A case for nuclear in SA

South Africa is right to procure nuclear power, and it will not cost R1 trillion or more as suggested in the media. By Kelvin Kemm.


South Africa is embarking on a major project to build nuclear power plants. This is the correct path to follow and is a well thought out, carefully-crafted project plan.

Unfortunately, “nuclear power” creates huge emotion in people. There is a rather — can I say it? — traditional anti-nuclear lobby which came about at the time of the rapid building of nuclear weapons. I feel that this original anti-nuclear sentiment from weapons has continued on to nuclear power plants.

There is no connection between nuclear weapons and nuclear power, except for the word “nuclear”. There is no connection between hot curry and hot cars either.

Nuclear weapons are designed to explode; to do that a country has to be able to enrich uranium beyond the 90% mark. It is this high-level uranium enrichment potential which is the major worry that much of the world has with the nuclear programmes of Iran and North Korea.

Power reactors run on enrichment figures of 5% to 10% — such uranium cannot be used for weapons. Nuclear reactors cannot explode like nuclear bombs. It is possible to have fires or floods in nuclear plants, like what happened in Chernobyl and Fukushima and, like in any industrial plant, fire and water can really mess up one’s day.

Both those reactors were old and obsolete by modern standards. The reactors of today are known as Generation Three and Generation Three-Plus.

These new reactor designs have been fundamentally altered to use the natural laws of physics as safety factors.

Fukushima had a waste fuel storage pond above ground. So, when it cracked due to the earthquake, the water ran out and the waste fuel overheated because the water had been a coolant. Gen III passive safety reactors are designed, for example, with tanks of water above them, so if the lights go out, as it were, the water will fall down naturally onto the reactor for long enough to provide emergency cooling.

Modern reactors have had extra features added, but they have also had other features simplified and some parts have been taken out to lessen complexity and so lead to a generally safer design.

Cost and construction

I am a nuclear physicist, not a finance expert, so I don’t claim to be an authority on international finance. However, so many simplistic things are stated in the media, and claimed by anti-nuclear activists concerning nuclear costs, that some sense needs to prevail.

South Africa has sensibly decided on a fleet approach to nuclear power. “Fleet” means having a mentality now of multiple construction and one skilled labour team moving from one site to the next, to provide continuity and to learn and improve on the job.

The country is planning for an additional 9,6GW of nuclear power, which translates into three nuclear power stations, each of which will have two or three nuclear reactors, depending on which type we decide to construct. I personally do not believe that 9,6GW is enough — we should have 12,5GW on the table now, but it is easy to increase in capacity the future if we have a fleet mentality to start with.

Henry Ford discovered that having a motorcar production line produced cars more cheaply than building one car at a time.

My next point: South Africa is not “buying” nuclear power plants from anybody.

Construction teams, composed mostly of South Africans, will build the nuclear power stations.

The construction will take place in collaboration with foreign companies, and we will build the nuclear power stations to some plan, provided by another country.

Built in South Africa

South Africa currently does this with motorcars. South Africans build and export German, Japanese and American cars all over the world. They are built to German, Japanese and American plans and standards, but the welding, bolting, painting and assembly is done by South Africans. Yes, there are foreign company engineers and representatives coming and going all the time, but they don’t actually make the cars.

The same will happen with nuclear power stations.

South Africa is not literally buying a nuclear power station by writing out one cheque and sending it to a foreign country, which will then arrive carrying a nuclear power station in a box.

So, it’s time we stopped talking of the cost of R1 trillion and asking whether we have the money. Firstly, the R1 trillion, which I have seen escalated in some media to R1,4 trillion “because the rand has gone down against the US dollar” is not the figure calculated by the nuclear technology folks.

We talk of a number in the ballpark of R650bn, but we have to see what proposals foreign companies and South African finance experts put on the table.

That is part of the bidding process; the South African government is not being secretive by not mentioning its own figure — it is doing what any sensible businessman would do by keeping its cards close to its chest while it sees what the other players bid.


Another potential spinoff from a fleet approach is the potential for localisation. A company is not going to gear up and tool up to produce specialist items for one nuclear plant.

But if the target is a carrot: a fleet of reactors, or even hundreds, then that is a different matter. Companies will invest the time and expense to train staff and install the specialist gear required. For an area such as the Eastern Cape, the job creation potential is massive.

Hundreds of reactors? Yes, hundreds. The world market currently has some 500 reactors. These consist of operating reactors and reactors under construction. If some local company builds, for example, a range of a few nuclear-compliant valves, then they can potentially supply these for maintenance and new reactor construction to the world.

Already the South African Nuclear Energy Corporation is exporting nuclear-grade fabricated components to Korea for nuclear fuel assemblies. The target of nuclear fabrication export is not a dream — it’s started. So let’s build on this opportunity.

The localisation target mentioned in public is 50%. I believe that’s quite possible, but some critics have said it’s more likely to be “10%, if you are lucky”. Meanwhile, the Russian nuclear company Rosatom has said: “Why not aim for 60%?” So 50% is totally reasonable.

There is no technical reason why this cannot be achieved. Good project management will be the determining factor.


