Supported by RES and Storage Systems. In 2018 International Symposium on Power Electronics, Electrical Drives, Automation and Motion 122 (SPEEDAM); IEEE, 2018; pp 395Œ400. https://doi.org/10.1109/SPEEDAM.2018.8445222. (83) Joos, G.; de Freige, M.; Dubois, M. Design and Simulation of a Fast Charging Station for PHEV/EV Batteries. In 2010 IEEE Electrical Power & Energy Conference ; IEEE, 2010; pp 1Œ5. https://doi.org/10.1109/EPEC.2010.5697250. (84) Yoon, H.; Kim, J.; Song, E. Design of a N ovel 50 KW Fast Charger for Electric Vehicles. J. Cent. South Univ. 2013, 20 (2), 372Œ377. https://doi.org/10.1007/s11771-013-1497-8. (85) Hossain, E.; Murtaugh, D.; Mody, J.; Far uque, H. M. R.; Haque Sunny, M. S.; Mohammad, N. A Comprehensive Review on Second-Life Batteries: Current St ate, Manufacturing Considerations, Applications, Impacts, Barriers & Potential Solutions, Business Strategies, and Policies. IEEE Access 2019, 7, 73215Œ73252. https://doi.org/10.1109/ACCESS.2019.2917859. (86) Denholm, P.; Nunemaker, J.; Gagnon, P.; Cole, W.; Denholm, P.; Nunemaker, J.; Gagnon, P.; Cole, W. The Potential for Battery Energy Storage to Provide Peaking Capacity in the United States. 2019, No. June. (87) Neubauer, J.; Pesaran, A. The Ability of Ba ttery Second Use Strategies to Impact Plug-in Electric Vehicle Prices and Serve Ut ility Energy Storage Applications. J. Power Sources 2011, 196 (23), 10351Œ10358. https://doi.org/10.1016/j.jpowsour.2011.06.053. (88) Sathre, R.; Scown, C. D.; Kavvada, O.; He ndrickson, T. P. Energy a nd Climate Effects of Second-Life Use of Electric Vehicl e Batteries in California through 2050. J. Power Sources 2015, 288, 82Œ91. https://doi.org/10.1016/j.jpowsour.2015.04.097. (89) HOMER PRO Version 3.7 User Manual; 2016. (90) He, L.; Zhang, S.; Chen, Y.; Ren, L.; Li , J. Techno-Economic Potential of a Renewable Energy-Based Microgrid System for a Sustainable Large-Scale Residential Community in Beijing, China. Renew. Sustain. Energy Rev. 2018, 93 (May), 631Œ641. https://doi.org/10.1016/j.rser.2018.05.053. (91) U.S. Internal Revenue Service. Residential Renewable Energy Tax Credit http://programs.dsireusa.org/system/program/detail/1235 (accessed Dec 26, 2018). (92) U.S. Internal Revenue Service. Business Energy I nvestment Tax Credit http://programs.dsireusa.org/system/program/detail/658 (accessed Dec 26, 2018). (93) Sengupta, M.; Xie, Y.; Lopez, A.; Habte, A.; Maclaurin, G.; Shelby, J. The National Solar Radiation Data Base (NSRDB). Renew. Sustain. Energy Rev. 2018, 89, 51Œ60. (94) Stackhouse, P. W. Surface meteorology and solar energy https://eosweb.larc.nasa.gov/. (95) DTE Electric Company. Rate Book for Electric Service. M.P.S.C. No. 1 - Electric. Detroit, Michigan 2014. 123 (96) Los Angeles Department of Water and Power. Rate ordinance http://ezweb.ladwp.com/UserFiles/Rate Summ aries/Electric Rates Summary (effective July 1, 2017).pdf (accessed Oct 29, 2017). (97) Arizona Public Service Electric Company. Rate Schedule R-3: Resi dential Service Saver Choice Max. 2017. (98) Portland General Electric Comp any. Schedule 7: Re sidential Service. P.U.C. Oregon No. E-18. Portland, Oregon 2017. (99) U.S. Energy Information Administrati on. Electricity: End-Use Prices: Residential. Annual Energy Outlook 2018: With Projections to 2050. Washington, DC 2018. (100) DTE Energy. DTE Energy Advanced Impl ementation of Energy Storage Technologies https://www.smartgrid.gov/document/DTE_Energy_Advanced_Implementation_of_Energy_Storage_Final_TPR. (101) Debnath, U. K.; Ahmad, I.; Habibi, D. Gr idable Vehicles and Second Life Batteries for Generation Side