/ 23 August 2015

Solar puzzle: We’ll follow the sun, but tomorrow may rain

Grid technology: Cities like Johannesburg are conventionally powered
South Africa’s buildings account for 40% of the country’s greenhouse gas emissions.

A tower rises up from Stellenbosch University’s Mariendahl experimental farm, against the backdrop of scenic mountains. Facing it like supplicants, more than a hundred mirrors stare at the tower, automatically tilting to catch the sun.

This is the Helio100 project, funded by the Technology Innovation Agency (TIA) and a test-bed for concentrating solar power (CSP) technology. For many in the renewable energy sector, CSP offers a solution to the perennial problem of renewables: What happens when the wind isn’t blowing and the sun is hidden behind clouds or by nightfall?

CSP is the solar equivalent of the Drakensberg Pumped Storage Scheme, in which water from a dam is pumped to a dam with a higher elevation: when electricity is needed, the water from the higher dam is released, to flow down to the lower dam, generating electricity as it goes. With CSP, the sun’s energy is stored – currently within a molten salt mixture – and then released when it is needed.

According to the International Energy Agency, “by 2050, with appropriate support, CSP could provide 11.3% of global electricity … In the sunniest countries, CSP can be expected to become a competitive source of bulk power in peak and intermediate loads by 2020, and of base-load power by 2025-2030.”

This is good news for a country like South Africa. The nation has one of the highest rates of direct normal irradiation, a measure of sunniness. But the difficulty is: How do you harness and store this solar energy?

Molten salt mixture
In a CSP plant, mirrors position themselves independently to best reflect the sun’s rays to a receiver on top of a tower, converting the solar energy (concentrated solar flux) into thermal energy (heat). This heat can either be stored, usually in a molten salt mixture, to be released later or used immediately to heat water to superheated steam to power turbines – similar to what already happens in a coal-fired power station, except that it uses the sun’s heat instead of burning coal for energy. This technology is different from photovoltaic (PV) technology – such as solar panels on a roof – which directly converts sunlight into electricity.

In 2009, Thomas Fluri – a post-doctoral researcher at Stellenbosch University, but now the head of the solar thermal power plants and high temperature group at the Fraunhofer Institute in Germany – determined the potential for CSP in South Africa by identifying areas that had enough sun to justify the deployment of the technology.

“The identified suitable areas could accommodate plants with a nominal capacity of 510.3GW [gigawatts] in the Northern Cape, 25.3GW in the Free State, 10.5GW in the Western Cape and 1.6GW in the Eastern Cape, a total potential nominal capacity of 547.6GW for the whole country,” he writes in an article published in the journal Energy Policy.

For context, this is an order of magnitude larger than the electricity produced by South African power utility Eskom’s fleet of 20 power stations.

The country has a natural abundance of the fossil fuel and relies heavily on coal, with the majority of its electricity coming from Eskom’s coal-fired power stations. This is one of the reasons why South Africa has the largest greenhouse gas emissions on the continent. Yet, after it signed a pledge at the United Nations’s 17th Congress of the Parties climate talks in 2009, the country is under pressure to reduce its emissions, after having committed to cutting its emissions by 34% by 2020, with certain provisos.

To achieve this goal, South Africa will have to include more renewable energies in its power mix. The country’s energy road map – the Integrated Resources Plan 2010 – makes provision for another 9.6GW of nuclear energy, 6.3GW coal, 11.4GW renewables and 11GW from other sources. An updated IRP2010 was mooted in 2013, but this has yet to be turned into a policy document, although in this updated document the portion of electricity generated by CSP grows from 600MW in the original IRP2010 to 3 300MW in the new version.

According to a 2013 report by German development agency Deutsche Gesellschaft für Internationale Zusammenarbeit, the Southern African Solar Thermal and Electricity Association and the South African department of trade and industry, “A significant benefit of CSP is that it has little environmental impact – a solar-only plant has almost no greenhouse gas emissions in operation and occupies a similar or smaller portion of land compared to PV.”

In their report prepared by consulting company Ernst & Young and energy company Enolcon, the writers say that CSP power stations can have a lifespan of up to 60 years, similar to coal and nuclear power stations.

Less mature technology
But CSP is still a less mature technology than most others. “Concentrating solar power is the new kid on the block,” says Helio100 project leader Paul Gauché, “which means it is still more expensive.” But Helio100, different to other CSP heliostat technologies, is nevertheless “showing signs of being the cheapest heliostat system in the world”.

There are four different types of CSP systems: parabolic troughs use bent mirrors in long lines to heat many kilometres of piping filled with a heat-absorbent liquid; sterling dish systems that look like large DStv dishes and point at the sun at all times, but have not yet been commercialised; similarly, the linear fresnel system – which look like the parabolic troughs but with flat mirrors – has not been commercialised; the final option is a solar tower surrounded by moving heliostats.

