Office of Technology, Policy, and Strategy Space-Based Solar Power Erica Rodgers, Ellen Gertsen, Jordan Sotudeh, Carie Mullins, Amanda Hernandez, Hanh Nguyen Le, Phil Smith, and Nikolai Joseph Reviewer(s): A.C. Charania, Tom Colvin, and Roger Meyers January 11, 2024 Report ID 20230018600 NASA Headquarters 300 E Street SW Washington, DC 20024 This report is intended for informational purposes only, and does not indicate a commitment or intention, implied or otherwise, by the government to engage in any activity or to enter into any agreement, contract or other obligation. The inclusion of information in this report does not constitute endorsement of any entity, or any products, services, technologies, activities, or Agency policy. The information contained in this report reflects solely the views and opinions of the authors.
For more information on the NASA Office of Technology, Policy, and Strategy to view this and other reports visit https://www.nasa.gov/offices/otps/home/index.html Executive Summary Space-Based Solar Power Purpose of the Study This study evaluates the potential benefits, challenges, and options for NASA to engage with growing global interest in space-based solar power (SBSP). Utilizing SBSP entails in-space collection of solar energy, transmission of that energy to one or more stations on Earth, conversion to electricity, and delivery to the grid or to batteries for storage. Experts in both the aerospace and energy sectors are debating the benefits of SBSP as more organizations globally begin SBSP technology development programs. Proponents claim SBSP could deliver large amounts of electricity at competitive prices and with fewer greenhouse gas (GHG) emissions than terrestrial renewable electricity technologies while accelerating development of the space economy. Skeptics say SBSP has no clear development path and would divert billions of dollars from known terrestrial solutions while damaging the environment. While it is generally understood that SBSP is cost prohibitive and technically infeasible today, this study assesses operating SBSP systems in 2050. Part of NASA’s mission is to innovate for the benefit of humanity – it is through this lens that the Agency weighs whether and how to support SBSP development. The study addresses the following questions: • Under what conditions would SBSP be a competitive option to achieving net zero GHG emissions compared to alternatives? • If SBSP can be competitive, what role, if any, could NASA have in its development? To answer these questions, we spoke with more than 30 stakeholders and subject matter experts across the aerospace and energy sectors, reviewed over 100 documents relating to SBSP, developed a model to characterize and estimate the costs and GHG emissions of SBSP under varying technological and economic conditions, and qualitatively assessed challenges to SBSP development. Using these data sources, we: 1. Generated first-order lifecycle cost and emissions estimates for first-of-a-kind, utility-scale SBSP and compared those with current renewable electricity production technologies, 2. Conducted sensitivity analyses to assess whether a competitive SBSP solution is feasible, 3. Conducted qualitative assessments of challenges, opportunities, and NASA’s role, 4. Discussed options for NASA’s engagement, and 5. Made follow-on study recommendations. January 11, 2024 Report ID 20230018600
ii Key Findings Question 1: Under what conditions would SBSP become competitive? System Designs We assessed two representative SBSP designs: Innovative Heliostat Swarm (Representative Design One, RD1) and Mature Planar Array (Representative Design Two, RD2), based on existing concepts. The SBSP designs serve simply as point designs for assessment purposes and should not be viewed as endorsements to or by NASA. RD11 and RD22 are broadly derived from historical, publicly available designs that include recent updates and provide enough data from which to perform a first-order analysis of this kind. RD1 generates power 99% of the year and collects solar radiation by autonomously redirecting its reflectors toward a concentrator to focus sunlight throughout each day. RD2 uses flat panels, with solar cells facing away from Earth and microwave emitters facing toward the Earth. RD2 generates power 60% of the year due to its limited capability to reposition itself or redirect solar radiation toward its solar cells. Each SBSP design is normalized to deliver 2 gigawatts (GW) of power to the electric grid to be comparable to very large terrestrial solar power plants operating today.3 Therefore, five RD2 systems are needed to deliver roughly the same amount of power as one RD1 system. The functional representation of each design is illustrated in Figure 1. Each SBSP design’s size (which is dominated by the area of its solar panels) and mass is significant. To provide context, consider two examples of space systems with significant mass and solar panel area: an aggregated mass, the International Space Station (ISS); and a distributed mass, a constellation of 4,000 Starlink v2.0 satellites4. The solar panel area is 11.5km2 for RD1 and 19km2 for RD2. The RD1 solar panel area is more than 3,000 times and 27 times greater than that of the ISS and Starlink constellation, respectively. The mass is 5.9Mkg for RD1 and 10Mkg for RD2. The RD1 1 John C. Mankins “SPS-Alpha Mark-III and an Achievable Roadmap to Space Solar Power,” 72nd International Astronautical Congress, October 15, 2021. 2 Susumu Sasaki et al. “A new concept of solar power satellite: Tethered-SPS” Acta Astronautica 60 (2006) 153-165 and Pellegrino et al. "A lightweight space-based solar power generation and transmission satellite." (2022) https://doi.org/10.48550/arXiv.2206.08373. 3 Voiland, Adam, “Soaking up Sun in the Thar Desert,” NASA Earth Observatory, January 26, 2022. https://earthobservatory.nasa.gov/images/149442/soaking-up-sun-in-the-thar-desert 4 The >4,000 Starlink satellites in orbit today are smaller than the v2 and include 4 different configurations, but offer us an example of the kind of upmass that is already approved for this and other existing satellite constellations. Other large constellations are comparable, but of existing constellations, Starlink has already delivered the most mass into orbit. Assuming a mass of 1250kg and solar array area of 105 m2 per Starlink v2 satellite. These systems were chosen because at the time of this report’s publication they represent the most massive single monolithic system in Earth orbit (ISS), and the most massive single distributed system (Starlink constellation).
