Dynamic one-dimensional model for biological methanation in a stirred tank reactor

Power-to-gas technology can facilitate the transition toward a renewables-based energy system by converting excess electricity to hydrogen and then into methane via methanation. Unlike traditional chemical methanation, biological methanation uses an aqueous solution of biomass (archaea), which consumes H-2 and CO2 to produce CH4. The process is limited primarily by the gas liquid mass transfer step. In addition to experimental research, modeling is often used to guide and expedite the development and scale-up of bioreactors from the laboratory to the pilot and commercial scales. Modeling has been used to optimize and test various operation conditions outside the range of experimentation. Estimations of gas-liquid mass transfer and the related stirring power are important for optimization and feasibility studies in the application of biological methanation to power-to-gas systems. Related published literature, however, is limited. In this study, a dynamic model for a continuously stirred biomethanation reactor was developed with novel approach that combines semi-fundamental modeling of gas liquid mass transfer, hydrodynamics, and biological reactions. The model was validated against existing experimental data and used in a sensitivity analysis of critical parameters, a scale-up study of a biomethanation reactor, and process dynamics studies. In each of the varying operational conditions, the model reproduced the trends observed in the experimental studies. The sensitivity analysis showed that biological parameters have a minimal effect on methane production. Conversely, the model is very sensitive to the gas-liquid mass transfer properties, such as the geometry of the impeller and reactor. The scaled-up study of biomethanation reactors with a CH4 production capacity of 56-508 Nm(3)/h revealed that the required stirring power is 0.7-1.1% from the electrolyzer power and decreases as the size of the reactor increases. High output quality ( similar to 98%) of the methane could be reached in each of the studied cases, and the overall efficiency of the power-to-methane process was roughly 50%. Dynamic simulations showed that the modeled process is tolerant to large gradients in the input parameters. After correctly setting the reactor- and process-specific parameters, the model can be used to perform scaled-up and dynamic studies of various reactor designs and different biomass solutions.