Redox flow batteries (RFBs) are often limited by their low energy density and high costs. The authors have developed a high-energy, low-cost all-liquid Polysulfide-Iodide Redox Flow Battery (PSIB) using highly soluble, inexpensive, and reversible polysulfide and iodide materials. Compared to the metal mixing or semi-solid methods typically used for high-energy RFBs, the all-liquid characteristic of PSIB is essential for practical scaling and development. This study has achieved one of the highest energy densities in all-liquid aqueous RFBs (43.1 Wh L−1 for both catholyte and anolyte), along with high Coulombic efficiency (93-95%) and stable cycle life. The material cost of the PSIB system per kilowatt-hour ($85.4 kWh−1) is significantly lower than that of the state-of-the-art vanadium-based flow batteries ($152.0-154.6 kWh−1). Additionally, this study demonstrates the excellent electrochemical reversibility of the polysulfide and iodide redox chemistry.Research Content

The authors first introduced the design concept of the PSIB flow battery. To achieve an all-liquid characteristic along with high energy density and low cost, highly soluble, inexpensive, and reversible polysulfide and iodide materials were utilized to obtain a high-energy, low-cost PSIB flow battery (Figure 1a). Figure 1b displays the reaction potentials and volumetric capacities of various aqueous redox pairs, indicating that I−/I3− and S2−/S22− both have higher volumetric capacities than other alternatives due to their high solubility (>6.0M). Furthermore, the potential difference between I−/I3− (0.536V vs. SHE) and polysulfide (S2−/S22−, −0.510V vs. SHE) is equivalent to a theoretical cell voltage of 1.05V. Figure 1c shows the cyclic voltammograms (CV) on a gold electrode for K2S2–KCl solution (blue) and KI–KCl solution (red). The oxidation and reduction peaks for the iodide-containing solution occur at approximately 0.46V vs. SCE, with a peak separation of 101mV; whereas the peaks for the polysulfide-containing solution occur at about −0.50V vs. SCE, with a peak separation of 381mV. Compared to the S2−/S22− pair, the I−/I3− pair exhibits faster reaction kinetics. Therefore, the authors developed nickel foam treated with polysulfide (SNF) to serve as the anode in the battery, promoting the kinetics of the polysulfide redox reaction.

Subsequently, the authors investigated the electrochemical characterization of the PSIB in static mode. Figure 2a presents the galvanostatic charge-discharge profiles of the PSIB with various polysulfide/iodide concentration combinations, revealing that all combinations achieved more than 85% of the theoretical capacity, indicating reversible and effective electrochemical energy conversion between iodide and polysulfide. Utilizing the highest concentrations (6M KI | 3.3M K2S2), with an excess volume of polysulfide, the system can achieve 46.0 Ah L−1 and 41.4 Wh L−1 for catholyte + anolyte; through capacity matching, it can achieve 52.9 Ah L−1 and 49.4 Wh L−1 for catholyte + anolyte. Furthermore, the discharge capacity increases proportionally with the concentration of iodide, consistent with the fact that iodide is the capacity-limiting side, being one-third the capacity of polysulfide at the same concentration and volume (Table 1). Additionally, the PSIB cell achieved a Coulombic efficiency (CE) of 93% and a voltage efficiency (VE) of 78% at 2M KI-2M K2S2, but the efficiency decreased at high electrolyte concentrations due to increased electrolyte resistance. The authors then assessed the cycling stability of the PSIB cell (Figures 2b and c). The PSIB cell showed a high capacity retention rate (>98%) and high Coulombic efficiency (>97%) after 50 cycles at 80% SOC. The Coulombic efficiency rapidly increased from 80% in the first cycle to over 95%, which can be attributed to the regulation of the carbon felt electrode, such as increased hydrophobicity/wettability after a few cycles, thereby reducing charge transfer and interfacial resistance. Figures 2d and e display the rate capability of the PSIB cell. When the current increased from 5 mA cm−2 to 20 mA cm−2, the discharge capacity did not significantly decrease, and the Coulombic efficiency of the PSIB cell remained above 96% (except for the first cycle), indicating that the PSIB system can be used at high current densities. Moreover, the PSIB cell with two membranes (4M KI-3M K2S2) had a higher Coulombic efficiency (97.7%) than the cell with one membrane (4M KI-3M K2S2, 93.8%), demonstrating that using two membranes can reduce crossover, thereby improving Coulombic efficiency. Additionally, as the current density increased from 5 mA cm−2 to 20 mA cm−2, the voltage efficiency decreased from 84.0% to 58.5%, and the charging voltage gradually increased during the cycling process.The authors also investigated the sources of overpotential. Reference electrodes were utilized in both the anode and cathode of the PSIB cell (Figures 3a and b). The four-electrode PSIB cell divides the overpotential into three parts: (1) the transmembrane voltage (related to the Ohmic resistance of the electrolyte and membrane); (2) the anode with the Hg/Hg2SO4 reference electrode (reflecting the kinetics and mass transport losses of the iodide/triiodide redox reaction); (3) the cathode with the Hg/HgO reference electrode (reflecting the kinetics and mass transport losses of the polysulfide redox reaction). As shown in Figure 3c, the transmembrane voltage (initially 60 mV, increasing to 74 mV after 10 cycles) is the primary voltage loss of the overall cell overpotential (~83 mV). On the other hand, the catholyte (~2 mV) and anolyte (~21 mV) reactions exhibit smaller overpotentials. In summary, the authors believe that improvements in membrane conductivity and stability will significantly enhance the rate performance and cycle life of the PSIB system.

