Computational screening of electrolyte materials: status quo and open problems

Development of novel approaches for designing advanced energy storage devices generally requires a fundamental understanding of the atomic structure of and processes in constituent materials. Correspondingly, the modern-day frameworks for computational design of electrolyte materials extensively employ either quantum calculations or atomistic simulation methods, or often, a combination thereof. Within the frameworks of recently devised approaches, these two computational methods are augmented by advanced machine learning techniques. Here, we focus on the recent developments in electrolyte materials design, with the emphasis on the computational design of liquid electrolytes. A particular attention is paid to the recent progress in building a unified framework for large-scale and high-throughput screening of liquid electrolyte material systems for battery applications. We address the status quo in the area and present a perspective on essential further efforts that have to be undertaken to construct realistic paradigms for intelligent design of electrolyte materials by testing physical and chemical properties of large sets of candidate chemical compounds and their combinations. The issues of further improvements in quantum calculations and atomistic simulation are also briefly addressed in the context of energy storage device applications.