Redox Flow Battery Simulation for Long-duration Energy Storage

Technical Challenge

Long-duration energy storage is critical to ensure reliable and sustained integration of intermittent wind and solar resources. Redox flow batteries (RFBs) offer numerous advantages, including decoupled energy and power, long service life, and safe, scalable design. Optimizing the performance of RFBs requires maximizing the allowable current drawn during discharge

while minimizing the pressure drop required to pump electrolyte across the stack. Both factors depend sensitively on flow channel design and geometry. Simulation of RFBs can help reduce development time and cost, de-risk scale-up, guide optimal design, and accelerate commercialization.

Veryst Solution

 

 

Veryst used COMSOL Multiphysics® to simulate the performance of three different flow channel designs to determine the impact of flow rate and current density on cell performance at room temperature. Serpentine, parallel, and interdigitated flow fields were simulated (Fig. 1) for a kW-scale cell (25 cm2 geometric area).

 

 

 

 

The three simulated flow configurations exhibit distinct velocity distributions (Fig. 2) that impact the cell pressure drop and mass-transfer-limiting current density (Fig. 3). The serpentine channels redirect flow across the porous electrodes, achieving good mass transfer (high limiting current density) at the expense of enhanced viscous dissipation (high pressure drop).

 

 

 

 

 

The parallel flow configuration mitigates viscous dissipation by dividing the flow across all 25 channels, thereby reducing the speed of the flow into each channel. However, at practical flow rates (>25 ml/min) the upstream channels are bypassed due to strong fluid inertia, resulting in uneven flow distribution and poor mass transfer. The interdigitated flow field achieves both low pressure drops and moderate limiting current densities by separating the inlet and outlet channels, forcing flow through the electrodes at relatively modest speed.

 

 

 

Multiphysics simulations of coupled fluid flow and electrochemistry in the porous electrodes were performed for the interdigitated flow cell to evaluate its polarization behavior during charge and discharge. Electrochemical kinetic parameters for an alkaline organic electrolyte chemistry were calibrated using cyclic voltammetry measurements from literature [1] (Fig. 4). These voltammograms predict an equilibrium cell potential of 1.2 V.

 

 

 

 

Cell overpotentials were simulated over a wide range of flow rates and current densities to map out the feasible operating space of the kW-scale RFB (Fig 5A). The partitioned channel design in the interdigitated flow cell enables control of the mass-transfer-limiting current density by varying the flow rate. Our simulations enable direct computation and comparison of the various voltage losses in the cell (Fig 5B), which limit the cell’s performance. These losses are due to ohmic (ionic and electronic), mass-transfer (concentration), kinetic (activation), and contact resistances.

Serpentine, parallel, and interdigitated flow fields
Figure 1: Simulated serpentine, parallel, and interdigitated flow configurations of a 25 cm² cell.
Velocity distributions in flow channels and electrodes.
Figure 2: Velocity distributions in the serpentine, parallel, and interdigitated flow cells for a flow rate of 50 ml/min. (A) In-plane distribution in the flow channels. (B) Through-plane distribution in the electrodes.

 

Cell pressure drop and limiting current density.
Figure 3: Pressure drop (top) and limiting current density (bottom) versus flow rate in the serpentine, parallel, and interdigitated flow cells.
Cyclic voltammograms of posolyte and negolyte.
​​​​Figure 4: Cyclic voltammograms of posolyte (ferrocyanide) and negolyte (anthraflavic acid) scanned at 100 mV/s. Electrochemical kinetic parameters used in the simulations were calibrated to experimental measurements from literature [1].

 

Cell polarization and relative voltage losses.
Figure 5: (A) Polarization plot for the interdigitated flow cell with an initial state of charge of 50%. (B) Pie chart of relative losses that contribute to the net cell overpotential.

Conclusion

Veryst’s simulations highlighted the performance tradeoffs among three different flow configurations for a kW-scale redox flow battery (RFB) cell. The interdigitated flow cell design achieves the optimal balance between flow resistance and mass transfer. Cell polarization is primarily limited by ionic conduction in the electrolyte. Simulations such as these can help accelerate timetables for RFB development and commercialization.

References
[1] K. Lin et al., Science 349.6255 (2015): 1529-1532

 

You may also like...

PEM Electrolyzer Simulation for Low-Temperature Hydrogen Production

Water electrolysis for hydrogen production is a key enabling technology for global decarbonization. In this case study, Veryst simulated the electrical current distribution and gas generation in a proton exchange membrane electrolyzer stack to identify potential process inefficiencies and recommend optimal operating conditions

Thermal Management in Battery Packs

Thermal management is crucially important for battery performance in consumer products, electric vehicles, and grid-level storage systems. In this case study, Veryst used multiphysics simulations to evaluate different thermal management strategies in prismatic and cylindrical battery packs.

Battery Pack Impact Simulation

From smartphones and cameras to wireless headphones and battery packs, portable electronics proliferate. Consumers expect excellent resilience to device drops, increasing pressure on manufacturers to test thoroughly and optimize their designs. Veryst utilized its unique expertise in accurately modeling complex materials, conducting high strain rate testing, and simulating impact events to simulate the drop impact of an external battery pack.

Bubble Entrapment in Microchannels

Bubbles trapped in microchannels can distort the fluid flow and impact the device performance. Veryst developed a multiphase CFD model to predict the effect of geometry and surface properties on the likelihood of bubble entrapment.

Scaling Yield and Mixing in Chemical Reactors

Scaling chemical reactions from the lab to pilot or production requires a detailed understanding of the physical system, which frequently involves heat transfer, mass transfer, reaction kinetics, and fluid flow. This case study illustrates how multiphysics simulations can support design decisions involved in scaling up chemical reactors.

Can we help? Just want to keep in touch?