A novel biomass sorption-enhanced chemical looping gasification process to achieve maximized hydrogen yield under autothermal equilibrium: experimental validation and DNN-augmented process simulation

Biomass-based sorption-enhanced chemical looping gasification (SECLG) is an innovative yet challenging technology for maximizing hydrogen yield autothermally. In this study, a novel autothermal SECLG process is proposed to maximize hydrogen yield. This process combines experimental validation with process simulation enhanced by deep learning neural networks. The results indicate for each Ca/C (CaO-to-C molar ratio), net heat decreases with increasing ratio of solids flowing from calciner to air reactor α, and decreasing ratio of solids flowing from air reactor to fuel reactor β. The data points where net heat = 0 for each Ca/C was extracted, which represents the condition for achieving autothermal equilibrium in the SECLG. Autothermal operation is achievable only for k (NiO-to-CaO molar ratio) ≥ 0.2 at Ca/C = 0.5. In contrast, at Ca/C = 1 and 1.5, autothermal equilibrium is observed across the entire range of k values. To maximize H2 yield under autothermal equilibrium for each Ca/C, value of α should be increased, while β decreased; maximum yield (83.51g H2/kg biomass) occurred at Ca/C = 1.5, k = 0.11, α = 0.99, β = 0.01. Increasing Ca/C enhances maximum H2 yield and significantly expands the autothermal operation regime to 93.3%. This study provides valuable insights into the design and optimization of SECLG process.