iii mass is more than 14 times greater and 1.2 times greater than that of the ISS and Starlink constellation, respectively. This study assessed lifecycle cost and emissions based on the following scenario: SBSP systems are developed on the ground in the 2030s and launched to low-Earth orbit (LEO), and then transferred to and assembled in geostationary orbit (GEO) in the 2040s. The SBSP systems are operated in GEO from 2050-2080, by transmitting energy to one or more stations on Earth. Maintenance, which entails developing, launching, and assembling new spacecraft modules, occurs between 2060-2080. SBSP systems disposal operations, which entail developing and launching debris removal spacecraft to GEO to transfer spacecraft modules to a graveyard orbit, occurs between 2060-2085. Lifecycle Calculations We developed a model to calculate the cost and GHG emissions for all aspects of the SBSP reference designs across the full lifecycle of development, assembly, operation, maintenance, and disposal. Including disposal or decommissioning of a system is a best practice when assessing its full lifecycle. The calculated lifecycle cost and GHG emissions are for first-of-a-kind systems delivering 2 GW of power to the electric power grid beginning in 2050. At the end of 2022, according to the Energy Information Administration (EIA), the United States had 1,160 GW of total utility-scale electricity-generation capacity.5 We calculated the lifecycle cost of electricity and lifecycle GHG emissions intensity for each representative SBSP design using common industry expressions: levelized cost of electricity (LCOE) and Economic Input Output – Life Cycle Analysis (EIO-LCA). The LCOE is the average cost of electricity over a generator’s lifetime and is a mainstay of energy sector analyses. The EIO-LCA is an established methodology for estimating first-order emissions intensity of economic activity. LCOE has several limitations. For example, it does not consider the variable value of energy at different locations or times.6 EIO-LCA also has limitations. This methodology often uses spend-based metrics to estimate emissions, which assumes a relationship between cost, efficiency, and emissions that may not always align with direct measurements of emissions by economic activity. All cost estimates are measured in Fiscal Year 2022 (FY22) dollars. EIO-LCA uses measured GHG emissions of producing goods and services by mass (like kilograms of steel), area (like square meters of solar cells), or cost (like dollars spent on services) by aggregating macroeconomic data. We then compared the LCOE and lifecycle GHG emission intensity (EIO-LCA) to alternative terrestrial renewable electricity production technologies using data from the National Renewable Energy 5 EIA, “Electricity explained,” last updated: June 30, 2023 https://www.eia.gov/energyexplained/electricity/electricity-in-the-us-generation-capacity-and-sales.php. 6 For more information on metrics, see NREL, “Competitiveness Metrics for Electricity System Technologies” 2021, https://www.nrel.gov/docs/fy21osti/72549.pdf.
iv Laboratory (NREL) to assess if the representative SBSP designs are competitive. We compared to NREL’s 2050 cost projections and NREL’s 2021 GHG emissions for nuclear fission, geothermal, hydroelectric, utility-scale solar photovoltaics with storage, and land wind without storage. We use 2021 emissions data because there are no projections for this data. We include land wind without storage for comparison because it has the lowest cost and lowest emission intensity of all electricity production technologies tracked by NREL.
v Figure 1. Functional Decomposition of SBSP Design Reference Systems. [Left] RD1. [Right] RD2.
vi Baseline Assessment We made assumptions across the full lifecycle of development, assembly, operation, maintenance, and disposal to calculate the cost and GHG emissions of first-of-a-kind SBSP designs. The study’s baseline assessment and sensitivity analyses (Table 1) incorporate three categories of assumptions regarding space capabilities: 1) beyond assumes certain capabilities will be available by 2050, 2) comparable uses today’s capabilities as a starting point; and 3) below covers the possibility that an existing capability does not perform to previously demonstrated levels when used in a novel SBSP system. We do not include novel architectures or recent advances in material science that may alter the specifications of a 2 GW SBSP system. These assumptions do not represent NASA’s position on the future aerospace industry and serve only as an analytical platform. Table 1. Key Input Parameters for Multiple Variable Sensitivity Analysis and Baseline Analysis. Green triangles pointing upward indicate an assumption beyond what has been achieved to date, yellow bars are achievable today, and red triangles pointing downward are below today’s capability (these are assumed given the first-of-a-kind nature of the SBSP systems studied).