When I was in Moscow recently, I spoke to the head of Rosatom, and he sees nuclear power as a world effort, something like a big club. That’s my view, too. A few months ago, I was invited to be guest speaker at a national nuclear event in Hanoi in Vietnam. The Vietnamese officials told me they looked to South Africa for inspiration, because we clearly know what we are doing and we understand their conditions.

They said most developed countries have no concept of pulling heavy loads through a hundred kilometres of jungle, on a dirt road. South Africans know all about those sorts of conditions. The head of nuclear from Indonesia was also in Hanoi and he agreed. He told me his country is a collection of islands — they can’t run cables in the air from island to island. He said South Africans are, however, used to those sorts of snags.

I have also received invitations to speak in Bolivia and Turkey and have been told that South Africa’s actions in nuclear power are an inspiration to them.

The environment

The scientific investigation team that carried out the environmental impact assessment (EIA) on potential sites, recommended that a location near Oyster Bay (outsider Jeffrey’s Bay) be the place where the first new nuclear power station be built.

I have toured all over the 4 000ha site, and it’s fantastic. A Chinese delegation that visited there recently reported to me that it is one of the best sites they have seen for nuclear anywhere in the world.

Site preparation can start virtually immediately — the moment the minister of environmental affairs puts her signature on the final EIA document.

The budget for the site preparation alone is tens of billions of rand.

The term “site preparation” refers to a list of actions such as expanding/reinforcing harbour facilities to bring large tonnage items ashore; building/widening roads to carry these loads; building/strengthening bridges; new roads to the site; digging down to bedrock to lay deep foundations; running water supply to the site; electrical supply; housing for workers; and much more.

Essentially, all of this will be done by South Africans. These individuals and their companies will pay tax and they will buy pipes, cables and so on from other companies across the country, which will pay tax. Money will flow to the fiscus immediately.

It is simply not the case that R1 trillion, paid in US dollars, is going out of the country.

Pride in performance

When construction on the actual reactor buildings starts, it will be South Africans pouring the concrete and laying the cables, connecting the water pipes. When construction arrives at the more intricate stage of the installation of cooling pumps, pipework, valves and many other assemblies such as electrical control circuits, then we will start to find out if companies prepared themselves to supply the required nuclear-grade pumps, valves and such.

We will need highly skilled coded welders and universities to be able to use x-rays to look inside a weld to see if the atoms landed up in the correct places, because that is what nuclear-grade welding means. We will need to build huge capacity in the ranks of skilled artisans. We are talking of dedicated people who really take pride in their work.

Obviously, the selected foreign companies will be involved. They will have engineers and planners on site, working side by side with South Africans. They will also have experts visiting facilities all over the country, to check on components and assemblies being manufactured or tested.

Certain components and assemblies will not be made in South Africa and will enter by sea or air, depending on their size and value. South African experts will go overseas to check on the fabrication there.

Meanwhile, throughout this whole process, the South African National Nuclear Regulator (NNR) will check that everything is done according to licensing laws and protocols. The NNR monitors the health and safety of the South African public by making sure that processes and parts conform to the standards as laid down by the NNR, which in turn collaborates with international bodies in mandating such specifications and processes.

  • Kelvin Kemm is a nuclear physicist, CEO of Nuclear Africa and a member of the ministerial advisory council on energy
  • This piece was first published on Moneyweb and is used here with permission

Share this article

  • OrgMas

    Very informative but this does gloss over some very real concerns. Nuclear risk is not only about explosion it is also about contamination. The fault at Fukushima was mentioned and yes the risk of some form of leakage is very low but the risks associated with the waste are not mentioned.

    Whilst most of the work will be local and will thus pour must economic benefits back into the country that does also expose us to further risk. Those complex welds are where the proverbial hits the fan. It is those welds and the people doing them that have cause the delays at Madupi. 4 years late in delivery now. When talking R650billion delays will push the cost to above the trillion mark. It is the local labour component that is the risk.

    In principle nuclear is one of the safest forms of power generation actually having a better safety record than all other forms including solar. But why only look at big reactors? Pebble beds are the safest type and small enough to deploy close to large industrial complexes with the need for costly infrastructure?

  • Andrew Fraser

    “My next point: South Africa is not “buying” nuclear power plants from anybody.” This is true, but not in a good way. South Africa (if it accepts the Russian bid) will not own the reactors at all, nor the power they produce. That will all belong to Rosatom, and South Africa will have to pay whatever Rosatom decides is fair for the power. Not a good situation to have a strategic asset controlled by an external nation state. Gwede’s concern about Fullbright Scholarships is nothing compared to this.

  • Andrew Fraser

    “We talk of a number in the ballpark of R650bn, but we have to see what proposals foreign companies and South African finance experts put on the table.” Looking at every large capital programme in recent history, there is always a overrun. Very easy to get from R650Bn to R1 Trillion. Only an 80% overrun.

    The Medupi build program was supposed to cost R80Bn (2007), latest official number is R105Bn (2014), and most third party estimates (and claims of leaked Eskom numbers) are in the R300Bn range. That would be more than a 200% overrun.

