The fuel-cladding interfacial friction coefficient in water-cooled reactor fuel rods

A central problem in the development of cladding failure criteria and of effective operational, design or material remedies is to know whether the cladding stress is enhanced significantly near cladding ridges, pellet chips or fuel pellet cracks; the latter may also be coincident with cladding ridges at pellet-pellet interfaces. As regards the fuel pellet crack source of cladding stress concentration, the magnitude of the uranium dioxide-Zircaloy interfacial friction coefficient μ governs the magnitude and distribution of the enhanced cladding stress. Considerable discussion, particularly at a Post-Conference Seminar associated with the SMIRT 4 Conference, has focussed on the value of μ, the author taking the view that it is unlikely to be large (< 0.5). The reasoning behind this view is as follows. A fuel pellet should fracture during a power ramp when the tensile hoop stress within the pellet exceeds the fuel's fracture stress. Since the preferred position for a fuel pellet crack to form is at the fuel-cladding interface midway between existing fuel cracks, where the interfacial shear stress changes sign, the pellet segment size after a power ramp provides a limit to the magnitude of the interfacial shear stresses and consequently to the value of μ. With this argument as a basis, the author's early work used the Gittus fuel rod model, in which there is a symmetric distribution of fuel pellet cracks and symmetric interfacial slippage, to show that μ < 0.5 if it is assumed that the average hoop stress within the cladding attains yield levels. It was therefore suggested that a high interfacial friction coefficient is unlikely to be operative during a power ramp; this result was used to support the view that interfacial friction effects do not play a dominant role in stress corrosion crack formation within the cladding. (orig.)