Search

Unique, rapid identification and quantification of the dominant aging modes in lithium-ion batteries (LiBs) with early and non-specialized test data is a significant scientific challenge. Leveraging synthetic-data, deep-learning (DL) techniques have great potential to enable fast and robust classification and quantification of battery aging modes that produce different patterns of cell aging. This study presents, for the first time, a synthetic–data-based DL modeling framework for rapid and automatic classification and quantification of battery-aging modes and resultant aging, with experimental validation of the technique. Availing synthetic dQ.dV−1 curves for roughly 26,000 initial conditions and aging modes, the framework classified the dominant aging modes for cells undergoing fast charge in less than 100 cycles. Upon classification, the framework quantified the evolution of the aging modes, which were often not uniform with cycling, for 22 Gr/NMC532 pouch cells tested up to 600 cycles at different charging rates (1C–9C).

Using a unique combination of advanced characterization techniques, we identify specific degradation mechanisms and quantify degradative species formed during fast charge cycling of lithium-ion battery pouch cells.

Extreme fast charging (XFC, ≤15 min charging time) of Li-ion batteries (LIBs) has been proposed as an immediate target to increase the commercial appeal of electric vehicles. However, XFC of LIBs i...

Enabling extreme fast charging (XFC, ≤10–15 min charging) requires a comprehensive understanding of its implications. While lithium plating is a key bottleneck for the anode, the full extent of limitations for the cathode are not well-understood, particularly in extended-cycle settings with well-defined battery designs and conditions. This article presents cycle-life implications of XFC on cathodes at multiple length scales, combining electrochemical analyses, degradation modeling, and post-test characterizations. The comprehensive test matrix includes 41 well-defined gr/NMC pouch cells under varied fast-charge rates (1–9C) and state-of-charges cycled up to 1000 times. Cathode issues remain minimal in early cycling, but begin to accelerate in later life, when distinct cracking is found and identified as a fatigue mechanism. The bulk structure of cathodes remains intact, but distinct particle surface reconstruction is observed; however, this shows less pronounced effect on cathode aging than does cracking.

Raw battery data and codes for battery aging mode classification. The raw dataset consists of44 NMC/Gr single layer pouch cells. The data provided include cycle-by-cycle capacity, Coulombic efficiency, end of charge voltage (EOCV), and end of discharge voltage (EODV). The codes provided will perform data pre-treatment, analysis, and aging mode classification.&nbsp

The desire for energy-dense and fast-charged battery technology in consumer electronics, electric vehicle, grid, and aviation applications is pushing the envelope from materials to cell and pack designs. However, some approaches could inherently decrease safety of the battery—thus requiring the development of advanced management and diagnostics. Safety of lithium-ion batteries (LiBs), particularly in multicell configurations, is highly variable and could evolve with use. Existing works primarily focus on cell life and safety diagnostics without considering module and pack-level uncertainties and sometimes imply that cell-level electrochemical diagnostics would work in modules or packs, a naive oversimplification. Using example case studies, we highlight the potential and challenges associated with extending single-cell diagnostics to multiple cells, note the existing gaps, and motivate the research, development, and support communities to devote efforts to fill the gap by developing diagnostics at these levels for current- and future-generation LiBs.

Realization of extreme fast charging (XFC, ≤15 minutes) of lithium-ion batteries is imperative for the widespread adoption of electric vehicles. However, dramatic capacity fading is associated with XFC, limiting its implementation. To quantitatively elucidate the effects of irreversible lithium plating and other degradation mechanisms on the cell capacity, it is important to understand the links between lithium plating and cell degradation at both the local and global (over the full cell) scales. Here, we study the nature of local lithium plating after hundreds of XFC cycles (charging C-rates ranging from 4C to 9C) in industrially-relevant pouch cells using spatially resolved X-ray diffraction. Our results reveal a spatial correlation at the mm scale between irreversible lithium plating on the anode, inactive lithiated graphite phases, and local state-of-charge of the cathode. In regions of plated lithium, additional lithium is locally and irreversibly trapped as lithiated graphite, contributing to the loss of lithium inventory (LLI) and to a local loss of active anode material. The total LLI in the cell from irreversibly plated lithium is linearly correlated to the capacity loss in the batteries after XFC cycling, with a non-zero offset originating from other parasitic side reactions. Finally, at the global (cell) scale, LLI drives the capacity fade, rather than electrode degradation. We anticipate that the understanding of lithium plating and other degradation mechanisms during XFC gained in this work will help lead to new approaches towards designing high-rate batteries in which irreversible lithium plating is minimized.