vii Beyond: We assume costs to launch a Starship7 and reuse each Starship, along with operations costs, are lower in 2050 than today. This is in part because autonomous capabilities are assumed for the representative SBSP designs. Comparable: We assume solar cell efficiency at the current state of the practice for GEO satellites because technological advances are difficult to predict beyond a few years. We assume an orbital transfer method leveraging refueling launches to reach GEO at the current state of the practice.8 We assume manufacturing curves and initial hardware costs at approximately the current state of the practice as a “starting point” for learning over the multi-decade manufacturing process. Manufacturing curves were selected based on analogous industries with similar production levels. Below: We assume a hardware lifetime below that of the current state of the practice for GEO hardware because the SBSP designs are first-of-a-kind systems requiring multiple refurbishment cycles. Our study found the following: The baseline lifecycle cost of electricity for RD1 is 0.61 $/kWh and for RD2 is 1.59 $/kWh. Launch is the largest cost driver (71% of RD1 and 77% of RD2) as 2,3216 launches are required to deliver 5.9Mkg of mass for RD1 and 3,960 are needed to deliver 10Mkg of mass for RD2. Most of these launches (12 of every 13) serve only to refuel payloads in LEO for transfer to GEO. Manufacturing is the second largest cost driver (22% for RD1 and 18% RD2) and includes initial spacecraft hardware development and manufacturing. Learning curves enable cost decreases over time as experience is gained through producing 1.5M spacecraft modules for RD1 and 2M spacecraft modules for RD2. In general, RD2 is more expensive than RD1 because more mass is involved; five RD2 systems are needed to generate roughly the same amount of power as one RD1 system. Figure 2 shows the comparison of the lifecycle baseline assessment to terrestrial renewable electricity production technologies, whose costs range from 0.02-0.05 $/kWh. The RD1 LCOE and RD2 LCOE are 12-31 and 32-80 times higher, respectively, than the 2050 projections for terrestrial alternatives. Therefore, our baseline analysis of SBSP designs does not return cost competitive results relative to terrestrial alternatives. For comparison, the average energy cost of a U.S. household in August 2022 was 0.167 $/kWh.9 7 Due to the size and mass of the representative SBSP designs, for purposes of this study, we used available data from Space Exploration Technologies Corporation’s (SpaceX’s) Starship launch vehicle. because at this time it is anticipated to be the largest super heavy launch vehicle with data available. It is important to note that multiple specifications for this vehicle are planned, and cover a range of payload capacities, fuel capacities, and more. The study’s use of data from Starship does not indicate any endorsement by NASA. 8 Blue Origin Fed’n, LLC; Dynetics, Inc.-A Leidos Co., B-419783 et al., July 30, 2021, 2021 CPD ¶ 265 at 27 n.13 9 U.S. Bureau of Labor Statistics (2023), Average energy prices for the United States, https://www.bls.gov/regions/midwest/data/averageenergyprices_selectedareas_table.htm.
viii The baseline lifecycle GHG emissions intensity for RD1 is 26 gCO2eq./kWh and for RD2 is 40 gCO2eq./kWh. For comparison, the U.S. electric grid in 2021 produced an average of 385 gCO2/kWh. 10 Launch is the largest driver and leads to 64% and 72% of the GHG emissions for RD1 and RD2, respectively. GHG emissions intensity for both RD1 and RD2 fall within the range of GHG emissions intensities (13-43 gCO2eq./kWh) for terrestrial renewable electricity production technologies. For comparison, the GHG emissions intensities of coal and natural gas are 486 gCO2eq./kWh and 1001 gCO2eq./kWh, respectively. 11 RD1 and RD2 emissions intensities do not include upper atmosphere effects of launch emissions, which are assumed to be worse than producing the same emissions on the surface of the Earth, and still under study by NASA and the academic community.12 Our baseline analysis indicates our SBSP designs may have similar lifecycle GHG emissions intensities to those of terrestrial alternatives, pending further studies launch emission effects in the upper atmosphere. Sensitivity Analyses We conducted sensitivity analyses on the assumptions that drive the lifecycle cost and GHG emissions intensity to evaluate what conditions could allow RD1 and RD2 to be cost competitive (Figure 3). We varied the following input parameters one at a time to assess their individual impact on lifecycle cost and emissions: launch costs, first unit manufacturing costs, manufacturing learning curves, hardware lifetime, solar cell efficiency, and orbital transfer methods. Lower launch costs or use of electric propulsion to transfer mass from LEO to GEO each resulted in the most significant reduction of LCOE to about 0.20 $/kWh for RD1 and to about 0.50 $/kWh for RD2. This decrease is not enough to make the representative designs cost competitive with terrestrial alternatives. Cost competitiveness can be achieved by varying multiple assumptions (Table 1) at the same time to provide a combination of cost and capability improvements beyond the advances already assumed in the baseline assessment. This favorable combination reduces the LCOE to 0.03 $/kWh for RD1 and 0.08 $/kWh for RD2, figures that are competitive with terrestrial alternatives. This combination also reduces the GHG emissions intensities (3.78 gCO2eq./kWh for RD1 and 4.33 gCO2eq./kWh for RD2) to values less than nuclear and wind-without-storage technologies. 10 EIA, (2022, November 25), How much carbon dioxide is produced per kilowatthour of U.S. electricity generation? https://www.eia.gov/tools/faqs/faq.php?id=74&t=11. 11 Ibid. 12 National Oceanic and Atmospheric Administration (2022, June 22), Projected increase in space travel may damage ozone layer, https://research.noaa.gov/2022/06/21/projected-increase-in-space-travel-may-damageozone-layer/.
