James McKenzie believes that “concentrator solar power” could be the answer to our environmental woes.

The Ivanpah Solar Power Facility in Nevada uses thousands of mirrors to focus sunlight onto a tank of water at the top of a tower, with the resulting steam being used to drive turbines. (Courtesy: BrightSource Energy)

Modern solar cells are so good at converting sunlight into electricity that today’s flat-panel photovoltaics (PVs) are cheap, efficient, long lasting and plentiful. As I mentioned last month, they economically outperform concentrator photovoltaics (CPVs), which use lenses or curved mirrors to focus the Sun’s rays onto tiny solar cells. Despite showing much early promise, CPVs today seem far too complicated and expensive to succeed.

There is, though, another kind of solar power I find exciting, which uses sunlight to warm up a heat-retaining liquid. The hot fluid can be used to boil water, with the resulting steam driving a turbine to generate electricity. Known as concentrator solar power (CSP), it is only really economic at large scale, but has one huge advantage. Because the liquids in CSP can be stored, the energy can be converted into electricity even when the Sun isn’t shining.

Concentrated solar power has one huge advantage: the energy can be converted into electricity even when the Sun isn’t shining

According to the US National Renweable Energy Laboratory, three main types of CSP have been tested over the years. First there are linear concentrator systems, which use long U-shaped mirrors to collect the Sun’s energy. Motors actively tilt the mirrors towards the Sun, focusing sunlight onto tubes (receivers) that run the length of the mirrors. The tubes, which contain the hot fluid, are usually placed in troughs lying along the focal line of the mirror, although sometimes a single tube is placed above multiple mirrors.

Another kind of CSP uses a mirrored dish rather like a large satellite dish. To minimise costs, the dish is not one single structure but is usually made up of many smaller flat mirrors. The curved surface directs and concentrates sunlight onto a thermal receiver, which absorbs and collects the heat. The hot fluid heats a gas in a version of the classic Stirling engine, moving pistons to create mechanical power, which drives a generator.

Finally there are “power tower systems”, which use a large field of flat, Sun-tracking mirrors. Known as heliostats, they focus and concentrate sunlight onto a receiver at the top of a tower. As with the other CSPs, a heat-transfer fluid heated in the receiver generates steam to drive a turbine. Most use molten salts as a heat-transfer fluid that can store energy for use at night or when it’s cloudy.

Mixed fortunes

There were more than 90 CSPs in operation around the world by the end of 2021, with Spain and the US accounting for over half the globally installed capacity of 6.4 GW. Examples include the 110 MW Crescent Dunes facility in Nevada (which uses molten salt) and the 394 MW Ivanpah project in California (which runs on water). The beauty of CSPs is that they are all highly efficient. Power-tower systems (where the fluid is at 250–565 °C) can convert up to 35% of all solar energy into electricity, while dish systems (running at 550–750 °C) can be up too 34% efficient.

According to a 2021 report from the International Renweable Energy Assocation (IREA), the levelized cost of electricity (LCOE) – a kind of average net cost – of CSPs has fallen sharply in recent years. In the decade to 2020, the global weighted average LCOE of newly commissioned CSPs plants tumbled by 70% from $0.361/kWh to $0.107/kWh. The reductions have been driven largely by the fact that these plants can operate at ever higher temperatures, which cuts storage costs and allows them to run for longer periods.

For plants commissioned between 2016 and 2020, the IRENA report found that about four-fifths had at least four hours of storage, while 35% had eight hours or more. This trend is expected to accelerate even further with IRENA already finding that the average storage time had risen from 4.7 hours for projects commissioned in 2020 to 17.5 hours for those commissioned the following year. The average CSP project size in 2021 was a respectable 110 MW.

Cost concerns

But CSP systems aren’t perfect. They use lots of water and it’s not cheap keeping the mirrors clean, which is vital if you want to keep their efficiencies high. And, of course, these systems only work well if there’s lots of sunlight, which makes them impractical in many parts of the world. What’s more, the cost of rival renewable sources of energy has been plummeting.

According to IRENA, the LCOE for solar PV has fallen by 88% since 2010 to $0.08/kWh, while onshore wind power has dropped by 68% during that time to $0.033/kWh and offshore wind has decreased by 60% to $0.075/kWh. Despite its recent drop in price, CSP remains much higher at $0.107/kWh. It makes CSP look uneconomic, although this headline figure includes the cost of storing the energy, which is omitted from the other LCOE numbers.

For all the promise, concentrated solar-power systems lag far behind conventional solar photovoltaic panels

Still, for all the promise, CSPs lag far behind conventional solar PV panels, which had a total installed capacity of 957 GW in 2021. Solar PVs are simply such an easy and scalable technology, provided of course that you can store the energy in batteries. Given the huge amount of work going into developing batteries for electric vehicles, I’m sure solar PV will give CSP with thermal storage a run for its money over the next 20 years.

As the IRENA 21 report concludes: “In the absence of strong global policy support for CSP, the market remains small and the pipeline for new projects unambitious”. I find that disappointing, given the remarkable fall in costs for CSP since 2010 and its ability to provide dispatchable power 24/7 in sunny areas at a reasonable price. Sadly, energy production is all about the economics – and CSP is always playing catch-up on cost.

But perhaps all is not lost. First, the batteries needed for solar PV might rocket in price as they require rare materials that might one day become costly. Second, research suggests that the heat stored in CSPs could be used to produce a range of “green fuels”, such as hydrogen or ammonia. Perhaps CSPs might one day even work in partnership with their nemesis – the flat solar PV panel – to produce electricity and fuel round the clock.