Unless markets put a premium on dispatchable renewable energy there is a danger that concentrated solar power will not achieve the cost cuts it needs to remain viable
COST REDUCTION
The concentrated solar power sector needs to more than double for the technology to remain competitive against photovoltaic systems equipped with batteries LIMITED MARKETS
At the moment only two markets in the world are rewarding the kind of dispatchable night-time generation that concentrated solar power is uniquely positioned to offer KEY QUOTE
You need to understand that CSP technology cannot be compared directly with intermittent renewables The start of the futuristic science-fiction film “Blade Runner 2049” sees a flying car soaring over a Californian landscape studded with concentrated solar power (CSP) plants. CSP advocates believe this vision could help achieve the mid-century decarbonisation goals, but others argue there is a danger CSP could be a forgotten technology by 2049, not just in California but around the world. For it to be competitive with photovoltaic (PV) solar and onshore wind, the amount of CSP capacity in the world has to more than double to 7 gigawatts (GW), according to research by the Energy Economics Group at the Vienna University of Technology. However, as of 2021, there are only two countries with auction programmes that could accommodate CSP. Unlike PV, which works by converting light directly to electricity, CSP uses mirrors to focus sunlight onto a heat exchanger. The heat exchanger heats a fluid, usually oil or molten salt, which is used to power a steam turbine or similar engine. This mode of operation has two important consequences for where in the world CSP can be a feasible option. The first is it does not work well in diffuse sunlight. CSP plants are not worth building outside of areas with high solar irradiance, such as deserts in Africa, America and the Middle East. The second is that, while you can produce electricity from a single PV cell, to make a CSP plant work you have to install enough mirrors (called heliostats) to drive a turbine. Furthermore, larger heliostat fields can collect more heat and improve the efficiency of the system, so there is a financial advantage in building big CSP plants.
ARGUMENTS FOR CSP
So why bother with CSP at all, when there is PV? The main reason CSP has emerged in recent years is the possibility of long-term storage. It can use tanks of molten salt as a thermal energy store. As salt is a commodity it can be bought cheaply and used in vast quantities, unlike the materials used for lithium-ion batteries, which could be prone to supply chain shortages. Lithium-ion batteries are also needed in the growing digitalisation and electric mobility programmes. Similar reasoning applies to the much-hyped green hydrogen sector. The world will need to produce massive amounts of the gas for applications such as transportation and steelmaking before it can afford to burn hydrogen for electricity. But it seems the only way for CSP to remain a valid weapon in the world’s decarbonisation arsenal is for lawmakers to recognise now the value that it could offer tomorrow. “You need to understand that CSP technology cannot be compared directly with intermittent renewables,” says Miguel Méndez Trigo, at Spanish CSP developer Abengoa. “CSP technology, thanks to its thermal storage [attributes], offers the possibility of generating large volumes of energy whenever it is needed, independent of whether the primary resource—solar radiation—is available or not at that moment.”
MAINSTREAM SOURCES
In the early 2000s, when both CSP and PV were relatively novel and expensive, both technologies were seen as having the potential to become mainstream sources of renewable energy. As a result, Spain invested heavily in solar from 2008 to 2014, overtaking the United States as the world’s largest CSP market. At the same time, though, many other countries were starting to only invest in PV. Because PV can be used at any scale and almost anywhere in the world, these investments added up quickly. Growing orders helped Chinese PV manufacturers achieve economies of scale, cutting the cost of products. Meanwhile, CSP cost reductions were slower, dependent on a limited number of large projects. Although the industry as a whole was able to reduce costs more quickly than that of any other form of renewable energy, the sheer scale of PV manufacturing rendered CSP less and less competitive. After Spain’s buildout ended in 2014, it took around three years for the global CSP market to see further significant growth.
