This article is a shorter version of the detailed calculations we made. By the way, we provide in-depth research services for SpaceTech, DeepTech, and Energy sectors. If you need one, please, reach out to our co-founder Denis Kalyshkin via denis@spaceambition.org. We’d be happy to help!
The dream of harnessing the boundless energy of the Sun from space is far from new. The original idea by Asimov took form in the 1970s by the US engineer Dr. Peter Glaser and ever since engineers proposed a variety of massive Space-Based Solar Power (SBSP) plants, envisioning colossal arrays in GEO orbit at altitudes of around 35,800 km, capable of beaming down gigawatts of microwave power. In an era of cheap and abundant fossil fuels, the sheer scale and cost of such undertakings rendered them economically unviable, more science fiction than practical energy solution.
However, the narrative has shifted dramatically. The looming shadow of global warming and the urgent global drive towards Net Zero emissions have injected fresh urgency into the SBSP concept. The compelling need to drastically reduce our carbon footprint offers a powerful argument for revisiting space-based solar, even if the initial price of electricity might be higher than terrestrial fossil-based energy. Yet, the stunning progress of Earth-bound solar power presented a seemingly compelling hurdle. Why endure the immense challenge and expense of launching panelsinto orbit when ground-based solarrapidly became cheaper, with simpler maintenance and no need for rocket launches [9:00]? For a while, SBSP dreams were once again relegated to the back burner.
But the game is changing again. A seismic shift in space launch economics is underway, spearheaded by revolutionary launch systems like Starship and New Glenn. Suddenly, the once-prohibitive cost of reaching orbit is falling. Some bold projections even suggest that space solar electricity could reach an astonishing £26($34)/MWh – a figure that approaches the most competitive terrestrial solar costs of around $29/MWh. Could this be the dawn of space-based solar power? Is it time for investors to take notice and stake their claim on this celestial energy frontier? Let’s delve into the numbers and separate hype from genuine potential.
Our reasoning is presented in the three sections ahead. Firstly, we introduce the framework we use: the levelized cost of energy (LCOE). We then calculate LCOE when applying Earth-standard, cost-effective solar panels, solely accounting for purchase and launch costs. Secondly, we analyze the three key economic drivers: launch cost per kilogram, solar panel mass per square meter, and solar panel cost per kilowatt. This analysis reveals that the combined costs of procurement and launch could become feasible in the future. Finally, we emphasize that the primary obstacle to space solar power now resides in the expenses associated with supporting orbital structures and microwave antennas. Math alert! Don't worry, it's not too scary.
Chinese design idea for a SBSP station. Credit: Xinbin Hou, CAST, China
Fundamental Cost Comparisons Using Levelized Cost of Energy
To make a meaningful comparison between energy sources – be they in space or on Earth, renewable or fossil – we need a common yardstick. We require a universal metric to fairly evaluate their economic viability. While numerous metrics exist, each with its own set of limitations and nuances, for initial, high-level assessments, simplicity is important. For preliminary investor evaluations, absolute precision isn't necessary; we need a tool for rough comparability. Once a handful of energy sources emerge as potentially competitive, then a deeper dive with more tailored, situation-specific metrics becomes essential.
The Levelized Cost of Energy (LCOE), measured in $ per Megawatt-hours, is a widely accepted cost framework. It calculates the total cost of energy production over a project's lifetime, encompassing all relevant expenditures: upfront Capital Expenses (CAPEX), one-time decommissioning costs, and ongoing Operational Expenses (OPEX), all annualized for clear comparison. Essentially, the LCOE boils down to the ratio of the Present Value of all Costs to the Present Value of all Energy Produced.
For those who prefer to skip the mathematical intricacies, it's sufficient to understand that the LCOE represents the minimum price per unit of energy a supplier must charge to recoup all costs and break even. The good news is that for long-duration projects, this seemingly complex formula can be simplified significantly (under certain assumptions), reducing to a more manageable form:
\(LCOE=\frac{d*CAPEX+OPEX}{E}\)
(Explore the footnote at the end for those keen on the simplification details!)
Here d is the discount rate expressed as a fraction, and E is the annual energy production, in Megawatt-hours. The discount rate is a number that calculates how much future money is worth in today's terms. A project with a higher discount rate means investors are more focused on getting their money back sooner and are less willing to wait for long-term returns. A lower discount rate suggests investors are more patient and willing to wait longer for the benefits. The choice of discount rate can significantly impact the calculated LCOE.
This simplification merely shifts the challenge to accurately estimating the remaining two key terms. But before tackling that, let's establish a theoretical lower bound for the LCOE of SBSP. Imagine for a moment we could utilize the same cost-effective solar panels (shortly PV, or photovoltaics) deployed on Earth andonly consider the expense of purchasing and launching them into orbit. We isolate these core CAPEX components, drawing back all other expenses – the vast orbital infrastructure, microwave transmission antennas and ground receiving stations, development, OPEX, insurance, and unforeseen contingencies – to a separate, encompassing term. This simplified "lower-bound" LCOE can be represented as the first two terms:
Here, the first component reflects the cost of purchase of solar panels per unit of power capacity – a crucial metric as nominal power output scales with panel area, directly linked to panel cost per square meter. However, cost per watt is a more readily understood industry standard, especially when comparing terrestrial solar panel costs, currently around $500/kW. Keep in mind that the peak PV power on Earth will be close to the nominal one whereas in space the peak PV power will be approximately 30% higher than the nominal due to the absence of the atmosphere.
