Chemical looping air separation for oxy-fuel power plants

Research Doctorate - Doctor of Philosophy (PhD)Oxy-fuel combustion, which refers to the combustion of coal in the presence of oxygen rather than air, is one of the key technology options among the portfolio of carbon capture and storage (CCS) technologies. Oxy-fuel combustion captures carbon dioxide in-situ, producing a CO2-enriched flue gas stream (95 vol% CO2) ready for storage. The main shortcoming of the oxy-fuel combustion, however, is the high cost and energy intensity of its oxygen plant. At present, the main technology option for the oxygen plant is cryogenic air separation which typically consumes about 20% of the total electricity produced in oxy-firing mode and, thus significantly reduces the thermal efficiency of the oxy-fuel power plant. The oxygen plant also accounts for 30% of the total capital investment. In view of the above, a comprehensive program of study on alternative air separation technology - chemical looping air separation (CLAS), which offers a cost effective method for large-scale oxygen production, has been systematically conducted in the current study. The present study aims to identify and characterise the suitable metal oxide oxygen carriers for CLAS applications. The broad objectives of the project have been achieved using a combined theoretical and experimental approach. A comprehensive review of the current state of oxygen carrier developments and utilisation was carried out. Other emerging air separation methods have also been reviewed in detail as part of this literature review. A thermodynamic method of identifying the feasible metal oxide oxygen carriers for high temperature air separation was developed in the present study. The energy cost associated with oxygen production using CLAS and its comparison with an advanced cryogenic air separation unit has been investigated in detail. The reaction mechanisms, underpinning the oxidation and oxygen release processes under conditions pertinent to CLAS, were also determined as one of the main objectives for the current study. The relevant experiments have been carried out using a variety of experimental setups, including thermogravimetric analysis (TGA), packed-bed, and interconnected circulating fluidised beds (ICFB) system. Numerous oxides of metal elements from the periodic table were systematically investigated as potential oxygen carrier candidates for CLAS based on a thermodynamic approach in the current study. The majority of the metal oxides exhibit the capability of releasing oxygen. However, most of them cannot be used in CLAS considering the oxygen equilibrium partial pressure (EPP) lower than 0.015, which may increase the use of sweep gas during the oxygen releasing process and correspondingly the energy consumption. Moreover, the likely carbonation and formation of hydrides on exposure to CO2/steam enriched conditions for metal oxides, such as the Ca and Na-based oxides, limit their application in the preparation of oxygen carriers. Only Mn3O4/Mn2O3, CoO/Co3O4, and Cu2O/CuO are found to be the most suitable oxidation pairs to transport oxygen in CLAS process. Furthermore, a special case for the integration of CLAS with oxy-fuel power plant, i.e., the direct use of clean CO2 flue gas as the sweep media during oxygen release for CLAS, was used for the operation optimisation based on these three systems. Results have shown CLAS has significant lower energy consumption for oxygen production, compared to the advanced cryogenic air separation. Furthermore, oxygen carriers of Mn-, Co-, and Cu-based metal oxides with Al2O3 or SiO2 as a support were prepared by the dry impregnation method with an exception of the spray dried CuO/MgAl2O4. The reactivity of these prepared oxygen carriers was determined through TGA in the temperature range of 800-950oC. Among these studies, CuO/SiO2 and Co3O4/Al2O3 exhibit a very fast reaction rate in both oxidation and reduction processes. Co3O4/Al2O3 has shown a higher rate of oxygen transport (ROT) for the oxygen release than CuO/SiO2 at the same temperature, but a lower ROT of oxidation. It was also noted that the Mn and Co-based oxygen carriers on both Al2O3 and SiO2 have thermodynamic limitations in high temperature oxidation under air. The CuO/MgAl2O4 carrier was found to have a very stable performance and the highest oxygen transport capacity (OTC), however, very low mechanical strength after test, less than 0.8 N. Oxygen content attaining the equilibrium partial pressure was achieved from CuO/SiO2 and CuO/MgAl2O4 in a packed bed reactor at the temperature of 800, 850, 900, and 950oC. The reduction and oxidation kinetics were determined for CuO/SiO2 and CuO/MgAl2O4. Avrami-Erofe'ev random nucleation and subsequence growth (A2) and phase boundary reaction (R2) mechanisms were found to fit the experimental results very well for both of these two oxygen carriers. In response to the sinter issue of the mono-metallic CuO/SiO2 oxygen carriers, attempts were made to add promoters to CuO/SiO2 including MgO, Fe2O3, NiO, Mn2O3, and Co3O4. The results indicated that the agglomeration of CuO can be supressed via the proper addition of secondary metal oxides, such as MgO, Fe2O3, NiO, and Co3O4. The best improvement was achieved by MgO with the added weight no less than 20.4 wt%. In addition, an interconnected circulating fluidised beds (ICFB) system was designed, built, and operated for CLAS. The CLAS process has been successfully demonstrated using a Cu-Mg bimetallic oxygen carrier. Oxygen product at a level very close to the equilibrium concentration was continuously obtained in this rig