ix The following combination of revised assumptions yields SBSP solutions that are cost competitive with terrestrial alternatives, with lower GHG emissions: • lower launch cost: $50M per launch, or $500/kg; $425/kg with 15% block discount • electric propulsion orbital transfer from LEO to GEO • extended hardware lifetimes: 15 years • cheaper servicer and debris removal vehicles: $100M and $50M, respectively • efficient manufacturing at scale: learning curves of 85% and below Our sensitivity analyses highlight the need for advances across a wide range of SBSP enabling capabilities.
x Figure 2. Comparison of SBSP Systems Cost ($FY22) and GHG Emissions Baseline Assessment with Terrestrial Alternatives
xi Figure 3. SBSP Systems Cost ($FY22) and GHG Emissions Reduction Sensitivities Results
xii Question 2: What role, if any, should NASA have? Question 1 provided a model for understanding the biggest cost drivers for SBSP: launch and manufacturing. To understand NASA’s potential role, the study qualitatively assessed challenges and opportunities for SBSP development. We reviewed technological, regulatory, and policy challenges, as well as technological and economic opportunities. The review found that SBSP enabling technologies have broad applicability to a wide suite of future NASA mission needs, from power beaming on the Moon, to autonomous operations for science and human exploration, to lightweight materials. NASA currently funds research and development activities in each of these areas, though some areas receive significantly more funding: In-space servicing, assembly and manufacturing received ~$280M in FY22, autonomy received ~$244M in FY22, while wireless power transmission investments are today limited to concept studies (<$1M).13 This study provides two main categories of options for NASA leadership to consider without making a specific recommendation: 1. Undirected organic development: NASA is working on almost all the enabling technologies for SBSP and may want to focus only on its own current and planned missions’ needs, limiting further involvement, upon request, to supporting U.S. organizations pursuing SBSP and maintaining awareness of SBSP advances around world. NASA could fund these areas without adding SBSP as a separate line item in its budget. That said, further study of potential benefits of SBSP to planned missions is warranted. 2. Pursue partnership opportunities to advance SBSP: NASA may find mutually beneficial returns from supporting external SBSP development given the relevance of enabling technologies to other agency missions. Moreover, these technologies, from autonomous operations to wireless power transmission, have many use-cases beyond NASA missions, and are being pursued by a broad set of public and private actors for many non-SBSP applications. This study also provides follow-on study recommendations regardless of option choice, including: 1. Building on the first order analysis, study cutting-edge SBSP systems using the most rigorous lifecycle emissions and cost assessments as performed by NREL on other electricity production technologies. 2. Perform a technical design trade evaluation of SBSP technologies for NASA mission applications, such as energy infrastructure on the Moon. 13 Brandon, E. (2019, April 10). Power Beaming for Long Life Venus Surface Missions. Retrieved June 2023, from https://www.nasa.gov/directorates/spacetech/niac/2019_Phase_I_Phase_II/Power_Beaming/ and Lubin, P. (2021, April 2). Moonbeam-Beamed Lunar Power. Retrieved from https://www.nasa.gov/directorates/spacetech/strg/lustr/2020/Moonbeam_Beamed_Lunar_Power/.
xiii Conclusion We performed a first-order lifecycle study of two representative SBSP designs for 2 GW utility-scale power generation that are presumed to begin operating in 2050 to determine 1) the conditions under which SBSP would be a competitive option to achieving net zero GHG emissions; and 2) assuming SBSP can be competitive, the role, if any, NASA could play in its development. We assumed baseline capabilities to develop, assemble, operate, maintain, and dispose of the SBSP systems are a mix of capabilities that are above, below, or comparable to capabilities demonstrated to date. We then compared the LCOE and lifecycle GHG emission intensity of the SBSP designs to terrestrial renewable electricity production technologies. Our findings indicate the SBSP designs may produce lifecycle GHG emissions per unit of electricity that are comparable to terrestrial alternatives, pending further studies of upper atmosphere effect of launch emissions. We find the SBSP designs are more expensive than terrestrial alternatives and may have lifecycle costs per unit of electricity that are 12-80 times higher. However, cost competitiveness may be achieved through a favorable combination of cost and performance improvements related to launch and manufacturing beyond the advancements assumed in the baseline assessment. NASA is developing technologies and capabilities to meet its future mission needs, such as in-space servicing, assembly, and manufacturing (ISAM) and autonomy, which are enablers for SBSP. NASA could maintain its focus on core Agency missions and technologies, while documenting their relevance to SBSP. NASA may also enhance coordination with U.S. and international partners on technology development with relevance to SBSP. We recommend regular reviews of global SBSP developments and focused analyses of SBSP designs that may enable NASA’s core missions.
