At the inspection three graphite bricks of the larger type shown in the figure above (bricks containing nuclear fuel);were found to have cracks due to what is known as keyway root cracking.
When graphite is irradiated in a reactor it initially shrinks. Material close to the fuel brick bore experiences a higher radiation dose than at its outside. As a result, the stresses at the bore are initially tensile, and so any cracks start at the bore, as shown in the figure (above). Later in the reactor life they become compressive; as shown in the figure, the keyway root cracks that have been seen start at the outside of the brick and show that for some bricks the stresses there have now reversed.
Keyway root cracking was expected to happen, and EDF Energy has undertaken theoretical and experimental studies to determine the number of bricks that can be cracked without reactor safety being compromised.
Quintessa is providing support to EDF Energy in evaluating the implications of the reactor inspections. Questions that need to be considered include:
- What conclusions can be drawn about the extent of keyway root cracking in the reactor from the inspections that have been made?
- How do we expect cracking to evolve over the next few years, and what do we expect to see at forthcoming inspections?
Quintessa has undertaken calculations using two sets of information:
- Just the observations at the most recent inspection and previous inspections at the Hunterston and its sister station Hinkley Point B. Using just the data on the observed cracks involves statistical methods and provides confidence that the current extent of cracking in the reactor provides no safety concerns.
- Our understanding of the processes that lead to cracking using information on the properties of the graphite incorporated in mathematical models.
Before the latest inspection Quintessa models were employed to make predictions of what would be seen, both in terms of the way that the brick shapes evolve and the expected numbers of cracked bricks. Overall the numbers of cracked bricks and the observed brick shapes were compatible with expectations, but some of the details were not. Two of the three bricks that experienced keyway root cracking were in a brick layer that was expected to see cracks later than observed. Possible reasons for this observation are being investigated and the models will be updated before the next inspections take place.
The figure above shows how a specified measure of the brick shape (referred to as a 'lambda factor') evolves as the reactor ages (the x-axis is the core burn-up, related to the total energy produced by the reactor). One of the bricks that experienced keyway root cracking is shown in black in this figure. This illustrates that bricks whose shape changes faster than average tend to be more at risk of keyway root cracking; this relationship does not provide a full explanation of why some bricks crack and others do not, but it is consistent with model calculations.
Understanding the evolution of keyway root cracking will be an important issue over the next few years. The inspection programme will gather more data and these methods will be a key part in setting out the inspection programme and eventually this analysis will support the Lifetime of AGRs.