Asset Manage ment in the Smart Grid. Int. J. Electr. Po wer Energy Syst. 2016, 82, 114Œ123. https://doi.org/10.1016/j.ijepes.2016.03.006. (102) Fu, R.; Feldman, D.; Margol is, R.; Woodhouse, M.; Ardani, K. U.S. Solar Photovoltaic System Cost Benchmark : Q1 2017 ; National Renewable Energy Laboratory: Golden, Colorado, 2017. (103) Lin, J.; Damato, G. Energy Storage Š a Cheaper and Cle aner Alternative to Natural Gas- Fired Peaker Plants Energy Storage; Berkeley, CA, 2011. (104) U.S Environmental Protection Agency. Electric Power Monthly with Data for August 2019; 2019. (105) U.S. Energy Information Administration. Capital Cost Estimates for Utility Scale Electricity Generating Plants; 2016. https://doi.org/10.2172/784669. (106) U.S. Energy Information Administration. Electric Power Monthly with Data for November 2017; 2017. (107) Bare, J. TRACI 2.0: The Tool for the Re duction and Assessment of Chemical and Other Environmental Impacts 2.0. Clean Technol. Environ. Policy 2011. https://doi.org/10.1007/s10098-010-0338-9. (108) PRé Sustainability. About SimaPro. (109) Wernet, G.; Bauer, C.; Steubing, B.; Rei nhard, J.; Moreno-Ruiz, E.; Weidema, B. The Ecoinvent Database Version 3 (P art I): Overview and Methodology. Int. J. Life Cycle Assess. 2016, 21 (9), 1218Œ1230. https://doi.org/10.1007/s11367-016-1087-8. (110) LTS. DATASMART LCI Packag e (US-EI SimaPro® Library) 124 https://ltsexperts.com/services/software/datasmart-life-cycle-inventory/ (accessed Nov 22, 2018). (111) Anctil, A.; Lee, E.; Lunt, R. R. Net Energy and Cost Benefit of Transparent Organic Solar Cells in Building-Integrated Applications. Appl. Energy 2020, 261 (December 2019), 114429. https://doi.org/10.1016/j.apenergy.2019.114429. (112) U.S. Environmental Protection Agency. Year 2016 Summary Tables EGRID. Washington, D. C. February 2018. (113) Kamath, D.; Shukla, S.; Arsenault, R.; K im, H. C.; Anctil, A. Evaluating the Cost and Carbon Footprint of Second-Life Electric Vehi cle Batteries in Residential and Utility- Level Applications. Waste Manag. 2020. https://doi.org/10.1016/j.wasman.2020.05.034. (114) Kamath, D.; Shukla, S.; Anctil, A. An Economic and Environmental Assessment of Residential Rooftop Photovoltaics with Second Life Batteries in the US. In Conference Record of the IEEE Photovoltaic Specialists Conference; 2019. (115) Shukla, S.; Kamath, D.; Arsenault, R.; K im, H. C.; Anctil, A. Economic and Environmental Benefit of Photovoltaics and Second Life Ba tteries for US Residential Microgrids. Prog. (116) Idaho National Laboratory. Plugged In: How Americans Char ge Their Electric Vehicles ; 2015. (117) Idaho National Laboratory. Electric Vehicle Charging Infrastructure Summary Report ; 2014. (118) U.S. Energy Information Administration. EIA-930 U.S. Electric System Operating Data ; 2019. (119) Uddin, M.; Romlie, M. F.; Abdullah, M. F.; Abd Halim, S.; Abu Bakar, A. H.; Chia Kwang, T. A Review on Peak Load Shaving Strategies. Renew. Sustain. Energy Rev. 2018, 82, 3323Œ3332. https://doi.org/10.1016/j.rser.2017.10.056. (120) Erol-Kantarci, M.; Mouftah, H. T. TOU- Aware Energy Management and Wireless Sensor Networks for Reducing Peak Load in Smart Grids. In 2010 IEEE 72nd Vehicular Technology Conference - Fall ; IEEE, 2010; pp 1Œ5. https://doi.org/10.1109/VETECF.2010.5594388. (121) U.S. Energy Information Administration. Annual Energy Outlook 2019: With Projections to 2050; U.S. Energy Information Administr ation: Washington, D. C., 2019. (122) Liu, Y.; Zhu, Y.; Cui, Y. Challenges and Opportunities towards Fast-Charging Battery Materials. Nat. Energy 2019, 4 (7), 