Current Status of Experimental and Theoretical Work on Sodium/Fuel Interaction (SFI) at Karlsruhe 'Simulation Experiment'

Experimental and theoretical work on sodium fuel interaction (SFI) is being carried out from many years at the nuclear research center of Karlsruhe (Germany) within the framework of the 'Projekt Schneller BrUter' (PSB). This paper concerns with the description and the discussion of the experimental and theoretical results already obtained. A test loop has been constructed in which experiments can be carried out with different geometries. The sodium can be either stationary or flowing and its initial temperature can go up to 700 deg. C. The fuel pins are heated electrically. Some experimental results under reactor conditions have already been obtained and are being analysed. Small scale experiments were performed in water and sodium with small amounts of Ag, Al, Au, Cu, Fe, Pb, Sn, Zn and stainless steel. Aims of these experiments are to analyse the general character of the thermal interaction between the hot and the cold material, to measure the Leidenfrost temperature and to investigate trigger mechanisms which may lead to thermal explosion. Two different codes have been developed at Karlsruhe. An extensive parametric analysis has been carried out with both codes. Results with the first code show that the total mechanical work produced decreases with the time scale of the vapour film layer around the fuel particles and increases with the speed with which the fuel breaks down into fragments and mixes with the liquid sodium. The time to empty a reactor channel is between 15 and 20 msec for values of the fragmentation and mixing time constants between 5 and 10 msec. The results obtained from the second code are basically consistent with those obtained from the first code. Here the effect of the thermal conductivity of the coolant and the propagation effects of the pressure within the reaction zone have been investigated. Some results of fuel coolant experiments in Na show that the mechanical energy has a maximum at a Na-temperature of about 720 deg. C. This can be explained by assuming that this energy is due to the collapse of vapour bubbles. A model is being developed based on this assumption. In addition work is being carried out in cooperation with the european nuclear research center at Ispra. Small quantities (o.5 gr) of various molten metals (Ag; Al; Au; Cu; Fe; Pb; Sn; Zn and stainless steel) were heated in a levitation coil and dropped in water contained in a plexiglass vessel (40 mm diameter). The water temperature was 20 deg. C. Temperature of the hot material was measured with thermocouples placed at the bottom of the vessel. Pressure was also measured and high speed films of the interaction were also taken. Experiments were carried out in air as well as in inert atmosphere (nitrogen). Analysis of the debris after interaction included microscopic, mechanical, metallographic and chemical analysis. The copper/water system was particularly investigated (melting point of copper 1083 deg. C). Three different outcomes of the copper experiments were observed A. The copper drop entered the water and fragmented before reaching the bottom of the vessel B. The copper drop reached the bottom of the vessel and after some time fragmented C. The copper drop reached the bottom of the vessel and did not fragment Outcomes 'A' and 'B' were observed for initial copper temperatures of 1400 deg. C and 1650 deg. C respectively. When the drop reached the bottom film boiling regime took place. In the case of outcome 'B' small jets coming from inside the particle were sometimes observed 1 to 10 msecs before fragmentation. Thermal explosion took place only when copper was heated in air. In the case of outcome 'C' the Leidenfrost phenomenon was investigated. Results were found to basically agree with existing analytical correlations