Redox Flow Batteries (RFBs) are emerging as a promising option for facing the stationary energy storage needs in electric grids as regards both power quality and energy management services [1], for advantages such as intrinsic reversibility, power/energy independent sizing, high round-trip efficiency at top power, room temperature operation, and extremely long charge/discharge cycle life [2]. The best-researched and already commercially exploited technology is all-vanadium chemistry [3], where important performance progresses have been recently achieved on laboratory small cells [4], which allow for higher power densities with more compact and less expensive cell stacks. But other chemistries are also under development, which can provide also high energy densities, thus opening the door to mobility applications, in a time when electric vehicles are blooming [5]. These programs aim at more compact and cheaper systems, which can take the technology to a real breakthrough in stationary and mobile applications. However some technological issues lay unresolved, which also burden RFBs’ potentials [6,7]. Redox flow batteries are similar to PEMFCs (proton exchange membrane fuel cells) in many aspects, and actually they use an ionomer as electrolyte and two porous layers as electrodes where reactants diffuse, working at a low temperature. Indeed, they are special types of fuel cell, where reactants circulate from two tanks to the cell electrodes, making the arrangement suitable for full reversible electrochemical reactions. As such, RFB systems are typically designed borrowing the fuel cell stack layout, which is based on bipolar plates interleaved between cells. This lay out that has been conceived with the aim of achieving compact, minimum volume designs, since early devices. However, its application to a system using electrolytes dissolved in a conducting liquid (e.g. a sulfuric acid solution) causes the circulation of shunt (i.e. parasitic) currents which affect the stack efficiency. Moreover, even if these losses, which depend on the square of the stack voltage, are acceptable at full power, when load current is near its maximum, their relative effect increases at lower loads, prejudicing the efficiency of the stack. Longer and thinner ducts can reduce shunt currents, but they involve higher pressure drops in the solutions and consequently a higher power demand for the circulating pumps. A give-and-take compromise design is generally adopted that balances shunt currents and pressure drops, which are the most important loss effect, apart from cell electrochemical losses. In this work we have developed an alternative inter-cell plate design, based on an alternate-series topology, that allows to minimize both shunt current and pressure drop losses, providing at the same time a very flexible operation scheme, proving quick load variations and null losses at no-load. The drawback of this scheme is a more severe specification for the solid-state dc/dc step-up converter, namely the power management system interfacing the stacks with the dc/ac converter and the grid. This issue has been addressed by mean of an original design, i.e. an isolated boost with coupled inductors (IBCI). The presentation will provide full details of the fluid-dynamic stack topology and will outline the basic operation concept of the dedicated dc/dc converter.
Search from 250 M+ research papersSearch from 250 M+ research papers
Open Access
Jan 1, 2018
Vida Dujmović + 1
