UCLA

In the field of solar thermochemical energy storage, ammonia synthesis/dissociation is feasible for practical use in the concentrating solar power industry. In ammonia-based solar thermochemical energy storage systems, the stored energy is released when the hydrogen (H2) and nitrogen (N2) react exothermically to synthesize ammonia (NH3), providing thermal energy to a power block for electricity generation. But ammonia synthesis has not yet been shown to reach temperatures consistent with the highest performance modern power blocks (~650 °C). The following dissertation addresses an ongoing investigation into the field of ammonia based thermochemical energy storage.

In the first part of the dissertation, the state of the art of Concentrating Solar Power (CSP) and Thermochemical Energy Storage (TCES) are reviewed. In the second part, a two-dimensional model is proposed to simulate heating supercritical steam in an ammonia synthesis reactor. Thirdly, the model is validated as the model predicted temperature profiles match with experimental measured results well. A sensitivity analysis is carried out for the model to study the effects of six input parameters on heat transfer and reaction kinetics. The results show the process is “heat-transfer-limited” and most sensitive to activation energy. The process is also very sensitive to inlet ammonia mass fraction. Improving heat transfer and decreasing inlet ammonia mass fraction are crucial to improve the capability of the reactor to heat steam. In the fourth part, some preliminary designs for ammonia synthesis systems are proposed. Parametric studies are made with the model for each component in the proposed systems. The systems are optimized and investigated to minimize the total tube wall volume per unit power. The result shows that improving heat transfer by small dimensions and subdividing the reactor into different sections for different steam temperature ranges are good for minimizing the tube wall volume for the system. Also, it is necessary to optimize the entire system simultaneously since the wall volume for each component is comparable. At last, a future study plan for model improvement and system optimization is proposed.