Stresses which induce both PCI (usually denotes combined mechanical and chemical pellet-cladding interaction) and PCMI (usually denotes pel­let cladding mechanical interaction) are caused by expansion of the fuel pellet against the cladding during power increases (Adamson et al, 2006/7; Strasser et al., 2010a). PCI failures are driven by a stress corrosion crack­ing (SCC) assisted component resulting from fission product release from the fuel, while PCMI failures are generally due to purely mechanical crack­ing, often enhanced by a reduction in cladding ductility due to formation of local hydrides at the clad outer surface. At the micro level, the PCI crack always starts at the cladding inner surface and propagates towards the outer cladding surface while the PCMI crack propagates from the outer to inner surface.

PCI is associated with local power ramps during reactor start-up or power manoeuvring (e. g. rod adjustments/swaps, load following) as shown sche­matically in Fig. 5.3, and is caused by the combination of cladding stress due to the power increase and the influence of iodine, caesium and cadmium released during the power increase in a susceptible material (Adamson et al., 2006/2007; Strasser etal., 2010a). This combination of stress, embrittling

fission products and susceptible material may result in SCC of the fuel clad­ding, as shown in Fig. 5.4.

PCI failures may occur in PWRs/VVERs and BWRs (Strasser et al, 2010a). The failure mechanism is much more prevalent in BWRs, since reactor power is controlled in part by control rod movements that subject the fuel to rapid power level changes. (The reactor power in both BWRs and PWRs is also reg­ulated by flow control.) In PWRs and VVERs, reactor power is not normally controlled by insertion and extraction of the control rods in the core; rather, reactor power is controlled by the boron concentration that is continuously decreased during operation to compensate for the decrease in reactivity. This type of reactor power control is much smoother than in the BWR case and, consequently, PCI failures are less common in PWRs. However, during reac­tor power increases, and specifically during a class II transient (anticipated operational occurrences, AOO), PCI failures may occur in a PWR.

To prevent PCI failures, it is necessary to remove at least one of the fun­damental conditions (tensile stress, sensitive material, aggressive environ­ment) which cause SCC. There are two principal types of remedy (Strasser et al., 2010a):

1. One is to develop reactor operation restrictions that will ensure cladding stresses are always below the PCI threshold stress during power increases. This is the main measure in avoiding PCI defects and the only measure used in PWRs. Operating rules (also called management recommenda­tions, or pellet-cladding interaction operating management restrictions (PCIOMRs)) to limit local power increases and ‘condition’ fuel for power ramping were implemented in BWRs during the late 1970s to mit­igate the PCI issue. The rules are usually a function of exposure and were developed by the different fuel vendors, so they differ between various fuel types. To establish and validate these rules, extensive power ramp tests were performed by the fuel vendors in experimental reactors.

2. The second remedy — design improvement — consists of two approaches: 2a. Cladding design

2a1. Development of radial cladding texture and small grain size that may increase cladding PCI resistance.

2a2. Development of the barrier/liner concept, initially with a ‘pure’ zirconium (Zr) metal barrier at the cladding inner diameter (ID). The barrier is soft and serves to reduce the local stress, hence giv­ing the cladding resistance to SCC. Later, fuel vendors realized that the Zr could be alloyed with Fe to improve the secondary degradation resistance in case of rod failure. The Fe in the Zr will dramatically improve the corrosion resistance of the liner/barrier but may reduce the PCI performance. Although this remedy has so far only been used in BWRs, it should be equally applicable to PWRs.

2b. Pellet design

2b1. Reducing the cladding local strains (and stresses) by shortening the pellet, chamfering the corners and eliminating the dishing.

2b2. Pellets with additives are being developed both for BWRs and PWRs that will increase the margins towards PCI failures (Adamson et al., 2006/7; Patterson, 2010). The additives of inter­est fall into two general categories, the first category involves materials that are essentially insoluble in the fluorite lattice and exists as a separate, grain boundary phase, for example, mix­tures of alumina and silica (aluminosilicates or Al-Si-O). The second category involves materials that are soluble in the cat­ion sub-lattice, such as chromia, or involve a mixture of soluble and insoluble materials, such as chromia and alumina. Although many other additives fall into both categories, attention is directed to the aluminosilicate additives and chromia-base dop­ants as they appear to be the closest to large-scale application.

2b3. Aluminosilicate additives consist of a mixture of SiO2 and Al2O3 and is offered by GNF. During the pellet sintering process, the additive forms a glassy phase that collects on the grain bound­aries. It appears that the Al-Si-O additive at the pellet grain boundaries will chemically react with I, Cs and Cd, thus prevent­ing these SCC-promoting elements from accessing the fuel clad inner surface (Matsunaga et al, 2009, 2010). Additive fuel has been irradiated in commercial and test reactors in the US and in Europe. Ramp tests under BWR conditions in the R-2 and Halden reactors of segmented additives rods from commercial reactors show excellent resistance to the PCI failure mechanism (Davies et al.,1999).

2b4. Chromia (Cr2O3) is the dopant of greatest commercial signif­icance in this class of additives. Two types of chromia-based additives are being offered. The first consists of Cr2O3 in UO2 as offered by AREVA (Delafoy et al, 2003). The sec­ond consists of Cr2O3 and Al2O3 in UO2 as offered by Westinghouse (Arborelius et al., 2005). Alumina is reported to be used in the second form to minimize the effects of chromium on the fission cross-section of doped pellets while enhancing grain growth. In both cases, chromia is expected to reside largely within grains as interstitial Cr3+ and as insoluble Cr2 O3 depending on the concentration and tem­perature. The alumina in the mixed Cr-Al-O dopant should exist as a grain-boundary phase as in the Al-Si-O additive. The cation dopants were developed to increase grain size to reduce fission gas release (FGR) at extended burnup(see for example Delafoy et al., 2007). In addition to improved FGR, chromia-based dopants are reported to improve PCI resistance. Information available in this area is less exten­sive for the chromia-based dopants than for the aluminosil­icate additives. However, ramp tests indicate that the resis­tance to PCI failures of fuel with chromia-based dopants are improved relative to standard fuel in cladding without PCI-resistant liners (Delafoy et al., 2007).