540Œ550. https://doi.org/10.1038/s41560-019-0405-3. (123) Carrilero, I.; Ansean, D.; Viera, J. C.; Fernandez, Y.; Pereirinha, P. G.; Gonzalez, M. Impact of Fast-Charging and Regenerative Br aking in LiFePO4 Batteries for Electric Bus Applications. In 2017 IEEE Vehicle Power and Propulsion Conference (VPPC); IEEE, 125 2017; pp 1Œ6. https://doi.org/10.1109/VPPC.2017.8331013. (124) Anseán, D.; González, M.; Viera, J. C.; García, V. M.; Blanco, C.; Valledor, M. Fast Charging Technique for High Power Lithium Iron Phosphate Batteries: A Cycle Life Analysis. J. Power Sources 2013, 239, 9Œ15. https://doi.org/10.1016/j.jpowsour.2013.03.044. (125) Richa, K.; Babbitt, C. W.; Gaustad, G.; Wang, X. A Future Perspective on Lithium-Ion Battery Waste Flows from Electric Vehicles. Resour. Conserv. Recycl. 2014, 83 (2014), 63Œ76. https://doi.org/10.1016/j.resconrec.2013.11.008. (126) Alamerew, Y. A.; Brissaud, D. Mode lling Reverse Supply Chain through System Dynamics for Realizing the Transition towards the Circular Ec onomy: A Case Study on Electric Vehicle Batteries. J. Clean. Prod. 2020, 254, 120025. https://doi.org/10.1016/j.jclepro.2020.120025. (127) Wang, Y.; Huscroft, J. R.; Hazen, B. T.; Zhang, M. Green Information, Green Certification and Consumer Perceptions of Re manufactured Automobile Parts. Resour. Conserv. Recycl. 2018, 128, 187Œ196. https://doi.org/10.1016/j.resconrec.2016.07.015. (128) Zhao, S.; Zhu, Q.; Cui, L. A Decision- Making Model for Remanuf acturers: Considering Both Consumers™ Environmental Preference and the Government Subsidy Policy. Resour. Conserv. Recycl. 2018, 128, 176Œ186. https://doi.org/10.1016/j.resconrec.2016.07.005. (129) Grosjean, C.; Miranda, P. H.; Perrin, M.; Poggi, P. Assessment of World Lithium Resources and Consequences of Their Geographic Distribution on the Expected Development of the Elec tric Vehicle Industry. Renew. Sustain. Energy Rev. 2012, 16 (3), 1735Œ1744. https://doi.org/10.1016/j.rser.2011.11.023. (130) Kushnir, D.; Sandén, B. A. The Time D imension and Lithium Resource Constraints for Electric Vehicles. Resour. Policy 2012, 37 (1), 93Œ103. https://doi.org/10.1016/j.resourpol.2011.11.003. (131) Sverdrup, H. U. Modell ing Global Extraction, Supply, Price and Depletion of the Extractable Geological Resources with the LITHIUM Model. Resour. Conserv. Recycl. 2016, 114, 112Œ129. https://doi.org/10.1016/j.resconrec.2016.07.002. (132) Vikström, H.; Davidss on, S.; Höök, M. Lithium Availability and Future Production Outlooks. Appl. Energy 2013, 110, 252Œ266. https://doi.org/10.1016/j.apenergy.2013.04.005. (133) Miedema, J. H.; Moll, H. C. Lithium Ava ilability in the EU27 for Battery-Driven Vehicles: The Impact of Recycling and Substitution on the Confrontation between Supply and Demand Until2050. Resour. Policy 2013, 38 (2), 204Œ211. https://doi.org/10.1016/j.resourpol.2013.01.001. (134) Sato, F. E. K.; Nakata, T. Recoverability Analysis of Critical Materials from Electric Vehicle Lithium-Ion Batteries through a D ynamic Fleet-Based Approach for Japan. 126 Sustainability 2019, 12 (1), 147. https://doi.org/10.3390/su12010147. (135) Zeng, X.; Li, J. Implications for the Ca rrying Capacity of Lithium Reserve in China. Resour. Conserv. Recycl. 2013, 80 (1), 58Œ63. https://doi.org/10.1016/j.resconrec.2013.08.003. (136) Ziemann, S.; Weil, M.; Sc hebek, L. Tracing the Fate of LithiumŒŒThe Development of a Material Flow Model. Resour. Conserv. Recycl. 2012, 63, 26Œ34. https://doi.org/10.1016/j.resconrec.2012.04.002. (137) Ziemann, S.; Müller, D. B.; Schebek, L. ; Weil, M. Modeling the Potential Impact of Lithium Recycling from EV Batteries on Lithium Demand: A Dynamic MFA Approach. Resour. Conserv. Recycl. 