Extreme fast charging (XFC) of commercial lithium-ion batteries (LIBs) in ≤10-15 minutes will significantly advance the deployment of electric vehicles globally. However, XFC leads to considerable capacity fade, mainly due to graphite anode degradation. Non-destructive three-dimensional (3D) investigation of XFC-cycled anodes is crucial to connect degradation with capacity loss. Here, we demonstrate the viability of simultaneous neutron and X-ray tomography (NeXT) for ex-situ 3D visualization of graphite anode degradation. NeXT is advantageous because of the sensitivity of neutrons to Li and H and X-rays to Cu. We combine the neutron and X-ray tomography with micron resolution for material identification and segmentation on one pristine and one XFC-cycled graphite anode, thereby underscoring the benefits of the simultaneous nature of NeXT. Our ex-situ results pave the way for the design of NeXT-friendly LIB geometries that will allow operando and/or in-situ 3D visualization of graphite anode degradation during XFC.

The research and development (R&D) community is continually pushing for high energy density Lithium-ion batteries (LiBs) by innovating materials, designs, and operating conditions.1 We expect these batteries to operate under a range of conditions (e.g., fast charging, fast discharging, low temperatures, etc.).1,2 Advanced diagnostics could play a key role in ensuring an extended life of these energy-dense LiBs while ensuring their safety. Therefore, a parallel effort is enabling advanced diagnostics and prognostics to ensure the safe operation of these batteries under the stated conditions throughout their service lives. Many researchers have proposed different performance and safety diagnostics of LiB in their early stages, often using small or specially designed cells through controlled tests.1 Extending these diagnostics to automotive grade cells in real life environments has not been performed extensively to understand their uncertainties. This gap in development between the early R&D stage to near the deployment stage is impeding their maturity; hence, delaying their adoption in their expected applications. In this presentation, we will discuss the unreliability issues of different electrochemical and electromechanical diagnostics related to LiBs under different safety critical conditions when cell size, age, and operating conditions changes. We will discuss the underlying issues of unreliability in different diagnostics with experimental data, followed by a discussion of several emerging safety diagnostics. Finally, the need for independent evaluation of LiB safety diagnostics along with some logistical pathways to do so will be discussed. References T.R. Tanim, E. J. Dufek, and S. V. Sazhin, MRS Bulletin, 46, 420-428 (2021) S. Ahmed et al., J. Power Sources, 367, 250–262 (2017).

A key challenge for Lithium-ion batteries (LiB) is the capability to manage battery performance and predict the battery’s lifetime, which are crucial information for optimum operation of electric-vehicles (EVs). In that regard, early detection of battery-aging phenomena and their implications on performance is necessary for maintaining warranty and avoiding safety-related liabilities. However, rapid diagnosis of the root causes of battery degradation has not yet been well-developed nor studied upon a broad mixture of cell designs and use cases. Here, a machine-learning (ML)- based framework is presented that aims at identifying aging modes or mechanisms during battery cycling. Cycle-by-cycle electrochemical signatures (e.g., end-of-rest voltages, columbic efficiency, and capacity fade) were analyzed then correlated with aging modes, such as loss of active materials in the cathode (LAMPE) or loss of Li inventory (LLI). Multiple battery designs and use conditions were analyzed to develop this ML framework, including 44 single layer pouch cells representing two cathode chemistries (NMC532 and NMC811), two electrode loadings (low and moderate), and five charging rates (1, 4, 4.5, 6, and 9C CC-CV profiles). Predominant aging behaviors depend on both the cell design and use. Aging behaviors can be classified into a combination of LAMPE and LLI. A classification accuracy of 90% was achieved using only the first 200aging cycles, and the percentage of LAMPE was predicted with 4.3% of error after 600 cycles. As seen here, the overall aging behavior of the battery was found to be dependent on cathode chemistry, electrode build, and usage conditions. This ML framework is promising for battery developers to save the time and effort that is typically associated with prolonged testing. The ML framework will also advise on innovative design strategies to counter specific degradation modes, thus significantly accelerating the development cycle for building longer lasting batteries. From the end user’s perspective, this aging-mode diagnosis framework enables the ability to fine-tune battery service scenarios to minimize degradations thereby improving both lifetime and safety in EVs or stationary storage systems. Figure 1

