Solar at night

Despite headlines about solar overtaking coal as the largest generator in the NEM, the latest example showed that solar actually topped coal for only 30 minutes. Coal still remains Australia’s dominant electricity generation technology and still provides over 50% of our energy. We rely heavily on coal overnight from about 6pm to 9am, yet it will be gone from the Australian electricity market within 10-15 years.

Australia urgently needs a replacement. While in some corners gas is touted as a replacement for coal, it won’t help us achieve emissions reductions targets and address a rapidly warming planet. If Australia is to meet its emission reduction targets, we need a coal replacement that is renewable, fully dispatchable, able to provide multiple hours of firm capacity and be available to meet our night-time power needs.

Australia’s day-time power needs are already well on the way to being supplied by low-cost solar PV, and the proportion of daytime solar PV will only increase over time. However, access to multiple hours of reliable overnight renewable capacity is the challenge. While wind can be a great night-time renewable resource in certain places, it is an intermittent form of generation. To provide firm capacity from 6pm in the evening to 9am the following morning, renewable energy storage with more than 12 hours duration is required.

AEMO has indicated that 45 GW of dispatchable storage capacity will be needed by 2050 to deliver 620 GWh of energy. However, given recent announcements about early retirements of coal fired power plants, more storage capacity may be required sooner. The need for dispatchable storage capacity will further increase if there is an increase in overnight demand due to rapid uptake of electric vehicles, or if there is an increase in domestic and international emissions reduction targets. If all three of these factors occur concurrently, as is expected, Australia will face significant challenges.

Australia’s energy ministers, in August this year, agreed to fast-track options to ensure that firm renewable capacity and associated investment incentives are established to manage the risks of the disorderly exit of coal.

The energy ministers also agreed to fast-track an emissions objective into the National Energy Objectives to ensure that sustainability becomes a key aspect of future system requirements. 

These decisions establish the need for firm renewable capacity, in the form of long duration intra-day storage, as a critical priority to replace coal fired generation.

While lithium-ion batteries would appear to be the logical choice to solve this storage challenge, they are only cost-effective for short durations, around one to four hours.

Technologies with the potential to deliver cost-effective intra-day storage (ie. 12-18 hours), as required to meet Australia’s overnight energy needs, include pumped hydro, concentrated solar thermal power (CSP), hydrogen storage, flow batteries and compressed air.

Of these, CSP is the one best suited to Australia’s hot, dry and sunny climate. Globally, there are now over 110 operational CSP plants, with a combined capacity of just under 7 GW, delivering 30,000 GWh per annum, which equates to 12 hours of daily generation. 

With another 30 CSP plants expected by 2025, CSP is already delivering nighttime energy with the reliability of daytime sunshine in sunny countries all around the world.

CSP systems have similar strengths to coal fired power plants, but with no emissions. The zero-emissions component is derived from CSP’s use of hundreds of heliostats – specially designed and tracked mirrors – that aim the reflected sun at a central collector. 

Rather than turning the sun’s energy directly into electrical current – as occurs with solar PV – they concentrate the sun’s heat up to 600˚C and then store that heat in large salt tanks. The hot salt is then used to create steam to drive a turbine, just like a coal-fired power station. 

CSP plants typically have 10-15 hours of thermal energy stored, allowing them to cover the crucial overnight electricity market which is currently serviced by coal. 

Another important feature of CSP, which makes it popular in transitioning utility grids, is the system-strength and inertia that it provides to the grid. This is thanks to the synchronous generation from the spinning turbine which creates frequency control and ancillary services (FCAS) at no charge. This negates the need for expensive battery projects, which are currently providing an increasing proportion of FCAS and which themselves require condensers to simulate synchronous power generation.

Regional employment can also benefit from CSP projects, which require a large workforce for construction, and highly skilled teams for ongoing operations. Construction typically requires over 600 people for two to three years, with ongoing operations requiring 20 people for a 100 MW plant. 

In essence, CSP projects employ about five times as many people as PV and wind projects. The jobs themselves are ideally suited to people who’ve worked in fossil fuel power plants, due to the similarities in operations.  

Like wind, solar and batteries, long-duration renewable energy storage projects take several years to permit, design, construct and commission (although they can be built much quicker than pumped hydro). Given the urgency of the decarbonisation challenge, decisions to establish incentives to create the right environment for private sector investment are required now.

Such incentives do not need to involve large government spending. What is typically required is an offtake agreement of suitable duration and a market recognition that multiple hours of firm, dispatchable renewable energy capacity has a higher value than intermittent renewable projects, at certain times of the day. This is particularly the case for firm capacity overnight.

The USD 862 million NOOR III CSP plant in rural Morocco produces 150 MW, with seven hours storage, and is underwritten by a 25-year Power Purchase Agreement (PPA). 

Direct comparisons between PPA costs of CSP and utility-scale solar PV are not easy. Typically, daytime solar PV PPAs are priced at less than half the cost of CSP. However, a PPA for PV, with 12 hours of battery storage for overnight use, would be ten times the price of a CSP PPA. 

Most PV PPA’s do not guarantee a night-time price, consequently the client pays the difference between the PV PPA price and the night-time market price. At present, this is not an issue in Australia because night-time power costs are low, thanks to coal. But this will change as coal exits the market and nighttime power prices become more expensive.

This question of PPAs is why the latest projects combine CSP and solar PV technologies, creating a combined daytime and nighttime PPA: PV during the day and CSP only at night. Dubai’s USD 3.9 billion Noor Energy 1 project integrates a 700 MW CSP plant with 15 hours of storage and a 250 MW solar PV system. 

It should be remembered that the current renewable costings that make solar PV look so cheap do not include the nighttime storage gap – a gap filled by CSP’s long duration storage. 

Solving the coal problem isn’t easy. But driving rapid deployment of proven CSP technology is one of the best answers we have if we are to successfully decarbonise, at the least cost, while keeping the lights on and the EVs charged up and ready to go each morning.

Author: Dominic Zaal, Director of the Australian Solar Thermal Research Institute (ASTRI).