Nuclear energy: the best option for South Africa?

May 30th, 2017, Published in Articles: EE Publishers, Articles: Energize


Eskom appears to be more concerned with building large new nuclear power stations than in signing power purchase agreements with independent power producers which use renewable energy sources. Energize caught up with energy analyst and managing director of EE Publishers, Chris Yelland, and asked his opinion of what generation technologies South Africa should opt for.

Chris, you are seen by many as an informed energy analyst and your views and opinion are highly regarded by people in the energy sector and the general public. However, there now seems to be a perception that you are opposed to nuclear energy. So where do you stand?

No, I am not in any camp – not in the renewables camp, and not in the nuclear camp. Being labelled in this way is a kind of personalisation of the issues that is unhelpful. It is a sign that the proponents or opponents are unable to address or answer the real issues rationally, and therefore resort to personalisation of the issues, labelling people and putting them into little boxes.

I am certainly not opposed to a nuclear new-build in South Africa on ideological or technology grounds. But there are real issues that both nuclear and renewable energy proponents must deal with.

What are these real issues that must be dealt with by the nuclear industry? Can you elaborate on them?

Firstly, there are public perceptions of political motives, political interference and corruption associated with mega-project procurements. There are widespread public perceptions that things happen in secret behind closed doors, that due process is not being followed, and that there are some rather sinister motives. Whatever we think of these perceptions, whether they are true or not, they actually need to be dealt with.

The high, upfront capital costs, and associated financing and affordability of such mega-projects, is an issue, and one really has to deal with this issue, because it is one of the big drawbacks of nuclear.

We must also fully understand the levelised cost of electricity (LCOE) from nuclear power over the economic lifetime of the plant, taking into account the overnight capital cost, interest during construction, the fixed and variable operating, maintenance and fuel costs, and the costs of decommissioning and waste disposal. The LCOE indicates the overall cost, in R/kWh of the electricity delivered from a nuclear power plant, in order to be able to compare it properly on a similar basis with other technologies.

Nuclear power stations take a long time to build – up to ten to twelve years per reactor – and mega-projects are prone to high cost and time overruns. These realities cannot simply be ignored.

South Africa needs flexibility in an uncertain and unpredictable world, where electricity demand is difficult to predict in the years ahead, and disruptive technologies are on the horizon. Technologies such as wind, solar PV and energy storage may change the rules of the game.

In an uncertain world, is it prudent to commit to a single technology, vendor country and vendor for a fleet of long lead-time mega-power projects for 80 to 100 years? Or would it be better to proceed with multiple, smaller projects with short and reliable lead-times, and lower price tags, that can be ordered and built flexibly to meet changing demand, economic circumstances and technologies?

These issues are not actually nuclear vs. renewables, but issues of inflexible mega-projects vs. smaller flexible projects. So it’s not a question of being anti-nuclear or pro-renewables. It’s a question of giving oneself enough flexibility to deal with the real world in the decades and century ahead.

In the past you were seen as pro-nuclear. Has something changed?

Yes, change is with us all the time.

Up to a decade or so ago, the only non-carbon emitting generation technologies that existed were nuclear and hydro. And up to only a couple of years ago, even though there were alternatives in the form of renewable energy, these were not the least-cost option. Up until two or three years ago, nuclear was, in fact, the least-cost, non-carbon emitting technology for South Africa.

But this has now changed. A tipping point was reached as the price of wind and solar PV energy came crashing down. All of a sudden there are now lower-cost alternatives to new nuclear and new coal power. Nuclear is no longer the least-cost option, and a blend of wind, solar PV, gas and pumped storage can deliver reliable, despatchable, baseload power at lower cost than new nuclear and even new coal power.

What new technology options exist, and what do you think is the correct technology mix for South Africa going forward?

As I see it, there are three broad technology options that we could look at going forward, plus of course blends of all three, as we already have coal, nuclear, diesel and hydro power in the mix.

Firstly, there is the big new-nuclear option for South Africa, to replace the old coal-fired power stations that have to be retired as they reach end-of-life. This option seems to be favoured by Eskom.

Secondly, there is the option of coal, more coal and still more coal power, to replace Eskom’s old, end-of-life coal-fired power stations. Some 80% of South Africa’s electricity currently comes from coal. We have been building coal-fired power plants for decades, and there’s an argument that as we have plenty of coal reserves, we should stick to what we know best, and use our natural coal resources going forward.

Thirdly there’s the option of wind, solar PV, gas and pumped storage. This is a low carbon option, just as nuclear is a low carbon option. But it is also an option to deliver reliable, despatchable baseload power in a flexible way at lower cost than the nuclear option. This is what is termed “flexible power”.