2018, 133 (January), 76Œ85. https://doi.org/10.1016/j.resconrec.2018.01.031. (138) Rooney, M.; Nuttall, W. J.; Kazantzis, N. A Dynamic Model of the Global Uranium Market and the Nuclear Fuel Cycle. Resour. Policy 2015, 43, 50Œ60. https://doi.org/10.1016/j.resourpol.2014.11.003. (139) Aslani, A.; Helo, P.; N aaranoja, M. Role of Renewa ble Energy Policies in Energy Dependency in Finland: System Dynamics Approach. Appl. Energy 2014, 113, 758Œ765. https://doi.org/10.1016/j.apenergy.2013.08.015. (140) Feng, Y. Y.; Chen, S. Q.; Zhang, L. X. System Dynamics Modeling for Urban Energy Consumption and CO2 Emissions: A Case Study of Beijing, China. Ecol. Modell. 2013, 252 (1), 44Œ52. https://doi.org/10.1016/j.ecolmodel.2012.09.008. (141) Robalino-López, A.; Mena-Nieto, A.; Garc ía-Ramos, J. E. System Dynamics Modeling for Renewable Energy and CO2 Emissions: A Case Study of Ecuador. Energy Sustain. Dev. 2014, 20 (1), 11Œ20. https://doi.org/10.1016/j.esd.2014.02.001. (142) Stasinopoulos, P.; Compston, P.; Newell, B. ; Jones, H. M. A System Dynamics Approach in LCA to Account for Temporal EffectsŠ a Consequential Energy LCI of Car Body-in- Whites. Int. J. Life Cycle Assess. 2012, 17 (2), 199Œ207. https://doi.org/10.1007/s11367- 011-0344-0. (143) Trappey, A. J. C.; Trappey, C. V.; Hsia o, C.-T.; Ou, J. J. R.; Chang, C.-T. System Dynamics Modelling of Product Carbon Footprin t Life Cycles for Collaborative Green Supply Chains. Int. J. Comput. Integr. Manuf. 2012, 25 (10), 934Œ945. https://doi.org/10.1080/0951192X.2011.593304. (144) International Organi zation for Standardization. ISO 14040-Environmental Management - Life Cycle Assessment - Principles and Framework ; 2006; Vol. 3. https://doi.org/10.1016/j.ecolind.2011.01.007. (145) Das, D.; Dutta, P. A System Dynamics Framework for Integrated Reverse Supply Chain with Three Way Recovery and Product Exchange Policy. Comput. Ind. Eng. 2013, 66 (4), 720Œ733. https://doi.org/10.1016/j.cie.2013.09.016. 127 (146) Georgiadis, P.; Athanasi ou, E. The Impact of Two-Product Joint Lifecycles on Capacity Planning of Remanufacturing Networks. Eur. J. Oper. Res. 2010, 202 (2), 420Œ433. https://doi.org/10.1016/j.ejor.2009.05.022. (147) Georgiadis, P.; Besiou, M. Environm ental and Economical Sustainability of WEEE Closed-Loop Supply Chains with Recycling : A System Dynamics Analysis. 2010, 475Œ493. https://doi.org/10.1007/s00170-009-2362-7. (148) Georgiadis, P.; Besiou, M. Sustainability in Electrical and Elect ronic Equipment Closed- Loop Supply Chains: A System Dynamics Approach. 2008, 16, 1665Œ1678. https://doi.org/10.1016/j.jclepro.2008.04.019. (149) Wang, Y.; Hazen, B. T. Consumer Pr oduct Knowledge and Inte ntion to Purchase Remanufactured Products. Int. J. Prod. Econ. 2016, 181, 460Œ469. https://doi.org/10.1016/j.ijpe.2015.08.031. (150) Wassenhove, L. N. Van. Produc t Reuse Economics in Closed-Loop. 2008, 17 (5), 483Œ 496. https://doi.org/10.3401/poms.1080.0051. (151) Asif, F. M. A.; Bianchi, C.; Rashid, A. ; Nicolescu, C. M. Performance Analysis of the Closed Loop Supply Chain. 2012, 1Œ21. (152) Gu, X.; Ieromonachou, P.; Zhou, L.; Tseng, M. L. Developing Pricing Strategy to Optimise Total Profits in an Electric Vehi cle Battery Closed Loop Supply Chain. J. Clean. Prod. 2018, 203, 376Œ385. https://doi.org/10.1016/j.jclepro.2018.08.209. (153) He, Y.; Zheng, Y.; Wu, Y.; Tian, X.; Jin, X. Tracking Flows of Sec ondary Vehicle Batteries in China. Resour. Conserv. Recycl. 