Abstract:Increasing interest in the automotive field for electrification has led to an increased demand for lithium-ion batteries (LIB). LIB remain the most popular and widely applied energy storage technology. The microstructure of the positive electrode of LIBs governs its electrochemical performance. For this reason, it is quite imperative to determine an optimal electrode microstructure design to achieve the best possible energy properties. Mechanistic simulation models can be a powerful tool in analyzing the effect of microstructural parameters on the battery cell performance. Not only do they negate costs for necessary materials, but they also allow analysis and optimization while avoiding time-consuming experiments. Doyle-Fuller-Newman (DFN) models, being pseudo-2-dimensional (P2D), are computationally less expensive than their higher dimension counterparts; however, they often apply the Bruggeman relation to compute effective parameters such as the electrode conductivity of the electrode. Despite its simple implementation (i. e. a correction factor to reflect effective parameters), the Bruggeman relation delivers inaccurate results. [1, 2] In this work [1], we coupled a DFN model with the structure surrogate model from [2], excluding the Bruggeman relation. The model predicts the effective transport parameters and thus the battery performance more accurately. With this coupled model, we investigated the effect of the cathode mass loading and density on the achievable volumetric discharge energy density. [1] Furthermore, we developed a numerical optimization tool, which can predict the optimal structural parameters for maximum energy density. Our findings suggested different optimal designs for different current densities; whereas higher mass loading was preferred at lower C-rates, it was less favorable at higher C-rates. The concentration profiles of lithium-ions across the cathode thickness at the end of discharge suggest transport limitations to be the main reason behind this. As a further step, we carried out an uncertainty analysis to determine the robustness of the optimal designs to structural fluctuations with respect to their energy density. Our findings provide a deeper understanding into the influence of structural parameters on the discharge performance of lithium-ion battery cells, using only simulative methods. Additionally, our methodology of numerical optimization and subsequent uncertainty analysis allows for a quick and simple method for determining optimal structural parameters and their robustness against fluctuations. This approach can assist to produce battery cell electrodes with an optimized design in industry as well as help simulative and experimental researchers to further investigate optimal electrode design. [1] Karaki et al., Energy Technology 2022 [2] Laue et al., E. Acta 2019 Figure 1

Model-Based Prediction of Composition of an Unknown Blended Lithium-Ion Battery Cathode

Capacity Estimation of Lithium-ion Battery Based on Electrochemical Model with Electrolyte Dynamics

Electrochemical Model Based State of Charge Estimation for Lithium Batteries with Backstepping Techniques

Chemometric and Machine Learning Analysis of Lithium Concentration and Solvation Behavior in Li-Ion Battery Electrolytes

Identification of stoichiometric and microstructural parameters of a lithium-ion cell with blend electrode

Unconstrained Machine Learning Screening for New Li‐Ion Cathode Materials Enhanced by Class Balancing

Fast State-of-Charge Estimation for Lithium-Ion Batteries Using a Simplified Electrochemical Model Without Initial State Restrictions

Pseudo-Two-Dimensional Modeling of Lithium-Ion Conversion Cathode Materials

Co-Estimation of State-of-Charge and State-of- Health for Lithium-Ion Batteries Using an Enhanced Electrochemical Model

Machine Learning Based Identification of Lithium-Ion Battery Cell Chemistries

Voltage Mining for (De)lithiation-stabilized Cathodes and a Machine Learning Model for Li-ion Cathode Voltage

Transfer Learning Enabled Deep Learning Model for the Prediction of Battery Performance from Electrolyte Formulations

Accelerating the transition to cobalt-free batteries: a hybrid model for LiFePO 4 /graphite chemistry

Simplification of pseudo two dimensional battery model using dynamic profile of lithium concentration

Using Reference Electrodes to Validate Parameterization of Individual Electrode Soh in the Single Particle Model with Electrolyte

Mathematical Toolbox for Joint Structural and Electrochemical Analysis of Lithium-Ion Battery Cells

A Hybrid Electrochemical Multi-Particle Model for Li-ion Batteries

Molecular-Scale Interfacial Model for Predicting Electrode Performance in Rechargeable Batteries

Data-Driven Approach to Learning Optimal Forms of Constitutive Relations in Models Describing Lithium Plating in Battery Cells

Improving Electrolyte Performance for Target Cathode Loading Using Interpretable Data-Driven Approach

Composition and state prediction of lithium-ion cathode via convolutional neural network trained on scanning electron microscopy images