Heat Transfer to Suspensions of Microencapsulated Phase Change Material Flowing Through Minichannels

The heat transfer to water-based suspensions of microencapsulated phase change material (MEPCM) flowing laminarly through rectangular copper minichannels was investigated both experimentally and numerically. The MEPCM-particles had an average size of 5 μm and contained as phase change material n-eicosane, which has a theoretical melting temperature of 36.4 °C. Water and suspensions with particle mass fractions of 10% and 20% were considered. While the experiments result in rather global values such as wall temperatures at certain points, suspension in- and outlet temperatures, and the pressure drop, the numerical simulations allow additionally a more detailed insight, for example, into the temperature distribution in the flowing suspension. The results show that MEPCM suspensions are only advantageous in comparison to water in a certain range of parameter combinations, where the latent heat is exploited to a high degree. The available latent heat storage potential, which depends on the particle fraction in the suspension and on the mass flow rate, has to be in the same order of magnitude as the supplied heat. Moreover, the mean residence time of the particles in the cooling channels must not be considerably shorter than the characteristic time for heat conduction perpendicular to the flow direction. Otherwise, the particles in the center region of the flow leave the cooling channels with still solid cores, and their latent heat is not exploited. Furthermore, the benefit of the added MEPCM particles depends on the inlet temperature, which has to be slightly below the theoretical melting temperature, and on the subcooling temperature after the heat supply, which has to be sufficiently low to guarantee that the entire phase change material solidifies again before it re-enters the cooling channels. The suspensions showed Newtonian behavior in the viscosity measurement. The actual pressure drop determined in the experiments is smaller than the pressure drop estimation based on the measured viscosities. The difference between the two values increases with increasing particle mass fraction. This shows that the particles are not evenly distributed in the flowing suspension, but that there is a particle-depleted layer close to the channel walls. This reduces the required pumping power, but makes it even more important to provide conditions, in which a sufficiently large amount of the supplied heat is conducted to the center region of the channels.