Redox Flow Battery Cost Modeling: Bridging Techno-Economic Analysis to Materials Selection Criteria

Abstract

Growing adoption of renewable energy technologies, combined with a desire to improve electrical grid efficiency, is driving the development of grid-scale energy storage devices. Redox flow batteries (RFBs) offer an attractive solution to grid-scale storage due to independent scaling of power and energy, long service life, and simple manufacturing.1 Despite these attractive features, RFBs have failed to reach widespread deployment due to high prices; in 2014, RFB prices exceeded $500 kWh-1.2 The United States Department of Energy (DOE) has established that decreasing RFB system price (including the battery, inverter, and installation) to below $150 kWh-1, in the near-term, could enable market penetration for 2 – 4 h grid-scale discharge applications.3 Despite existing high prices, recent studies indicate that RFBs could achieve the aggressive DOE target by minimizing reactor and materials costs.2 Techno-economic modeling provides a powerful tool for evaluating the price performance of energy storage systems by relating system price to materials properties, electrochemical performance, and component cost parameters. Prior techno-economic studies on RFBs have considered cost reductions afforded by decreasing manufacturing costs2,4,5 and improving reactor performance,6 but no studies systematically evaluate cost and performance tradeoffs from varying electrolyte materials selection. To address this literature gap, we develop a techno-economic model that accounts for the reactor, electrolyte, balance-of-plant, and additional cost contributions to RFB price. In particular, a detailed electrolyte cost model, accounting for the constituent active material, salt, and solvent costs, allows for cost-driven electrolyte materials selection. After solidifying the techno-economic model, we quantify cost constraining variables for both aqueous (Aq) and nonaqueous (NAq) RFBs, which span a broad design space with a wide array of materials options. AqRFBs take advantage of low area specific resistance (ASR) cells and inexpensive electrolytes employing water as the solvent and inorganic salts (e.g., H2SO4, NaCl). One drawback to AqRFBs is that the typical electrochemical stability window of aqueous electrolytes (≤ 1.5 V) limits energy density. Contrastingly, NAqRFBs offer the possibility of much higher energy densities by utilizing nonaqueous solvents with wide electrochemical windows (3 – 4 V), but suffer from high solvent and salt costs, as well as high reactor ASR. Through our techno-economic analysis, we find that AqRFB prices are sensitive only to variations to cell voltage and active material cost and molecular weight, while NAqRFB prices are sensitive to nearly every model parameter because NAqRFBs exhibit similar cost contributions from both the reactor and electrolyte. Finally, this work explores the available electrolyte materials design space mapped out by the techno-economic model. Design maps, such as that presented in Figure 1, illustrate tradeoffs in RFB constituent costs and performance to achieve a RFB with a $100 kWh-1 battery price (excluding inverter and installation costs). The analysis culminates in a set of suggested pathways to realistically achieve the near-term DOE target and even decrease the battery price to $80 kWh-1and below. Beyond the immediate application to RFBs, this work establishes a framework for cost-conscious electrochemical device research. Our calculated design maps translate system-level constraints to quantitative guidelines for materials selection, bridging the gap between abstract cost models and tangible experimental goals. The design maps also highlight alternative development pathways that may be unaddressed or underexplored. We hope that this methodology will apply to other electrochemical systems where cost is a primary design constraint. Acknowledgments We gratefully acknowledge the financial support of the Joint Center for Energy Storage Research (JCESR) and the National Science Foundation Graduate Research Fellowship Program. We also thank the JCESR Flow Chemistry Sprint team for their continued input and feedback.