The second component breaks down the launch cost element: the cost to launch payload to space per kilogram, multiplied by the total mass of PV. All other, significant cost drivers are bundled into the separate, harder-to-quantify term, marked as “+...”.
How the main cost drivers work
Now it's time to write down the annual energy production E as the product of the PV area S and the annual solar energy per unit of area e, and overall efficiency 𝜂:
\(E=e*S*\eta,\)
To get:
\(LCOE=\frac{d}{e*\eta}*(PVcost ($/kW)*(PV nominal\, power(kW/m^2))+(launch\,cost($/kg))*(PV mass (kg/m^2))+...\)
Here PV nominal power, measured in kW/m2 is the solar irradiance in space (1.36 kWh/m2), normalized to the best possible atmospheric conditions (because remember, nominal power expresses the power PVs would yield on Earth), resulting in roughly 1 kWh/m2. This simplified equation reveals that the lower limit of SBSP energy cost hinges on 3 fundamental economic drivers in this reduced problem: launch cost per kilogram, solar panel mass per square meter, and solar panel cost per nominal kilowatt.
With just these three core factors, we can perform some illuminating back-of-the-envelope calculations.
Consider terrestrial solar with a discount rate of 5%. If we eliminate launch costs entirely (setting the second term to zero) and place panels on Earth, the sun doesn't shine 24/7. Average sunlight hours cause e to be something close to 2,5 MWh/m2/year, varying with latitude and climate. Terrestrial utility-scale solar panel efficiency, typically not higher than 30%, further reduces annual energy capture. Applying these terrestrial conditions to our simplified formula yields a cost figure near $33/MWh if PVs are purchased at $500/kW, fairly close to the benchmark LCOE figures reported by entities like Lazard for best ground-based solar.
Now, let's propel those panels into orbit, assuming the same discount rate. Suddenly, sunshine becomes continuous – 24 hours a day, 365 days a year, yielding e=11.9 MWh/m2/year! However, orbital operations introduce new inefficiencies. The process of collecting solar energy, converting it to microwaves, transmitting it through the atmosphere, and reconverting it to grid-compatible electricity on Earth inevitably incurs 63% losses (a rather mere estimate). Overall system efficiency drops to approximately 11%. Nonetheless, the factor of longer sunshine wins and the energy cost becomes lower than $33/MWh calculated above:
But to factor in the launch cost, let's optimistically assume a revolutionary launch cost of just $100 per kilogram of payload to orbit. This figure is almost twice as high as a super-optimistic Musk’s assumption of a $1,000,000 launch cost to deliver 21 metric tons to a GTO (Geostationary-Transfer Orbit, kind of halfway between LEO and GEO. SBSP will never be placed in LEO due to the space debris problem, but it could be MEO, GEO, or other higher orbits, depending on the project). We'll also, very optimistically, assume that space-rated solar panels, despite their need for radiation hardening and specialized design, miraculously cost the same per watt as their terrestrial counterparts (again, to establish a lower-bound estimate), and weigh 1kg/m2 (this can be even lower if solar concentrators are used). Crunching these optimistic numbers gives us:
\(\frac{d}{e*\eta}*(launch\,cost($/kg))*(PV mass (kg/m^2))=$3.5/MWh\)
\(LCOE=$17.5/MWh+ $3.5/MWh+...=$21/MWh+...\)
As we can see, this lower limit cost of SBSP is surprisingly comparable to (and, importantly, lower than) ground-based solar energy. Crucially, we haven't yet accounted for the third, catch-all cost term, encompassing the significant expenses we deliberately omitted for this lower-bound calculation.
The first term explains numerically what the proponents of SBSP like to sell: 24/7 sunshine is important, even despite the energy transmission losses: it reduces LCOE by a factor of ~2. But, you pay a price due to the launch cost (the second term), which was prohibitively high 10 years ago, but now there is a hope it will be tolerable soon.
Within this concise post, we cannot fully dissect the complexities of space-based structures, antenna development, or the massive project development costs baked into that “+...” third term. However, this simplified analysis provides a crucial framework for understanding the key cost drivers. You now have a roadmap for the next level of investigation. Deeper dives, like existing detailed SBSP cost models, factor in these complexities – though even those models grapple with inherently uncertain future cost projections.
Nevertheless, our reduced-problem analysis underscores the pivotal role of the three main drivers of SBSP economics: the launch costs, PV mass, and the PV cost. A shift from a $100 per kilogram launch cost to a more realistic near-term $1,000 per kilogram, for example, would instantly inflate the total cost by an additional $31.5/MWh. If the radiation-resistant PVs will cost twice as much as the ones used in the terrestrial utility-scale solar plants, this adds another $17.5/MWh. These two factors combined will result in the minimum theoretical cost being $70/MWh, which already makes SBSP non-competitive.
