Supercooling, Crystallization and Melting within Emulsions and Divided Systems: Mass, Heat Transfers and Stability

The aim of this chapter is to give a view about the conditions of crystallization and melting of liquid dispersed within emulsions submitted to a temperature scanning or stored at a fixed temperature. Basic emulsions defined as droplets dispersed in a quasi non miscible liquid and more sophisticated ones as mixed and multiple will be considered. It will be shown that as far the phases are concerned with the state changes are dispersed or divided, the crystallization occurs when the emulsion or the divided system is cooled below, sometimes far below, the melting temperature corresponding to the solid-liquid equilibrium which is observed, on the contrary, without any delay upon heating. The statistical character of the crystallizations of the droplets, driven by the laws of nucleation, will be explained and the probability of crystallization will be defined. Pure dispersed material and binary solutions will be considered. Experiments obtained by using Differential Scanning Calorimetry (DSC) show that the supercooling depends on different parameters such as the volumes and composition of the droplets, and the number and the nature of the freezing-melting cycles. It appears that the smallest the system volume, the highest the concentration, the lowest is the crystallization temperature. Furthermore, for certain dispersed organic compounds, at the crystallization of very supercooled droplets, the appearance of metastable unknown crystalline phases is observed. The temperature and latent heat of transformation between these phases or with classical stable phases are determined and schemes representing the Gibbs free energy explain the succession of possible phase changes.

In this Chapter 2 mass transfers occurring within emulsions are described. Different kinds of mass transfer are considered. First it is the one resulting of the presence of yet frozen droplets and still supercooled droplets due to a difference in the chemical potentials of ice and liquid water. This kind of mass transfer has been referred as solid ripening. Another mass transfer leading to the formation of solid is the introduction in the emulsion of a material that could form gas hydrate The compound CCl3F has been tested. By using the technique DSC and X-Rays diffraction it was possible to detect information about the formation and dissociation of the hydrate. Another kind of mass transfer is the composition ripening. It is observed when the compositions of the different phases present in the emulsions are different. Mass transfer in mixed emulsions made of two populations of droplets having different composition either, urea+water and pure water or tetradecane and hexadecane have been studied. Mass transfer induces a change in the composition of the droplets and therefore a change in the freezing and melting temperatures determined by a cooling and a heating of the mother emulsion maintained at ambient temperature. Calculating the energies involved, the percentage of compound transferred can be determined. Similar mass transfers are observed in multiple emulsions. The proposed model for the transfer shows that the transport is enhanced by the presence of micelles. All these transfers must have an influence on the further stability of simple or mixed emulsions after the mass transfer as bigger droplets, and solids are present.

In this chapter, we describe the crystallizations of the supercooled droplets of the emulsion but we take into account the corresponding released energy. It is sufficient to warm up the emulsion at a temperature sufficiently high to be in a range where the nucleation rate vanishes stopping the crystallizations. So, we observe a local self-regulation of the temperature. On the contrary, the melting of the crystallized droplets occurs when the local temperature induces the thermodynamic equilibrium. These phases transforms are shown whether on a cylinder about 500 cm3 in volume or by calorimetry experiments on smaller samples. Models are presented to explain in all cases the observed phenomena. They are based on the resolution of the equation of the heat conduction with a heat source proportional to the probability of crystallization, given by the nucleation laws, and to the number of the remaining unfrozen droplets. In the particular case of a high concentrated emulsion of water, it is demonstrated the possibility of seedings increasing the number of crystallizations. Different methods are presented to determine the nucleation rate function and, in particular, a characterization by an inverse method. We describe also the case of emulsions flowing in a pipe. We mainly treat the case of the pure substances, but the case of binary solutions is approached.

In this chapter, we present experimental results and modelings for a divided system constituted by a tank filled by spherical nodules heated or cooled by a fluid which passes through it. These nodules contain a phase change material as pure compounds or eutectics. The industrial application concerns the latent heat storage of the thermal energy. We present examples in the field of the sub-ambient storage, but our investigations might be extrapolated at other different levels of temperature. First, we describe the kinetics of phase transforms in the case of individual nodules and the role of an enhancement of the heat conductivity by adding divided carbon. We will notice some similitude with the emulsion because the supercooling phenomenon is also present in spite of the higher volume of phase change material. In the tanks, we will describe a kind of self-regulation. A simplified model, in case of the vertical flows, permits the description of the energy storage either in charge or in discharge mode and explains the self-regulation. The difference with the emulsions is that, due to a smaller supercooling, we must take into account the temperature field inside the nodule. The influence of characteristic parameters such as the flow rates of the fluid or the final inlet temperatures is detailed. The case of the horizontal tank where natural convection is concerned is evoked. Finally the case of incomplete discharge modes, often programmed, is described.