Graphite core ageing

Over time during reactor operation, the graphite bricks age and their properties change due to interaction with the radiation environment and the reactor coolant. This can lead to the graphite losing weight and the development of cracks in the graphite bricks, both of which are well-known phenomena and have been the subject of significant interest by the industry, academics and the regulators for many decades.

We require the licensee, EDF, to demonstrate through their safety case that they have adequate understanding of the graphite behaviour, to justify safe operation of the core in a clear, evidence-based manner. We require EDF to clearly define conservative limits of operation based on the extent and adequacy of their understanding of graphite core ageing

Challenges

The last operating Magnox reactor ceased operation at the end of 2015. Three of the UK’s AGR fleet ceased operation by 2022, but the remaining four AGRs will continue to account for around 10% of the UK electricity supply for a number of years. The graphite core of these reactors cannot be replaced and ageing mechanisms such as weight loss and cracking can change the mass, dimensions and material properties within the core. As such, they pose unique challenges to EDF and to us as the regulator.

As well as moderation, the fundamental safety requirements of an AGR core include allowing free movement of control rods, free movement of fuel and directing the flow of coolant gas to ensure adequate cooling of the fuel and core structure. Essentially, significant weight-loss and cracking may compromise these safety requirements.

Weight-loss

During operation, the graphite slowly loses weight due to oxidation caused by the reactor's carbon dioxide coolant gas. Loss of weight affects both the mechanical properties of the graphite brick, and reduces its effectiveness as a moderator. Weight-loss is potentially a life limiting condition for the reactors, although we believe that most of the AGRs will have their life limited by the progression of cracking.

Cracking

Four types of cracking have been observed in graphite bricks:

1. Keyway root cracking - this is considered to be the likely phenomenon that will ultimately limit the lifetime of most of the AGRs. The origin of keyway root cracking is caused by the graphite at the outer surface of the bricks moving into tension due to changes in the internal stress of the brick. This mechanism can only occur later in life as it is dependent on the total amount of irradiation received by the graphite. It can consequently progressively crack many bricks across the core.
2. Bore cracking - Cracks originating at the surface of the brick bore or channel (closest to the fuel) are a consequence of early life ageing behaviour, when the tensile stresses at the surface of the brick are at their greatest. This mechanism has only affected a small number of bricks and inspection data has shown that the number of cracks is not increasing at a greater rate than expected, supporting the claim that they are a result of early life tensile stresses at the bore.
3. Cracking of outer graphite blocks - Four of the Advanced Gas-cooled Reactors in the UK have a unique design, which includes outer shielding graphite blocks surrounding the graphite core. This design, which is specific to Heysham 2 and Torness power stations, has a number of functions, including directing the carbon-dioxide coolant and shielding the other components outside of the core. In 2015, EDF Energy discovered cracking in a small number of blocks within this shielding outer layer in one of the four reactors. Subsequent inspections confirmed the presence of this phenomenon in the other three reactors. EDF Energy conducted investigations and analysis of the issue and produced a revised safety case. We are satisfied that the cracks do not present a challenge to the safety of the graphite core.
4. Cracking of the seal-ring groove wall – the Heysham 2 and Torness reactors have a unique design which includes grooves in the ends of the fuel bricks. The purpose of these grooves is to hold a seal-ring which aligns the fuel channels to aid free movement of fuel as well as aiding the direction of coolant flow. A consequence of keyway root cracking is that additional cracking of the seal-ring groove wall can occur. This may lead to the generation of graphite debris in the fuel channel which would pose a potential risk to the coolant flow paths and to the free movement of fuel.

It is the responsibility of the licensee, EDF, to demonstrate to us that the graphite core can continue to operate safely as it ages, and it undertakes an extensive programme of testing and analysis to support its safety case for operation. Also, EDF periodically shuts down each reactor as it is required to carry out inspections and remove samples of the graphite to determine the level of weight-loss and cracking.

Limits

Within its operational safety cases EDF sets safety limits on weight-loss and cracking for each reactor core based on extensive research and regular surveillance and analysis of the graphite behaviour.

A requirement of the nuclear site licence is for the licensee to determine and justify the operating limits of its nuclear plant. This means that the limits can change depending on the licensee's understanding of the graphite behaviour. They may obtain information through surveillance and associated research, to revise the limits for weight loss and cracking stated within the safety cases for the reactors.

However, any proposed change to those limits must be presented to us through a robust safety justification, which demonstrates that it is safe to adjust the limits, based on the evidence provided.

Following our own detailed assessment, which can include discussions with academic experts, if we are satisfied with the evidence and the safety justification, we will agree to operation with the revised limits.

Our role

We require EDF to demonstrate through their safety case that they have an adequate understanding of the material changes to the graphite and their rate of progression, to justify safe operation of the core in a clear, evidence based manner. We also require that there are clear safety margins between the worst case core condition predicted during a period of operation and that which has been shown to be safe, with no “cliff edge” change of behaviour. They are required to conduct regular inspections to understand the changes in the core, also remove samples of the graphite to conduct suitable examinations or experiments. Subsequent analysis of these results is used as the basis of the case for continued operation.

At the end of each statutory reactor outage, our inspectors conduct detailed assessments of the licensee's safety case that supports the proposed return to service, following the inspections of the graphite and supporting research, considering all of the evidence provided. If there are any safety concerns, we would not permit the return to service of the reactor until they have been satisfactorily addressed.