Enabling extreme fast charging (XFC, charging in 10 to 15 minutes) in a lithium-ion battery (LiB) could play a key role in subsiding consumer’s range anxiety and spur the widespread adoption of electric vehicles (EVs).1,2 Such a high rate of charging induces unique aging modes in LiBs, thereby requiring a comprehensive understanding to enable effective solution strategies to minimize the negative effects of life and performance. This presentation will present a comprehensive understanding of the dominating aging modes and mechanisms of XFC in low- and moderate-loading Gr/NMC LiBs. We will discuss the major limitations of XFC in LiBs by first using experimental and modeling results followed by a comprehensive electrochemical analysis of cycle life aging implications for different charging conditions (e.g., 1C to 9C rate conditions). We will then discuss the aging mechanisms using comprehensive post-testing as well as multimodal and multi-scale microscopy techniques. Solid state diffusion in the negative electrode is not a key limiting factor for the fast charge conditions evaluated. Inadequate Li+ transport through the electrolyte primarily causes performance and distinct aging phenomena in LiBs. Eliminating the Li+ transport limitation within the electrolyte can offer a distinct increase in material utilization, avoiding Li deposition. Under such circumstances, the cathode could degrade in distinct ways depending on the particular NMC (e.g., NMC532 vs. NMC811) variants. NMC811 experiences a greater subsurface crystallographic degradation and interfacial degradation and displays similar extents of sub-particle cracking as compared to NMC532 under comparable charging conditions. Surprisingly, the NMC811 maintains superior electrochemical performance despite the more aggressive degradations. We found the better cycle life performance of NMC811 to be related to its inherently better solid state diffusion, electronic conduction, and radially oriented grain architecture. References S. Ahmed et al., J. Power Sources, 367, 250–262 (2017). Y. Liu, Y. Zhu, and Y. Cui, Nat. Energy, 4, 540–550 (2019).

Physics-based methods and tools for rapid classification, quantification, and forecasting of lithium-ion battery aging modes and life Sangwook Kim,1 Zonggen Yi,1 Ross R. Kunz,1 Eric J. Dufek,1 Tanvir Tanim,1 Kevin L. Gering1, Bor-Rong Chen,1 Peter Weddle, 2 Kandler Smith, 2 1 Energy and Environmental Science and Technology, Idaho National Laboratory, Idaho Falls, Idaho 83415 USA 2 Center for Energy Conversion & Storage Systems, National Renewable Energy Laboratory, Golden, CO 80401, USA 242nd ECS Meeting, Atlanta, GA, Oct. 9 - 13, 2022 Symposium: A03 – Lithium Ion Batteries Lithium-ion batteries play a central role in powering electric vehicles, stationary energy storage systems, and consumer electronics. Lately, machine learning and deep learning (DL) has been successfully used to gain insights into battery degradation during battery operation and predict lifetime in battery research and development community. Fast and robust classification and quantification of battery aging (e.g., Loss of Lithium Inventory (LLI) and Loss of active Material (LAM)) and accurate long-term forecasting of battery life enable more proactive planning of battery management and preemptive actions of modified operating conditions to achieve safe operations and prolong battery life. Here, we present the development of Incremental Capacity (IC)-DL framework for fast charging conditions as a diagnostic tool for aging mode classification and quantification (Figure 1). The classification and quantification of dominant battery aging modes is conducted using a synthetic-data-based DL modeling framework. Over 6000 initial conditions and 26,000 different aging conditions are generated using IC model to train DL. We applied trained DL classification and quantification algorithm to 22 Gr/NMC532 pouch cells with a different loading and charging protocol tested up to 600 cycles. Besides the analysis of reference performance test data at C/20, cycle-by-cycle data at higher C-rate is analyzed. This IC-DL framework as a diagnostic tool is also used in different battery chemistries, such as Gr/NMC811and LTO/LMO. IC-DL framework enables unique, rapid identification and quantification of the dominant aging modes at different C-rates in different battery chemistries. Additionally, a prognostic tool, Sigmoidal Rate Expression (SRE)-type mathematics are employed to evaluate the capacity loss and aging mode (i.e., LLI). SREs are robust engines that contain three variables that capture the thermodynamic and kinetic “thumbprint” of the mechanism progression within the context of a batch reactor scenario [1].We show two different methods by which SRE parameters can be early assessed based on quantified values from IC-DL framework; (1) extrapolative techniques using specialized functions to determine SRE parameter convergence and (2) a technique based on deep learning and Monte Carlo framework. Overall results from both methods confirm that we can predict capacity loss and LLI at the end of test (i.e., 600 cycles) within 1-2% absolute error using three weeks of testing data (or 125 cycles). We believe the IC-DL framework combined with SRE-base prognostics will reduce lithium-ion batteries development cycle as well as shorten the turnover time for validation, fulfilling the demands of rapidly growing battery market. References K. Gering, Electrochimica Acta, 228(20), 636-651 (2017) Figure 1

