Private power

The South African Government is aiming to double the country's energy output in the next 17 years as it seeks to end a heavy reliance on coal
and susceptibility to power outages. To create the necessary generation capacity, the government had proposed a building programme that included six nuclear stations; however, neither it nor South African energy utility Eskom could fund this. Instead, independent generation schemes funded by private investment capital are to fill the gap. In May 2011, the country's Department of Energy (DoE) published its Integrated Resource Plan (IRP) outlining the new capacity required for the next 20 years and how much should be generated by each type of technology.

Quick to build, creating local jobs and helping reduce the country's per-capita CO2 emissions, renewable energy makes up a significant chunk (17.8GW by 2030) of the IRP via the Renewable Energy Independent Power Producer Procurement Programme (REIPPPP) (see 'Renewable Energy Independent Power Producer Programme', page 54). This is a competitive bidding scheme that sets a maximum cost per kilowatt hour for each generation technology. Successful bidders then supply electricity to Eskom's Single Buyer Office at the agreed rate. Five REIPPPP rounds are planned, with the first having taken place in 2012 and the last due in 2016. The first two rounds allocated 2,440MW to 48 preferred bidders. A third round, already underway, will take the total to 3,725MW.

The REIPPPP has, however, encountered a rocky road thus far, starting with the DoE's abandonment of its initial feed-in tariff approach. There was also a lack of competition in round one, and the extra competition in round two led to lower bids. Solar PV dropped by 40% from R2.75/kWh to R1.65/kWh, wind fell by 22% (R1.15/kWh to R0.9/kWh) and concentrated solar power (CSP) by 6% (R2.68/kWh to R2.51/kWh).

There have been no takers for the biomass and biogas allocations, either, something Yousuf Haffejee, Eskom's head of single buyer office, puts down to the ceiling price not being close enough to market rates. Financial closure for the first round was also delayed twice from its original June 2012 date, eventually taking place in December 2012 (the second-round financial close also slipped to the end of April 2013). According to the DoE, the government needed to arrange financial support for Eskom to cover its obligations under the power purchase agreements (PPAs).

"There were some concerns about Eskom as a secure off-taker," says Scott Burger, analyst at GTM Research. "I think the newness and scale of a market that has gone from nothing to almost 1.5GW of solar alone in a very short period has caused a lot of understandable nervousness among investors."

Delays have also affected the third round. Bids due by October 2012 were first postponed until May of this year, then 19 August. The government blamed several factors, including the need to incorporate the lessons learned in previous rounds into round-three bidding and having to answer the many questions from bidders.

"The financial close for round one took place on a single day and that became a strain on the financing banks," says Chris Ehlers, business
director at Saudi Arabian-based ACWA Power, referring to one of the lessons the government had to take on board, although spreading the close over multiple days can result in uncertainty over the final PPA exchange rate, making life harder for external investors.

Ehlers adds: "Going forward, foreign banks will have to come to take some of the share. There is huge interest from foreign banks and that is something we would be happy to support."

Concentrated solar power

With more than 1GW each in the first three rounds of the REIPPPP, wind and solar are the two main renewable technologies, with CSP getting 200MW.

The 50MW Bokpoort plant was the only CSP project in the second round. Originally planned as 75MW, the 150MW awarded to the KaXu Solar
One (100MW) and Khi Solar One (50MW) projects meant Bokpoort's capacity had to be reduced to meet the 200MW limit. Generating more than 200GWh annually, the plant uses parabolic trough technology and has the largest storage capacity of any CSP project of comparable size to date: nine hours (1,300MWh of thermal energy) using molten salt technology. A joint-venture company co-owned by ACWA Power and local developer Solafrica will build and then operate the plant. Commercial operation is planned for Q3 2015.

ACWA has strived to meet South Africa's localisation and black economic empowerment (BEE) requirements, partnering with a local BEE party, Invest in Africa Energy. ACWA is funding equity stakes for investors such as the Kurisani Youth Development Trust as well as a local community trust. The total 10% community shareholding far exceeds the minimum 2.5% required, and the joint-venture partners will also invest $2 million annually in the community.

Bokpoort will create around 50 long-term positions, with most of the 650 construction jobs going to local workers. Parts such as pressure vessels,
storage tanks, piping and cabling will be sourced locally and make up 35-40% of the development. More specialised CSP equipment such as parabolic mirrors, heliostats and receiver pipes are less easy to manufacture locally, and the viability of producing assemblies such as CSP collector structures very much depends on the future amount of CSP to be built under the REIPPPP.

Meanwhile, the low cost of PV modules since 2011 has made it tougher for CSP to compete. According to GTM Research, in high solar irradiation areas, and in ideal conditions, prices of around $0.20/kWh for CSP with storage versus $0.12/kWh for PV are now typical.

"Many CSP manufacturers are now looking at alternative markets like desalination, although countries like Saudi Arabia, Morocco and the UAE are still initiating aggressive programmes to build out CSP," says Burger.

According to Ehlers, ACWA's approach of using third-party EPCs to build CSP plants leads to significant cost reductions and he predicts further drops in CSP tariffs, with Fresnel technology promising the lowest generating prices.

"Our business model is completely different to traditional 'one-stop-shop' CSP developers," he says. "You can't put cost pressure on your own EPC like we can on the open market. Our tariff has been 19% lower than the next best bidder."

CSP may struggle to get near PV's energy cost, but Bokpoort's storage capability enables it to match supply to South Africa's peak demand profile. Storage is responsible for Bokpoort's extra capital cost: R90 million a megawatt compared with R76 million for the two round one CSP projects. Ehlers says that Bokpoort's kilowatt hour rate is "20% rather than 6% lower than round one if you take the storage capex into account".

"There's currently no premium in the programme for this," says Ehlers. "It's a big issue for us and we are talking to the government about it. It's not fair that we don't see any benefit from our major investment in storage. It would help make a better case for future investment."

"In California, PV developers get a higher cost for energy generated at peak times," says Burger. "My assumption would be they would build the storage costs into the flat-rate tariff they agreed. It comes down to how ACWA negotiated the contract initially."

However, CSP will certainly feature in future REIPPPP rounds. In late 2012, the government announced a further 3,200MW of renewable generation
capacity, including 400MW of CSP. Speaking to Engineering News in early April 2013, DoE spokesperson Thandiwe Maimane said the department was "mindful of the need to allocate additional megawatts for CSP". Eskom's Haffejee has also said there is a need to "create some flexibility in renewable resource categories and allocation".

Despite the delays and teething troubles, there's little doubt that the REIPPPP has been a success: investment in rounds one and two totals €7.3 billion. Many challenges remain, but with South Africa's excellent solar resources, both CSP and the REIPPPP look like having a bright future.

"Confidence is definitely improving - this has been a learning process for them all," says Burger. "Given the nascent nature of these markets, this is very common. The third round will go better than the first."