2019, 142 (November 2018), 34Œ36. https://doi.org/10.1016/j.resconrec.2018.11.016. (154) Zhang, L.; Liu, Y.; Pang, B.; Sun, B.; K okko, A. Second Use Value of China™s New Energy Vehicle Battery: A View Based on Multi-Scenario Simulation. Sustainability 2020, 12 (1), 341. https://doi.org/10.3390/su12010341. (155) Wang, X.; Gaustad, G.; Babbitt, C. W.; Baile y, C.; Ganter, M. J.; Landi, B. J. Economic and Environmental Characterization of an Evolving Li -Ion Battery Waste Stream. J. Environ. Manage. 2014, 135, 126Œ134. https://doi.org/10.1016/j.jenvman.2014.01.021. (156) Ahmadi, L.; Fowler, M.; Young, S. B.; Fras er, R. A.; Gaffney, B.; Walker, S. B. Energy Efficiency of Li-Ion Battery Packs Re-U sed in Stationary Power Applications. Sustain. Energy Technol. Assessments 2014, 8, 9Œ17. https://doi.org/10.1016/j.seta.2014.06.006. (157) Debnath, U. K.; Ahmad, I.; Habibi, D. Quantifying Econo mic Benefits of Second Life Batteries of Gridable Vehicles in the Smart Grid. Int. J. Electr. Power Energy Syst. 2014, 63, 577Œ587. https://doi.org/10.1016/j.ijepes.2014.05.077. (158) Han, X.; Liang, Y.; Ai, Y.; Li, J. Economic Evaluation of a PV Combined Energy Storage Charging Station Based on Cost Es timation of Second-Use Batteries. Energy 2018, 165, 128 326Œ339. https://doi.org/10.1016/j.energy.2018.09.022. (159) Martinez-Laserna, E.; Sarasketa-Zabala, E. ; Stroe, D.-I.; Swiercz ynski, M.; Warnecke, A.; Timmermans, J. M.; Goutam, S.; Rodriguez, P. Evaluation of Lithium-Ion Battery Second Life Performance and Degradation. In 2016 IEEE Energy Conversion Congress and Exposition (ECCE); IEEE, 2016; pp 1Œ7. https://doi.org/10.1109/ECCE.2016.7855090. (160) IEC. IEC 61427-1: Secondary Cells and Batteries fo r Renewable Energy Storage- General Requirements and Methods of Test; Geneva, Switzerland, 2013. (161) Ferreira, S. R.; Rose, D. M.; Schoenwal d, D. A.; Bray, K.; Conover, D.; Kintner-meyer, M.; Viswanathan, V. Protocol for Uniformly Measuri ng and Expressing the Performance of Energy Storage Systems; 2014. (162) Goldie-Scot, L. A Behind the Scen es Take on Lithium- ion Battery Prices https://about.bnef.com/blog/behind-scenes-take-lithium-ion-battery-prices/ (accessed Sep 1, 2019). (163) Moore, S.; Kamath, D.; Anctil, A. Pub lic Opinion and Anticipated Adoption of Second- Life, Repurposed Home Battery Systems. Prog. (164) Cole, W.; Gates, N.; Mai, T.; Greer, D.; Das, P. 2019 Standard Scenarios Report: A US Electric Sector Outlook; 2020; Vol. NREL/TP-6A. (165) Murray, D.; Glidewell, S. An Analysis of the Operational Costs of Trucking: 2019 Update ; 2019. (166) U.S. Bureau of Labor Statis tics. Employment Cost Index Summary https://www.bls.gov/news.release/eci.nr0.htm# :~:text=Compensation costs for private industry,increase was also 2.8 percent.&text= Compensation costs for state and,3.0 percent in March 2019. (accessed Jun 6, 2020). (167) Nelson, P. A.; Ahmed, S.; Gallagher, K. G.; Dees, D. W. Modeling the Performance and Cost of Lithium-Ion Batteries for Electric-Drive Vehicles, Third Edition ; Argonne, IL (United States), 2019. https://doi.org/10.2172/1503280. (168) Allen, F. L.; O™Neill, W.; Allen, F. L.; O™Neill, W. The Automobile Revolution ; 2019. https://doi.org/10.4324/9781351305327-8. (169) Bloomberg New Energy Finance. Lithium-Ion Battery Recycling: 2 Million Tons by 2030 ; 2019. (170) Canals Casals, L.; Amante García, B.; Cr emades, L. V. Electric Vehicle Battery Reuse: Preparing for a Second Life. J. Ind. Eng. Manag. 2017, 10 (2), 266. https://doi.org/10.3926/jiem.2009. (171) Li-Cycle. Li-Cycle: Opportunity for Closing the L oop in the Lithium-Ion Battery Bupply Chain; 2020. 