  • William Stucke

    Hi Kelvin. Long time no speak!
    Yes, I agree that the word “nuclear” doesn’t mean that there is a significant risk of a reactor exploding. However, both sides in the Cold War deliberately chose reactor types that facilitated fuel enrichment and ignored safer, simpler, cheaper designs. Weapons were everything. The USSR sponsored the anti-nuclear groups in the West and were extremely effective in their stated goal: Make nuclear power unaffordable to the West.
    Now that the Cold War is over, there is no need to use the old designs, as you correctly point out. However, there’s also no need for the anti-nuclear lobby. Who’s going to tell them that in a persuasive way?

  • William Stucke

    > It is those welds and the people doing them that have cause the delays at Madupi. 4 years late in delivery now
    Only partly true. It was one of the factors. An astounding failure in project management, a lack of penalties for non-performance, and an unruly labour force all played important roles.

  • William Stucke

    > But why only look at big reactors? Pebble beds are the safest type and small enough to deploy close to large industrial complexes with the need for costly infrastructure?
    We have already made the mistake of “Bigger is better”. Medupi and Kusile are some of the largest power stations of their type in the world. The justification was “economies of scale”. As we have clearly seen, an inability to manage such “super projects” has led to those savings being a mirage. Indeed, they are costing considerably more than four or six smaller stations with the same output would have – not taking into account the billions lost by the economy due to the delay, load-shedding, uncertainty, etc.
    As far as the Pebble Bed Modular Reactor (PBMR) project is concerned, that’s a sad tale. Firstly, it lost its way and didn’t keep its eye focussed on one objective, but vacillated between generating power, CHP or industrial heat. Secondly, it got the chop from a government without the foresight to see how important small modular, cheap and intrinsically safe reactors are.
    So sad. We were well on the way to being world leaders in yet another field, and threw away the opportunity.

  • OrgMas

    Fair enough. I was too brief. You raise all the issues. One can only hope that they don’t repeat the errors.

  • OrgMas

    The anti nuclear lobby is now a quasi religion. They will not be dissuaded

  • OrgMas

    Indeed a great loss. When I first had the concept explained to me I was both fascinated and convinced. What a loss.

  • Greg Mahlknecht

    Great to see an informed article from someone who’s not grinding a political axe or arguing economics (economists seem to be the only profession apart from weathermen that can be wrong 90% of the time and still keep their jobs).

    My father was a metallurgist and provided consulting and parts for a number of nuclear power stations around the world (including some the work for Koeberg); As the author points out, we certainly have the technical knowledge here to do a lot ourselves as we are world class in a number of the required disciplines.

    I’m a believer in renewable energy sources, but this country simply isn’t ready for it yet – to implement, it seems to need a level of discipline and willingness to educate the citizens that our government simply hasn’t provided.

  • 小杜 (xiao du)

    I’ll just leave this here –

    “At the end of last year, during the budget review process, the DoE failed to account for Necsa (SA Nuclear Energy Corporation), which did not report and asked to be exempt from reporting to parliament.”

    “Necsa has major liabilities for its failure to address nuclear contamination,” she said. “There is no information on the extent of the contamination, the amount of the liability and the timeline to fix this. Now Necsa is saying that the government must pay and that it is not liable.

  • syco

    Well, personally, I would like to see the
    government/Eskom get a move on and get these nuclear plants built as soon as
    possible. I really don’t see why a bunch of ‘Tree Huggers’, who represent a
    small minority, should hold up the implementation of this important step. The
    supply of plentiful, uninterrupted electricity is key to foreign investment and
    the growth of our country. (Decent, affordable Broadband would also help). There are important business decisions tha thave to be made and obviously a certain amount of secrecy/discretion has toexercised. It is a given that ‘connected’ people will be salivating with the many opportunities for lining their pockets and the media can play an important role here by shining a ‘bright light’ on anything that doesn’t seem right but
    all this drama about Nuclear Power is getting a little stale. There is irrefutable
    proof that emissions from coal powered electricity generation causes medium/long term harm to more people than Nuclear energy ever has so let’s get on with it! (Have you ever been at a coalmine as the workers come up from the shafts and heard them coughing and wheezing – not to mention the coal dust covering their bodies or met anyone living near a coal powered power station or coal mine and asked them how they feel)?
    Just to put some perspective to the irrational inborn fear that people have
    about ‘Nuclear’ – some years ago I was sitting across the table from a senior
    Russian KGB agent (Colonel …) and we were discussing the well publicised
    widespread protests that were going on in America about the fact that a NASA
    deep space probe (Cassini) was going to be nuclear powered and I asked him why one
    never heard about any similar anti-nuclear protests in Russia where nuclear was
    so widely used. He laughed and said that they considered the misinformation
    program that they launched in the USA and Britain in the 50’s and 60’s, to
    brainwash people there about the dangers of all things nuclear, to be one of
    their greatest triumphs as it had slowed nuclear development in the West and allowed them (the Russians) to push ahead
    unhindered and go on to lead the nuclear ‘race’. (Their
    space probes also extensively used SNAP (Small
    Nuclear Applied Power)