xiv Table of Contents Executive Summary............................................................................................................. i Purpose of the Study...................................................................................................................... i Key Findings...................................................................................................................................ii Question 1: Under what conditions would SBSP become competitive? ................................................... ii Question 2: What role, if any, should NASA have? .................................................................................... xii Conclusion .................................................................................................................................. xiii Table of Contents ............................................................................................................ xiv 1.0 Introduction................................................................................................................... 1 1.1 Background ..............................................................................................................................1 1.2 Study Questions .......................................................................................................................5 2.0 Methodology Overview................................................................................................. 5 2.1 Cost Estimations.......................................................................................................................6 2.1.1 Functional Decomposition of SBSP Systems ...............................................................................................6 2.1.2 Concept of Operations ................................................................................................................................6 2.1.3 Levelized Cost of Electricity.......................................................................................................................10 2.2 GHG Emissions Intensity .........................................................................................................12 3.0 Results......................................................................................................................... 13 3.1 Summary of SBSP System Costs and GHG Emissions ................................................................17 4.0 Sensitivity Analyses ................................................................................................... 18 4.1 Launch ...................................................................................................................................18 4.1.1 Direct Launch to GEO ................................................................................................................................18 4.1.2 Reduced Launch Costs...............................................................................................................................18 4.1.3 Electric Propulsion Orbital Transfer ..........................................................................................................19 4.1.4 Spacecraft Hardware Life ..........................................................................................................................20 4.2 Manufacturing .......................................................................................................................20 4.2.1 Initial Hardware Costs ...............................................................................................................................20 4.2.2 Learning Curve ..........................................................................................................................................21
xv 4.2.3 Solar Cell Efficiency ...................................................................................................................................21 4.3 Combining Sensitivities...........................................................................................................21 4.4 Making SBSP Systems Competitive with Terrestrial Renewables ..............................................22 5.0 Challenges and Opportunities ................................................................................... 24 5.1 Challenges to Operational System Development .....................................................................24 5.1.1 Large-scale ISAM Capability Challenges ....................................................................................................24 5.1.2 Large-scale Autonomous Distributed Systems..........................................................................................25 5.1.3 Power Beaming .........................................................................................................................................25 5.2 Challenges to Reducing System Costs ......................................................................................25 5.2.1 Launch costs ..............................................................................................................................................25 5.2.3 Manufacturing at scale..............................................................................................................................26 5.2.4 Launch cadence.........................................................................................................................................26 5.3 Regulatory and Other Challenges ............................................................................................27 5.3.1 Active Debris Removal ..............................................................................................................................27 5.3.2 Spectrum Allocation ..................................................................................................................................27 5.3.3 Orbital Slot Allocation ...............................................................................................................................27 5.3.4 Security......................................................................................................................................................27 5.4 Ongoing Improvements to SBSP Technology Needs .................................................................28 5.4.1 ISAM ..........................................................................................................................................................28 5.4.2 Autonomous Distributed Systems.............................................................................................................28 5.4.3 Power Beaming .........................................................................................................................................28 5.5 Ongoing Improvements to SBSP Economic Needs ....................................................................29 5.5.1 Electric Propulsion Orbital Transfer ..........................................................................................................29 5.5.2 Alternative Launch ....................................................................................................................................30 5.5.3 Mass Manufacturing .................................................................................................................................30 5.5.4 Advanced Materials ..................................................................................................................................30 5.6 Architecture Opportunities .....................................................................................................30 6.0 Options for NASA to Consider................................................................................... 30 6.1 Option 1: Undirected and Organic Development .....................................................................31 6.2 Option 2: Pursue Partnership Options to Advance SBSP...........................................................31
xvi 7.0 Conclusion and Recommended Further Study ........................................................ 32 Appendix A: Representative Design Details................................................................... 35 Appendix B: Methodology ............................................................................................... 37 Overview .....................................................................................................................................38 Approach to Cost Calculations ......................................................................................................40 Functional Decomposition of SBSP Systems ......................................................................................................40 ConOps Phases ...................................................................................................................................................41 Levelized Cost of Electricity................................................................................................................................55 Results ................................................................................................................................................................59 Approach to GHG Emissions Calculations ......................................................................................62 GHG Emissions ...................................................................................................................................................62 Climate Comparisons .........................................................................................................................................72 Sensitivity Analyses......................................................................................................................72 Launch ................................................................................................................................................................73 Manufacturing....................................................................................................................................................74 Solar Cell Efficiency ............................................................................................................................................75 Multiple Variable Sensitivity Analyses ...............................................................................................................75 Appendix C: Acknowledgements .................................................................................... 78 Appendix D: Acronyms & Key Terms.............................................................................. 79 References ........................................................................................................................ 85
1 1.0 Introduction This report describes for NASA senior-level consideration the relative costs and greenhouse gas (GHG) emissions of space-based solar power (SBSP) systems to assess whether SBSP is a feasible option for achieving net-zero GHG emissions compared to alternative renewable sources of electricity production. Our assessment considered two reference SBSP system designs operating in geostationary orbit (GEO) – the lower technology readiness level (TRL) Innovative Heliostat Swarm, (hereafter referred to as Representative Design One, or RD1) and the higher TRL Mature Planar Array (Representative Design Two, or RD2) – and compared costs for their development, assembly, operation, maintenance, and disposal. We also compared the relative GHG emissions of each system by conducting material decompositions for an Economic Input Output – Life Cycle Assessment (EIO-LCA). 1.1 Background In response to climate change, organizations around the world are pursuing a range of policies called net zero. According to the United Nations (UN), “net zero means cutting greenhouse gas emissions to as close to zero as possible, with any remaining emissions re-absorbed from the atmosphere, by oceans and forests for instance.” There is growing U.S. and international policy and legislation on net zero. As of 2021 over 70 countries had set net-zero targets (United Nations, 2023). The U.S. submitted a long-term strategy to the UN in November 2021, officially committing to net zero emissions by 2050 at the latest (United States Department of State and the United States Executive Office of the President, 2021). The electric power sector accounted for 25% of U.S. GHG emissions in 2020, according to the U.S. Environmental Protection Agency (EPA), as shown in Figure 2 (EPA, 2023). The sector encompasses the generation, transmission, and distribution of electricity. Carbon dioxide (CO2) makes up 80% of GHG from the U.S. electricity sector. According to the U.S. Energy Information Administration (EIA), CO2 emissions by the U.S. electric power sector in 2021 were about 1,545 million metric tons (MMmt), or about 31% of the 4,970 MMmt of total U.S. energy-related CO2- emissions (EIA, 2023). These emissions primarily result from electricity generation using coal and natural gas, which are non-renewable energy sources (see Figure 2 inset). In 2021, 40% of U.S. electricity production came from renewable and nuclear sources as shown in Figure 3. The International Energy Agency estimates that to reach net-zero, the world will need to reduce its use of fossil fuels from 80% of the total today to slightly over 20% by 2050 (Bouckaert, et al., 2021). However, the EIA projects that by 2050, 44% of U.S. electricity will still come from fossil fuels (EIA, 2022).
