Observations of athermal gas release from water-reactor fuel at extended burnup

An experimental fuel, which was transient-tested in the Third Risø Fission Gas Project, released gas at an unusually low power of about 11 kW/m during the ramp. The percentage of gas released, 6–7% in the instrumented tests, was small compared with the 35–41% release obtained when the same fuel was transient-tested to 40 kW/m. The evidence suggests that the gas that was released had left the UO2 matrix during the base irradiation and had been stored for some considerable time in large closed pores in the fuel. It is assumed that release occurred when some of these pores were vented by cracks initiated in response to the high ramp rate. The main conclusion is that in low density fuel, fission gas may collect in the fabrication pores during steady-state irradiation and be released during load following if cracking occurs.