Thermal runaway (TR) has become a critical safety concern with the widespread use of lithium-ion batteries (LIBs) as an energy storage solution to meet the growing global energy demand. This issue has become a significant barrier to the expansion of LIB technologies. Addressing the urgent need for safer LIBs, this study developed a comprehensive model to simulate TR in cylindrical 18650 nickel cobalt manganese (NMC) LIBs. By incorporating experiments with LG^®-INR18650-MJ1 cells, the model specifically aimed to accurately predict critical TR parameters, including temperature evolution, internal pressure changes, venting phases, and mass loss dynamics. The simulation closely correlated with experimental outcomes, particularly in replicating double venting mechanisms, gas generation, and the characteristics of mass loss observed during TR events. This study confirmed the feasibility of assuming proportional relationships between gas generation and the cell capacity and between the mass loss from solid particle ejection and the total mass loss, thereby simplifying the modeling of both gas generation and mass loss behaviors in LIBs under TR. Conclusively, the findings advanced the understanding of TR mechanisms in LIBs, providing a solid foundation for future research aimed at mitigating risks and promoting the safe integration of LIBs into sustainable energy solutions.

Thermal runaway (TR) has become a critical safety concern with the widespread use of lithium-ion batteries (LIBs) as an energy storage solution to meet the growing global energy demand. This issue has become a significant barrier to the expansion of LIB technologies. Addressing the urgent need for safer LIBs, this study developed a comprehensive model to simulate TR in cylindrical 18650 Nickel Cobalt Manganese (NMC) LIBs. By incorporating experiments with LG^®-INR18650-MJ1 cells, the model specifically aims to accurately predict critical TR parameters, including temperature evolution, internal pressure changes, venting phases, and mass loss dynamics. The simulation closely correlated with experimental outcomes, particularly in replicating double venting mechanisms, gas generation and the characteristics of mass loss observed during TR events. The study confirmed the feasibility of assuming proportional relationships between gas generation and cell capacity, and between mass loss from solid particle ejection and total mass loss, thereby simplifying the modeling of both gas generation and mass loss behaviors in LIBs under TR. Conclusively, the findings advanced understanding of TR mechanisms in LIBs, providing a solid foundation for future research aimed at mitigating risks and promoting the safe integration of LIBs into sustainable energy solutions.

The persistent reliance of the cement sector on fossil fuels for CaO production and the high energy demand for DAC sorbent regeneration pose a significant barrier to reaching sustainable industrial production. Herein, a novel process concept is presented, establishing an integrated, fully electrified $${\rm CO}_2$$-loop driven by gliding arc plasma technology. Atmospheric $${\rm CO}_2$$ is captured in an alkaline absorption unit, where the pH value serves as a reliable performance parameter for system predictability. Subsequent liquid-solid conversion solidifies the captured $${\rm CO}_2$$ as $${\rm CaCO}_3$$, achieving a high yield with high phase purity. Rotating gliding arc plasma experiments successfully demonstrate the electric calcination for CaO regeneration and the $${\rm CO}_2$$ release to close the carbon and calcium loop. With the ability of simultaneous plasma-based $${\rm CO}_2$$ splitting the process enables the use of the resulting CO as a precursor for synthesis gas. The observation of internal temperature gradients facilitates the definition of a suitable reaction volume for calcination. XRD and TGA/DSC analyses confirm the formation of pure Ca-products, validating the concept. This work provides a foundational material-flow validation for this novel DAC Power-to-X pathway, substituting fossil fuels for calcination and DAC-operation entirely with electrical energy. The process offers a crucial opportunity for the decarbonization of the construction and transport sectors, with future work focusing on optimizing reactor design for continuous operation.

Accurate parameter estimation is essential for improving the predictive capability of electrochemical battery models used for battery analysis, simulation, and management. However, estimating electrochemical model parameters remains computationally demanding because of the large parameter space and the nonlinear interactions between model parameters. This study proposes a sequential hybrid optimization framework that combines the population-based Cuckoo Search Algorithm (CSA) with surrogate-based Bayesian optimization for electrochemical battery model parameterization within the Python Battery Mathematical Modelling (PyBaMM) framework. CSA is first employed to estimate the complete parameter set, after which Bayesian optimization selectively refines parameters identified as exhibiting the largest estimation errors. Compared with the individual optimization methods, the hybrid framework achieves lower parameter estimation errors while substantially reducing computational time relative to CSA alone. These results demonstrate that sequential selective refinement provides an efficient and reliable approach for electrochemical battery model parameterization and supports accurate offline calibration of physics-based battery models.