The authors also investigated the impact of flow rate on the cell performance and its practical scalability. Due to the reduction in mass transfer losses, the voltage efficiency and energy efficiency of the PSIB cell in continuous flow mode were significantly improved. The authors studied the effects of flow rate and current density on the charging/discharging curves (Figure 4b) and efficiency (Figure 4c) of the PSIB. The results indicated that the voltage efficiency achieved at a flow rate of 4 mL min−1 (ranging from 88.7% to 75.2% at 5 to 15 mA cm−2) was higher than that achieved in static mode (ranging from 84.0% to 62.4% at 5 to 15 mA cm−2). This improvement can be attributed to the reduced mass transfer losses at the electrode surface due to the flow of the electrolyte. However, further increasing the flow rate from 4 mL min−1 to 10 mL min−1 did not improve the voltage efficiency, suggesting that the mass transfer losses at 4-10 mL min−1 under 15 mA cm−2 are similar, meaning that the increase in flow rate has a limited impact on cell performance when the mass transfer resistance and redox reaction rates are balanced. Additionally, the PSIB system achieved a high Coulombic efficiency (CE) (~95%) after cycling (over 160 hours, Figure 4c), further demonstrating the stable and efficient cycling of PSIB cells in continuous flow mode.

The authors also studied the PSIB system with higher iodide/triiodide concentrations at 100% State of Charge (SOC), where the volume of the anolyte was designed to match the capacity of the catholyte. As shown in Figure 4d, the PSIB was operated for the first time at a current density of 37 mA cm−2 (the highest density used in this study) and a flow rate of 4.6 mL min−1, achieving an energy density of 43.1 Wh L−1 for catholyte + anolyte (100% SOC). The flow battery cycled between 25 and 37 mA cm−2 and showed stable cycling in subsequent cycles in continuous flow mode, with the Coulombic efficiency of the highest concentration PSIB at 100% SOC ranging between 85% and 93%.

Finally, the authors further investigated the cycling stability of the PSIB at various current densities and States of Charge (SOC). As shown in Figure 5, the capacity remained stable across different current densities (5 mA cm−2 to 25 mA cm−2) and SOC ranges (16% to 80% SOC), with the Coulombic Efficiency (CE) sustained above 90%. The stable cycling of the highest concentration PSIB battery indicates that the proposed PSIB has the potential to evolve into the next generation of redox flow batteries.Research Discussion and Conclusion

Figure 6 compares the theoretical and experimental energy densities of PSIB with the state-of-the-art vanadium-based RFBs, as well as the estimated chemical costs per kilowatt-hour. Firstly, the theoretical energy density of PSIB is estimated to be 80.0 Wh L−1 for both catholyte and anolyte, which is higher than the previously reported high-energy-density all-liquid aqueous flow batteries (for example, the theoretical energy densities of the state-of-the-art hybrid acid VRB and sulfate VRB are 50.3 Wh L−1 and 26.8 Wh L−1 for both catholyte and anolyte, respectively). Optimization of the electrode and cell configuration in the PSIB system in the future will narrow the gap between its theoretical and achieved energy densities. Secondly, the cost of the active materials used in PSIB ($0.29 kg−1 for sulfur and $8 kg−1 for iodine) is cheaper than the active materials used in vanadium-based flow batteries ($24 kg−1 for V2O5), significantly reducing the cost of the flow battery. The estimated cost of active materials per kilowatt-hour for the PSIB system is $85.4 kWh−1, which is notably lower than that of the hybrid acid VRB system ($152 kWh−1) under similar laboratory-scale current and power densities (PSIB: 37 mA cm−2 and 27 mW cm−2; hybrid acid VRB: 25 mA cm−2 and approximately 35 mW cm−2).

In summary, the authors have presented an all-liquid polysulfide/iodide redox flow battery that achieves a high energy density (43.1 Wh L−1 for both catholyte and anolyte) and significantly lower material costs per kilowatt-hour ($85.4 kWh−1) compared to the state-of-the-art vanadium-based redox flow batteries ($152.0-154.6 kWh−1). Future work involving membrane development and stabilization of triiodide/polysulfide intermediates will further improve the PSIB system. With its proven energy density, inherently low material costs, and benign chemical properties, the all-liquid PSIB offers a promising solution for high-energy-density and low-cost energy storage applications.

In summary, the authors have demonstrated an all-liquid polysulfide/iodide redox flow battery that achieves a high energy density of 43.1 Wh L−1 for both the catholyte and anolyte. Compared to the state-of-the-art vanadium-based redox flow batteries, which have a cost range of $152.0 to $154.6 per kWh, the material cost for the PSIB system is significantly reduced to $85.4 per kWh. Future work will focus on membrane development and the stabilization of triiodide/polysulfide intermediates, which will further refine the PSIB system. With its proven energy density, inherently low material costs, and favorable chemical properties, the all-liquid PSIB presents a promising solution for energy storage applications that require high energy density and low cost.

更多内容：
国内首个！《硫铁液流电池通用技术条件》团体标准发布
中性多硫化物/铁氰化物氧化还原液流电池
通过普鲁士蓝固体提升能量密度的硫铁液流电池
用于硫铁电池的连续离子层吸附和反应沉积的硫化铜电催化剂