As I have said, there can and will be a blends of all of the above. So, what do I think is the optimal mix? Well, it is actually not important what I think. What is important is that there should be a rational, scientifically-based, transparent integrated resource planning (IRP) process, involving all relevant stakeholders and the public.

We should define upfront the process and methodology to be used. If we commit to this methodology and follow the defined process, we must accept the outcomes and the answers, even if they are not exactly what we expected. So my view is: let the scientists, engineers and planners do their work properly without political interference.

The IRP process is widely used throughout the world, and presents a rational approach to a complex problem. So it’s not a question of what I think. It’s not a question of my gut feelings, or my personal views. What counts is that we do this whole thing in a rational, scientific and properly planned way.

So, if coal and more coal power is not the way forward, many thousands of people employed in coal mining and coal transportation can expect job losses resulting from a move away from coal. Is this a good thing for South Africa?

In my view, the decline of the coal sector is inevitable, as the world moves away from coal to a cleaner, low-carbon future, both locally and globally.

We live in a global village, and South Africa simply cannot continue to burn coal regardless of the consequences to water use, pollution, health and climate change. The world is expecting us to move to cleaner options, and South Africa has made international commitments to do just this. We need to plan ahead and address these matters going forward.

Job losses in the coal mining and coal road transport sectors are inevitable as Eskom decommissions its aging coal fleet and replaces this with cleaner technologies. In the next 30 years, a significant part of the coal fleet will be decommissioned. So what do we replace it with? More dirty coal? Or do we look at cleaner options such as renewable energy, gas, hydro and/or nuclear?

So the question should be: How do we deal with the socio-economic consequences, and the need to develop a competitive, inclusive and growing overall economy, and to replace smoke-stack industry jobs with better, higher value-adding jobs in a new, modern and clean economy?

A number of people speak about distributed generation rather than centralised generation. What do you think is the future of rooftop solar PV systems in South Africa?

The growth of rooftop solar PV in domestic, commercial and industrial applications has not been considered in the Draft IRP 2016 at all, and yet is a growing and inevitable reality, both globally and in South Africa.

The Department of Energy, Eskom and municipal electricity distributors ignore this growing alternative and supplement to conventional grid electricity at their peril. This is potentially a huge disruptor to the traditional business models of power utilities.

Customers are choosing cleaner and cheaper sources of energy to reduce both their costs and dependency on public utilities. Thus I expect very significant growth in this market as solar PV and battery storage prices continue to drop, while the price of grid electricity continues to rise.

Utilities have to sit up and take note. Otherwise they may find themselves in a death spiral, where rising costs of grid power drive their customers away to alternatives. As people move to these alternatives in greater numbers, so the costs of the new alternative technologies come down due to increasing economies of scale. At the same time, in a vicious circle, this further pushes up the price of grid power, as utilities try to recover their fixed cost structure from declining kWh sales volumes.

This really needs to be taken seriously. It has happened in other parts of the world, and it’s not unthinkable that it could happen in South Africa too.

Send your comments to


  • Coetzee Rudi

    Hi Chris
    Can we afford to wait for a more modern approach such as Thorium Reactors? I believe that this concept will be available 10 years?

    • Ben Franklin

      Thorium reactors have been around for over half a century

      • Coetzee Rudi

        Yes, agree but I’m referring to the present research being carried out by China and India.

        • Ben Franklin

          India has been seriously looking at thorium reactors since at least the 1960s. Molten salt Th232-U233 reactors have been extensively researched at Oak Ridge NL, and showed substantial promise. Yet was discontinued.

          It seems doubtful that these risky technologies can now be resurrected after so many decades. Properly assessing and addressing the risks is hugely expensive.

          See eg for a credible scenario with spent fuel that could threaten 8 million in the USA, yet is not acted on by the US NRC because of the huge costs of doing so.

          • Coetzee Rudi

            Interesting, thank you. What is the Steenkamp Thorium Ltd (100MW) power plant then for?

          • Ben Franklin

            There are many concepts for new nuclear technologies. Ask Bill Gates, for example. Which if any of these will deliver power, remains to be seen. Throughout the 1960s to 1980s there was much talk about the LMFBR.

            When the French built their 1.2 GW SuperPhenix, it appeared on the front covers of Scientific American, National Geographic Magazine . . . with statements from the most prominent nuclear reactor scientists & engineers predicting that this would completely “transform” the electric power industry.

            Yet that same hyped (and hyperexpensive) reactor was a dismal flop, and over its brief lifetime had an availability factor of a mere 7.8%.