A quantitative link between crack evolution in lithium-ion positive electrodes and the degrading performance on cells is not yet well established nor is any single technique capable of doing so widely available. Here, we demonstrate a widely accessible high-throughput approach to quantifying crack evolution within electrodes. The approach applies super-resolution scanning electron microscopy (SEM) imaging of cross-sectioned NMC532 electrodes, followed by segmentation and quantification of crack features. Crack properties such as crack intensity, crack width and length are quantified as a function of charge rate (1C, 6C, and 9C) and cycle number (25, 225, and 600 cycles). Hundreds of particles are characterized for statistical confidence in the quantitative crack measurements. The data on crack evolution is compared to electrochemical data from full cells and half cells with the NMC532 positive electrodes. We show that while crack evolution strongly correlates with capacity fade in the first 25 cycles, it does not correlate well for the following hundreds of cycles indicating that cracking may not be the dominant cause of capacity fade throughout the cycle-life of cells.

Battery performance is strongly correlated with electrode microstructure and weight loading of the electrode components. Among them are the carbon-black and binder additives that enhance effective conductivity and provide mechanical integrity. However, these both reduce effective ionic transport in the electrolyte phase and reduce energy density. Therefore, an optimal additive loading is required to maximize performance, especially for fast charging where ionic transport is essential. Such optimization analysis is however challenging due to the nanoscale imaging limitations that prevent characterizing this additive phase and thus quantifying its impact on performance. Herein, an additive-phase generation algorithm has been developed to remedy this limitation and identify percolation threshold used to define a minimal additive loading. Improved ionic transport coefficients from reducing additive loading has been then quantified through homogenization calculation, macroscale model fitting, and experimental symmetric cell measurement, with good agreement between the methods. Rate capability test demonstrates capacity improvement at fast charge at the beginning of life, from 37% to 55%, respectively for high and low additive loading during 6C CC charging, in agreement with macroscale model, and attributed to a combination of lower cathode impedance, reduced electrode tortuosity and cathode thickness.

Summary Broad use of global or spatially averaging measurements over a cell to characterize highly localized Li plating phenomena in lithium-ion batteries during fast charging has created a disconnect between measurements and the underlying causes. Consequently, the field is missing a clear path to implementing fast charging as well as to expand into extreme fast charging (XFC). Aiming to bridge these gaps, we present a detailed look into local detection of Li plating and the consequent cycle life implications for electrodes and cells under XFC by utilizing electrochemistry and high-energy X-ray diffraction. Significant heterogeneity in Li plating during XFC results in accelerated and non-uniform cycle life losses, in contrast to the prevailing acceptance that C rate is correlated to Li plating for XFC. This behavior is triggered by local electrode heterogeneity, which has yet to be identified and is not apparent in volume-averaged quantifications. A better understanding of these multiscale local electrode heterogeneities is crucial for identifying pathways to enable XFC.

Strong demand for higher energy batteries for extended range electric vehicles has driven battery technology development toward Li metal batteries. One limitation of Li metal batteries is cycle life. At odds are the amount of electrolyte to fulfill high energy and the amount needed to achieve long cycle life. Here we employ an electrochemical model and design of experiment to understand impacts on cycle life and failure as electrolyte content is reduced to levels needed to achieve reasonably high specific energy. With lean electrolyte, a gap exists in achieving long cycle life with a reduction by a factor of 6.5 when electrolyte amount is reduced from 37.0 g Ah−1 to 6.0 g Ah−1. Two primary failure routes are identified with differences between lean and high electrolyte quantities. The results suggest that reporting and performing experiments for high energy cell designs requires that amount of electrolyte be considered a critical parameter.

