Power plant design

PBMR: hot or not?

1 April 2009

On 5 February PBMR (Pty) announced that it was planning to look beyond electricity generation for its pebble bed reactor (see statement). Could this be the beginning of the end of the South African PBMR project? By Steve Thomas

In 1998, Eskom, the publicly owned South African electric utility, announced it was launching a programme to develop the Pebble Bed Modular Reactor (PBMR). In February 2009, PBMR (Pty), the company set up to develop and supply the reactors effectively abandoned the design in favour of a PBMR design primarily for non-power applications. The question the South African public must ask now is whether this was a very risky gamble with public money that should have been abandoned years ago, or whether it was a visionary attempt to develop a technology that could play an important role in world electricity supplies and which might yet pay off. The research summarised on pages 16-20 about the first PBR?in Jülich, Germany, suggests that the technological problems the PBMR has run into should have been foreseen no later than 2002. About 80% of the South African public money spent on this technology was incurred after this date.

The PBMR was presented as an ‘unpolished gem’, abandoned in the wake of the Chernobyl disaster, but which Eskom could develop into a world-beating product. In 1998, Eskom forecast that a demonstration plant would be built and at least 10 commercial orders per year would be placed worldwide from 2004 onwards. Ten years later, the programme was 20 years late, the estimated cost of the demonstration plant was more than 10 times over budget, and there were no customers.

But it had been clear since 2002 that the project was going badly wrong. The only firm customer for the PBMR had withdrawn, international partners were reluctant to invest, costs were escalating and lead-times slipping. The helium-driven gas turbine was problematic and the overall design was proving difficult to finalise. Since 2002, PBMR (Pty) has claimed that it was within six months of submitting a complete design for assessment by the South African regulator, the National Nuclear Regulator (NNR). This step was never achieved.

PBMR’s February announcement contained little detail on why the decision had been taken, and Eskom and the government remain silent. However, the reasons appear likely to be a mixture of commercial and technical issues.

The commercial issues – was the design economical and would anyone buy it – are relatively accessible to the layman and have been debated, but the technical issues raised by the Jülich report were a shock to anyone not deeply involved in the technology. The option to reduce helium gas temperatures from 950°C to about 800°C that use of the reactor for non-power applications allows, suggests this was a significant element in the decision to abandon the use of a gas turbine and focus on non-power applications. This raises the question of when those involved in the development of the PBMR became aware of this issue and whether decision-makers continued to allow funding for the project despite the existence of a problem that had the potential to derail the project.

PBMR (Pty) has taken the German prototype plant, the AVR, as its reference plant and the AVR has been portrayed to the South African public as an unqualified success. A German nuclear scientist, Peter Pohl, told the South African TV programme Carte Blanche “what was achieved is unique, in temperature, in burn up, in reliability – it’s just fantastic.”

The AVR only operated for about four reactor years at gas temperatures of greater than 900°C. The official final report* on the AVR states: “At an average hot gas temperature of 950°C the highest fuel temperatures were around 1300°C. [T]hese values do not represent a serious load because of the full ceramic design of the fuel element.”

The new Jülich report presents a different picture: “the end of life contamination reached several percent of a single core inventory, which is some orders of magnitude more than precalculated and far more than in large LWRs. A major fraction of this contamination is bound on graphitic dust and thus partly mobile in depressurization accidents, which has to be considered in safety analyses of future reactors.”

The new report claims that the assumption in the final report on the AVR – that the high temperatures were the result of poor quality fuel – was not proven and that other factors, as yet unknown, were probably involved. According to PBMR (Pty), the maximum fuel operating temperature should not exceed 1130°C. If the large temperature variations observed in the AVR are a guide, this assumption is far too optimistic.

The Jülich report also recommends that a gas-tight containment is required for any commercial pebble bed plants and that a large programme of R&D be conducted as well as a re-evaluation of the safety case.

Who knew what, and when?

While the full extent of the problems is only now becoming clear, the high temperatures experienced at AVR were known about 20 years ago. Eskom spent five years from 1993 onwards examining the PBMR before the development programme was announced and this period must have included an evaluation of AVR experience.

Even if the South African authorities had assumed the temperature problems were the result of poor fuel quality, the seriousness of the issue should have become clear in 2002. In 2000, PBMR (Pty) was joined in the project by US utility Exelon, which agreed to pilot the PBMR through the safety regulatory process in the USA. By the start of 2002, the NRC was asking searching questions on core temperatures. In April 2002, Exelon withdrew from the company.

The process of regulatory approval in the USA was closed in September 2002, without regulatory approval being granted. Exelon’s reasons for withdrawal were vague and it is not clear how far the issues that NRC raised contributed.

Eskom and PBMR (Pty) would have been aware of the NRC’s questions. It subsequently emerged that in 2002, Eskom was discussing withdrawal from the project at Board level. But it was decided that Eskom’s withdrawal was politically infeasible. How far these discussions were informed by the issues the NRC was raising is also not known. However, Eskom’s support for the PBMR seems to have been weak since then.

The South African NNR would have followed the US process very carefully. As a newly-founded regulatory body with no experience of licensing a reactor, much less a first-of-a-kind design, input from the world’s most experienced regulatory body would have been invaluable. Documents submitted by PBMR (Pty) to the NRC were authored by Maurice Magugumela, then an employee of PBMR (Pty), who, in 2005 became the CEO of the NNR.

