Fabrication, characterization, and use of core-shell composite phase change materials for thermal energy storage

Abstract

It is challenging to utilize low-grade heat, such as the waste heat from data center cooling, because the location, time, and temperature of the low-grade heat supply can differ significantly from those of the heat demand. Thermal energy storage using phase-change materials (PCMs) could buffer the mismatch. This thesis conducts a multi-scale study on a type of novel PCMs, i.e., core-shell structured PCMs, for thermal energy storage. Core-shell particles of n-octadecane encapsulated in porous silica (n-OD@SiO₂) are used to resolve the detailed microstructure of the core-shell PCM particles, provide a microstructure-based account of supercooling and the associated synthesis method to control supercooling, translate the particulate thermal and structural characteristics into packed-bed heat-transfer and storage performance, and finally evaluate the use a packed-bed core-shell structured PCMs in low-temperature heat recovery from data-center. On a particle level, varying the emulsion and interfacial-reaction conditions can systematically modify core-shell microstructures. Among all the n-OD@SiO2 particles synthesized, it was found that particle diameters spanned 70.6–559.9 nm, calculated shell porosities were between 0.58 and 0.88, and the deep supercooled fraction of the cooling enthalpy varied from 9 to 52%. Using thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC), the active n-OD and retained liquid from synthesis were identified. A three-region interpretation of the supercooling was obtained, n-OD adsorbed on the SiO₂ skeleton and n-OD confined in the shell both crystallizes at higher temperatures through shell-induced heterogeneous, whereas n-OD in the core interior crystallizes only at deep supercooling through homogeneous nucleation. Based on the measured microstructures and compounds of the n-OD@SiO2 particles, as well as the thermal properties such as supercooling temperatures and phase states of the particles, a phase-dependent pore-network model was constructed to study the effective thermal conductivity (keff) and apparent thermal energy storage density of n-OD@SiO2 core-shell PCM packed beds. The predicted keff values of 0.088-0.102 W m⁻¹ K⁻¹ agreed with measurements on pressed pellets (0.099-0.106 W/m K) within 3.8-11.1%, with the closest agreement (-3.8%) for the porosity-matched sample, outperforming classical effective-medium estimates. Parametric analysis showed that a thicker shell increases keff but reduces the available latent heat, whereas a more porous shell does the opposite, and particle size has only a weak effect under consistent geometric scaling. Bed-scale design is therefore a trade-off between heat transfer and storage capacity rather than conductivity maximization. At system level, the core-shell PCM packed bed is used for thermal energy storage from data centre waste heat. At the cold-climate benchmark (climate zone 5A), direct use of data-center heat dominates the benefit, cutting annual heat-pump electricity by 77.5 MWh (39.9%) relative to an ambient-source reference and raising the seasonal COP from 2.39 to 3.97. Adding the CPCM packed-bed storage saves a further 8.6 MWh, i.e., 7.4% of the electricity of the direct-use configuration without storage, for a cumulative saving of 44.3%. Supercooling reduces this benefit: assigning latent heat to the 9 °C release pathway instead of the 25 °C pathway lowers the storage-side source temperature and increases the heat-pump lift. At the measured deeply supercooled fraction of 0.36, the storage saving decreases from 9.2 to 8.6 MWh, and to 7.5 MWh if all latent heat is deeply supercooled, although the storage remains beneficial in every tested case. In summary, the thesis has the following contributions on multi-scales. First, it identifies and measures the parameters that govern core-shell PCM performance, e.g., the particulate compounds, the shell thickness and porosity, and the temperature-resolved heat release distribution. Second, using the experimental measurements, a core-shell PCM packed bed model is constructed and validated, and the effective thermal conductivity and thermal energy storage ratios are analyzed. Third, a data centre waste heat utilization system using a core-shell PCM packed bed is evaluated, and the advantages of thermal energy storage and the disadvantages of supercooling are understood.