The tricky art of storing heat from the sun

The poster boy for concentrated solar power with thermal energy storage is Torresol Energy’s 19.9MW Gemasolar power plant in Andalusia, Spain. Photo credit: SENER

The poster boy for concentrated solar power with thermal energy storage is Torresol Energy’s 19.9MW Gemasolar power plant in Andalusia, Spain. Photo credit: SENER

A Chilean renewable energy tender that closed in October could be the first in the world where energy storage becomes the deciding factor for success. Great news for the industry? Perhaps. But actually it’s not so easy to say.

The bidding for a concentrated solar power (CSP) plant in Chile’s Atacama Desert, backed by a USD$20m government grant and more than $86m in alternative funding, is conditional on a minimum three-hour thermal energy storage (TES) facility.

And sources close to the tender, the entries for which are still being considered by the Chilean government, have confirmed that if there is a tie between different offers on all the main eligibility criteria then the amount of storage will determine which project wins. This might sound unusual, but in CSP it’s not something that would raise eyebrows.

After all, CSP players have increasingly been embracing storage in recent years. For those new to the subject, CSP involves focusing sunlight onto a point to create enough heat to drive a turbine. The critical word here is ‘heat’: unlike photovoltaics (PV), CSP does not produce electricity directly from sunlight. That means it can only be used in areas with high solar irradiation. It’s also generally more expensive than PV.

Furthermore, plants cannot be built panel by panel, so the upfront investment is higher and the return on that investment takes longer.

The ace up CSP’s sleeve

Why would anyone bother, then? Well, it turns out CSP has an ace up its sleeve. And that ace is energy storage. Using an appropriate transfer medium, heat that is not needed immediately to power the turbine can be stored for later use, for instance when clouds pass by or even overnight. The heat sink can also ensure a constant power output under varying weather conditions.

The poster boy for CSP with TES is Torresol Energy’s 19.9MW Gemasolar power plant in Andalusia, Spain. It can store enough heat to power its turbine for 15 hours after sundown, effectively giving it a round-the-clock power production capability. Plants like Gemasolar can thus sidestep renewable energy’s biggest bugbear: intermittency. That can make them worth the extra cost and effort.

Certainly, ever since Gemasolar came on the scene a growing number of regulators have been adding ‘only with storage’ to their CSP plant tender requirements. Besides Chile, TES has been specified in CSP programmes underway in Saudi Arabia and South Africa.

Storage is also a key feature of the world’s largest solar plant, the 377MW Solar Electric Generating System that BrightSource Energy brought online this summer in California’s Mojave Desert.

Just what energy storage needs?

And Areva, which is already pushing ahead with serious plans for energy storage elsewhere, has confirmed it will be offering TES as standard in its grid-connected CSP projects after developing the necessary technology with Sandia National Labs last year. Hang on, though: storage making a renewable energy source worth the extra money? Being fitted as standard?

Isn’t this exactly the kind of thing the energy storage industry needs? Well yes, but there’s a rub. Within CSP, TES in practice means using a salt, such as a sodium nitrate and potassium nitrate combination, above its melting point. Molten salt has got a lot going for it. It holds onto heat well, isn’t toxic or flammable, and is made of easy-to-come-by elements.

Plus it is already produced commercially for a number of industrial processes. There are technical challenges associated with handling it, of course, but they are not insurmountable. However, it’s not the sort of thing you can use to replace the fuel in your car. Nor would it be much use for storing excess power on the electricity grid. In fact, the application of molten salt in energy storage is pretty much restricted to CSP.

This means that CSP in its present form cannot benefit from the potential cost reductions that could arise from widespread adoption of other energy storage technologies, such as batteries. Nor can it help contribute to that wider adoption and cost reduction. Instead, its economics are to an extent driven by the vagaries of the global market for molten salt, within which CSP occupies a relatively small niche.

A limited benefit for the wider industry

Hence, other than helping to demonstrate the general benefit of being able to store renewable power, CSP’s use of TES is largely irrelevant for the energy storage industry. And for now CSP remains the renewable generation source least likely to benefit from new developments in non-TES-based energy storage technology.

This could all change in the future: the Australian CSP developer Graphite Energy, for example, is already commercialising a technology that stores heat using graphite instead of molten salt. Meanwhile other teams have proposed using CSP to power industrial reactions aimed at yielding fuel sources such as hydrogen or ammonia.

At the moment, though, most of these alternative approaches amount to little more than drawing-board speculation. Furthermore, CSP’s primary output of heat instead of electricity means that it could in practice be forever relegated to using energy storage technologies that are incompatible with those in operation elsewhere across the power system.

Indeed, given the cost differential between PV and CSP, some observers feel the latter might never achieve mainstream acceptance as a grid power source and may instead work best as a source of process heat for industrial applications.

If this is the case, it will be ironic that the first renewable power source to fully integrate storage turns out to be the one least able to incorporate the energy storage industry’s technical advances.

Written by Jason Deign

Let us know what you think. Please leave a comment.