How machine learning can extend electroanalytical measurements beyond analytical interpretation.

Electroanalytical measurements are routinely used to estimate material properties exhibiting current and voltage signatures. Analysis of such measurements relies on analytical expressions of material properties to describe the experiments. The need for analytical expressions limits the experiments that can be used to measure properties as well as the properties that can be estimated from a given experiment. Such analytical relations are essentially solutions of the physics-based differential equations (with properties as coefficients) describing the material behavior under certain specific conditions. In recent years, a new machine learning-based approach has been gaining popularity wherein the differential equations are numerically solved to interpret the electroanalytical experiments in terms of corresponding material properties. Since the physics-based differential equations are solved, one can additionally estimate underlying fields, e.g., concentration profile, using such an approach. To exemplify the characteristics of such a machine learning assisted interpretation of electroanalytical measurements, we use data from the Hebb-Wagner test on a magnesium spinel intercalation host. As compared to the traditional analytical expression-based interpretation, the emerging approach decreases experimental efforts to characterize relevant material properties as well as provides field information that was previously inaccessible.

Effect of crystallite geometries on electrochemical performance of porous intercalation electrodes by multiscale operando investigation.

Lithium-ion batteries are yet to realize their full promise because of challenges in the design and construction of electrode architectures that allow for their entire interior volumes to be reversibly accessible for ion storage. Electrodes constructed from the same material and with the same specifications, which differ only in terms of dimensions and geometries of the constituent particles, can show surprising differences in polarization, stress accumulation and capacity fade. Here, using operando synchrotron X-ray diffraction and energy dispersive X-ray diffraction (EDXRD), we probe the mechanistic origins of the remarkable particle geometry-dependent modification of lithiation-induced phase transformations in V2O5 as a model phase-transforming cathode. A pronounced modulation of phase coexistence regimes is observed as a function of particle geometry. Specifically, a metastable phase is stabilized for nanometre-sized spherical V2O5 particles, to circumvent the formation of large misfit strains. Spatially resolved EDXRD measurements demonstrate that particle geometries strongly modify the tortuosity of the porous cathode architecture. Greater ion-transport limitations in electrode architectures comprising micrometre-sized platelets result in considerable lithiation heterogeneities across the thickness of the electrode. These insights establish particle geometry-dependent modification of metastable phase regimes and electrode tortuosity as key design principles for realizing the promise of intercalation cathodes.

Unconventional Charge Transport in MgCr2O4 and Implications for Battery Intercalation Hosts.

Ion transport in solid-state cathode materials prescribes a fundamental limit to the rates batteries can operate; therefore, an accurate understanding of ion transport is a critical missing piece to enable new battery technologies, such as magnesium batteries. Based on our conventional understanding of lithium-ion materials, MgCr2O4 is a promising magnesium-ion cathode material given its high capacity, high voltage against an Mg anode, and acceptable computed diffusion barriers. Electrochemical examinations of MgCr2O4, however, reveal significant energetic limitations. Motivated by these disparate observations; herein, we examine long-range ion transport by electrically polarizing dense pellets of MgCr2O4. Our conventional understanding of ion transport in battery cathode materials, e.g., Nernst-Einstein conduction, cannot explain the measured response since it neglects frictional interactions between mobile species and their nonideal free energies. We propose an extended theory that incorporates these interactions and reduces to the Nernst-Einstein conduction under dilute conditions. This theory describes the measured response, and we report the first study of long-range ion transport behavior in MgCr2O4. We conclusively show that the Mg chemical diffusivity is comparable to lithium-ion electrode materials, whereas the total conductivity is rate-limiting. Given these differences, energy storage in MgCr2O4 is limited by particle-scale voltage drops, unlike lithium-ion particles that are limited by concentration gradients. Future materials design efforts should consider the interspecies interactions described in this extended theory, particularly with respect to multivalent-ion systems and their resultant effects on continuum transport properties.

Electric-Field-Induced Spatially Dynamic Heterogeneity of Solvent Motion and Cation Transference in Electrolytes.

