Francisco Fernandez

Redox flow batteries (RFBs) have emerged as promising candidates for large‐scale energy storage owing to their decoupled energy and power densities. Electrolyte formulation plays a key role in determining the electrochemical performance of RFBs, influencing interfacial kinetics and transport within porous electrodes. Herein, we develop a multiscale modeling framework to investigate the coupled effects of the electrolyte concentration (methyl viologen (MV) and NaCl) and the fibrous 3D microstructure on the porous transport and electrochemical kinetics in organic RFBs using a kinetic Monte Carlo model. We link this interfacial kinetics to the advection–diffusion transport in stochastically generated carbon felt electrode microstructures through lattice Boltzmann method simulations via an interpolation‐based coupling. We use two performance descriptors, the fiber‐averaged rate of potential loss and convective area fraction, to quantify the kinetics and transport effects, along with their interplay. While increasing MV concentration enhances the interfacial kinetics, it limits the transport within the electrode microstructure. The interfacial kinetics shows to be nontrivially governed by the supporting electrolyte. A surface area–permeability trade‐off appears across microstructures with different electrolyte concentrations. This multiparadigm modeling workflow provides mechanistic insights into porous transport and interfacial electrochemical kinetics, enabling the coupled optimization of microstructures and electrolytes in RFBs.

Forecasting battery lifetime and degradation is valuable but difficult because lithium-ion batteries age through complex, non-linear processes. Physics-based models are often preferred for their mechanistic insight into key aging modes. This work introduces a 3D+1D electro-thermal-aging model coupling multiple degradation phenomena to capture lithium inventory loss (LLI), loss of active material (LAM), and impedance increase at the cell level. The model integrates a DFNbased electrochemical component with a thermal model, alongside an aging component considering solid electrolyte interface (SEI) growth, lithium plating, and transition-metal dissolution, migration, and deposition as side reactions. We calibrated the model using data from an industrial graphite/LFP cell cycled under three CCCV protocols from 10 °C to 60 °C and C/3 to 6C. The spatially resolved 3D model predicts the nonlinear capacity "knee" and sudden-death behavior, linking them to interconnected knees of LLI and impedance increase caused by localized transport limitations, matching experimental degradation rates. Our findings indicate that knee point and sudden death result from gradual yet abrupt lithium loss due to continuous lithium plating and impedance rise restricting lithium availability. This study emphasizes the need to explore how aging processes influence the knee point, helping forecast and extend battery lifespan.

Optimization of electrochemical energy storage and conversion devices often depends on the electrode microstructure properties, of which its 3D representation provides the most complete link between manufacturing process and device performance. Deep generative models can address slow experimental throughput and high computational costs of traditional microstructure resolution methods, yet most literature focuses on Generative Adversarial Networks, prone to mode collapse issues. This work applies a generative diffusion model to synthesize 3D‐resolved multiphase microstructures of lithium ion battery and proton exchange membrane fuel cell electrodes, sourced from experimental imaging and computer‐based simulations. Generated volumes are validated against training data based on microstructural descriptors, including volume fraction, specific surface area, and tortuosity factor. Moreover, the standard cubic sub‐volume augmentation strategy used forgenerative model training fails to capture depth‐dependent features in thick and heavily processed microstructures, such as calendered electrodes. A depth‐aware augmentation pipeline is proposed to incorporate full electrode thickness into training samples, improving through‐thickness tortuosity factor accuracy in the generated electrodes. Furthermore, 3D‐resolved multiphysics simulation on the generated microstructures reproduce the performance behavior of their training counterparts with high accuracy. These results establish a diffusion model‐based framework as a viable means to generate commercially relevant electrode data, intended for computational optimization and design of next‐generation energy devices.

The performance of lithium-ion battery (LIB) electrodes is significantly influenced by manufacturing parameters, particularly during the slurry drying stage. Slurry drying is one of the key steps which involves high energy consumption and implies high cost in the battery production. It often triggers the undesirable migration of carbon additives and binder throughout the electrode thickness. This migration results in reduced mechanical adhesion, increased internal resistance and a significant decline in long-cycle life. Modelling in this case is a necessary tool in predicting the impact of drying conditions on the electrode properties ^[1,2].

The commercialization of the sodium-ion battery is gaining interest due to the now-proven advantages over lithium-ion batteries in low-temperature, high-power, and safety-focused applications. Additionally, Na chemistry offers a more environmentally friendly and ethical alternative due to its predominant materials from sustainable and abundant sources. However, some limitations exist in terms of increasing energy density as hard carbon (HC), the commercially favorable active material (AM) for the negative electrode, faces various obstacles in terms of electrode densification [1]. In the traditional wet processing method of manufacturing electrodes, densification is typically achieved by calendering the electrodes after the coated electrode slurry has been dried. The factors that affect calendering include HC particle morphology, electrode composition, particle size distribution, distribution of the different electrode components, and orientation. These factors are also highly influenced by all the previous manufacturing process parameters, such as the solid content of the slurry, drying rate, and coating thickness, among others.

With sodium‐ion batteries (SIBs) entering the energy market, optimizing their performance has become a key research focus. At the electrode level, the parameters selected during their manufacturing process influence their electrodes microstructure and, consequently, the electrochemical performance. Here, we address different manufacturing conditions of hard carbon (HC) negative electrodes by varying the solid content in the slurry (35 and 40 % SC) and the calendering degree applied to the electrodes (uncalendered and 30 % calendered). The 3D microstructure of each sample is acquired using a focused ion beam (FIB) and scanning electron microscopy (SEM) technique, from which the real shape of HC particles is extracted and used to generate input microstructures for a discrete element method (DEM) calendering model designed to address the mechanical electrode behavior and its effects on current collector deformation. Also, the 3D electrode microstructures are used in a finite element method (FEM) model to obtain the electrochemical performance for C‐rates ranging from C/50 to C/2, and to compare these results with experimental ones. Furthermore, the DEM predicted microstructures are also injected into the FEM model to validate the electrochemical performance and compare it with the performance of microstructures reconstructed with FIB‐SEM. We demonstrate a novel approach combining experimentation and advanced mesoscopic modeling, to address how manufacturing parameters influence the performance of HC electrodes, providing an important guideline for optimizing their production for SIB applications.

Optimizing the manufacturing process of Lithium‐Ion Batteries (LIB. Finding efficient approaches that accelerate and replace time‐consuming, material scrap‐expensive trials‐and‐error optimization methods is a key area of research. This work presents a comprehensive LIB electrode manufacturing framework that combines physics‐based simulations with deep learning. This framework efficiently simulates the manufacturing process of LIB electrodes as a function of their formulation. This framework takes the form of a surrogate manufacturing model able to predict the impact of manufacturing parameters on the electrode microstructure and properties. The model is based on a regressor‐inspired variational autoencoder method. The analysis of the data and the predicted electrode functional metrics demonstrates the consistency of the approach with an electrode manufacturing model developed on the basis of physics. The reported framework holds significant promise in paving near real time optimization of LIB electrode manufacturing and supporting the optimization of battery cell design in pilot lines.