Investigating the Effects of Operation Variables on All-Vanadium Redox Flow Batteries Through an Advanced Unit-Cell Model

Abstract

Next-generation redox flow batteries will benefit from the progress of macroscopic continuum models that enable the optimization of new architectures without the need of expensive fabrication and experimentation. Despite previous attempts, there is still need for robust and thoroughly validated models. Here, a steady-state two-dimensional unit-cell model of an all-vanadium redox flow battery is presented. The model integrates state-of-the-art descriptions of the fundamental physical phenomena, along with new features such as local mass transfer coefficients for each active species, precise sulfuric acid dissociation kinetics, and experimental data of the electrochemical parameters and electrolyte properties. The model is validated at different states of charge and flow rates using polarization, conductivity and open circuit voltage measurements. Then, the contribution of operating conditions on battery performance is studied by analyzing its separate effect on the various phenomena that affect cell performance, such as local pore mass transfer limitations, parasitic hydrogen evolution reactions, crossover and self-discharge fluxes. The resulting model is a reliable tool that can be used to assess the relevance of these coupled phenomena that take place simultaneously within the reaction cell. This important information is critical to optimize cell components, reactor design and to select optimal operating conditions.

Exploring Temperature Effects in All-Vanadium Redox Flow Batteries through a Validated Unit-Cell Model

“…Zero heat flux (i.e., adiabatic) conditions are imposed at the current collectors, ∇T • ⃗ n = 0. The other boundary conditions remain the same as in the isothermal model [40]. A uniform electric potential is applied at the current collectors with a voltage difference V cell between the anode and cathode.…”

“…In-situ battery tests were carried out in the same experimental setup used for the isothermal campaign described in [40]. The installation was placed inside a climatic chamber to control and ensure a constant system temperature (Figure S1), with measurements taken at constant temperature increments of 10 • C from 5 • C to 45 • C. A 3 cm × 3 cm filter-press single cell operating in flow-through configuration was used for the electrochemical tests (Figure 1a).…”

“…The non-isothermal model proposed here is an extension of the isothermal model of a unit-cell all-vanadium redox flow battery presented in our previous work [40]. The isothermal model was validated at a constant temperature of 25 • C for different flow rates and SOCs.…”

“…where C * j and C s j refer to the species concentrations in the bulk space and at the fibers surface, respectively. The concentrations at the fiber surfaces are computed assuming that the mass transfer flux of the vanadium species j = {V 2+ , V 3+ , VO 2+ , VO + 2 } at the electrode-electrolyte interface is modeled by a linear Nernst diffusion layer [40]. The local mass transfer coefficient, k m,j , is modeled by the semi-empirical correlation proposed by Deepak et al [45] k m,j = 7…”

“…Having recognized the need of advanced non-isothermal models to accelerate the development of RFBs, in this work we extend our previous isothermal model for vanadium RFBs (VRFB) [40] to a fully validated non-isothermal model. State-of-the-art descriptions of all the physicochemical processes are included, along with temperature-dependent chemical kinetics, electrolyte properties, mass transfer coefficients, and sulfuric acid dissociation equilibrium.…”