In flow batteries, energy is produced by passing solutions of 'electroactive' materials — often, metal salts — through an electrochemical cell. A non-metallic electroactive material opens the way to large-scale energy storage. See Letter p.195
The adoption of intermittently available renewable energy sources, such as solar energy and wind power, to more than 20% of total energy capacity will require electric-energy storage systems to be deployed1. For grid-scale applications and remote generation sites, cheap and flexible storage systems are needed, but presently the options are either limited to a specific geographic location (such as pumping water from a reservoir to an elevated level as a source of potential energy) or expensive (for example, conventional batteries, flywheels and superconductive electromagnetic storage)2. On page 195 of this issue, Huskinson et al.3 report a major advance in the development of economical energy storage: a 'flow' battery that uses only water-soluble, non-metallic materials as the electrode components. These materials could lower the cost of flow batteries, while increasing the energy density.
Flow batteries require two soluble electroactive components — compounds that take part in an electrochemical reaction at an electrode. These components are separated by an ion-conducting membrane in an electrochemical cell, in which chemical energy is converted to electricity (and vice versa). In contrast to the stationary electroactive materials of conventional batteries, the electroactive components in flow batteries are pumped through the cell in a flow of liquid, and are stored outside the cell in separate tanks (Fig. 1). This design allows for individual optimization of the amount of energy stored (which is controlled by the size of the storage tanks) and the power generated (which depends on the size of the electrochemical cell or stack of cells).
Because the electroactive materials in flow batteries are stored separately, the possibility that they will react violently with each other is almost completely eliminated, making these devices much safer than conventional batteries. They also have more flexible layouts and are potentially cheaper. Unfortunately, the choice of electroactive materials for flow batteries is limited to a small selection of metal redox systems (with a few exceptions for cathode materials), and by the low solubility of these metal salts — typically, in water. The solubility problem prevents high energy densities from being achieved4.
Huskinson et al. overcome the solubility problem by using as the electroactive components soluble, organic, redox-active materials known as quinones in place of metals. Water-insoluble quinones were proposed5 as electrode materials in 1972, but the use of this class of compound as the energy-storage components of a flow battery is new. The authors found that the chemical reduction of their quinones to form hydroquinones in water at an electrode is very fast, which is a prerequisite for high-power battery discharge.
Redox potentials and the solubilities of metal complexes can be tuned by modifying the ligands bound to the metal atoms. With quinones, these properties can be modified by changing the chemical groups attached to the aromatic rings of the molecules. This offers much wider scope for modification than is possible for metal systems, because the chemical groups are closer to the redox centre than ligands are to metals, and so their effect is more pronounced. In addition, having negatively charged electroactive species — quinones to which negatively charged groups are attached, such as those used by Huskinson et al. — should help to reduce one of the major practical problems associated with flow batteries, namely the crossover of these materials through the negatively charged ion-conducting membrane. However, this approach has a downside, because any improvements in electrochemical properties and/or solubility will be associated with an increase in the molecular weight of the electroactive species, and will therefore reduce the energy density of the battery.
Huskinson and co-workers coupled the liquid quinone/hydroquinone system at the anode of their battery with a bromine/bromide system at the cathode. This cathode system has previously been used in a zinc–bromine flow battery6 and in a hydrogen–bromine regenerative fuel cell7 (a variant of a flow battery). The bromine/bromide cathode provides good energy density at a reasonable cost, although it is corrosive and environmentally unfriendly. When the authors tested a small version (2 square centimetres) of their flow battery, they found that it gave a respectable power density (600 milliwatts per square centimetre) and good current efficiency (the efficiency with which charge is transferred to allow a targeted electrochemical reaction to occur).
The new findings open the way to inexpensive energy storage, but there is a long way to go to develop a practically useful flow battery. In particular, several issues must be addressed before this chemistry can be used in grid-scale energy storage. The authors studied only quinone reduction, so the reverse reaction — the oxidation of hydroquinones — should also be investigated. If the reverse reaction is as fast as quinone reduction, then quinones could potentially be used in high-power devices. The effect of the electroactive-species concentration, and of impurities in the quinones, on the cell's performance and ability to be used through many charge–discharge cycles must be evaluated. If high-purity quinones are needed, it could noticeably increase the cost.
Bromine crossover through the membrane should also be considered seriously. Even if bromine does not react with compounds in the anode system, such crossover will reduce battery capacity and energy efficiency (the ratio of electrical-energy output to input), which should be measured as a function of cycle number. Scaling up from a small single cell to an industrial-sized, multi-cell stack may be challenging, and integrating the various components of a large-scale device into a working battery might also be difficult. For stationary energy storage, a long life (more than 10,000 cycles) is key to keeping costs down, so the number of cycles demonstrated in the paper (15) is far from that needed.
Nevertheless, Huskinson and colleagues' results are promising, and may serve as the basis for a new flow-battery technology. If long-term capacity and energy-efficiency retention can be demonstrated, and if practically useful batteries can indeed be prepared cheaply, then this technology will be suitable for a wide array of energy-storage applications.
Denholm, P., Ela, E., Kirby, B. & Milligan, M. The Role of Energy Storage with Renewable Electricity Generation (Natl Renewable Energy Lab, 2010).
Soloveichik, G. L. Annu. Rev. Chem. Biomol. Eng. 2, 503–527 (2011).
Huskinson, B. et al. Nature 505, 195–198 (2014).
Weber, A. Z. et al. J. Appl. Electrochem. 41, 1137–1164 (2011).
Alt, H., Binder, H., Klempert, G., Köhling, A. & Sandstede, G. J. Appl. Electrochem. 2, 193–200 (1972).
Leung, P. et al. RSC Adv. 2, 10125–10156 (2012).
Cho, K. T. J. Electrochem. Soc. 159, A1806–A1815 (2012).
Wang, W. et al. Adv. Funct. Mater. 23, 970–986 (2013).
About this article
Chemically stable anion exchange membranes based on C2-Protected imidazolium cations for vanadium flow battery
Journal of Membrane Science (2021)
Advanced Energy Materials (2020)
Advanced Materials Technologies (2020)
Facile fabrication of sulfonated poly(aryl ether sulfone)/polybenzoxazine crosslinked membrane for vanadium flow battery application
Polymer Bulletin (2020)
Binding Zinc Ions by Carboxyl Groups from Adjacent Molecules toward Long‐Life Aqueous Zinc–Organic Batteries
Advanced Materials (2020)