close
EPR Reactor Vessel

Nuclear reactor vessel woes – the story so far and why it's significant

In the coded wording of the nuclear industry, reactor vessel ‘anomalies’ were reported earlier this month following tests on these critical components of the EPR reactors similar to those planned by EdF for the Hinkley Point project.

But what are these anomalies and why are they important? Tony Roulstone explains the potential knock-on effects to the UK's EPR programme - and why they could be very serious.

In an EPR nuclear power station, the reactor vessel contains the nuclear core which generates the nuclear heat which is then converted to electricity in the rest of the plant.

There could be no more significant part of a nuclear plant. The reactor vessel is huge, almost 13m tall and weighing over 500tonnes - designed for 176bar and 350 degrees C.

" EdF has announced that it will continue with construction at Flamanville in France, but work on both Chinese reactors at Taishan will stop before installing fuel in the core."

A leak from this vessel must be ‘incredible’. It was a clash in 1970s and 80s between Lord Marshall, Chairman of the UK Atomic Energy Authority (and subsequently the Central Electricity Generating Board) and government chief scientific adviser Sir Alan Cottrell about this ‘incredibility of failure’ argument that was central to whether the UK would change from its gas cooled reactors to pressurised water reactors, such as at Sizewell and those proposed for Hinkley.

Gas-cooled reactors had pre-stressed concrete vessels which were regarded as leak-proof. The concern with the new reactors was that a steel vessel could fracture, allowing coolant to escape and result in melting of the highly-rated reactor core.

In the end, the ‘incredibility of failure’ argument was built on three legs:

  1. Choice of material, its strength and ductility,
  2. Protection against changes in steel properties caused by welding, or irradiation,
  3. Practical elimination of material defects, from which cracks might grow.

All three legs of the argument need to be in place for it to stand.

The latest reported problems with EPR vessels attack this argument at its source: material properties. The steel for reactor vessels is very specialised, with high strength and importantly, good ductility (fracture toughness) to avoid the possibility of brittle fracture.

High fracture toughness of steel requires low levels of carbon. In this case, the upper limit for carbon content is 0.22%. The reported anomalies relate to large sections of the top and bottom heads where the carbon content is well above the limits at 0.3%. This is linked with the measured toughness value which is well below the specification.

Reactor vessels are made by forging large ingots of steel in sections that are welded or bolted together. Only the largest forges in the world can handle these ingots, which exceed 100 tonnes in size.

"Additional delays to these three projects now seem inevitable. The potential knock-on effects to the UK EPRs could be very serious."

For this reason, the main sections of the vessel [such as that used at Flamanville], including the coolant nozzles, are forged by Japan Steel. In fact the whole vessel for the first EPR at Olkiluoto in Finland was also forged in Japan. The bottom and top heads of the EPR vessels under question were forged by Creusot-Loire, a subsidiary of Areva from locally provided steel.

It seems that Areva made six reactor vessels plus a test vessel for the EPRs in France, China (2), UK (2) and US or India. The test vessel was retained to provide qualification evidence for the safety cases. Carbon anomalies were measured in this test vessel, which is considered typical of the series. Similar results are expected in the other vessels, possibly originating from the initial steel ingots.

The vessels for the French and Chinese EPRs have been installed, deep in the containment structures and welded into the primary cooling circuit.

So now, what are the options for EdF and Areva? They could:

  • Argue that the test vessel results (anomalies) are not found in the actual reactor vessels,
  • Make the case that the region where carbon is high is not significant to vessel integrity,
  • Replace either the whole vessel, or just the upper head, which is more sensitive.

Further materials tests are planned over the next six months, which are likely to include examination of the installed vessels. EdF has announced that it will continue with construction at Flamanville in France, but work on both Chinese reactors at Taishan will stop before installing fuel in the core.

Additional delays to these three projects now seem inevitable. The potential knock-on effects to the UK EPRs could be very serious.

Tony Roulstone in a member of the University of Cambridge Nuclear Energy Centre and he is a former MD of Rolls-Royce Nuclear. He is a Fellow of the Institution of Mechanical Engineers, a Member of the Nuclear Institute and of the Institution of Engineering and Technology, member of the Board of the Advanced Manufacturing Institute, Sheffield and a member of the acumen7 network.