Cost Projections of Synthetic Active Materials for Redox Flow Batteries

Abstract

Inexpensive electrical-energy storage is critical for successful transformation of the electric grid [1]. Redox-flow batteries (RFBs) possess compelling attributes for cost-effective grid-scale storage [2]. Aqueous-soluble organics and organometallics have recently received attention as high-performance, non-flammable, non-toxic, and non-corrosive energy carriers for RFBs [3]. The primary goal of this work is to develop an improved understanding of whether these types of active materials can meet the price targets set by the U.S. Department of Energy (DOE) for stationary electricity storage. To achieve this goal, we use a bottom-up approach to determine if cost of manufacturing and sales price targets can be met with select families of organic and organometallic molecules. We also identify desirable characteristics of RFB active materials and their production processes which are critical for achieving economic feasibility.Synthetic organic and organometallic active materials for flow batteries offer a large design space in terms of solubility, redox potential, kinetic activity, stability, and other performance attributes. In addition to these performance metrics, price is an additional key metric for any new flow battery active material. Here we develop cost and price analyses (expressed on the $/kg and $/kWh basis) for synthetic organic negative electrolytes based on anthraquinones, and positive electrolytes based on ferrocyanides. We focus on two key economic metrics: the cost of manufacturing (which is helpful for comparing among production process options) and the estimated sales price (which is the full accounting of what a product will be sold for and is the metric which a redox flow battery company must consider). Plant construction capital costs for the synthetic active materials considered here are estimated using analogy with an existing plant where sufficient information is available or a step-counting method when no such information is available. We find that simple derivatives of anthraquinones, e.g., anthraquinone disulfonic acid (AQDS)[4],[5], which have simple production processes and use no organic solvents in their production, have an estimated sales price of ~25 $/kWh (assuming 1.5V cell potential) at an annual production volume (40-50 kMTA) equivalent to supply a flow battery capacity of ~10 GWh/y (which would require > $100M in capital to build a dedicated factory), while more elaborate functionalization and additional processing results in higher values. We also considered contract manufacturing of AQDS at ~200 MWh/y equivalent capacity, which results in an estimated sales price of ~40 $/kWh, a relatively modest increase compared to a large dedicated plant, and reflective of simple processing. We find that ferrocyanides produced in a variety of ways, and with several different cations (Na+, K+, NH4 +), also result in estimated sales prices of ~25 $/kWh for a dedicated production facility producing sufficient material for ~10 GWh/y. For reference, the sales price today of both of the Li-ion active materials (graphite plus the metal oxide cathode) is ~50 $/kWh, while that of a vanadium flow battery is about 50 to 60 $/kWh. The fact that even low-cost precursors and simple processing result in a couple price of ³50 $/kWh for synthetic reactants with a 1.5V cell potential has several clear implications for the pursuit of new synthetic reactants, including a focus on: (1) developing couples with a cell potential as high as possible, (2) developing redox active organics with low equivalent weights, preferably <100 g/equiv, and certainly less than the hundreds of g/equiv for some of the actives analyzed here, (3) analyzing the existing production of all raw material and intermediate inputs and whether they can all be procured at low expense, (4) avoiding use of significant amounts of organic solvents, which drive up processing costs even if they can be recycled. This analysis highlights the price advantages of elemental actives (e.g., transition metal salts, halide salts) that have low equivalent weights, very low-cost precursors, and simple processing. It is also clear that integration with existing production of upstream precursors to the desired active materials is the preferred route to cost-effective actives. Acknowledgements The information, data, or work presented herein was funded in part by the Advanced Research Projects Agency-Energy (ARPA-E), U.S. DOE, under Award Number DE-AR000994.