Aqueous solutions of super reduced polyoxotungstates as electron storage systems

Due to the increasing energy density demands of battery technology, it is vital to develop electrolytes with high electron storage capacity. Polyoxometalate (POM) clusters can act as electron sponges, storing and releasing multiple electrons and have potential as electron storage electrolytes for flow batteries. Despite this rational design of clusters for high storage ability can not yet be achieved as little is known about the features influencing storage ability. Here we report that the large POM clusters, {P5W30} and {P8W48}, can store up to 23 e− and 28 e− per cluster in acidic aqueous solution, respectively. Our investigations reveal key structural and speciation factors influencing the improved behaviour of these POMs over those previously reported (P2W18). We show, using NMR and MS, that for these polyoxotungstates hydrolysis equilibria for the different tungstate salts is key to explaining unexpected storage trends while the performance limit for {P5W30} and {P8W48}, can be attributed to unavoidable hydrogen generation, evidenced by GC. NMR spectroscopy, in combination with the MS analysis, provided experimental evidence for a cation/proton exchange process during the reduction/reoxidation process of {P5W30} which likely occurs due to this hydrogen generation. Our study offers a deeper understanding of the factors affecting the electron storage ability of POMs and provides insights allowing for further development of these materials for energy storage.

Elemental Analyses: Element analyses for K, Li, P and were performed on a Leeman inductivity-coupled plasma (ICP) spectrometer and the content of carbon, nitrogen and hydrogen were determined by the microanalysis services using an EA 1110 CHNS, CE-440 Elemental Analyzer.
Single Crystal X-Ray Diffraction: A suitable single crystal was selected and mounted onto a rubber loop using Fomblin oil. Single-crystal datasets and unit cells for compound Li-{P5W30} were collected at 150(2) K on a Rigaku XtaLAB Synergy R HyPix-Arc 150 diffractometer equipped with a graphite monochromator (λMo-Kα = 0.71073 Å) of a micro-focus sealed X-ray source (50 kV, 24.0 mA). Data collection and reduction were performed using the CrysAlisPro software package and structure solution and refinement were carried out with SHELXT-2018/3 and SHELXL-2018/3 via WinGX. 1 Most of the non-hydrogen atoms were anisotropically refined. Corrections for incident and diffracted beam absorption effects were applied using analytical numeric absorption correction 2 on multifaceted crystal models. CCDC  Gas Chromatography Analysis: Gas chromatography (GC) headspace analysis was performed using an Agilent Technologies 7890A GC system by optimised auto-sampling injection of gas from the headspace of the polyoxometalates holding tank into the GC. The column used was a 30 metre-long 0.320 mm widebore HP-molesieve column (Agilent). the carrier gas was Ar. The GC oven temperature was set to 27 °C and the front inlet was set to 100 °C.
After cooling, 9 g of KCl (0.12 mmol) was added to the pale-yellow solution and the solution was stirred for 30 min. Then, the pale-yellow solid was separated by centrifugation.
Recrystallization was carried by dissolving this solid in 30 mL of deionized water (100 °C, heated in an oil bath). Colourless block crystals were formed in next few days (normally within three days) and collected by filtration. Purer crystals were obtained by recrystallization one more time from 20 mL of deionized water at 100 °C (heated in an oil bath), and the crystals were collected by filtration. The structure of product was confirmed by single crystal XRD unit cell check on multiple crystals and purity was confirmed by 31 P NMR (-10.05 ppm in D2O, Figure S1). The crystal water in the structure was confirmed by TGA ( Figure S2 Figure S5). The structure was confirmed by single crystal XRD and 31 P NMR (-10.51 ppm in 1M D2SO4, Figure S6), and the composition and purity were checked by ICP and elemental analysis (Table S2). The crystal water was analysed by TGA ( Figure S7). Yield: 1.5 g. FTIR (ATR, 1300-500 cm-1): 1162 (sh), 1084 (sh), 1021 (sh), 986 (sh), 902 (sh), 717 (br) ( Figure S8). The structure of this sample was further confirmed by ESI-MS spectrum ( Figure S9 and Table S3)

