Document Type



Doctor of Philosophy


Mechanical Engineering

First Adviser

Oztekin, Alparslan

Other advisers/committee members

Oztekin, Alparslan; Neti, Sudhakar; Jaworski, Justin; Jain, Himanshu; Singh, Dileep


A numerical heat transfer analysis of encapsulated phase change material (EPCM) capsules was conducted by employing the enthalpy-porosity and VOF methods simultaneously to capture the complex multi-phase heat transfer that occurs within the capsules. The results of the numerical methods employed were validated by comparing the final shape of the solid phase change material (PCM) predicted to that seen within sectioned experimental capsules. The validated methods were used to study the effect that an internal void space has on the heat transfer within an EPCM capsule. Its effect is immediately noticeable as the isotherms no longer form the concentric rings predicted by the unsteady diffusion equation since the void acts as an insulator reducing the conduction rate in the upper portion of the capsule. Additionally, the increased melting rate resulting from convection in the molten PCM further reshapes the solid-liquid interface. In contrast, the solidification process is conduction-dominated and limited by the thermal conductivity of the chosen PCM resulting in considerably longer solidification times. The impact of an internal void on the overall heat transfer was further examined by considering three limiting cases of an upper void, central void, and random void distributions where the upper void is positioned opposite to the orientation of the gravitational vector. Since the PCM for the central void distribution is in direct contact with the entire capsule shell, it has the highest heat transfer rate during the initial melting stages leading to it having a melting time that is 22% and 39% faster than the random and upper void distributions. In an ideal world, one would like to keep the void located as close to the center of the capsule as possible. The vastly different evolution of the solid-liquid interface for the three cases considered highlights the profound impact an internal void has on the temporal and spatial evolution of the solid-liquid interface. The results for a single EPCM capsule were extended by evaluating the performance of a pilot-scale EPCM-based latent heat thermal energy storage (TES) system. The capsules sequentially showed the same evolution of the melting front within the capsules over the course of the charging process. The numerical results were compared to experimental recorded values for the temperature in the furthest downstream EPCM capsule. Agreement within 3% was seen during the initial solid sensible heating stages, however as the capsules began to melt the discrepancy increased due to a poor estimate for the values of the latent heat of fusion of NaNO3. This resulted in a 8% faster predicted melting time. However, this has minimal effect on the overall energy storage of the system due to the large operational temperature range applied in the current experiment. 65% of the 22 MJ of energy release by the heat transfer fluid (HTF) was stored in the 17.7 kg of NaNO3; 20% of this energy can be attributed to latent heat energy storage. Therefore, the system is able to store a large fraction of energy supplied by the HTF with a significant contribution from latent heat. Furthermore, if the operational temperature range were smaller the fraction of latent heat energy storage would have been significantly larger. The second law analysis of an example TES system was conducted to determine the benefit of s system employing a multiple PCMs. As expected, the latent heat-based systems were able to store more energy and exergy with comparable efficiencies than systems that rely on sensible heat only. Furthermore, when the overall cycle performance is examined, systems with multiple PCMs perform better than corresponding single PCM-based systems. While for the operating conditions and PCMs chosen the 2-PCM system was superior, great care is required during the design of an EPCM-based TES system as the difference between the melting point of the PCMs and the inlet temperatures during charging and discharging greatly affect the performance of the system. Lastly, experimental evaluations of the use of metallic oxides as new novel PCMs were conducted. In particular, the eutectic compounds in the Na2O-B2O3 system were investigated as they offer higher energy storage densities at melting temperatures comparable to the previously investigated nitrate and chloride salts. However, the material that was formed during the preparation of the samples was not a eutectic compound due to sodium evaporation and therefore did not melt congruently. Once the initial composition of the material was determined, the discrepancy between the theoretical and experimental energy storage were shown to be within the ±2% uncertainty of the calorimetry system. While these initial results are promising, further research is required before these metallic oxides can be integrated into EPCM-based latent heat TES systems as a superior alternative to the currently employed nitrate and chloride salts.