Lithium ion (Li-ion) cells are increasingly being used in high power applications such as Hybrid Electric Vehicles. This paper investigates how the life of lithium ion power cells can be increased by increasing their temperature. End of life (EOL) in power applications is often defined as when the battery is no longer able to provide the required charge/discharge power because the battery voltage exceeds the maximum/minimum allowable voltages associated with the battery's chemistry. This paper shows that battery life can be increased by step-wise temperature increases whenever the battery voltage exceeds a voltage limit when the EOL is reached. Experiments are conducted on commercially available lithium iron phosphate batteries for six months using a charge sustaining HEV cycle. Two, 1 °C temperature increases extend the life of the cells by 2000 cycles (3 months of continuous cycling).

This study investigates the effect of 50-kW (about 2C) direct current fast charging on a full-size battery electric vehicle's battery pack in comparison to a pack exclusively charged at 3.3 kW, which is the common alternating current Level 2 charging power level. Comparable scaled charging protocols are also independently applied to individual cells at three different temperatures, 20 °C, 30 °C, and 40 °C, to perform a comparative analysis with the packs. Dominant cell-level aging modes were identified through incremental capacity analysis and compared with full packs to gain a clear understanding of additional key factors that affect pack aging. While the cell-level study showed a minor impact on performance due to direct current fast charging, the packs showed a significantly higher rate of capacity fade under similar charging protocols. This indicates that pack-level aging cannot be directly extrapolated from cell evaluation. Delayed fast charging, completing shortly before discharge, was found to have less of an impact on battery degradation than conventional alternating current Level 2 charging.

In this article, we develop a 3D, continuum-level damage model implemented on statistically generated Li x Ni0.5Mn0.3Co0.2O 2 (NMC 532) secondary cathode particles. The primary motivation of the particle-level model is to inform cathode-particle design through detailed exploration of the influence of secondary and primary particle sizes on the damage predicted during operation, and determine charging profiles that reduce cathode fracture. The model considers NMC 532 secondary particles containing an agglomeration of anisotropic, randomly oriented grains. These brittle, Ni-based cathodes are prone to mechanical degradation, which reduces overall battery cycle life. The model predicts that secondary-particle fracture is primarily due to non-ideal grain interactions and high-rate charge demands. The model predicts that small secondary-particles with large grains develop significantly less damage than larger secondary particles with small grains. The model predicts most of the chemo-mechanical damage accumulates in the first few cycles. The chemo-mechanical model predicts monotonically increasing capacity fade with cycling and rate. Comparing to experimental results, the model is well suited for capturing initial capacity fade mechanisms, but additional physics is required to capture long-term capacity fade effects.

Heterogeneous battery performance is a critical issue for maximization of cell lifetime capacity and safety. Using high energy synchrotron X-ray diffraction, the influence of charge rate, voltage limit, uneven stack pressure, and gas generation on the lithium transport properties was quantified in single-layer graphite/LiNi0.5Mn0.3Co0.2O2 pouch cells. A freshly formatted cell tracked in operando during initial fast charge cycles indicated variable position-dependent performances, while lateral mapping showed a significant fast charge (6C) heterogeneity compared to slow charge (C/2). Pressure effects were non-dominant compared to charge rate. Maps of previously aged and rested cells indicate that lateral heterogeneity slowly equilibrates at rest, but regenerates upon further cycling at fast charge rate. Furthermore, an unformatted cell was mapped at charge and discharge during its first formation cycle to analyze the effect of byproduct gases on the heterogeneous lithium transport. Gas was observed as randomly interspersed “bubbles” which locally hindered lithium intercalation and caused significant heterogeneity. Electrode architectures and charging protocols that promote homogeneous intercalation are critical for predictable high-performance and long-life batteries.

The battery technology literature is reviewed, with an emphasis on key elements that limit extreme fast charging. Key gaps in existing elements of the technology are presented as well as developmental needs. Among these needs are advanced models and methods to detect and prevent lithium plating; new positive-electrode materials which are less prone to stress-induced failure; better electrode designs to accommodate very rapid diffusion in and out of the electrode; measure temperature distributions during fast charge to enable/validate models; and develop thermal management and pack designs to accommodate the higher operating voltage.