129 (172) Olivetti, E. A.; Ceder, G.; Gaustad, G. G.; Fu, X. Lithium-Ion Battery Supply Chain Considerations: Analysis of Potential Bottlenecks in Critical Metals. Joule 2017, 1 (2), 229Œ243. https://doi.org/10.1016/j.joule.2017.08.019. (173) Kochar, A. The Future of Lith ium-Ion Batteries and Resource Recovery. Li-cycle blog . November 2018. (174) International Energy Agency. Global EV Outlook 2020: Entering the Decade of Electric Drive? Glob. EV Outlook 2020 2020, 273. (175) U.S. Geological Survey. Nickel; 2020. (176) U.S. Geological Survey. Manganese; 2020. (177) U.S. Geological Survey. Lithium; 2020. (178) U.S. Geological Survey. Copper; 2020. (179) U.S. Geological Survey. Aluminum; 2020. (180) U.S. Geological Survey. Iron and Steel Scrap in January 2020; 2020. (181) U.S. Geological Survey. Cobalt in January 2020; 2020. (182) Bare, J. Tool for Reduction and Asse ssment of Chemicals a nd Other Environmental Impacts (TRACI). TRACI 2.1. 2015. (183) Mohr, M.; Peters, J. F.; Ba umann, M.; Weil, M. Toward a Cell chemistry Specific Life Cycle Assessment of Lithium ion Battery Recycling Processes. J. Ind. Ecol. 2020, 24 (6), 1310Œ1322. https://doi.org/10.1111/jiec.13021. (184) Ecoinvent. Ecoinvent 3.6 Dataset Documentation. Ecoinvent 2019. (185) Ford, A. Modeling the Environment: An Introduc tion to System Dynamics Modeling of Environmental Systems ; Island Press: Washington D.C., 1999. (186) World Nuclear Association. Nu clear Power Reactors https://www.world- nuclear.org/information-library/nuclear-fuel-cycle/nuclear-power-reactors/nuclear- power-reactors.aspx (accessed Nov 10, 2019). (187) Brook, B. W.; Alonso, A.; Meneley, D. A.; Misak, J.; Blees, T.; van Erp, J. B. Why Nuclear Energy Is Sustainable and Has to Be Part of the Energy Mix. Sustain. Mater. Technol. 2014, 1Œ2, 8Œ16. https://doi.org/10.1016/j.susmat.2014.11.001. (188) Price, R.; Barthel, F.; Blaise, J.; Mcmurray, J. Forty Years of Uranium Resources , Production and Demand in Perspective; 2006. (189) Edenhofer, O.; Madruga, R. P.; Sokona, Y.; Seyboth, K.; Matschoss, P.; Kadner, S.; Zwickel, T.; Eickemeier, P.; Ha nsen, G.; Schlömer, S.; et al. Renewable Energy Sources 130 and Climate Change Mitigation: Special Repo rt of the Intergovernmental Panel on Climate Change; 2011. https://doi.org/10.1017/CBO9781139151153. (190) International Atomic Energy Agency. Energy, Electricity and Nuclear Power Estimates for the Period up to 2050; 2019. (191) U.S. Nuclear Regulatory Commis sion. In Situ Recovery Facilities https://www.nrc.gov/materials/uranium-recovery/extraction-methods/isl-recovery-facilities.html. (192) World Nuclear Association. In Situ L each (ISL) Mining of Uranium https://www.world- nuclear.org/information-library/nuclear-fuel-cycle/mining-of-uranium/in-situ-leach- mining-of-uranium.aspx (accessed Jun 6, 2020). (193) Mudd, G. M. The Future of Yellowcake: A Global Assessment of Uranium Resources and Mining. Sci. Total Environ. 2014, 472, 590Œ607. https://doi.org/10.1016/j.scitotenv.2013.11.070. (194) Parker, D. J.; McNaughton, C. S.; Sparks , G. A. Life Cycle Greenhouse Gas Emissions from Uranium Mining and Milling in Canada. Environ. Sci. Technol. 