A Multiscale Framework for Predicting Electrode Slurry Rheology and Stability

Thesis Abstract

This study addresses the challenge of reliably predicting the rheology and stability of electrode slurries used in lithium-ion battery manufacturing, where inconsistent flow and sedimentation impede coating quality and electrode integrity. Despite advances in slurry formulation, there remains a gap in integrative frameworks that connect nanoscale particle interactions, mesoscale microstructure evolution, and macroscale rheological response. The aim is to develop a multiscale framework that predicts slurry rheology and stability under varied composition, solids loading, and processing conditions, enabling optimization of slurry formulations and coating processes. Specific objectives are (1) to quantify how particle size distribution, solid content, binder type, and solvent properties influence rheological behavior across scales; (2) to establish interscale couplings between nanoparticle agglomeration, paste yield stress, and flow stability using coupled DEM-CFD and population balance models; (3) to calibrate and validate the framework against experimental data from model and industrial slurries; (4) to identify processing windows that minimize sedimentation and phase separation while maintaining workable viscosity; and (5) to provide actionable guidelines for formulation and processing that reduce defect rates in electrode coatings. A mixed-methods approach is employed, integrating experimental measurements with computational modeling. The population under study comprises representative electrode slurry systems with graphite and silicon-based active materials, polyvinylidene fluoride (PVDF) binder, and carbon black in N-methyl-2-pyrrolidone (NMP) solvent, prepared at solids loadings of 40–70 vol%. Experimental data are drawn from 150 slurry samples produced under controlled variations of particle size distribution, binder-to-carbon ratio, and solvent quality. Rheological characterization includes steady shear, oscillatory rheology, and thixotropy tests using a rotational rheometer (TA Instruments HR-3) with parallel-plate geometry. Sedimentation stability is assessed via analytical centrifugation and optical coherence tomography (OCT) for time-resolved height and density stratification. Microstructural characterization employs focused ion beam-scanning electron microscopy (FIB-SEM) and X-ray computed microtomography to quantify particle networks, porosity, and agglomerate statistics. The multiscale framework comprises (i) a nanoscale interaction model capturing van der Waals and electrostatic forces governing agglomeration; (ii) a mesoscale discrete element method (DEM) coupled with computational fluid dynamics (CFD) to predict slurry flow, viscosity, and stability as a function of particle packing and binder distribution; (iii) a macroscale rheological constitutive model informed by the hierarchical structure, integrating yield stress and thixotropy parameters; and (iv) a system-level optimization module that maps formulation variables to processing windows for coating and drying. Model parameters are estimated via Bayesian inference and Markov chain Monte Carlo (MCMC) to quantify uncertainties. Validation is conducted against 60 independent slurry samples and industrial coating trials, with performance metrics including predicted viscosity at shear rates of 10–1000 s?1, yield stress, thixotropic recovery, and sedimentation onset time. Anticipated findings indicate that multi-scale coupling markedly improves prediction accuracy of viscosity (R2 > 0.92) and sedimentation onset (RMSE < 5% of initial height) relative to single-scale models. The framework is expected to reveal critical thresholds where binder distribution and particle fines content synergistically suppress agglomeration while maintaining processable rheology. The study contributes to knowledge by providing a rigorously validated, transferable multiscale model that unites colloidal interaction theory, granular dynamics, and continuum rheology for battery slurries, enabling design of robust formulations and coating processes with reduced defect densities. The results will offer practical guidelines for selecting particle size distributions, binder formulations, and solids loading to achieve target rheology and stability, along with sensitivity analyses that highlight the most influential parameters. The main conclusion is that a coherent multiscale framework can predict both rheological performance and long-term stability of electrode slurries, facilitating accelerated development cycles and higher-quality electrode coatings. Recommendations include adopting multiscale modeling as a standard tool in slurry formulation, extending the framework to incorporate temperature effects during drying, and applying the approach to other particulate suspensions in energy storage manufacturing.