Autonomous Discovery of Battery Electrolytes with Robotic Experimentation and Machine-Learning

Abstract

<div> <div> <div> <p>Innovations in batteries take years to formulate, requiring extensive experimentation during the design and optimization phases. We approach the design of a battery electrolyte as a black-box optimization problem. We report here the discovery of a novel battery electrolyte by a robotic electrolyte experiment guided by machine-learning software. Motivated by the recent trend toward super-concentrated aqueous electrolytes for high-performance batteries, we utilize Dragonfly - a Bayesian machine-learning software package - to search mixtures of commonly used lithium and sodium salts for super-concentrated aqueous electrolytes with wide electrochemical stability windows. Dragonfly autonomously managed the robotic test-stand, recommending electrolyte designs to test and receiving experimental feed- back in real time. Within 40 hours of continuous experimentation, Dragonfly discovered a novel, high-performing aqueous sodium electrolyte that a human-guided design process may have missed. This result demonstrates the possibility of integrating robotics with machine-learning to rapidly and autonomously discover novel battery materials.</p></div></div></div>

Perspective—Combining Physics and Machine Learning to Predict Battery Lifetime

“…[72][73][74] Optimization of a LIB over materials, additives, or system parameters can be facilitated in a closed-loop process where ML plays the role of a surrogate model which chooses the optimal set of design or use variables. 3,6,27,75 Attia et al 6 demonstrated ML-driven closed-loop optimization of fast-charging protocols, where a forecasting model was used to reduce the required experimental cycling time by a factor of 30. 26 Optimization could also use a multi-fidelity setting, 76 where ML chooses from performance measurements of two fidelities; namely, PB model vs cycling experiment, to reduce cost.…”

High‐Throughput Experimentation and Computational Freeway Lanes for Accelerated Battery Electrolyte and Interface Development Research

“…Dave et al built a system named “Otto” to enable HT automated formulation and characterization for liquid aqueous battery electrolytes. [ 88,89 ] Compared to traditional low throughput experiments, this system allows much faster formulation of 140 electrolytes within 40 h. In addition, the machine learning method coupled with automated evaluation of the acquired datasets enables inverse material design. The optimal electrolyte was found to be a novel dual‐anion sodium electrolyte that exhibits a wider electrochemical stability window than the baseline sodium electrolyte.…”

Toward Better and Smarter Batteries by Combining AI with Multisensory and Self‐Healing Approaches

“…A third and equally critical step is the translation of the model predictions to implementable procedures for autonomous orchestration of the materials synthesis robotics [59] and manufacturing processes. While recent examples, for example, from Venkat Viswanathan et al who have demonstrated the ability to autonomous direct the testing of 140 electrolyte formulas and yielding non-intuitive optimum, [60] have shown a path forward, the complexity of the challenges facing BIG-MAP and BATTERY 2030+ will also depend on transformative breakthroughs in a number of other additional areas.…”