Hydrogen as well as the syngas (the mixture of CO and H
2) are important industrial raw materials widely used in chemical synthesis, ammonia production and the petroleum industry. The conventional steam methane reforming (SMR) produces about 75% of world’s total hydrogen but emits about 7 kg of CO
2 per kg H
2 on average, which is responsible for around 3% of the worldwide CO
2 emissions. Since the greenhouse gas emission is becoming more and more problematic, it is necessary to introduce new technologies that are environmentally friendly, low in energy consumption, and high in hydrogen efficiency. Chemical Looping Reforming (CLR) is an emerging technology that potentially reduce the CO
2 emissions and minimize the energy losses in syngas production. CLR shares the same principle as chemical looping combustion (CLC) in which the metal oxygen carrier (OC) is alternatingly oxidized and reduced using a dual-fluidized bed reactor that act as air reactor (AR) and fuel reactor (FR), respectively. The OC particles is oxidized by air in AR, while in FR, the OC is reduced by the methane and syngas while acting as catalyst for methane reforming. Accurate predication of the methane conversion and syngas composition is crucial for the design of industrial scale CLR unit. In this work, a detailed 1-D model coupling fluidized bed hydrodynamics based on Kunii-Levenspiel two-region model with intrinsic reaction kinetics including catalyst deactivation was developed. The back-mixing of solids in BFB is accounted in this model. This model can be used for process optimization and for the prediction of the performance of commercial scale a-CLR unit. An autothermal chemical looping reforming (a-CLR) unit that equivalent to 50 commercial SMR pipes is presented as an example in this paper. The required reactor design to reach a target reactor performance was studied. The effects of the main operating conditions on the performance of both the AR and the FR were analyzed. These results confirm the feasibility of the chemical looping reforming process with reactors of reasonable size for industrial applications.
We highlighted that the main difference between a-CLR and CLC is that, in the a-CLR, the oxygen carrier (OC) is mainly at reduced state in the FR, while it is only slightly oxidized (about 5% Ni is oxidized) in the AR and the autothermal operation can be guaranteed. Conventional well-mixed reactor assumption is not valid due to the low NiO-CH4 ratio in the a-CLR unit. Thus, 1-D axial dispersion model should be used to model solid phase. The required oxygen to CH4 ratio to maintain autothermal operating is about 1.14 if heat is sufficiently recovered from the off-gas of both the AR and the FR. The temperature difference between two reactors has less influence on the required oxygen to CH4 ratio to maintain autothermal operation while the small difference with respect to oxygen to CH4 ratio is resulted from energy loss of the off-gas of both reactor. The operating temperature of FR and interfacial transfer of the methane from the bubble phase to the emulsion phase limit the conversion of CH4, while the former also affects the conversion of OC in AR due to higher oxidation rate at higher temperature. With appropriate reactor design, near equilibrium conversion can be achieved.