2 Figure 3. Share of U.S. Greenhouse Gas Emission by Economic Sector in 2020, (EPA, 2023). Inset. Share of U.S. CO2 Emissions from Electric Power by Technology in 2022 (EIA, 2023). Some experts have noted that SBSP is a renewable energy alternative that could contribute to netzero goals, though SBSP is not featured in any of the net zero pathways considered by the most recent International Panel on Climate Change (P.R. Shukla, 2022). An SBSP system collects solar energy in space, converts that to microwave or optical laser energy, and transmits that energy to the Earth. A ground station receives the energy, converts it to electricity, and delivers it to the power grid for use. The rate and intensity of worldwide research into SBSP has seen significant growth: The number of publications on the topic nearly doubled from 2018 to 2022, with most of the research concentrated in China, the U.S., the European Union (EU), Japan, and Russia (NASA Library, 2023).
3 Figure 4. U.S. Electricity Generation by Technology. [Left] Share of U.S. Electricity Generation by Technology in 2021, (EIA, 2022). [Right] Projected share of U.S. Electricity Generation by Technology in 2050, (EIA, 2022). Many countries have ongoing SBSP studies, design concepts, and technology development activities including the U.S., the United Kingdom, the EU, Japan, China, South Korea, and Australia. In the U.S., for example, the California Institute of Technology (Caltech) completed the first successful electricity beaming demonstration from space to ground in June 2023 (Caltech, 2023). In general, this work is funded and conducted by academic, commercial, and government communities motivated by economic development, net zero, and national policy goals. Figure 4 displays a map of current international SBSP activity. The SBSP concepts described in this report focus on civil applications of SBSP to deliver electricity to the power grid. SBSP is being pursued by different organizations for different use-cases. The resulting national benefits could extend beyond immediate fiscal returns or near-term GHG emissions reductions.
4 Figure 5. International SBSP Activities. SBSP studies, design concepts, and technology developments are funded around the world for economic development, net-zero goals, and national goals. Countries with non-space-based power beaming efforts are not included.
5 1.2 Study Questions The idea of SBSP is not new to NASA, which conducted feasibility studies first in the 1970s (NASA & DoE, 1980) and again in the 1990s (Mankins, A fresh look at space solar power: New architectures, concepts and technologies, 1997). These studies found it prohibitively expensive to develop, launch and assemble, operate, maintain, and dispose of SBSP systems ($1T estimate in then-year dollars for an SBSP technology demonstration in the 1970s and $250B estimate in then-year dollars for the first commercial kilowatt (kW) of power in the 1990s). The context of SBSP development has changed significantly in the last three decades, however, prompting this study. Public and private actors across the international community are motivated to develop SBSP for economic development, net-zero goals, and global leadership. The study seeks to answer two questions: • Under what conditions would SBSP be a competitive option for achieving net zero GHG emissions compared to alternatives? • If SBSP can be competitive, what role, if any, could NASA have in its development? Alternative renewable electricity production technologies that we compared to SBSP include nuclear fission, geothermal, hydroelectric, utility-scale solar photovoltaics with storage, and land wind without storage. The study includes nuclear power even though it is not usually grouped with renewables because it is considered “non-emitting” by the EPA (EPA, 2023). Given increasing investment and attention to SBSP worldwide, this study is intended to inform NASA decision-making regarding any potential Agency role in SBSP development. Therefore, we present options for consideration for senior leaders. 2.0 Methodology Overview To determine the feasibility of SBSP we estimated the cradle-to-grave costs and GHG emissions of two system designs: RD1 (Innovative Heliostat Swarm) and RD2 (Mature Planar Array) based on existing concepts with updated technology assumptions on mass, efficiency, and launch capacity. The lifecycle cost estimates were used to calculate the levelized cost of electricity (LCOE) for each system for comparison to terrestrial renewable alternatives using data from the National Renewable Energy Laboratory (NREL). GHG emissions were estimated using a hybrid mass- and spend-based Economic Input Output-Life Cycle Analysis (EIO-LCA) and compared to terrestrial alternatives using NREL data.
