7 August 2020 in The Journal of Chemical Physics
There is significant interest in improving the performance of batteries to increase electrification of transportation and aviation. Recently, performance improvements have been in large part due to changes in the composition of the cathode material family, LiNixMnyCo(1−x−y)O2 (e.g., 111–622–811). Despite the importance of these materials and tremendous progress with density functional theory (DFT) calculations in understanding basic design principles, it is computationally prohibitively expensive to make this problem tractable. Specifically, predicting the open circuit voltage for any cathode material in this family requires evaluation of stability in a quaternary phase space. In this work, we develop machine-learning potentials using fingerprinting based on atom-centered symmetry functions, used with a neural network model, trained on DFT calculations with a prediction accuracy of 3.7 meV/atom and 0.13 eV/Å for energy and force, respectively. We perform hyperparameter optimization of the fingerprinting parameters using Bayesian optimization through the Dragonfly package. Using this ML calculator, we first test its performance in predicting thermodynamic properties within the Debye–Grüneisen model and find good agreement for most thermodynamic properties, including the Gibbs free energy and entropy. Then, we use this to calculate the Li-vacancy ordering as a function of Li composition to simulate the process of discharging/charging of the cathode using grand canonical Monte Carlo simulations. The predicted voltage profiles are in good agreement with the experimental ones and provide an approach to rapidly perform design optimization in this phase space. This study serves as a proof-point of machine-learned DFT surrogates to enable battery materials optimization.
19 May 2020 in ACS Energy Letters
23 December 2019 in The Journal of Chemical Physics
Density functional theory (DFT) calculations are routinely used to screen for functional materials for a variety of applications. This screening is often carried out with a few descriptors, which use ground-state properties that typically ignore finite temperature effects. Finite-temperature effects can be included by calculating the vibration properties, and this can greatly improve the fidelity of computational screening. An important challenge for DFT-based screening is the sensitivity of the predictions to the choice of the exchange correlation function. In this work, we rigorously explore the sensitivity of finite temperature thermodynamic properties to the choice of the exchange correlation functional using the built-in error estimation capabilities within the Bayesian Error Estimation Functional-van der Waals (BEEF-vdW). The vibrational properties are estimated using the Debye model, and we quantify the uncertainty associated with finite-temperature properties for a diverse collection of materials. We find good agreement with experiment and small spread in predictions over different exchange correlation functionals for Mg, Al2O3, Al, Ca, and GaAs. In the case of Li, Li2O, and NiO, however, we find a large spread in predictions as well as disagreement between experiment and functionals due to complex bonding environments. While the energetics generated by the BEEF-vdW ensemble is typically normal, the complex mapping through the Debye model leads to the derived finite temperature properties having non-Gaussian behavior. We test a wide variety of probability distributions that best represent the finite temperature distribution and find that properties such as specific heat, Gibbs free energy, entropy, and thermal expansion coefficient are well described by normal or transformed normal distributions, while the prediction spread of volume at a given temperature does not appear to be drawn from a single distribution. Given the computational efficiency of the approach, we believe that uncertainty quantification should be routinely incorporated into finite-temperature predictions. In order to facilitate this, we have open-sourced the code base under the name dePye.
15 December 2020 in Journal of Physics: Energy
Machine-learning potentials are accelerating the development of energy materials, especially in identifying phase diagrams and other thermodynamic properties. In this work, we present a neural network potential based on atom-centered symmetry function descriptors to model the energetics of lithium intercalation into graphite. The potential was trained on a dataset of over 9000 diverse lithium–graphite configurations that varied in applied stress and strain, lithium concentration, lithium–carbon and lithium–lithium bond distances, and stacking order to ensure wide sampling of the potential atomic configurations during intercalation. We calculated the energies of these structures using density functional theory (DFT) through the Bayesian error estimation functional with van der Waals correlation exchange-correlation functional, which can accurately describe the van der Waals interactions that are crucial to determining the thermodynamics of this phase space. Bayesian optimization, as implemented in Dragonfly, was used to select optimal set of symmetry function parameters, ultimately resulting in a potential with a prediction error of 8.24 meV atom−1 on unseen test data. The potential can predict energies, structural properties, and elastic constants at an accuracy comparable to other DFT exchange-correlation functionals at a fraction of the computational cost. The accuracy of the potential is also comparable to similar machine-learned potentials describing other systems. We calculate the open circuit voltage with the calculator and find good agreement with experiment, especially in the regime x ≥ 0.3, for x in Li
x
C6. This study further illustrates the power of machine learning potentials, which promises to revolutionize design and optimization of battery materials.
11 December 2019 in Journal of the Electrochemical Society
Layered Li(Ni,Mn,Co,)O2 (NMC) presents an intriguing ternary alloy design space for optimization as a cathode material in Li-ion batteries. In the case of NMC, however, only a select few proportions of transition metal cations have been attempted and even fewer have been adopted on a large scale. Recently, the high cost and resource limitations of Co have added a new design constraint and high Ni-containing NMC alloys have gained enormous attention despite possible performance trade-offs. Although the limited collection of NMC cathodes have been successful in providing the performance needed for many applications, specifically electric vehicles, this concern around Co requires further advancement and optimization within the NMC design space. Additionally, it is not fully understood if this material space is a disordered solid solution at room temperature and any arbitrary combination can be used or if there exist distinct transition metal orderings to which meta-stable solid solutions will decay during cycling and affect performance. Here, we present a high fidelity computational search of the ternary phase diagram with an emphasis on high-Ni, and thus low Co, containing compositional phases to understand the room temperature stability of the ordered and disordered solid solution phases. This is done through the use of density functional theory training data fed into a reduced order model Hamiltonian that accounts for effective electronic and spin interactions of neighboring transition metal atoms at various lengths in a background of fixed composition and position lithium and oxygen atoms. This model can then be solved to include finite temperature thermodynamics into a convex hull analysis to understand the regions of ordered and disordered solid solution as well the transition metal orderings within the ordered region of the phase diagram. We also provide a method to propagate the uncertainty at every level of the analysis to the final prediction of thermodynamically favorable compositional phases thus providing a quantitative measure of confidence for each prediction made. Due to the complexity of the three component system, as well as the intrinsic error of density functional theory, we argue that this propagation of uncertainty, particularly the uncertainty due to exchange-correlation functional choice is necessary to have reliable and interpretable results. We find that for the majority of transition metal compositions of the layered material, specifically medium to high-Ni content, prefer transition metal ordering and predict the collection of preferred compositions in the ordered region.
28 August 2023 in ECS Meeting Abstracts
Sustainable route to cement production is critical for achieving net-zero goals. In this talk, we will discuss a novel scheme for direct electrochemical synthesis of calcium silicates. We will discuss connections between the charging of convention Li-ion cathodes and electrochemical cement production.