A heliostat – which comes from the Greek word helios for sun and stat for stationary – behaves like a sunflower: it tracks the movement of the sun, moving to face the gigantic nuclear reaction in the sky and reflecting its energy. Helio100 has 120 rectangular, monolithic mirrors. “Our pilot is going to be close to 300m2 of mirror,” claims Gauché – that reflect the sun’s energy on to the top of the 12m tower.

heliostat system in Stellenbosch
The heliostat system in KaXu or (pictured) Stellenbosch doesn’t require big engineering earth works. And it’s possibly also the cheapest system of its kind in the world. (Abengoa Solar)

“Helio100 is a culmination of a big systems engineering project,” explains Gauché, who is also founding director of Stellenbosch University’s Solar Thermal Energy Research Group. The heliostats are effectively smart robots that “know the angle between the sun and the tower, depending on the time of day, and know where the sun is with respect to the tower. They each know this independently,” he says.

But he says what makes the Helio100 system “unique” is it doesn’t require “big civil engineering earthworks. It has a unique structural frame that can be plonked down. That’s our fancy word,” he says, “‘plonkability'”.

Meanwhile, on the roof of Stellenbosch University’s engineering building, Cebo Silinga, a researcher from the Centre for Renewable and Sustainable Energy Studies, shows me one of the early frame prototypes. Fortunately, the day is overcast, otherwise I would be blinded by the sheer quantity of reflected light. As it is, I am forced to squint to take in the large area littered with numerous solar water heaters and student projects. In the far corner, just in front of the control room, a large frame – a triangular base, with a scaffold rising from it – sits inert, cordoned off by red-and-white poles. Heliostats, each 1.8m by 1.2m and covered with brown canvas, rise out of it – at its centre and around the edges.

Above the frame, more heliostats are attached to the roof of the control building, all of them staring blindly – they are also covered with brown canvas – towards an 18m-high lattice tower on the rooftop lab.

Silinga points out the heliostats, stating that it was a petrochemical company Sasol-funded project to create a moveable foundation for heliostats and more on the roof. Gauché says: “The end result is a heliostat field, providing just over 40m2 of mirror area … christened the ‘Sasol Helio40 System’.”

Gauché believes that these foundations are one of the features that makes Helio100 unique. “There is virtually no ground preparation,” he says. “And you can move it.” The heliostat mirrors are also very light and can be “lifted with two hands”. “The entire heliostat field has been developed to be manufactured by mostly unskilled labour and installed by only two people,” says Gauché.

He uses the example of a mine using this CSP system: “Mines can’t guarantee that they can operate for 30 years, which is how long it can take to pay off a CSP plant. This system is smaller and lower risk – you can just pick up your plant.’

But the Helio100 project is not yet on a commercial scale, Gauché says. “Helio100 is in the technology development phase, one step prior to industrialisation and commercialisation.”

This is mainly thanks to R15-million from the Technology Innovation Agency, a government agency tasked with funding innovation and technology development. However, Gauché says 10 to 15 master’s and doctoral students – funded by the National Research Foundation and industry – have graduated as a result of work done on the project.

“Several other related campus projects are approaching similar maturity, with the intention of developing more components toward an entire solar system,” he says.

heliostat system in KaXu
The heliostat system in KaXu. (Abengoa Solar)

In 2015, South Africa inaugurated its first commercial solar thermal electric plant, called KaXu Solar One (KaXu means “open skies” in the Nama language). The plant – the result of a public-private partnership with Spanish company Abengoa, the KaXu Community Trust and the Industrial Development Corporation with a total price tag of R7.9-billion – has been delivering up to 100MW of power to the national grid since its inception. This is enough to power about 80 000 homes.

The faces of the 120 parabolic troughs look as though you have taken the long sides of a sheet of paper and pulled them towards each other: a shiny tilted half-pipe, with the bottom half reflecting the deep blue of the Northern Cape’s cloudless skies, the top half mirroring the desert sand.

The colour inside the 1 200 mirrored parabolic trough collectors changes as they track the sun, transferring that energy first into receiving tubes filled with heat-absorbing liquid and then fed through large pipes to a turbine, where the heat-filled liquid causes water to steam, turning the turbines to generate electricity. And Eskom, through its renewable energy independent power producer procurement programme, has agreed to buy the electricity that the station will produce.

KaXu’s solar field covers about 1km2, and will have the storage capacity to provide electricity for about 2.5 hours after sunset, or before dawn. And storage – aside from the environmental benefits – is one CSP offering that has governments and power producers drooling slightly.

In the Ernst & Young/Enolcon report the authors write: “CSP, with thermal energy storage, has an advantage over other renewable technologies due to its predictability in dispatch. It can support peak periods, which is when power is most needed on the grid system. Power available during peak periods is considered to be of higher value than power outside of peak hours.” In South Africa, that peak period is between 5pm and 9pm every day.

