Modeling the drying process in hard carbon electrodes based on the phase-field method

TL;DR

This work develops a full-field multiphase-field framework to model pore emptying during drying of battery electrodes, incorporating two-phase flow, capillarity, wetting, and an evaporation mechanism. By applying a diffuse-domain formulation to embed solid phases and an evaporation term in the phase-field evolution, the approach resolves complex microstructures derived from SEM images of hard carbon electrodes. Validation on canonical wetting and capillarity tests confirms the method’s accuracy and convergence, and application to SEM-based HC-B and HC-C microstructures reveals that microstructural details and capillary dynamics strongly influence drying, breakthrough times, and solvent-front evolution. The study finds correlations between the capillary number and drying metrics, and shows that tuning the contact angle via surface engineering can optimize drying, with implications for binder distribution and electrode performance. Future work will extend to 3D microstructures and binder transport, enabling more comprehensive optimization of electrode drying processes.

Abstract

The present work addresses the simulation of pore emptying during the drying of battery electrodes. For this purpose, a model based on the multiphase-field method (MPF) is used, since it is an established approach for modeling and simulating multiphysical problems. A model based on phase fields is introduced that takes into account fluid flow, capillary effects, and wetting behavior, all of which play an important role in drying. In addition, the MPF makes it possible to track the movement of the liquid-air interface without computationally expensive adaptive mesh generation. The presented model is used for the first time to investigate pore emptying in real hard carbon microstructures. For this purpose, the microstructures of real dried electrodes are used as input for the simulations. The simulations performed here demonstrate the importance of considering the resolved microstructural information compared to models that rely only on statistical geometry parameters such as pore size distributions. The influence of various parameters such as different microstructures, fluid viscosity, and the contact angle on pore emptying are investigated. In addition, this work establishes a correlation between the capillary number and the breakthrough time of the solvent as well as the height difference of the solvent front at the time of breakthrough. The results indicate that the drying process can be optimized by doping the particle surface, which changes the contact angle between the fluids and the particles.

Figures (16)

• Figure 1: Schematic representation of the drying process of the battery electrodes.
• Figure 2: Validation case of an evaporating fluid in a capillary.
• Figure 3: Initial setup for the rise of a fluid in a capillary.
• Figure 4: Rise of a liquid in a capillary: Dimensionless rise height $\hat{h}$ versus the Bond number for simulation results and the analytical solution. Additionally, the effect of the grid refinement $n_{\text{x}}$ on the rise height $\hat{h}$ is shown as an inlet at a specific Bond number of $\mathit{Bo}=0.05$.
• Figure 5: Simulation results for different pore structures, with the initial condition in the top row and time of breakthrough in the bottom row. The fluids, referred to as Fluid 1 (air) and Fluid 2 (liquid), are shown in white and blue, respectively. The solid phase is displayed in gray.