6 Limitations of the study include: • This is a first-order assessment of notional systems: Cost and GHG emissions estimates are not exhaustive, and outputs are heavily influenced by assumptions about a technology with no historical data points built and operated in an unknown future. • The reference designs assessed are not representative of all proposed SBSP architectures: For example, systems in lower orbits have been proposed but are not assessed in this study. For a detailed methodology, broken out by each step of the analysis and including results, please see Appendix B. 2.1 Cost Estimations To estimate costs, we: 1) defined SBSP systems based on six key functions, 2) divided the SBSP system lifecycle into five concept of operations (ConOps) phases, generating cost estimates for each phase, and 3) used ConOps costs to determine the LCOE for comparison to other renewable energy technologies. 2.1.1 Functional Decomposition of SBSP Systems Generating electricity using SBSP systems involves six functions: collect solar energy in space, convert (in space) energy to microwave or optical energy, transmit that energy to Earth, receive the transmitted energy at one or more ground stations, convert (ground) that energy to electricity, and deliver electricity to the grid for consumption or to batteries for storage. This study assesses two representative SBSP designs: 1) the RD1 (Innovative Heliostat Swarm) concept, which uses a concentrator to improve its capacity factor, and 2) the RD2 (Mature Planar Array) concept, a less complex system that collects solar energy on one side and emits this energy as microwaves on the other. Figure 5 describes each reference design in terms of the six key functions; more detailed information on each concept is presented in Appendix A. 2.1.2 Concept of Operations The ConOps for each reference design is broken into five lifecycle phases: develop, assemble, operate, maintain, and dispose. Including disposal or decommissioning of a system is a best practice when assessing its full lifecycle. Figures 6a and 6b provide a visual summary of each ConOps phase. We estimate the cost of each SBSP reference designs by ConOps phase. Appendix A contains a detailed breakdown of each phase and all relevant parameters. Appendix B shows the mapping of ConOps phases to each functional step.
7 Figure 6a. Functional Decomposition of SBSP Design Reference Systems
8 Figure 6b. ConOps for the Innovative Heliostat Swarm Design Reference One System
9 Figure 6c. ConOps for the Mature Planar Array Design Reference Two System
10 Baseline assumptions are derived from a mix of current and projected technologies and costs. Among the key assumptions in the baseline assessment is that Space Exploration Technologies Corporation’s (SpaceX’s) Starship launcher, which is currently in testing, will be commercially available at $100M per launch. We make no claim as to the reliability of this assumption given the early stages of Starship development, but rather make this assumption because on a per kilogram basis, this represents a similar decline in launch prices from today as has occurred in the past 10 years. We include a 15% block buy discount because of the very large number of launches required to deploy an SBSP system. We assume one payload-laden Starship in LEO requires refueling by 12 separate Starship propellant tankers to reach GEO (Blue Origin Fed’n, LLC; Dynetics, Inc.-A Leidos Co., 2021).14 Based on subject matter experts (SME) input, we also assume 100 reuses of Starships used for refueling, and that payload-carrying Starships are single-use. SpaceX has been able to conduct two Falcon 9 launches in a week, so we assume this launch cadence for the Assemble ConOps phase. The Aerospace Corporation provided estimates of manufacturing learning curves, first-unit costs for each hardware element, component and system lifetimes, as well as module assembly time, all of which were reviewed, augmented, and incorporated by the study team. Finally, literature on SBSP concepts informed system specifications of the RD1 and RD2 designs. Estimated time to assemble a fully developed SBSP system in our baseline assessment is 7.4 years for RD1 and 12.6 years for RD2. Key input parameters are shown in Table 2. For a complete accounting of inputs, assumptions, and calculations, please see Appendix B. 2.1.3 Levelized Cost of Electricity Using our cost estimates, we calculated the LCOE measured in $/kWh for each reference design to compare overall costs to other renewable electricity production technologies. LCOE is commonly used by the energy sector for comparative analyses. LCOE is calculated by estimating the lifecycle cost (using different units for different categories) of the SBSP system and dividing that by lifecycle kWh production by the system. Figure 7 provides an overview of the LCOE calculation. 14 This estimate was derived using publicly available information about the initial human landing system for lunar exploration to be developed by SpaceX and modifying that based on the assumption that reaching GEO orbit would not require as many refueling launches as would reaching cislunar space. Blue Origins Fed’n, LLC; Dynetics, Inc.-A Leidos Co., B-419783 et al., July 30, 2021, 2021 CPD ¶ 265 at 27 n.13. This data was used because the Starship launch vehicle is anticipated to be the largest super heavy launch vehicle with available data and should not be construed as an endorsement by NASA. It is important to note that multiple specifications for this vehicle are planned, and cover a range of payload capacities, fuel capacities, and more.