The Next Generation lifeline

PBMR (Pty) is pinning its hopes on the US government’s Next Generation Nuclear Plant (NGNP) programme. However, the changes it recently announced seem to disqualify the new design from consideration. The specification requires: outlet temperature in the range of 850-950ºC; production of both electricity and hydrogen using an indirect cycle with an intermediate heat exchanger to transfer the heat to either a hydrogen-production demonstration facility or a gas turbine.

The PBMR has to compete with two other technologies for funding. Even if it is successful and the apparently unrealistic time-scales being projected are met, the completion of a demonstration plant, if one is ultimately built, is more than a decade away. Overall, the chances of the NGNP programme offering a salvation route for the PBMR look slim.

Throughout the project, all of the South African institutions involved in the PBMR – Eskom, PBMR (Pty), the South African government and the NNR – have failed to keep the public informed on how its money was being spent, while parliament has shown no appetite to carry out its role of oversight. This should not be acceptable where such large sums of public money are involved. There are ample opportunities in South Africa for public spending that would give a high probability of a good pay-off, for example in health and education. What is needed is an independent inquiry to establish who knew what and when. If the project was continued after it was clear the chances of success were minimal, those involved should be held accountable and new procedures introduced which offer better control of public spending.

Author Info:

Steve Thomas, Professor of Energy Policy, PSIRU, Business School, University of Greenwich, 30 Park Row, London, SE10 9LS, UK

Related Articles
PBMR (Pty)’s perspective
Pebble bed reactor – safety in perspective
South Africa’s nuclear plans on hold
MHI to collaborate on PBMR development

History of the South African PBMR project

High temperature pebble-bed gas-cooled reactors have been under development in Germany, primarily through the German government's Julich research centre, since the 1950s. A prototype plant at Julich, AVR, which produced 15MW of electrical power, operated from 1967-88. This was followed up by a demonstration plant, THTR-300 (300MW), which went critical in 1983. It last generated power in 1988 after numerous technical problems had been encountered and was formally closed in 1989. The AVR was generally portrayed as highly successful and later design concepts were based on the AVR rather than the THTR. Subsequently, the THTR-300 designer, ABB, joined forces with Siemens, which had also designed a similar size plant producing a new design. The joint venture company, HTR, now owned by Areva NP and Toshiba/Westinghouse following various mergers and takeovers, promoted this design for a couple of years before giving up active promotion of it although continuing to sell technology licenses.

The HTR design was taken up by South Africa in 1993 by the publicly-owned electric utility, Eskom. HTR cooperated with Eskom who produced a design which, unlike its predecessors, was expected to generate electricity using a gas turbine driven directly by the helium coolant gas rather than via a heat exchanger and a conventional steam circuit. In 1998, these efforts were made public and Eskom then forecast that following completion and operation of a demonstration plant, commercial orders would be possible from 2004 onwards. The electrical output has been uprated from 110MW in 1998 to 165MW.

In 2000, a new company, PBMR (Pty), was set up to complete a feasibility phase. This company remained 100% owned by Eskom, but other investors were brought in to fund this phase. This funding would allow them to take up shares in proportion to their investment in a successor company to PBMR (Pty) that would complete a demonstration phase in which a demonstration plant would be built and operated, and commercial orders would be placed.

The investors in PBMR (Pty) have varied over time but the main ones were: British Nuclear Fuels (BNFL), which promised 22.5% of the funding for the feasibility phase in 2000; Exelon, a US electric utility that took 12.5% in 2000; and the South African government's Industrial Development Corporation (IDC), which took 25%, also in 2000. 10% was reserved for an Economic Empowerment Entity with Eskom expected to take the remaining 30%. However, Exelon withdrew in April 2002. While it did pay some of the money promised, it is not clear whether it fulfilled the full 12.5% quota. BNFL ran into financial problems in 2002 and stopped funding in 2003 paying only 15% of the cost of the feasibility phase. Although unreported at the time, IDC reduced its stake to 13% in 2002. The Economic Empowerment Entity stake was never taken up. So Eskom's contribution was about 60%, rather than the 30% planned. South African public money, which was expected to pay for 55% of this phase, actually funded about 75%.

The feasibility phase appears to have been completed in March 2004, by which time PBMR (Pty) stated that ZAR1.4 billion ($141 million) had been spent, of which it seems about ZAR1 billion ($101 million) had come from South African public money. Since 2004, PBMR (Pty) has been funded entirely by the South African government and Eskom. The successor company was never set up, despite continuous reports that this was imminent. Despite claims that Toshiba/Westinghouse held 15% of the project, none of the partners in the feasibility phase ever agreed to fund the demonstration phase. The project has, in some respects, been in limbo since the feasibility phase was completed in 2004. Nevertheless, a number of contracts have been placed. For example, by December 2007, four contracts had been signed with German companies Nukem for fuel and SGL Carbon for supply of graphite. UK firm Frazer-Nash has engineering consultancy contracts for the reactor pressure vessel safety and graphite core design. The value of these contracts has not been published. The largest contract is probably with Spain's Equipos Nucleares S.A. which has a contract worth a total of ZAR312 million ($31 million) for design and delivery of components for the plant's main power system pressure boundary, including the pressure vessel. No definitive list of the contracts that have been placed and their values exists.

In February 2009, PBMR (Pty) issued a statement on what it called "a change in product strategy". The South African government has not committed any money to support the company's work after the end of fiscal 2010. According to the Department of Public Enterprises, the ministry that is managing the project, by that time the government will have provided ZAR10 billion ($1 billion).

External weblinks
Nuclear Engineering International is not responsible for the content of external internet sites.


Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.