Abstract The ARTEMIS experimental program is devoted to the study of corium/concrete interaction performed using simulating materials. The aim of the first series of tests was to study the phenomenology of the corium/concrete interaction, in particular at the interface between the liquid pool and the concrete, versus the heat dissipated in the pool and gas velocity. During this first series of tests, six 1D tests were performed. This interpretation of the experimental results shows that, if gas velocity and heat flux are sufficiently low, a stable crust, made of solid accumulations, builds up at the beginning of the interaction whereas the pool remains liquid. These solid masses are not melted or dissolved afterwards. They are not in thermodynamic equilibrium with the pool. In addition, the quantity of these solid masses depends on the initial superheat of the pool and has a significant impact on the progress of the interaction. If the conditions of crust build-up are not satisfied, the pool is in a semi-solid state. This report identifies the validity limits of a corium/concrete interface description based on the equilibrium phase segregation model and has led to the establishment of the basic features of a new model, suitable for this test series. This consists in considering the solid masses built up at the beginning of the interaction only as thermal resistance. This model has been successfully tested for the 5 reliable tests, using the MEDICIS module of the ASTEC code. The correlation between the experimental results and the calculations confirms this interpretation.