11 Table 2. Key Input Parameters Key Input Parameter Value Source and Rationale Starship launch cost ($) $100M, 15% block buy discount 2013-2022 36% launch cost decrease, Falcon Heavy cost of $1500/kg to 2030s. Scale of launches may offer discount. Reuses of launch vehicle (does not include payload Starships traveling to GEO) 100 SpaceX states most components of Falcon 9 may be reused 100 times, but some elements must be replaced after 10 uses. New Glenn claims 25 reuses. 100 reuses = 4 times the state of practice. Orbital transfer method 12 refuel launches in LEO 1 month refuel time 2 months travel to GEO Blue Origin Fed’n, LLC; Dynetics, Inc.-A Leidos Co., supra.. Does not consider cryogenic boiloff. First order delta-v. Launches per year 104 Assuming two launches per week. Falcon 9 currently launching about 1.5x/week. Manufacturing learning curves 85% for servicer 75% for modules 90% for debris vehicles The Aerospace Corporation estimates based on aggregate manufacturing sector data. First unit costs ($) $1B for servicer $1M for SBSP modules $500M debris vehicles The Aerospace Corporation estimates based on satellite and solar cell industry and OSAM-1 costs. Solar cell efficiency 35% NASA assessment of Smallsat technology. Hardware lifetime (years) 10 The Aerospace Corporation estimates. System lifetime 30 The Aerospace Corporation estimates. Initial system upmass (kg), number of modules 5.9M (RD1), 1.46M 10M (RD2), 2M Inferred from Mankins, Sasaki, Pellegrino. Ground rectenna 6km diameter (RD1) 4km diameter, 5 sites (RD2) The Aerospace Corporation assessed costs of analogs: solar power plants and large antenna arrays. Operations cost 1.2M/month Assumes autonomous operations capability. Significantly less than today ~500k/year per satellite. Assembly time per module (minutes) 40 (RD1), 38 (RD2) The Aerospace Corporation estimates based on Orbital Express and ISS analogs.
12 Figure 7. Cost Calculations for SBSP Systems. Refer to Figure 6 for ConOps Phase activities. 2.2 GHG Emissions Intensity To estimate GHG emissions in carbon dioxide equivalents (CO2eq), we 1) used material decompositions of each reference design, and 2) provided a comparison to other renewable energy technologies, drawing from NREL emission data (National Renewable Energy Laboratory, 2023). We estimated the GHG emissions of each design in three steps: 1. Estimate the material composition of each design, in kilograms (kg) or square meters (m2) 2. Cite authoritative sources on the emissions intensity of delivering components and materials, in kgCO2eq per kg, m2, or thousands of dollars (kUSD). 3. Estimate the lifecycle emissions intensity using a hybrid mass and spend-based Economic Input Output-Life Cycle Analysis (EIO-LCA). EIO-LCA uses aggregate data on sectors of the U.S. economy to quantify the GHG emissions that can be attributed to specific sectors and activities. Our analysis uses the aggregated metrics provided by the International Aerospace Environmental Group (IAEG) (International Aerospace Environmental Group, 2023), including datasets from Carnegie Mellon University and the U.S. Department of Defense (DoD) on the GHG emissions intensity of activities measured in kilograms of carbon equivalents (kgCO2eq.) per kUSD, kg, or m2. This spend-based approach is used where material decomposition does not provide adequate coverage of post-processing and assembly work. It is important to remember that because the EIO-LCA model relies on aggregated economic transactions and their recorded GHG emissions, there is an assumed relationship between cost,
13 efficiency, and reduced emissions, though it is possible to reduce costs without mitigating GHG emissions of manufacturing. The resulting GHG emissions estimates were then compared to other renewable energy technologies. Figure 8. Calculations for SBSP GHG Emissions. Refer to Figure 6 for ConOps Phase activities. Results of the initial baseline cost and emissions estimates were assessed to identify cost and climate drivers. We then conducted sensitivity analyses to determine the effect of incremental changes in these drivers. 3.0 Results The study provides rough-order cost and GHG emissions estimates for the RD1 (Innovative Heliostat Swarm) and RD2 (Mature Planar Array) SBSP systems broken down by ConOps phase: Develop, Assemble, Operate, Maintain, and Dispose. Cost estimates for each ConOps phase by system are shown in Figure 9. For a detailed table of costs please see Appendix B. Both systems have a ~2 gigawatt (GW) capacity. The total estimated cost for each system is: RD1, $276B; and RD2, $434B. For both systems Maintain comprises over 50% of the overall cost. Assemble costs comprise about 25% of total cost for both systems. The most impactful cost element is launch, representing 71% and 77% of total cost for RD1 and RD2, respectively. Dispose, Develop and Operate are, in descending order, the next most expensive phases, but combined are less than Assemble for each reference design. The largest costs in Develop are for research and development (R&D), manufacturing and integration of all spacecraft hardware and systems, and program support services. Costs in Operate are primarily in the ground system; RD2 requires five ground rectennas where RD1 requires one. Dispose is unique in that it is the only ConOps phase where launch is included but is not the primary cost driver. For Dispose, the continuous operation of the Active Debris Removal (ADR) fleet for years is the largest cost.
RkJQdWJsaXNoZXIy MTU5NjU0Mg==