While electric fields primarily result in migration of charged species in electrolytic solutions, the solutions are dynamically heterogeneous. Solvent molecules within the solvation shells of the cation will be dragged by the field while free solvent molecules will not. We combine electrophoretic NMR measurements of ion and solvent velocities under applied electric fields with molecular dynamics simulations to interrogate different solvation motifs in a model liquid electrolyte. Measured values of the cation transference number (t_{+}^{0}) agree quantitatively with simulation-based predictions over a range of electrolyte concentrations. Solvent-cation interactions strongly influence the concentration-dependent behavior of t_{+}^{0}. We identify a critical concentration at which most of the solvent molecules lie within solvation shells of the cations. The dynamic heterogeneity of solvent molecules is minimized at this concentration where t_{+}^{0} is approximately equal to zero.

Probing spatial coupling of resistive modes in porous intercalation electrodes through impedance spectroscopy.

In porous intercalation electrodes, coupled charge and species transport interactions take place at the pore-scale, while often observations are made at the electrode-scale. The physical manifestation of these interactions from pore- to electrode-scale is poorly understood. Moreover, the spatial arrangement of the constituent material phases forming a porous electrode significantly affects the multi-modal electrochemical and transport interplay. In this study, the relation between the electrode specification, resultant porous microstructure, and electrode-scale resistances is delineated based on a virtual deconvolution of the impedance response. Relevant short- and long-range interactions are identified. Without altering the microstructural arrangement, if the electrode thickness is increased, the resistances do not scale linearly with thickness. This dependence is also probed to identify the fundamental origins of thick electrode limitations.

Elucidating the Molecular Origins of the Transference Number in Battery Electrolytes Using Computer Simulations.

The rate at which rechargeable batteries can be charged and discharged is governed by the selective transport of the working ions through the electrolyte. Conductivity, the parameter commonly used to characterize ion transport in electrolytes, reflects the mobility of both cations and anions. The transference number, a parameter introduced over a century ago, sheds light on the relative rates of cation and anion transport. This parameter is, not surprisingly, affected by cation-cation, anion-anion, and cation-anion correlations. In addition, it is affected by correlations between the ions and neutral solvent molecules. Computer simulations have the potential to provide insights into the nature of these correlations. We review the dominant theoretical approaches used to predict the transference number from simulations by using a model univalent lithium electrolyte. In electrolytes of low concentration, one can obtain a quantitative model by assuming that the solution is made up of discrete ion-containing clusters-neutral ion pairs, negatively and positively charged triplets, neutral quadruplets, and so on. These clusters can be identified in simulations using simple algorithms, provided their lifetimes are sufficiently long. In concentrated electrolytes, more clusters are short-lived and more rigorous approaches that account for all correlations are necessary to quantify transference. Elucidating the molecular origin of the transference number in this limit remains an unmet challenge.

Quantifying selective solvent transport under an electric field in mixed-solvent electrolytes.

Electrolytes in lithium-ion batteries comprise solvent mixtures, but analysis of ion transport is always based on treating the solvents as a single-entity. We combine electrophoretic NMR (eNMR) measurements and molecular dynamics (MD) simulations to quantify electric-field-induced transport in a concentrated solution containing LiPF6 salt dissolved in an ethylene carbonate/ethyl methyl carbonate (EC/EMC) mixture. The selective transport of EC relative to EMC is reflected in the difference between two transference numbers, defined as the fraction of current carried by cations relative to the velocity of each solvent species. This difference arises from the preferential solvation of cations by EC and its dynamic consequences. The simulations reveal the presence of a large variety of transient solvent-containing clusters which migrate at different velocities. Rigorous averaging over different solvation environments is essential for comparing simulated and measured transference numbers. Our study emphasizes the necessity of acknowledging the presence of four species in mixed-solvent electrolytes.

Machine learning accelerates identification of lithiated phases in X-ray images of battery hosts.