Synthesis of K28Li5H7[P8W48O184]·92H2O (3)
The original synthetic procedure of synthesizing K28Li5H7[P8W48O184]·92H2O was reported in the literature 5 . However, we have found that the former method has two important obstacles, namely, yield and prolong time of crystallisation. Therefore, Cronin group has developed new synthetic protocol for the preparation of KLi-{P8W48} derivatives, which was followed for this work. As with the preparation of K12H2[α-P2W12O48]·24H2O, fresh starting material was used to ensure the optimum yield.
To 200 mL H2O in a 250 mL beaker, was added 6.0 mL CH3CO2H, followed by lithium acetate dihydrate (9.0 g, 88 mmol). After 5 minutes of vigorous stirring to allow the lithium acetate to dissolve, LiCl (4.24 g, 100 mmol) was then added, before again allowing 5 minutes to dissolve. for the rate of addition. Finally, the pH was altered to exactly 5.00 with glacial acetic acid.    [NaP5W30O110] 14cluster displays a round plate shape and has a symmetry of D5h point group ( Figure S13). A fivefold axis passes the Na atom and vertical to plate equatorial plane defined by the five P atoms. The lithium salt compound 2 Li14[NaP5W30O110].38H2O crystallises in monoclinic system C2/m space group. Half {P5Mo30} cluster was found in the asymmetric unit.

Crystallographic analysis of Li-{P5W30}
The cluster is completed with the other half cluster generated by a crystallographic mirror plane, which is also the horizontal (equatorial) plane symmetry (σh) of the cluster D5h point group.
The Na atom inside the cluster is disordered over two positions each with half occupancy. The

Electrochemical tests 4.1 Cyclic voltammetry
All the electrochemical data were collected by using a Bio-logic SP-150 potentiostat. A homemade electrochemical cell was used for cyclic voltammetry (CV) measurements. A glassy carbon electrode, carbon felt and Hg/HgSO4(sat. K2SO4) were used as the working electrode, counter electrode, and reference electrode respectively. All the solutions were degassed with Ar for at least half an hour to remove the oxygen before CVs were collected. The scan rate is 10 mV s -1 .

Flow-cell system for reduction/oxidation of {P5W30} and {P8W48}
Figure S14: Flow cell device used for the reduction and oxidation of {P5W30} and {P8W48}.
The two-parts flow cell system comprised a mixture of commercial and custom-made components. It was assembled in Figure S14, in similar way described in our previous paper. 6 POCO graphite plates (thickness = 3mm, channel width = 1 mm, channel depth = 1 mm, landing between channels = 1 mm) were purchased from balticFuelCells GmbH and were Such a device has a very low ohmic polarization resistance of around 50 to 100 mΩ.
Galvanostatic electrolysis method with a suitable current density was applied to reduce the polyoxometalate solution. The voltage plateaus in the potential verse charge curves during the reduction (and oxidation) processes correspond to the thermodynamic potentials for the different amount of electrons put into this system.

K-{P5W30}
The                 {P8W48} has the lowest utilization rate of W, only part of W in the structure are redox active and participate in the reduction/oxidation process. Figure S26: Equilibrium position of H3O + and Li + during reduction and reoxidation and how this affects NMR shift observed.

NMR and ESI-MS analysis of reduction/reoxidation
The performance limit of Li-{P5W30} was analyzed by NMR, MS and GC.  The reduced sample was taken out immediately after charging finish and protected with Ar to avoid oxidation by air, before injection it was also diluted with degassed methanol. The original and re-oxidized one was also after dilution with degassed methanol.

Concentration effect of Li-{P2W18} and Li-{P5W30} and LiNH4-{P8W48}
The various concentration of the non-reduced polyoxotungstate solution was investigated by 1 H, 7 Li and 31 P NMR study. All the reduced samples were taken out immediately after charging finish and protected with Ar to avoid oxidation by air. For D2SO4 ones, samples were directly taken and sealed with Ar for NMR tests. For H2SO4 ones, a few drops of D2O (Vsample solution: VD2O= 500 uL : 10 uL) were added for locking in NMR measurement, and then sealed with Ar.

ESI-MS analysis
The reduced sample was taken out immediately after charging finish and protected with Ar to avoid oxidation by air, before injection it was also diluted with degassed methanol. The original and re-oxidized one was also after dilution with degassed methanol. The peaks for original one has been analysed in supplementary Table 3. Through assignment of the peaks (Table S12 and S13), we found after reduction {P5W30} was protonated and after reoxidization, there are more Li + coordinated in the structure than in the original status (Table   S9).     injected into the mass spectrometer. The reduced sample was taken out immediately after charging finish and protected with Ar to avoid oxidation by air, before injection it was also diluted with degassed methanol. The original and re-oxidized one was also after dilution with degassed methanol.   injected into the mass spectrometer. The reduced sample was taken out immediately after charging finish and protected with Ar to avoid oxidation by air, before injection it was diluted with degassed methanol. The original and re-oxidized one was also after dilution with degassed methanol.