New battery technology will be crucial to the electrification of transportation and aviation 1, 2 , but battery innovations can take years to deliver. For battery electrolytes, the many design variables present in selecting multiple solvents, salts, and their relative ratios [3][4][5][6][7] mean that optimization studies are slow and laborious, even those restricted to small search spaces. The key challenge is to lower the number and time-cost of experiments needed to formulate an electrolyte for a given objective.

Developing high-energy and efficient battery technologies is a crucial aspect of advancing the electrification of transportation and aviation. However, battery innovations can take years to deliver. In the case of non-aqueous battery electrolyte solutions, the many design variables in selecting multiple solvents, salts and their relative ratios make electrolyte optimization time-consuming and laborious. To overcome these issues, we propose in this work an experimental design that couples robotics (a custom-built automated experiment named "Clio”) to machine-learning (a Bayesian optimization-based experiment planner named "Dragonfly”). An autonomous optimization of the electrolyte conductivity over a single-salt and ternary solvent design space identifies six fast-charging non-aqueous electrolyte solutions in two work-days and forty-two experiments. This result represents a six-fold time acceleration compared to a random search performed by the same automated experiment. To validate the practical use of these electrolytes, we tested them in a 220 mAh graphite∣∣LiNi0.5Mn0.3Co0.2O2 pouch cell configuration. All the pouch cells containing the robot-developed electrolytes demonstrate improved fast-charging capability against a baseline experiment that uses a non-aqueous electrolyte solution selected a priori from the design space.

High-concentration aqueous electrolytes have shown promise as candidates for a safer, lower-cost battery system. Ionic conductivity is a key property required in high performing electrolytes; the Advanced Electrolyte Model (AEM) has previously shown great accuracy in predicting ionic conductivity in highly-concentrated non-aqueous electrolytes. This work provides extensive experimental data for mixed and highly concentrated aqueous electrolyte systems, rapidly generated via a robotic electrolyte testing apparatus. These data demonstrate exceptional accuracy from AEM in predicting conductivity in aqueous systems, with the accuracy being maintained even in highlyconcentrated and mixed-salt regimes.

No 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>