The true bottleneck is the antenna cost
The paradigm is shifting. Solar panels themselves are becoming increasingly inexpensive commodities, and launch costs are no longer going to be the insurmountable barrier they once were. Even with less wildly optimistic launch cost scenarios, the combined cost of purchase and launch of solar arrays becomes manageable. The real bottleneck for Space-Based Solar Power now lies in the monumental upfront development costs and the price of the supporting orbital structures and, most critically, the gargantuan, gigawatt-scale microwave transmitting antenna and ground receiving infrastructure, all hidden in the “+...” term. Developing these sophisticated systems to be both highly efficient and cost-effective will be the central engineering and economic challenge.
Current antenna technology capable of reliably and safely transmitting gigawatts of microwave energy from a high orbit to a precise point on Earth simply doesn't exist at this scale. Building such systems, initially, will inevitably be exceptionally expensive. Before SBSP can become economically competitive at a global scale, substantial time, dedicated research, and iterative development are essential to drive down the cost. Perhaps the most prudent path forward is not an immediate, all-in commitment to a full-scale SBSP plant. Instead, we should focus on developing and demonstrating key SBSP technologies in more manageable,nearer-term applications. Lunar exploration, in-situ resource utilization on the Moon or Mars, and orbital manufacturing platforms could serve as vital proving grounds, enabling incremental technology maturation and cost reduction, paving the way for the eventual realization of affordable, large-scale Space-Based Solar Power.
We advocate for meaningful space exploration that benefits everyone on Earth, from consumers to investors. Read our other articles on spaceeconomics. If you like this and other articles and have ideas for similar simple economics estimates, drop us a line at denis@spaceambition.org.
CAPEX(n): Capital Expenditures in year 'n', encompassing initial one-time costs for engineering, procurement, equipment construction, and decommissioning. Expressed in Present Value.
OPEX(n): Operational Expenditures in year 'n', representing recurring annual costs such as operations and maintenance, refueling, security, and insurance. Expressed in Present Value.
The energy produced(n): Average energy output (electricity or heat) delivered to the distribution system (grid or pipelines) in year 'n', measured in MW·h. Expressed in Present Value.
n: temporal index, typically set as annual increments.
i: Annual compound inflation rate (expressed as a fraction, e.g., 0.02 for 2%). The inflation rate ('i') allows for adjusting cost representation: we set 'i' to zero to yield "real" or constant dollar values.
d: Annual discount rate (expressed as a fraction, e.g., 0.02 for 2%), reflecting the time value of money and risk premium for long-term projections.
Assuming stable and uniform energy production and OPEX throughout the whole lifetime of the power plant: Energy produced(n)=const=E, OPEX(n)=const=OPEX,
and CAPEX being distributed in the first year only, our model reduces to:
Two things. The wireless power transfer is not a new technology, it sits in every communication device globally (ie your cell phone in your pocket). The beam forming and aiming is already demonstrated at scale with cell phone towers. Obviously SBSP is bigger, but simpler since the antennas are not 2-way data transfer devices. Second, the firming costs of terrestrial solar are non-trivial - every market with significant solar penetration sees higher total cost of energy (LCOE + firming) with many markets at $350+/MWh. SBSP is closer than anyone suspects - able to achieve total cost of energy lower than any other source on the planet.
The value of continuous (baseload) and dispatchable power is overlooked in this article. Imperial College London has completed a study of the UK energy system with and without SBSP. It shows that for every 2 GW of SBSP, there are annual savings of over £1Bn, and these savings scale with increasing SBSP capacity. It reduces the need to over build wind and solar capacity, reduces the need for other backup and storage, and greatly reduces the need to expand the grid transmission system. It also provides export revenues without costly interconnector cables.
As a standalone baseload energy technology, SBSP will be highly competitive but as part of the whole energy system its value is massive in delivering abundant affordable and reliable energy.
Two things. The wireless power transfer is not a new technology, it sits in every communication device globally (ie your cell phone in your pocket). The beam forming and aiming is already demonstrated at scale with cell phone towers. Obviously SBSP is bigger, but simpler since the antennas are not 2-way data transfer devices. Second, the firming costs of terrestrial solar are non-trivial - every market with significant solar penetration sees higher total cost of energy (LCOE + firming) with many markets at $350+/MWh. SBSP is closer than anyone suspects - able to achieve total cost of energy lower than any other source on the planet.
The value of continuous (baseload) and dispatchable power is overlooked in this article. Imperial College London has completed a study of the UK energy system with and without SBSP. It shows that for every 2 GW of SBSP, there are annual savings of over £1Bn, and these savings scale with increasing SBSP capacity. It reduces the need to over build wind and solar capacity, reduces the need for other backup and storage, and greatly reduces the need to expand the grid transmission system. It also provides export revenues without costly interconnector cables.
As a standalone baseload energy technology, SBSP will be highly competitive but as part of the whole energy system its value is massive in delivering abundant affordable and reliable energy.