2016, 50 (17), 9746Œ 9753. https://doi.org/10.1021/acs.est.5b06072. (195) Haque, N.; Norgate, T. The Greenhouse Ga s Footprint of In-Situ Leaching of Uranium, Gold and Copper in Australia. J. Clean. Prod. 2014, 84 (1), 382Œ390. https://doi.org/10.1016/j.jclepro.2013.09.033. (196) Doka, G. Life Cycle Inventory of Generic Uranium In-Situ Leaching; 2011. (197) Farjana, S. H.; Huda, N. ; Mahmud, M. A. P.; Lang, C. Comparative Life-Cycle Assessment of Uranium Extraction Processes. J. Clean. Prod. 2018, 202, 666Œ683. https://doi.org/10.1016/j.jclepro.2018.08.105. (198) Mudd, G. M. Critical Review of Acid in Situ Leach Uranium Mining: 1. USA and Australia. Environ. Geol. 2001, 41 (3Œ4), 390Œ403. https://doi.org/10.1007/s002540100406. (199) Taylor, G.; Fa rrington, V.; Woods, P.; Ring, R.; Mollov, R. Review of Environmental Impacts of the Acid In-Situ Leach Uranium Mining Process; 2004. (200) Zhao, L.; Deng, J.; Xu, Y.; Zhang, C. Mineral Alteration and Pore-Plugging Caused by Acid in Situ Leaching: A Case Study of the Wuyier Uranium Deposit, Xinjiang, NW China. Arab. J. Geosci. 2018, 11 (22), 707. https://doi.org/10.1007/s12517-018-4064-7. (201) U.S. Department of Energy. Restoring America™s Competitive Nuclear Energy Advantage ; 2020. (202) Committee on Uranium Mining in Virginia. Uranium Mining in Virginia: Scientific, Technical, Environmental, Human Health and Safety, and Regulatory Aspects of Uranium 131 Mining and Processing in Virginia ; National Academies Press: Washington, D.C., 2012. https://doi.org/10.17226/13266. (203) IAEA. Uranium in Situ Leaching. In IAEA-TECDOC-; 1992. (204) Uranium One. Willow Creek: Mineral Resources and Mineral Reserves ; 2016. (205) World Nuclear Association. US uran ium mining and exploration https://www.world-nuclear.org/information-library/country-profiles/countries-t-z/us-uranium-mining.aspx (accessed May 28, 2020). (206) Arcadis. Willow Creek ISR Project -Christensen Ranch, Mine Units 2-6: Response to Comments from U.S. Nuclear Regulatory Comm ission Staff Technical Evaluation Report ; 2015. (207) U.S. Nuclear Regulatory Commission. Willow Creek Uraniu m Recovery Project https://www.nrc.gov/info-finder/materials/uranium/licensed-facilities/is-christensen-ranch.pdf (accessed Mar 15, 2020). (208) Tweeton, D.; Peterson, K. Uranium Leaching Lixiviants. Bureau of Mines . 1981, pp 1Œ35. (209) Mudd, G. Uranium Mi ning & CO2 Accounting. In AusIMM Uranium Conference, Perth, Western Australia; 2011. (210) van Leeuwen, J. W. S. Energy from Uranium; 2006. (211) U.S. Nuclear Regulatory Commission. In-S itu Leach Uranium Recovery and Alternatives. In Generic Environmental Impact Statement for In-Situ Leach Uranium Milling Facilities (NUREG-1910); 2009; Vol. 1. (212) Graves, D. H.; Cutler, S. E. Preliminary Economic Assessment of the Lost Creek Property, Sweetwater County, Wyoming; 2013. (213) Carmen, C.; Kunin, R. Recovery of Ur anium from Carbonate Leach Liquors Using Weak- Acid Cation Exchange Resins. React. Polym. 1986, 4, 77Œ89. (214) U.S. Department of Transportation; Bureau of Transportati on Statistics and U.S. Department of Commerce; Census Bureau. Value, Tons, and Ton-Miles of Freight Shipments within the U.S. by Domestic Es tablishments https ://www.census.gov/programs-surveys/cfs.html (accessed Jun 5, 2019). (215) Geffen, C. A. An Analysis of the Risk of Transporti ng Uranium Ore Concentrates by Truck ; 1981. (216) Dones, R. Ecoinvent 3.5 Dataset Docume ntation: Nuclear Electri city Production BWR. 1Œ 7. (217) Warner, E. S.; Heath, G. A. Life Cycle Greenhouse Gas Emissions of Nuclear Electricity 132 Generation: Systematic Review and Harmonization. J. Ind. Ecol. 2012, 16 (SUPPL.1), 73Œ 92. https://doi.org/10.1111/j.1530-9290.2012.00472.x. (218) World Nuclear Association. Operable Reactors http://www.world- nuclear.org/information-library/facts-and-figures/reactor-database.aspx. (219) World Nuclear Association. The Nuclear Fuel Cycle http://www.world- nuclear.org/information-library/nuclear-fu el-cycle/introduction/nuclear-fuel-cycle- overview.aspx. (220) Laurent, G.; Izart, C.; Le chenard, B.; Golfier, F.; Marion, P.; Collon, P.; Truche, L.; Royer, J.