The ability to charge battery electric vehicles (BEVs) on a time scale that is on par with the time to fuel an internal combustion engine vehicle (ICEV) would remove a significant barrier to the adoption of BEVs. However, for viability, fast charging at this time scale needs to also occur at a price that is acceptable to consumers. Therefore, the cost drivers for both BEV owners and charging station providers are analyzed. In addition, key infrastructure considerations are examined, including grid stability and delivery of power, the design of fast charging stations and the design and use of electric vehicle service equipment. Each of these aspects have technical barriers that need to be addressed, and are directly linked to economic impacts to use and implementation. This discussion focuses on both the economic and infrastructure issues which exist and need to be addressed for the effective implementation of fast charging at 400 kW and above. In so doing, it has been found that there is a distinct need to effectively manage the intermittent, high power demand of fast charging, strategically plan infrastructure corridors, and to further understand the cost of operation of charging infrastructure and BEVs.

To achieve a successful increase in the plug-in battery electric vehicle (BEV) market, it is anticipated that a significant improvement in battery performance is required to increase the range that BEVs can travel and the rate at which they can be recharged. While the range that BEVs can travel on a single recharge is improving, the recharge rate is still much slower than the refueling rate of conventional internal combustion engine vehicles. To achieve comparable recharge times, we explore the vehicle considerations of charge rates of at least 400 kW. Faster recharge is expected to significantly mitigate the perceived deficiencies for long-distance transportation, to provide alternative charging in densely populated areas where overnight charging at home may not be possible, and to reduce range anxiety for travel within a city when unplanned charging may be required. This substantial increase in charging rate is expected to create technical issues in the design of the battery system and the vehicle's electrical architecture that must be resolved. This work focuses on vehicle system design and total recharge time to meet the goals of implementing improved charge rates and the impacts of these expected increases on system voltage and vehicle components.

Battery thermal barriers are reviewed with regards to extreme fast charging. Present-day thermal management systems for battery electric vehicles are inadequate in limiting the maximum temperature rise of the battery during extreme fast charging. If the battery thermal management system is not designed correctly, the temperature of the cells could reach abuse temperatures and potentially send the cells into thermal runaway. Furthermore, the cell and battery interconnect design needs to be improved to meet the lifetime expectations of the consumer. Each of these aspects is explored and addressed as well as outlining where the heat is generated in a cell, the efficiencies of power and energy cells, and what type of battery thermal management solutions are available in today's market. Thermal management is not a limiting condition with regard to extreme fast charging, but many factors need to be addressed especially for future high specific energy density cells to meet U.S. Department of Energy cost and volume goals.

The temperature and capacity of individual cells affect the current distribution in a battery pack. Non uniform current distribution among parallel-connected cells can lead to capacity imbalance and premature aging. This paper develops models that calculate the current in parallel-connected cells and predict their capacity fade. The model is validated experimentally for a nonuniform battery pack at different temperatures. The paper also proposes and validates the hypothesis that temperature control can reduce capacity mismatch in parallel-connected cells. Three Lithium Iron Phosphate cells, two cells at higher initial capacity than the third cell, are connected in parallel. The pack is cycled for 1500 Hybrid Electric Vehicles cycles with the higher capacity cells regulated at 40°C and the lower capacity cell at 20°C. As predicted by the model, the higher capacity and temperature cells age faster, reducing the capacity mismatch by 48% over the 1500 cycles. A case study shows that cooling of low capacity cells can reduce capacity mismatch and extend pack life.

