Nuclear: will we learn?

Einstein: ” ... doing the same thing over and over and expecting different results”. In 2018, SA’s nuclear tender had to be abandoned because it could not be financed.

In 2010, the pebble bed modular reactor programme was ended after 14 years with no orders in sight. There followed numerous reports of imminent calls for tenders and calls for information from vendors, all leading nowhere until a court judgment in 2017 that the SA nuclear programme was illegal put a temporary end to this.

But the SA government has released a new call for information from reactor vendors. If the outlook for the nuclear industry had suddenly got better, this would be understandable. But since 2017, everything has got worse.

Of the four French evolutionary power reactors under construction in Europe, Olkiluoto (Finland) and Flamanville (France) are still two to four years from completion after 13 to 15 years of construction, and costs at Hinkley Point (UK), which only recently started construction, have increased by about 15% since 2017.

Of the two Westinghouse AP1000 projects in the US, one was abandoned in 2017 because costs and schedules were out of control. The executives involved are now facing criminal prosecution. At the other, costs are ballooning and it was reported 80% of components failed their operational tests.

The Korean design, previously seen as cheap on the basis of an order for UAE, is suffering major component quality problems in UAE and Korea, and after eight years of construction is still not ready. This design would need major and costly safety upgrades for it to be licensable in Europe.

China has yet to win export orders for its reactors and the CAP1400 design of the vendor, SNPTC, allocated to SA by the Chinese government has yet to start construction in China despite the vendor saying it was ready to start building for the past six years.

In Europe, there is increasing concern that allowing Chinese companies to own major infrastructure would compromise national security and this may derail a UK reactor project. Whether a Russian design would be politically acceptable, given the findings of the commission of inquiry into state capture in SA, is a moot point.

The latest rabbit out of the hat from the nuclear industry is small modular reactors (SMRs) producing up to 300MW — compared with large reactors discussed above, which produce more than 1,000MW. However, these are a decade or more from being demonstrated technically or economically. The design nearest deployment is US NuScale, which produces 60MW. It has been under development for 15 years but still has no firm customers and, in the form it will be sold, is years away from approval by safety authorities.

Rolls-Royce has been promoting its SMR design under development since 2017. At 450MW, it is hardly small. Rolls-Royce’s selling point is its experience of supplying reactors for UK submarines, but these are typically about 20MW and not suitable for use on land. The design will not be ready until about 2030 at the earliest, even if the British government is willing to meet RollsRoyce’s demands that it meet at least half the estimated £500m development costs and commit to buying at least 15 of them.

China has continued to pursue the pebble bed concept using the same concept as SA, but after eight years of construction at the first site, yet another delay to 2021 was announced this week.

Steve Thomas

Emeritus energy professor, University of Greenwich

YOUR LETTERS: Letters of more than 300 words will be edited for length. Letters should be sent by e-mail to busday@bdfm.co.za or faxed to

(011) 280-5505. E-mailed letters are preferred. Anonymous correspondence will not be published. Writers should include a daytime number.