-J.; Filippov, L. Numerical Modelling of Column Experiments to Investigate In-Situ Bioleaching as an Alternative Mining Technology. Hydrometallurgy 2019, 188 (July), 272Œ290. https://doi.org/10.1016/j.hydromet.2019.07.002. (221) Kaksonen, A. H.; Deng, X.; Bohu, T.; Zea, L.; Khaleque, H. N.; Gumulya, Y.; Boxall, N. J.; Morris, C.; Cheng, K. Y. Prospec tive Directions for Biohydrometallurgy. Hydrometallurgy 2020, 195 (May), 105376. https://doi.org/10.1016/j.hydromet.2020.105376. (222) Willem, J.; Leeuwen, S. Van. Uranium Mining & Milling. Ecologist 2006, 36 (5), 044. https://doi.org/10.2307/3928149. (223) OECD Nuclear Energy Agency; In ternational Atomic Energy Agency. Uranium 1988: Resources, Production and Demand; Paris, 1988. (224) OECD Nuclear Energy Agency; In ternational Atomic Energy Agency. Uranium 2007 : Resources , Production and Demand Nl Ea D; 2008. (225) OECD Nuclear Energy Agency; In ternational Atomic Energy Agency. Uranium 2011: Resources, Production and Demand; 2012. (226) Federal Ministry of Agriculture Regions and Tourism (BMLRT) of Austria. Data Section https://www.world-mining-data.info/?World_Mining_Data (accessed Feb 2, 2020). (227) Li, J.; Wang, Y.; Tan, X. Research on th e Classification Method for the Secondary Uses of Retired Lithium-Ion Traction Batteries. Energy Procedia 2017, 105, 2843Œ2849. https://doi.org/10.1016/j.egypro.2017.03.625. (228) ANSI/CAN/UL 1974: Standard for Evaluation for Repurposing Batteries. 2018, No. ANSI/CAN/UL 1974, 3Œ4. (229) Wang, X.; Gaustad, G.; Babbitt, C. W. Targeting High Value Metals in Lithium-Ion Battery Recycling via Shredding and Size-Based Separation. Waste Manag. 2016, 51, 204Œ213. https://doi.org/10.1016/j.wasman.2015.10.026. (230) D™Adamo, I.; Gast aldi, M.; Rosa, P. Recycling of E nd-of-Life Vehicles: Assessing Trends and Performances in Europe. Technol. Forecast. Soc. Change 2020, 152 (November 2019), 133 119887. https://doi.org/10.1016/j.techfore.2019.119887. (231) Garrido-Hidalgo, C.; Ramirez, F. J.; O livares, T.; Roda-Sanchez, L. The Adoption of Internet of Things in a Circular Supply Ch ain Framework for the Recovery of WEEE: The Case of Lithium-Ion Electric Vehicle Battery Packs. Waste Manag. 2020, 103, 32Œ44. https://doi.org/10.1016/j.wasman.2019.09.045. (232) Nassar, E.; Tokimatsu, K.; Aziz, M. Potent ial Distributions of Electric Vehicle Secondary Used Batteries for Frequency Regulation in Europe. In Energy Procedia ; Elsevier B.V., 2019; Vol. 159, pp 394Œ399. https://doi.org/10.1016/j.egypro.2018.12.070. (233) Ilankoon, I. M. S. K.; Tang, Y.; Ghorbani , Y.; Northey, S.; Yellishetty, M.; Deng, X.; McBride, D. The Current State and Future Directions of Percolation Leaching in the Chinese Mining Industry: Chal lenges and Opportunities. Miner. Eng. 2018, 125 (June), 206Œ222. https://doi.org/10.1016/j.mineng.2018.06.006. (234) EPA. eGRID https://www.epa .gov/energy? resources/egrid/index.html. (235) Jungbluth, N.; Stucki, M.; Frisch knecht, R.; Büsser, S. Photovoltaics. Authors Updat. Autor Überarbeitung Niels Jungbluth, ESU-serv ices Autorin Überarbeitung Lucia Ciseri Autoren Bearb. 2010. (236) Bureau of Transporta tiom Statistics. 2012 Commodity Flow Survey Final Estimates https://www.rita.dot.gov/bts/press_releases/bts056_14. (237) Battery University. T ypes of Lithium-ion Batteries https://batteryuniversity.com/learn/article/t ypes_of_lithium_ion (accessed Sep 9, 2020). (238) Cameco Corporation. Cameco 2015 Annual Report; 2016. (239) Lenntech. Dowex Ion Exchange Resins: Fundamentals of Ion Exchange .