Towards Self-Driving Labs for Better Batteries: Accelerating Electrolyte Discovery with Automation and Bayesian Optimization

Jackie T. Yik,Carl Hvarfner,Jens Sjölund,Erik J. Berg,Leiting Zhang

DOI: https://doi.org/10.26434/chemrxiv-2024-mqb6s

2024-11-28

Abstract:The integration of automation and data-driven methodologies offer a promising approach to accelerating materials discovery in energy storage research. Thus far, in battery research, coin-cell assembly has advanced to become near fully- automated but remains largely disconnected from data-driven methods, which have been primarily developed for computational or multi-fidelity datasets. To bridge the disconnect, this work presents a self-driving laboratory framework designed to accelerate electrolyte discovery by integrating automated coin-cell assembly, galvanostatic cycling of LiFePO4||Li4Ti5O12 full cells, and Bayesian optimization for selecting subsequent experiments based on prior results. The integration of Bayesian optimization highlights machine-intelligent experimental decision-making enabling closed-loop experimentation-analysis workflow for self-driving laboratories. The study focuses on an organic-aqueous hybrid electrolyte system comprising four co-solvents dimethyl sulfoxide, trimethyl phosphate, acetonitrile, and water—and two salts, lithium perchlorate and lithium bis(trifluoromethanesulfonyl)imide (LiTFSI). Using this framework, high- performing electrolyte formulations were identified, with water content emerging as a key performance factor. Online electrochemical mass spectrometry provided insights into hydrogen evolution as a significant side reaction, while revealing limited effects due to co-solvent interactions. The results highlight the potential of combining Bayesian optimization with autonomous experimentation for improving efficiency in materials research, while contributing new insights into electrolyte behavior for next-generation aqueous lithium-ion batteries.

Chemistry

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to accelerate the discovery process of battery electrolytes through automated experiments and Bayesian optimization techniques. Specifically, the research aims to develop a self - driving laboratory framework that can integrate automated button - cell assembly, constant - current charge - discharge tests of full - cells, and a Bayesian optimization method for selecting subsequent experiments based on previous results. The goal of this framework is to improve the performance of organic - aqueous hybrid electrolyte systems, especially to explore the key performance factors in electrolyte systems composed of four co - solvents (dimethyl sulfoxide, trimethyl phosphate, acetonitrile, and water) and two salts (lithium perchlorate and lithium bis(trifluoromethanesulfonyl)imide).

The core problem of the research is that current battery testing methods and efficient experimental methods have remained almost unchanged for decades. Although machine - learning techniques and data - driven methods have made certain progress in battery material modeling and property prediction, these techniques are mainly limited to computational - level applications and have not been effectively combined with the actual experimental process yet. In addition, although the cost reduction and technological progress of laboratory automation equipment have made automation in small - scale research environments possible, how to combine these automation techniques with data - driven methods to accelerate the research of next - generation batteries remains a challenge.

For this reason, this paper proposes a robotic platform combined with Bayesian optimization for screening electrolytes in full - cells, demonstrating the feasibility of optimizing new - type electrolyte systems under the condition of limited knowledge of component interactions. In this way, the research not only accelerates the discovery process of electrolytes but also provides valuable insights for electrolyte design in sustainable aqueous lithium - ion batteries.