On Friday 1 June 2018, the Minister of Energy, Jeff Radebe, hosted an Energy Sector Round Table Stakeholders Engagement. I followed the discussions intently, and especially the opening remarks from the honourable Minister was forward thinking and optimistic.
In terms of nuclear energy, the Koeberg nuclear power plant was praised for its contribution of electricity to the national grid (around 5%) as well as the SAFARI-1 research reactor for being one of the world’s largest contributors of medical isotopes used for cancer treatments.
The opportunity for exploring shale gas in South Africa as a resource to support economic growth as well as contribute energy to our energy sector has been highlighted. Another bid window (bid window 5) was also announced, with a total installed capacity of 1,800MW at an estimated cost of R50 billion. A commitment has also been made to finalise the Integrated Resource Plan by August 2018, as well as to make a draft available for public consultation well before this date.
In reality, our single biggest challenge moving into the future is to mitigate carbon emissions in a cost-effective way, shale gas, however cleaner than burning coal, does not fit into that picture.
As someone who is working with nuclear energy and has quantified the potential it holds for industrialising economies, I believe that nuclear energy will feature in the new Integrated Resource Plan. However, during the engagement, Prof Anton Eberhard eluded to the fact that the energy sector is a highly contested space.
On the other hand, I believe that all energy sources should be able to contribute on equal footing and pitting renewable energy against coal, or nuclear is not helpful in developing energy solutions of the country, an energy mix of all energy sources is crucial for South Africa.
It did however cause me to wonder, how nuclear energy would stack up against the often-lauded Renewable Energy Independent Power Producer Programme with the most recent power purchase agreements signed (bid window 3.5 and 4) on 4 April 2018.
To do this, I decided to look at the current REIPPP power purchase agreements in place and compare that to the 9,600MW nuclear build proposal that was often called too expensive and controversial by proponents of the project.
The REIPPP programme with the fine print
To have a fair value comparison, we also need to understand the nature of these energy sources. Most of the REIPPP energy sources are intermitted energy sources, in other words, the availability of these sources is determined by the cycles of nature, and not by the grid control centre. For this reason, the installed capacity, and the average supplied electricity onto the grid vary widely.
According to the National Energy Regulator of South Africa, “Monitoring Renewable Performance of Power Plants, Issue 10, from September 2017, the following capacity factors was achieved: Wind: 31%, Solar PV: 23%, Solar CSP: 25%, Small Hydro: 63%. According to the draft IRP, the capacity factor for Landfill Gas is estimated at 80% and Biomass at 70%.
This means that on average, these percentages of the installed capacity available over a year. This results in an average capacity factor of 28% for the REIPPP bid window 1 through 4 and an average capacity of 1773,6MW from the installed 6328MW. 88,4% of this 1773,6MW consist of intermitted solar and wind. Table 1, depicts the details of these bid windows.
Table 1: REIPPP contribution to generation capacity to the grid
According to the Department of Energy the total value of these projects are R201,8 billion (bid window 1-3) and R58 billion (bid window 3,5 and 4) with a total of R258,8 billion for the available 1,773MW capacity.
These plants will typically last for between 15 and 25 years. Additional systems and cost is necessary to shift the supply to meet the demand profile of the country (storage). It is also important that these figures are committed to, guaranteed by treasury and will be paid off by Eskom over a 20 year period.
Bid window 1 through 3 contributed a total of 32,532 job years since inception in 2011, this equates roughly to 8,133 full time jobs, since the start of these projects. For bid window 3,5 and 4 provide roughly 2,338 full time jobs based on this experience.
It has been claimed however that it will produce, 58,000 jobs, a figure highly questionable based on operating experience and also the main source of some controversy as unions tried to block the signing of the agreements with legal action in February 2018.
The Nuclear build programme, without the controversy
Nuclear energy on the other hand, has its own limitations, its availability is often called dispatchable baseload.
This means that although it is always available, economically, it would not be wise to ramp the power up and down, as the cost of unserved energy is often higher than the cost of served energy. In 2017 the Koeberg Nuclear Power Plant achieved a capacity factor of 92%. For this reason, other energy sources are often supplementing it.
Taking the proposed 9,600MW of new build nuclear and dividing it into four nuclear power plants (or bid windows) consisting of 2 x 1,200MW reactors each, we see each window supplying 2,208MW of dispatchable baseload onto the grid. This programme is depicted in Table 2.
Table 2: Nuclear new build generation capacity potential to the grid
The average build times of these type of reactors is about 8 years although we see this number coming down to 5 years in countries with that has built up experience in new-build projects. This compares well to the REIPPP time frames.
