Specimen Size Effects in the Determination of Nuclear Grade Graphite Thermal Diffusivity

Understanding the effects of radiation on nuclear grade graphite requires complex experiment design with specific specimen geometries. These geometries are not always optimal for the various testing and measurement standards used to characterize the graphite before and after irradiation. In this work, geometry constraints of specimens for laser flash determination of thermal diffusivity are investigated. The laser flash technique depends on the assumption of one dimensional adiabatic heat conduction between two of the specimen surfaces. As test temperatures rise above 400°C, radiation heat loss from all surfaces becomes significant and must be accounted for through modeling. Experiments were performed using industrial grade AXF-5Q graphite and nuclear grade graphite, PCEA, NBG-18, and Gilso, to understand the limits of the common radiation heat loss correction models used for laser flash diffusivity testing. From this data, a maximum specimen diameter to thickness ratio is presented. This testing limitation, along with the limited volume available in irradiation experiments, drives the specimen diameter and thickness to be minimized. As the specimen thickness is reduced, the significance of the graphite inhomogeneity and high thermal conductance influence on measurement uncertainty must be considered. Due to the relatively high thermal conductivity of graphite, thin specimens can result in low thermal transit times. Furthermore, as the thickness of the sample approaches two or three times the size of filler grain material used in the graphite, the thin specimen may not represent the homogeneous bulk material. This effect as well as ambiguity in the exact arrival time of the laser energy due to finite laser pulse widths is investigated for a lower limit on specimen thickness.