(108b) High-Throughput and Data-Driven Strategies for the Design of Deep Eutectic Solvent Electrolytes.

Deep eutectic solvents (DES) are composed of two or more materials that when mixed possess a lower melting point than either of the individual components. In recent years, both DES and their Ionic Liquid (IL) counterparts have been proposed as electrolytes for various energy storage applications due to their adequate ionic conductivities, particularly for Redox-Flow Batteries (RFBs). A strong emphasis has been placed on RFB materials that are low-cost and environmentally friendly. In this regard, DES have an advantage over ILs because they are composed of cheap, widely available materials with an inexpensive synthesis method. In addition, many of these materials are biodegradable and non-hazardous. However, a major challenge in finding optimum candidates is the overwhelming design space for these materials. The number of possible DES combinations rivals or even exceeds the number of estimated 10¹⁸ different possible ILs. Current experimental efforts and methods have been slow to make a significant dent in the design space. While computational methods have become a powerful tool to provide further insight where experiments may fall short, a common theme amongst the literature is the lack of data available to make significant strides in developing generalized predictive models for DES. The development of ideal DES electrolytes will require optimization and functional design strategies, along with new methods that can tackle the vast chemical space.

Here, we present our approach to the design of DES electrolytes using high-throughput experimentation and data driven strategies. DES are synthesized by first dissolving the starting components in concentrated stock solutions using volatile solvents. An open-source, automated liquid handling robot then prepares solutions of the components at various molar compositions in multi-well plates. The samples undergo a series of evaporation steps to remove the volatile solvent followed by a final step under vacuum. Liquid samples at room temperature are identified, and the eutectic composition is determined using an in-house constructed and open-source thermal infra-red imaging system able to determine melting points in high-throughput at a fraction of the time compared to traditional methods. Electrochemical characterization to determine conductivities and potential windows is performed in a high-throughput manner using multi-well plates with screen-printed electrodes connected to a standard potentiostat. Finally, approaches to using data science, cheminformatics and engineering metrics to develop design of experiments is presented as a strategy to further probe the chemical space for these materials.