Although disputed as being excessive by nuclear experts, the 9600MW nuclear build programme has often been quoted as having a price tag of R 1 trillion (R 1000 billion). If we accept this figure (Egypt recently signed 4800MW contract with Rosatom for $30 billion – R 780 billion for 9600MW), this translates to R250 billion for 2208MW equivalent capacity dispatchable – as compared to the already committed R258 billion for the 1773,6MW equivalent intermitted electricity from the REIPPP programme. Again, nuclear energy is competing very well in deed. What makes this even more competitive, is the fact that a nuclear plant is designed to last for 60 years.
An interesting comparison is also to see what intermitted bid window 1 through 4 would cost if it was to supply an equivalent of the 9,600MW nuclear on the grid (albeit not yet dispatchable) – this figure comes to R1,406 trillion, 40% more expensive than the dispatchable nuclear source.
Granted, nuclear power requires nuclear fuel and as a result the fuel and O&M cost is higher than that of renewable energy sources. To put this into context, the Egypt contract with Rosatom includes an additional $30 billion for the entire 60-year life cycle cost, including fuel supply and repatriating the spent fuel to Russia.
That is a total of $60 billion (R1,56 trillion for 9,600MW) full 60-year life cycle cost. Of course, this would be R390 billion for the full 60-year life cycle of 2 unit 2,400MW nuclear power plant, or R330 billion for a bid-window 1 through 4 equivalent (and dispatchable!) supply.
According to the NIASA Skills report, with the exclusion of the site indirect labour, the Nuclear Steam Supply System (NSSS) vendor and subcontractor staffs, the contract staff and the vendor start-up personnel, the total requirement for a 9,600MW nuclear build is 32 800 persons during the construction phase, with an additional 36000 indirect jobs being created. For a single 2 unit plant, this becomes, 8,200 direct and 9,000 indirect jobs.
It is important to note, that should such a large project be conducted, the construction phase will last for more than 2 decades and some workers might move from one sight to the next. Typically, each plant will need around 1500 full-time staff during operation.
With the Medupi coal plant coming to an end, a number of layoffs are in the foreseeable future with no new mega-project on the horizon, the engineering and construction industry is facing real challenges. A nuclear new build project could benefit from these developed skills.
Looking forward into the future
At the risk of being called out for looking backward (the power purchase agreements for bid window 1 through 4 will last for 20 years into the future), it is also important to look at the future, and the new proposed bid-window, bid-window 5. Assuming the same average capacity factor for bid window 5 as that of bid window 1 through 4 of 28%, the procured 1,800MW capacity will supply 504,5MW equivalent onto the grid, again mostly intermitted sources. This is depicted in Table 3.
Table 3: Proposed bid-window 5 capacity contribution to the grid
|Bid window 5|
|Load Factor||Installed Capacity( MW)||Equivalent Capacity (MW)|
With an estimated cost of R50 billion, this results in a 30% reduction from bid window 1 through 4, and a cost of R951 billion for an equivalent of the 9,600MW nuclear on the grid (albeit not yet dispatchable). The additional cost for shifting the supply to match the demand profile, still need to be factored in.
To do this, a low carbon option is to look into battery storage, this option has recently received a lot of attention after the installation of a 100MW/129MWh battery by Tesla in Southern Australia at a cost of R500 million – the biggest battery ever built. To spread the 28% load factor over a 24-hour period, a 142MW/2,443MWh battery will be required, without providing additional generation capacity.
To do this for bid-window 1 through 4 a 2,507MW/33,588MWh battery will be required, hardly feasible with today’s technology. Never say never. More feasible would be a smaller battery would provide welcome stabilisation during short intermittencies, but not be able to shift supply.
What the future could be
With all these figures, and more importantly, supply configurations available, we now have an opportunity to setup an energy mix to take advantage of all of them to move the country moving forward while addressing the most important aspect - reducing carbon emissions in a cost-effective way, while creating jobs and growing the economy.
Currently the biggest weakness of the REIPP program is the intermitted nature of the supply sources, it will always result on over and under supply during varying conditions and load profiles. Adding a dispatchable low-carbon emitting energy source such as nuclear to the foundation of the mix, this will reduce the load on natural gas during low supply times, while the grid controllers will have the option to run the most economical generating mix during high supply times.
If nuclear power is measured on its merits, it will always provide high value, electricity to the market, not to mention that an inflated R1 trillion nuclear project still undercuts the current REIPP program with 40% in capital cost.
About the author
Dr. Anthonie Cilliers Pr.Eng
National Coordinator: South African Network for Nuclear Education, Science and Technology (SAN-NEST); Regional Coordinator: AFRA NEST; Honorary Research Fellow: University of the Witwatersrand