Santos et al. (2022) propose a machine learning-based approach to identify various lithiated phases across lengthscales in X-ray images of battery particles, thus enabling automatic interpretation of such information in much bigger datasets and creating opportunities to unravel previously inaccessible scientific understanding.

Toward Bottom-Up Understanding of Transport in Concentrated Battery Electrolytes.

Bottom-up understanding of transport describes how molecular changes alter species concentrations and electrolyte voltage drops in operating batteries. Such an understanding is essential to predictively design electrolytes for desired transport behavior. We herein advocate building a structure-property-performance relationship as a systematic approach to accurate bottom-up understanding. To ensure generalization across salt concentrations as well as different electrolyte types and cell configurations, the property-performance relation must be described using Newman's concentrated solution theory. It uses Stefan-Maxwell diffusivity, ij , to describe the role of molecular motions at the continuum scale. The key challenge is to connect ij to the structure. We discuss existing methods for making such a connection, their peculiarities, and future directions to advance our understanding of electrolyte transport.

How Machine Learning Will Revolutionize Electrochemical Sciences.

Electrochemical systems function via interconversion of electric charge and chemical species and represent promising technologies for our cleaner, more sustainable future. However, their development time is fundamentally limited by our ability to identify new materials and understand their electrochemical response. To shorten this time frame, we need to switch from the trial-and-error approach of finding useful materials to a more selective process by leveraging model predictions. Machine learning (ML) offers data-driven predictions and can be helpful. Herein we ask if ML can revolutionize the development cycle from decades to a few years. We outline the necessary characteristics of such ML implementations. Instead of enumerating various ML algorithms, we discuss scientific questions about the electrochemical systems to which ML can contribute.

Stochasticity at Scales Leads to Lithium Intercalation Cascade.

Porous intercalation electrodes are synonymous with the promise of lithium-ion batteries toward electromobility. These electrodes exhibit stochastic geometrical features spanning different length scales. The implication of microstructural inhomogeneity on the lithium intercalation dynamics is hitherto unknown. Starting from three-dimensional (3D), X-ray tomograms of intercalation electrode microstructures, we characterize the microstructural variability in porous intercalation electrodes. Furthermore, a physics-based analysis of electrochemical response reveals that the stochastic features can cause preferential lithiation fronts.

In Operando Detection of the Onset and Mapping of Lithium Plating Regimes during Fast Charging of Lithium-Ion Batteries.

Existing in operando methods for detection of plated lithium can only detect the presence of plating after the charge is complete and irreversible damage has already occurred. In this work, the characteristic potential minimum on the graphite electrode during high rate lithiation is proposed and assessed as an in operando technique for detecting the onset of lithium plating. While other studies have shown that rapid self-heating of a cell can cause this type of "voltage overshoot", we confirm through temperature-controlled coin cell experiments that such a voltage profile can also be caused by the occurrence of severe lithium plating. In cells which demonstrated voltage overshoot, macroscopically observable lithium plating films were present on the graphite electrodes upon disassembly, resulting in very poor single-cycle Coulombic efficiency. The significance of this voltage characteristic is confirmed through direct observation of the onset of lithium plating in an in situ optical microscopy cell. We observe that the growth of large metallic lithium deposits within the porous electrode structure can cause swelling and cracking of the graphite electrode, suggesting loss of active material due to mechanical electrode degradation as an important consequence of severe lithium plating.

Controllable Electrode Stochasticity Self-Heats Lithium-Ion Batteries at Low Temperatures.

Cold start is an ineluctable stipulation for electric vehicle operation under low-temperature extremes. It has typically been addressed through cell-level heating strategies. We advocate an electrode-level strategy leveraging pore-scale manifestation of thermal metastability that promotes self-heating. Appropriate controllable stochastic characteristics of porous electrodes are delineated, which contribute to the proposed solution at low temperatures. This approach is most conducive to high-energy-density Li-ion cells and devoid of extrinsic overheads.