THERMAL ENERGY STORAGE
What happened in the meantime is that CSP developers established a new way to make the technology competitive again: storing some of the sun’s heat to use later in the day. The first CSP plant to use thermal energy storage was Solar Two, a ten-megawatt (MW) pilot project commissioned in 1996 in California. Its solar field was used to heat up two tanks of molten salt that were able to power a turbine for up to three hours after the sun went down. For the three years it was in operation, Solar Two remained something of a curiosity. But in the 2010s, as countries started to install large quantities of PV and wind power, CSP developers became aware that thermal energy storage could act as a long-duration battery. This realisation led to a shift in plant design. New plants were installed with the intention of having multiple hours of thermal energy storage as standard. Developers started to build hybrid plants where PV could provide power during the day while CSP stored energy exclusively for use at night. These models worked well in markets such as South Africa, where policymakers recognised it was worth paying a premium for renewable energy that could be delivered in the evening.
DISPATCHABLE CLEAN ENERGY
In Spain and Chile, regulators are proposing power auctions in 2021 that will reward plants that can deliver dispatchable clean energy overnight. The question for developers is whether CSP could provide this—or whether the technology could be edged out by PV and traditional batteries. As with the original contest between CSP and PV, the answer depends on scale. CSP’s levelised cost of energy is currently between $126 and $156 per megawatt-hour, according to 2020 figures from Lazard, a US-based financial advisory firm, compared to between $31 and $42 per megawatt-hour for utility-scale crystalline silicon PV. Molten salt, however, is much cheaper than lithium-ion batteries for delivering large amounts of energy over a longer period of time. This currently makes CSP with thermal energy storage the cheapest way to provide overnight renewable energy in sunny climates. But the cost of batteries is falling rapidly thanks to the economies of scale of battery manufacturing for the automotive industry. The Energy Economics Group study says it could already be cheaper to use PV and batteries for up to two or three hours of night-time power delivery. That is less than a third of the average 9.3 hours of discharge time for molten salt storage attached to modern CSP plants. But unless CSP plants can get cheaper too, then further battery cost reductions will see CSP’s night-time market opportunity being eroded. The Energy Economics Group study team believes CSP needs to cut its costs by about a quarter to stay competitive. “With 5.8 GW of CSP projects being operational worldwide and a learning rate of 20%, CSP has to add around 7 GW of global capacity to achieve this medium cost scenario,” it says.
PROCUREMENT PLANS
This “seems feasible for a large country or world region,” the authors add. And 7 GW is within reach based on Chile’s procurement plans alone. Although the country has not specified a target for CSP, its demand for night-time generation could add up to 10 GW of capacity by 2050. Meanwhile, Spain is aiming to procure 5 GW of clean dispatchable overnight power up until 2030, to replace generation capacity lost as nuclear reactors close down. It is important to note that neither country is specifically targeting CSP in its procurement plans. Jonathan Walters, at E3G, a climate change think tank, does not see this as a problem. “Auctions should not ask for CSP,” he says. “They should ask for dispatchable renewable energy, or at least for night-time renewable energy. This will let technology show what can be done most cheaply.”
LIMITED CAPACITY
The challenge for CSP is that Chile and Spain are only putting limited amounts of night-time capacity out to tender in the initial stages. Their combined solicitations will add up to less than 500 MW of capacity this year and few markets other than Chile and Spain are looking to reward overnight production at present, limiting the opportunities for new CSP plant development. In contrast, IHS Markit, a London-based market analyst firm, expects there to be 10 GW of new battery installations worldwide this year. In the race to scale and cut costs, CSP does not seem like much of a winner. That, in turn, does not bode well for CSP taking most of the night-time capacity on offer in Chile and Spain over the coming years. There is a possibility it will gradually lose out to PV and batteries as the price differential between the two asset classes grows. None of this would matter if PV and batteries could offer a reasonable alternative to CSP for long-term bulk energy storage. However, some experts believe that CSP may yet have a key role to play in the energy transition—just that we cannot see it yet. Decarbonisation today is proceeding perfectly well because we do not yet need great quantities of storage and can still rely on fossil fuels for backup, says Julián Blanco, of the Plataforma Solar de Almería, a Spanish state-backed CSP research hub. “Right now, the closure of coal and nuclear plants can easily be replaced with gas,” he says. “Getting to a decarbonisation level of 80% or 90% is not a problem. But that last part, from 85% or 90% to 100%, is a problem. This is where CSP would be more than justified.”
TEXT
Jason Deign