Ambipolar zinc-polyiodide electrolyte for a high-energy density aqueous redox flow battery

Redox flow batteries are receiving wide attention for electrochemical energy storage due to their unique architecture and advantages, but progress has so far been limited by their low energy density (~25 Wh l−1). Here we report a high-energy density aqueous zinc-polyiodide flow battery. Using the highly soluble iodide/triiodide redox couple, a discharge energy density of 167 Wh l−1 is demonstrated with a near-neutral 5.0 M ZnI2 electrolyte. Nuclear magnetic resonance study and density functional theory-based simulation along with flow test data indicate that the addition of an alcohol (ethanol) induces ligand formation between oxygen on the hydroxyl group and the zinc ions, which expands the stable electrolyte temperature window to from −20 to 50 °C, while ameliorating the zinc dendrite. With the high-energy density and its benign nature free from strong acids and corrosive components, zinc-polyiodide flow battery is a promising candidate for various energy storage applications.

Supplementary Fig. 10. Mass spectrometry analysis of (a) pristine and (b) EtOH-added catholyte at fully charged condition. The presence of ZnI 3 and molecular triiodide confirms our NMR and DFT-based analysis. It should be noted that the solutions are diluted more than 1000 times to obtain high resolution spectra. Hence the peak intensity may not represent the respective ion concentration in the electrolyte.
Supplementary Fig. 11. Voltage profiles of a flow cell test on a 2.5M ZnI 2 electrolyte with and without ethanol. The addition of ethanol (10 vol% ) will decrease the conductivity of electrolyte, which will then lead to the reduced voltage efficiency (VE) from 88.3% to 84.5% under the same current density (10 mA/cm 2 ) and operating temperature (25 o C).

Estimation of energy density and definition of theoretical capacity
In ZIB systems, during the charge process, zinc ions at the negative side are electroplated onto the surface of the carbon fibers within the graphite felt electrode as the zinc metal. Meanwhile, the zinc ions are also functioning as charge carriers across the membrane. Therefore, the same amount of zinc ions transfer through membranes from positive to negative half-cell to balance the charge due to the negligible transportation of iodide ions through the membrane. The XPS analysis in Supplementary Fig. 2 for a used Nafion membrane after charge shows the presence of zinc metal cations in the water channels. The negligible presence of iodine elements (compared with zinc ion concentration) clearly proves that fully hydrated zinc cation (i.e., [Zn.6H 2 O] 2+ ) is the main charge transport species during the cycling process. During discharge the reverse process takes place. As a result, the amount of zinc ions will decrease at the positive side and remain constant at the negative side with increasing the state of charge (SOC). In addition to the above analysis, experimental flow cell charge/discharge cyclings with varied anolyte volumes deliver identical energy output. Therefore, it is concluded that the anolyte in a ZIB flow cell does not directly participate in the redox reactions that contribute to the overall cell energy and power delivery. The energy density of a ZIB flow cell is solely dependent on the concentration of the catholyte. Therefore, the volume of effective active species is defined as the volume of cathode electrolyte (n = 1 in Eqn.1). As a result, the theoretical capacity is calculated based on that all the iodides convert into triiodide in catholytes on charge.
To further explain this design, we carried out two different flow cell tests with anolytes/catholytes volume ratio of 1:3 and 2:3, respectively, while the catholyte volumes are the same 75 mL in both flow cell tests. The concentration of ZnI 2 at 0% SOC at both sides is 3.5 M.
As shown in Supplementary Fig. 14, the charge/discharge voltage curves for both cells are nearly identical, confirming our design that the system energy density does not depend on the anolyte volume, but solely on the catholyte, which, on the system level, provide an opportunity to minimize the system volume.

Characterization
In order to confirm the presence of I 3 and I 5 in ZIB electrolytes, the commercially available Tetrabutylammonium Triiodide (TBA-I 3 , TCI, 99%) and prepared Tetrabutylammonium Pentaiodide were measured using Raman spectroscopy for comparison, respectively. to an Exactive mass spectrometer (Thermo Fisher Scientific). In the negative ion mode, polyiodide species dominate the spectrum. All the mass spectra were recorded at a mass resolution of 25000.
The X-ray photoelectron spectroscopy (XPS) analyses were performed using a Kratos Axis Ultra spectrometer (Kratos Analytical, Manchester, UK) equipped with a monochromatized Al X-ray source (10 mA, 15 kV). The Nafion membrane is washed with DI water prior to measurement to remove surface adsorbed electrolyte molecules.
The morphologies of zinc dendrites on the surface of graphite felts (GFs) were characterized by a FE-SEM (JEOLJSM-7600F).