A tethered platform hovering at an altitude of 20 kilometers would operate in the stratosphere, above most clouds and weather. At that altitude, a platform covered with photovoltaic cells would be able to collect considerably more sunlight than a ground-based solar collector. A company called stratosolar has developed a proposal to use either concentrated solar collectors or straight photovoltaic cells to generate electricity. Such schemes could potentially bring the cost of solar power down to 1 cent per kilowatt-hour. In an interview with Sander Olson, StratoSolar President Edmund Kelly describes what will be needed to make this concept a reality.
Question: What is StratoSolar? It’s a proposal to collect solar power at an altitude of 20 kilometers, where the troposphere gives way to the stratosphere. The stratosphere is a nearly ideal location for collecting solar power, since at that altitude the atmosphere has absorbed little of the sun’s rays. There are no clouds and low winds. There is day and night, but that’s totally predictable. Question: StratoSolar has researched two versions, one using Concentrated Solar Power (CSP) and the other involving floating platforms of straight photovoltaic cells. How does the CSP system work? The basic idea is a huge tethered concentrator at 20 km that tracks the sun. It continually collects light during the day and sends it to the ground through a light pipe. Some of the energy is stored as heat for 24-hour power production. Each square meter of a 2 km dish collects six times as much sunlight as a ground based heliostat mirror. In regard to converting sunlight into electric power, it is as efficiently as the best thermal plants. It is a fairly complicated system involving light pipes, heat storage, turbine generators and so forth. Question: What will it cost? Based on the cost of materials, plastic, aluminum, hydrogen and steel wire or Kevlar and a ten-year payoff the CSP version we expect it to eventually fall to around 1 cent per kilowatt-hour when the technology matures. Thermal storage is remarkably low cost, in this design no more than a tenth of a cent per kWh. Even an early version looks like it will initially generate electricity for 6 cents per kilowatt-hour. At 20km, a tethered platform gets the benefits of abundant sunlight and a benign environment. A tethered platform will get nearly constant sunlight from dawn till dusk, whereas on the ground a platform will only get peak sunlight for a brief period. Question: Can’t ground-based photovoltaic provide all of the earth’s energy needs? Photovoltaic technology has made considerable progress during the past two decades, but it is not anywhere close to being competitive with coal, in terms of cost per kilowatt-hour. In ground installations, the cell only receives sunlight on sunny days, and even then only intermittently. More efficient and less expensive cells can ameliorate these problems, but those advantages would also make our concepts even more viable. Therefore, we have developed an inherently better solution, one that can be deployed as far north as Stockholm. Question: Is the CSP approach is the more sophisticated of the two approaches? We believe that the most efficient way to get solar energy is by floating large solar collectors at an altitude of 20 Kilometers. However, the technology for concentrated solar power is newer, trickier than for photovoltaics, and involved greater R&D and higher risks. Therefore, for that reason we developed the photovoltaic system, which is simpler and easier to implement. The PV system floats a platform using hydrogen gas, also at 20 kilometers. The photovoltaic (PV) system should result in costs of 8 cents per kilowatt-hour, but with the continuing improvements in PV panels, we can reduce the cost even further. Question: What R&D costs are we talking about to fully develop the CSP and PV systems? How much electricity could one platform provide? The CSP systems contain many elements and the minimum system is large. This means the R&D cost to build a complete functioning system is several hundred million dollars. It will ultimately require billions of dollars to perfect this technology and to develop the infrastructure. However, the first platforms will generate revenue which will help fund the R&D. The PV platforms contain fewer elements and are smaller initially and modular, so the initial R$D cost is much lower than CSP, in the region of tens of millions of dollars for the first operational platform, and they can be easily scaled up to larger systems without additional R&D. We have calculated that 30 of these larger PV platforms could provide all of the daylight electricity needs of all of California. California could achieve energy independence within a decade. Question: Given sufficient funding, how long would it take to launch the first CSP platform? We would require at least five years to get a full-scale CSP platform up and running, since significant R&D is needed to make that concept viable. Within three years, we should be able to create a small platform that confirms the feasibility of the approach. The photovoltaic approach is modular, and therefore can be scaled up faster. Question: How confident are you that a 20 kilometer tether could be made sufficiently strong? What materials will be required for the floating platform? We will use Kevlar or UHMWPE for the tether, which have sufficient strength to support the platform without breaking. For a high voltage line, the tether will be very narrow in diameter. For the platform there is no need for exotic materials – the required materials are mostly aluminum, Mylar, and structural fabric. Question: What sort of efficiency losses would be incurred getting the power down to the ground? Low. Not different from 20 km of power lines. Question: At what latitudes can tethered platforms operate? Our platforms are designed to operate at between latitudes 30 and 60, which pretty much covers the industrial zones. We don’t recommend deploying these platforms over cities, at least initially. However, you could place them close to cities, obviating the need for long power cables and efficiency losses. Therefore, these platforms could operate over Germany, and collect substantially more energy than ground-based solar cells in a desert. Question: To what extent will weather be a problem? What about static electricity or hydrogen safety issues? With the CSP, there is a possibility of catastrophic risk from a major hurricane or tornado. That is one of the reasons that we developed the PV system. The platform itself will be above most clouds and so only the cables will have to endure weather. The PV system is designed to be able to withstand any expected wind, and will in general be unaffected by weather. We have also specifically designed the platforms to be unaffected by electrical discharges. In a properly designed system, the risks from using hydrogen are negligible. For example, surround the hydrogen bags with nitrogen. Question: To what extent can the costs of conventional ground-based solar power be reduced? During the past decade, packaging costs have come down substantially, and the costs for packaging the cell and the electronics have been reduced. There are limits as to how much lower costs can go – fixed costs for land and paving that won’t go down. A conventional solar station on the ground requires square miles of land and huge amounts of material. So there exists a bottom limit to costs. By contrast, the stratosolar approach has costs that are only 1/3 of those costs of a conventional solar station. Our system is the only feasible way to generate solar power that is cost-competitive with coal. Question: Could this technology be used to eventually provide all of the earth’s energy needs? Within a few decades, there could be thousands of platforms in operation. That sounds like alot, but it would have these platforms are surprisingly easy to build, and are actually quite safe to operate. Building and operating thousands of these platforms is a feasible and cost-effective proposition, even without any subsidies. Question: How much progress could be made with CSP and PV within the next decade? Within a decade, we could have PV deployed on a very large scale. CSP will take longer, but we could see the multiple CSP platforms operating within ten years. PV is on a technology development curve, and if we could develop an efficient electricity storage system, we might not even need CSP.