Identifying the dominating aging modes and mechanisms uniquely with higher confidence using early life-cycle data is a significant scientific challenge. In general, lithium-ion batteries experience different aging modes depending on use conditions—e.g., loss of lithium inventory and loss of active material in both electrodes, loss of power capability, etc.—due to different relevant aging mechanisms. The existing body of research almost exclusively uses extended-calendar or cycling-life data along with different post-testing to identify and quantify different dominating aging modes and mechanisms. Early identification of battery aging modes/mechanisms will enable accurate forecasting of battery state-of-health along with the root cause and shorten the design and testing cycle for new battery chemistries and cell designs for emerging applications. In this presentation, we present a deep learning (DL) modeling framework to enable early identification and quantification of dominating cell aging modes using battery incremental capacity (IC) information at different aging states (Figure 1). The DL model utilizes coin cell data with altered offset and N:P ratio simulating different aging conditions to train and construct two categories of DL models, serving two consecutive steps in the hierarchical learning framework: (i) a classification model to identify the dominating aging modes, e.g. LLI + LAMPE or LLI + LAMNE, (ii) a regression model to quantify the percentages for dominating aging modes, e.g., percentages of LLI and LAM. The deep learning models are built using a 1-D convolutional neural network structure. The deep learning model’s performance is verified with experimental Gr/NMC pouch cells data with different loadings and utilization conditions. It is demonstrated that the classification model can identify dominating aging modes successfully for current experimental validation sets within 50 cycles, and the quantification model can further determine their percentages of LLI and LAM under aging modes. Figure 1

Electric vehicles’ (EVs’) more-limited driving range and longer charging times—compared with gasoline vehicle refueling—cause range anxiety that inhibits mass adoption of EVs. Extreme fast charging (XFC) is the ability to charge batteries in a length of time (i.e., ~10 minutes or less) comparable to that required to refuel gasoline vehicles [1]. The cell performance and aging mechanisms under XFC conditions are quite distinct from those seen under slow charging conditions. Among the various XFC challenges, Li-plating on the anode is the cause of poor cell performance, with severe capacity fade, dendrites growth and internal shorts that could lead to catastrophic safety issues [2]. Literature reports on electrochemical (EC) Li detection, mostly at sub-zero temperatures and moderate-to-high charging rates, show Li plating displaying higher-favorability, reliable plating-driven mixed potential dynamics [3-5]. Those EC signatures have not been evaluated extensively for fast-charging conditions. Our study focuses on the sensitivity and reliability of EC Li-plating signatures at varying XFC conditions with long-term cycling and aging. The global EC signatures investigated were coulombic efficiency (CE), incremental capacity (dQ/dV), differential voltage (dV/dt), and end of relaxed charge (EOC) voltage on moderate loading Li/graphite and graphite/NMC532 coin cells and graphite/NMC532 single-layer pouch cells under XFC conditions at 25°C [6]. The sensitivity and reliability of investigated signatures vary with cell-to-cell and cycling conditions. For example, the sensitivity and reliability of dQ/dV is good with moderate C-rate lithiation and reasonable de-lithiation (C/5) conditions. With increasing C-rate (e.g., 6C), dQ/dV sensitivity decreases even under the same discharge (C/5) conditions. The dV/dt signatures at various states-of-charge with 6C-lithiation conditions varies with cell-to-cell and cycling conditions. In all cases, the dV/dt signature appears during initial cycling at different relaxation times, intensities, but within few cycles, the signal strength decreases and/or disappears even though CE and EOC EC signatures confirm continuous Li plating on the graphite electrode. The EC Li-detection signatures in half-cells, full cells and full pouch cells align with each other within the experimental error. Long-term and continuous reliability of Li-plating detection with EC signatures is still an issue for XFC conditions. Ahmed, S., et al., Enabling fast charging – A battery technology gap assessment. Journal of Power Sources, 2017. 367: p. 250-262. Santhanagopalan, S., P. Ramadass, and J. Zhang, Analysis of internal short-circuit in a lithium ion cell. Journal of Power Sources, 2009. 194(1): p. 550-557. Campbell, I.D., et al., How Observable Is Lithium Plating? Differential Voltage Analysis to Identify and Quantify Lithium Plating Following Fast Charging of Cold Lithium-Ion Batteries. Journal of The Electrochemical Society, 2019. 166(4): p. A725-A739. Petzl, M. and M.A. Danzer, Nondestructive detection, characterization, and quantification of lithium plating in commercial lithium-ion batteries. Journal of Power Sources, 2014. 254: p. 80-87. Schindler, S., et al., Voltage relaxation and impedance spectroscopy as in-operando methods for the detection of lithium plating on graphitic anodes in commercial lithium-ion cells. Journal of Power Sources, 2016. 304: p. 170-180. Tanim, T.R., et al., Electrochemical Quantification of Lithium Plating: Challenges and Considerations. Journal of The Electrochemical Society, 2019. 166(12): p. A2689-A2696.