The heat and mass transfer processes taking place in self-heated pools growing in soluble, gas-releasing solid beds have been investigated. Simulating experiments have been used to examine the effects of power density, pool-to-bed density ratio, and volumetric gas release ratio on the evolution of pool shape. A computer program, UWMCCI, has been developed and used to compare the experimental data to predictions of both gas-film- and gas-bubbling-type heat transfer models. The program has also been used parametrically to compare these two types of models at prototypical gas release rates. The following conclusions are drawn:

  1. The use of gas bubbling models will result in significantly larger pool dimensions than those predicted using film-type models.
  2. Current gas-bubbling-type models will overpredict the heat transfer coefficients even at superficial gas velocities within their valid range since they do not account for mass transfer effects. More precisely controlled simulation experiments combining heat and mass transfer effects are needed. Experiments aimed at determining the lower limit of applicability of gas-film-type models are also needed.
  3. The extreme sensitivity of the heat transfer coefficient predicted by gas-film-type models requires accurate knowledge of the transient history of the pool chemical composition and physical properties.