Materials by Design: Tailored Morphology and Structures of Carbon Anodes for Enhanced Battery Safety.

Next-generation Li-ion battery technology awaits materials that not only store more electrochemical energy at finite rates but also exhibit superior control over side reactions and better thermal stability. Herein, we hypothesize that designing an appropriate particle morphology can provide a well-balanced set of physicochemical interactions. Given the anode-centric nature of primary degradation modes, we investigate three different carbon particles-commercial graphite, spherical carbon, and spiky carbon-and analyze the correlation between particle geometry and functionality. Intercalation dynamics, side reaction rates, self-heating, and thermal abuse behavior have been studied. It is revealed that the spherical particle outperforms an irregular one (commercial graphite) under thermal abuse conditions, as it eliminates unstructured inhomogeneities. A spiky particle with ordered protrusions exhibits smaller intercalation resistance and attenuated side reactions, thus outlining the benefits of controlled stochasticity. Such findings emphasize the importance of tailoring particle morphology to proffer selectivity among multimodal interactions.

Quantifying Transport, Geometrical, and Morphological Parameters in Li-Ion Cathode Phases Using X-ray Microtomography.

The charge/discharge capabilities of Li-ion cathodes are influenced by the meso-scale geometry, transport properties, and morphological parameters of the constituent phases in the cathode: active material, binder, conductive additive, and pore. Electrode processing influences the structure and attendant properties of these constituents. Thus, performance of the battery can be enhanced by correlating various electrode processing techniques with the charge/discharge behavior in the lithium-ion cathodes. X-ray microtomography was used to image samples obtained from pristine Li(Ni1/3Mn1/3Co1/3)O2 (NMC) cathodes subjected to distinct processing approaches. Two sample preparation approaches were applied to the samples prior to microtomography. Casting the samples in epoxy yielded only the cathode active material domain. Encapsulating the sample with Kapton tape yielded phase contrast data that permitted segmentation of the active material and combined carbon/binder and pore regions. Geometrical and morphological details of the active material and the secondary phases were characterized and compared between the varied processing approaches. Calendered and ball-milled samples exhibited distinct differences in both geometry and morphology. Drying modes demonstrated variation in the distribution of the secondary and pore phases. Applying phase contrast capabilities, the processing-morphology relationship can be better understood to enhance overall battery performance across multiple scales.

Electrochemistry Coupled Mesoscale Complexations in Electrodes Lead to Thermo-Electrochemical Extremes.

Thermo-electrochemical extremes continue to remain a challenge for lithium-ion batteries. Contrary to the conventional approach, we propose herein that the electrochemistry-coupled and microstructure-mediated cross talk between the positive and negative electrodes ultimately dictates the off-equilibrium-coupled processes, such as heat generation and the propensity for lithium plating. The active particle morphological differences between the electrode couple foster a thermo-electrochemical hysteresis, where the difference in heat generation rates changes the electrochemical response. The intrinsic asymmetry in electrode microstructural complexations leads to thermo-electrochemical consequences, such as cathode-dependent thermal excursion and co-dependent lithium plating otherwise believed to be anode-dependent.

Secondary-Phase Stochastics in Lithium-Ion Battery Electrodes.

Lithium-ion battery electrodes exhibit complex interplay among multiple electrochemically coupled transport processes, which rely on the underlying functionality and relative arrangement of different constituent phases. The electrochemically inactive solid phases (e.g., conductive additive and binder, referred to as the secondary phase), while beneficial for improved electronic conductivity and mechanical integrity, may partially block the electrochemically active sites and introduce additional transport resistances in the pore (electrolyte) phase. In this work, the role of mesoscale interactions and inherent stochasticity in porous electrodes is elucidated in the context of short-range (interface) and long-range (transport) characteristics. The electrode microstructure significantly affects kinetically and transport-limiting scenarios and thereby the cell performance. The secondary-phase morphology is also found to strongly influence the microstructure-transport-kinetics interactions. Apropos, strategies have been proposed for performance improvement via electrode microstructural modifications.