            Against some modern wind farms which have measured lifetime capacity factors around 48%. See eg the 400 MW Anholt 1 and the 209 MW Horns Rev 2 cases . Or some SA wind farms with capacity factors well over 40%.

  • Evert Swanepoel

    I think Chris has analyzed the situation accurately.The cheaper rooftop solar becomes, the more people will go off-grid and the resulting drop in revenue will have dire consequensec for the Municipalities and Eskom.

  • Neville Morrison

    Neville Morrison – As always Chis you are 100% on target

  • Ben Franklin

    An excellent balanced view.

    The 450 MW Tamarugal CSP (concentrating solar power), with its 13 hours of energy storage capacity will generate power for 24 hours per day at $63/MWu = $0.063/kWu = R0.82 per kWu (@R13/$)

    Such dispatchable CSP should be considered an alternative to pumped hydro storage — especially in arid regions where pumped hydro is not always available.

  • Alan Cargill

    The cost of Ingula was $3.5bn. I find it interesting and misleading that pump storage schemes are referenced in terms of there maximum power output, ie 1.3GW. Ingula is essentially a battery and so should be rated in terms of WHrs, ie approximately 2GWhrs (I believe). This makes the cost of Ingula about 10 times the present cost of Li ion batteries per Whr. If we consider that the cost of batteries is continuing to drop, and the cost of pump storage will continue to rise (for instance as the best sites are used) then that factor of 10 is only going to become larger.

    There are advantaged and disadvantages withe both types of storage. The most significant on Li batteries being a lower life. Pump storage loses on two other important measures, round cycle efficiency and charge time.

    All things considered, as an engineer, I cannot see any reason why we would build another pump scheme when Li batteries are becoming available at grid scale.

    Looking further towards CSP,

    • Ben Franklin

      Most relevant comment. If the cost & capacity (GWh) figures for Ingula are correct, then its cost is $1.75/Wh = $1750/kWh = R22 750 @R13/$. Far above the cost of Li-ion storage batteries capable of 5 000 to 10 000 charge-discharge cycles.

      Li-ion batteries are moreover more compact, and are site-independent. Thus they may be located near PV or wind farms, or near load centres (cities or industries).

  • Alan M

    I think you will wait a long time for a Li ion or Li iron battery as big as an Ingula machine, of which there are four of over 300MW each. Eskom, currently in a surplus capacity mode, should be looking at long lead time capacity build aimed at replacing coal powered generators, like nuclear plant, and not signing up wind and solar farms that feed in poor quality electricity with no ability to clear grid faults or supply sudden surges like starting furnaces and similar heavy user loads, which a grid for mining, beneficiation, rail traction, and heavy industry needs. Wind and solar farms of a mere 100MW are too small to call utility scale in Eskom’s grid, and have to feed in from behind high impedance step-up transformers through overcurrent-limiting inverters at low distribution voltages like 132kV. Wind and solar farms have short lead times, and can be ordered much later than the substantially sized and firm capacity generators Eskom will need first.

    Eskom will need to replace coal power with nuclear to achieve economies of scale with steam generators that do not produce mountains of waste each year, as coal energy does. A Koeberg unit produces only 25 tons of spent fuel yearly, an asset that needs to be stored for future re-processing, not like coal ash dams that contain radioactive material and toxic matter that can be leached into underground water sources. A Koeberg station is equivalent to half a Medupi or Kusile, each of which will produce several hundred thousand tons of ash waste yearly.

    Combustion turbine powered units are only economical when used in combined cycle generation with steam turbine units, which are subject to the same cooling water needs as their bigger nuclear cousins.

    To save water the new firm capacity plant should be coastal, to use sea water for condensing the turbine exhaust steam. Inland plants that use dry cooling for condensing exhaust steam suffer from derating when ambient temperatures rise.

    It will be advisable to locate firm capacity bulk generating plant close to heavy load centres, as far as possible, to minimise losses in the high voltage transmission lines.

    Leave the wind and solar farms their niche market, micro-grids with no economic value to the national grid and hence no mains electricity. And don’t allow more renewable energy powered electricity unrestricted access to the grid whenever their ‘free’ energy is delivered. If they need back-up from spinning reserve plant the cost of grid services Eskom, and hence its paying customers, have to provide is simply too expensive, and if the spinning reserve plant could supply the electricity anyway, the renewables should be curtailed as their added cost-to-grid is wasteful and parasitic expenditure to Eskom and customers, resulting in tariff increases and claw-back claims.

Subscribe to our leading email newsletters


CLICK for other EE Publishers information products