1 February 2019 in Chemical Engineering Science
23 July 2020 in Nano Letters
1 September 2020 in Journal of Materials Chemistry A
Lithium metal is one of the most promising anode materials for next-generation batteries. However, the growth of lithium dendrites and the high reactivity of lithium cause serious safety and cyclability problems. A comprehensive understanding of the lithium deposition behavior, especially during the initial deposition stages, is critical for its performance improvement. The influences of current density and capacity on the nucleation size, site distribution, and growth pattern were discussed via in situ electrochemical AFM (EC-AFM) and COMSOL simulation, suggesting that lithium deposition initially follows reaction- and then transport-limited mechanisms. In addition, the solid electrolyte interface (SEI) during lithium deposition was systematically analyzed by conductive-AFM, AFM force probing and X-ray photoelectron spectroscopy (XPS) depth profiling to give a comprehensive picture of the electronic, mechanical and chemical properties of SEI film under realistic conditions.
23 July 2020 in Advanced Functional Materials
1 March 2023 in Journal of the Electrochemical Society
An efficient solver for the pseudo-2-dimensional (P2D) model of Li-ion batteries is developed based on the open-source computational fluid dynamics platform OpenFOAM. Species and charge conservation equations are solved in a segregated manner: potential and concentration of electrolyte and solid electrode are solved sequentially. The non-linearity and interlinkage are handled by an iterative procedure. In the P2D model, solving Li-ion (de)intercalation in solid particles implicitly is time-consuming. The Picard method is employed for calculating solid concentration explicitly without the need to solve the discretized equations system. In one time step, Li-ion concentration of future time is recursively approximated until convergence. The explicit solving of solid concentration in combination with the overall iterative procedure makes solving of the P2D model efficient. Nonlinear source terms are linearized whenever possible to retain stability. Adaptive time-stepping is devised, grounded on the derivative of open circuit voltage (OCV) with regard to Li-ion concentration at the particle surface. The developed solver is validated in detail with respect to COMSOL. Performance under various operation conditions, such as discharge of rate 10 C, hybrid charge-discharge cycle, and urban dynamometer driving schedule (UDDS), demonstrates its robustness.
13 July 2018 in Transport in Porous Media
Liquid penetration into thin porous media such as paper is often simulated using continuum-scale single-phase Darcy’s law. The underlying assumption was that a sharp invasion front percolates through the layer. To explore this ambiguous assumption and to understand the controlling pore-scale mechanisms, we have developed a dynamic pore-network model to simulate imbibition of a wetting phase from a droplet into a paper coating layer. The realistic pore structures are obtained using the FIB-SEM imaging of the coating material with a minimum resolution of 3.5 nm. Pore network was extracted from FIB-SEM images using Avizo software. Data of extracted pore network are used for statistically generating pore network. Droplet sizes are chosen in the range of those applicable in inkjet printing. Our simulations show no sharp invasion front exists and there is the presence of residual non-wetting phase. In addition, penetration of different sizes of droplets of different material properties into the pore network with different pore body and pore throat sizes are performed. We have found an approximately linear decrease in droplet volume with time. This contradicts the expected \(\sqrt{t}\)-behavior in vertical imbibition that is obtained using macroscopic single-phase Darcy’s law. With increase in flow rate, transition of imbibition invasion front from percolation-like pattern to a more sharper one with less trapping of non-wetting phase is also reported. Our simulations suggest that the single-phase Darcy’s law does not adequately describe liquid penetration into materials such as paper coating layer. Instead Richards equation would be a better choice.