Convenient low pressure At the moment, most solar thermal storages use molten salt sourced from Chile. According to SolarReserve, an American renewable energy provider, molten salt is the ideal heat-capture medium because “it maintains its liquid state even above [about 540°C], allowing the system to operate at low pressure for convenient energy capture and storage”.

If the medium were to turn into a gas at these temperatures, it would significantly increase the pressure in the system, which is not designed for extremely high pressures.

“After passing through the receiver [which is where the sun beams are focused], the molten salt then flows down the piping inside the tower and into a thermal storage tank, where the energy is stored as high-temperature molten salt until electricity is needed,” states SolarReserve. This company is present in South Africa, and is responsible for the PV Letsatsi Solar Power project, just outside Bloemfontein.

But, according to research coming out of Stellenbosch University’s Centre for Renewable and Sustainable Energy Studies, there could be another, more cost-effective option for South Africa: rocks.

Centre graduate Kenneth Allan, in a paper published in Solar Energy Materials & Solar Cells, investigates the use of a rock bed to store the thermal energy collected using CSP. Air is blown into and out of the rock bed to move the energy into and out of it. Rock has a number of benefits, including the low-cost factor and that many rocks have a higher threshold temperature than molten salt, which is about 530°C. This temperature cap restricts the amount of energy that can be stored.

An energy storage system also needs to be able to cope with wear and tear. “If a packed bed undergoes one full charge-discharge cycle every day, 365 days a year, in 10 years, the storage will undergo 3 650 full cycles,” Allan writes in collaboration with other authors. “Any rock that weathers rapidly and forms dust or sand … that fills the air passages in the bed is undesirable.”

Cooking and cooling rock in a kiln
So Allan collected rock samples from throughout the Northern Cape, and tested their thermal characteristics by cooking and cooling them repeatedly – nearly a thousand times apiece – in a kiln.

The study found that dolerite, also known as ysterklip in South Africa, where it is relatively abundant, was the most promising rock, but that more research was needed to characterise it as a storage solution. Gauché says the pack-bed system looks promising, and is “an order of magnitude [10 times] cheaper than molten salt systems”. But it is not part of the Helio100 system – yet.

“Helio100 will initially just be a central receiver collector system – heliostats, tower, cooling system, control. We’re planning to add storage as the next phase,” says Gauché. “We have two patented rock-packed bed ideas that we are using to design a prototype storage system.”

But, even with these innovations, the question remains: How do you get CSP-generated electricity on to the grid?

Although the Northern Cape is one of the best places in the world to generate electricity from solar power, it is also far away from the country’s populous economic areas. In fact, it is one of the largest and most sparsely populated areas in the country, which means that the areas are vast, but the infrastructure is scanty.

However, Fluri states in his article about the potential of CSP in South Africa that “it is usually assumed that CSP plants would be located in the Northern Cape … because of its excellent solar resources and its vast, sparsely populated areas. As a result of this assumption, policymakers in other provinces are likely to take little interest in CSP.”

But his study, which identified areas in South Africa that “get sufficient sunshine, are close enough to transmission lines, are flat enough, their respective vegetation is not under threat and they have a suitable land-use profile”, also found parts of the Western Cape, Free State and the Eastern Cape suitable for CSP.

Plan for bad weather
In fact, a number of studies argue that a network of geographically dispersed CSP plants is preferable to them being located in just one region, because there is a contingency plan for bad weather.

“You can build renewable energy systems within spitting distance of the power substations,” Gauché says. “We don’t need to be in Pofadder. That’s the beautiful thing about the solar resource in this country – it is highly distributed.” He cites the Cape corridor transmission line, an electricity pipeline that runs from the coal-fired power stations in the north of the country to the Cape in the south: “It’s a diagonal across the country, on the line of how the cold front weather patterns work.”

The idea is that renewables on the northern end of the line could add electricity to the grid while the south is covered in cloud. As the cloud front moves north, the southern renewables, which would then be exposed to the sun, could power the grid.

But the main problem for CSP is research and the investment that funds it. Because the technology is nascent, it requires further investment to take it to the point where it is fully commercially viable.

Gauché, in conjunction with fellow academics at Stellenbosch University and the University of the Western Cape, writes in an article published in the journal Development Southern Africa: “CSP is still entering the commercialisation phase and, as the learning curve sets in, the cost is expected to decrease significantly… Local economic benefit requires a certain minimum annual capacity to be committed for a sustainable period of time. Such a commitment could trigger an internal self-sustaining CSP industry that realises all of the preceding value propositions.”

This is an edited extract from Sarah Wild’s book Innovation: Shaping South Africa Through Science, which will be published in September by the Gordon Institute of Business Science and PanMacmillan South Africa