Coherent and Compact Zinc Electrodeposition Enabled by Compressing the Electric Double Layer of the Deposits

The porous hexagonal-platelet Zinc (Zn) deposits exacerbate the chemical corrosion and deteriorate the reversibility of the Zn electrodes in aqueous electrolytes. Based on the Derjaguin-Landau-Verwey-Overbeek (DLVO) theory, to turn the messy Zn deposits into agglomerate ones, the challenge is to weaken the electric double layer repulsive force, which is the main reason preventing the dense Zn deposits, between the electrodeposited Zn particles. Here, we proposed a strategy to compress the electric double layer and regulate the forces between the electrodeposited Zn particles by introducing inert charges to the surface of the Zn deposits. The results of the electron microscopies revealed dense and coherent electrodeposition of Zn, indicating that the van der Waals attraction between the deposits becomes governing during electrodeposition. Such results could be attributed to the adsorbed inert charges on Zn deposits decrease the net charges and weaken the electric double layer repulsive force. This design enables the Zn||Zn cells a long-term plating/stripping stability for > 1200 h, a high average Coulombic Efficiency of 99.9% for > 2100 h, and steady charge/discharge responses even under a draconian deep-discharge condition of 80% depth of discharge of Zn (DOD Zn ). In addition, the Zn||VS 2 full cells demonstrate significantly improved electrochemical reversibility

Despite the advantages of aqueous Zn-based batteries (e.g., Zn-MnO2, Zn-Br2, and Zn-Air batteries), including high safety, low cost, and nontoxicity, the sustained chemical corrosion and low reversibility of Zn electrodes encumber their practical applications. In aqueous electrolytes, the electrodeposition of hexagonal close-packed Zn metal has a strong propensity to form hexagonal platelets, which causes the deposited Zn a non-planar and flaky morphology [1][2][3] . Such a porous structure is bound to exacerbate the chemical corrosion during the repeated plating and stripping of the Zn-metal phase due to the increased exposure of Zn electrodes to the electrolytes. The loose Zn flakes also cause the loss of electrical contact between the deposits and substrates and further deteriorates the reversibility of Zn electrodes 4 .
What is worse, the dendritic Zn flakes may easily pierce the separator and lead to short circuits of batteries 5 .
To induce uniform Zn deposition with high reversibility, several strategies have been proposed: (1) constructing artificial interface layers to restrict the Zn crystal growth and isolate the Zn electrodes from aqueous electrolytes 6-8 ; (2) using substrates with a low lattice mismatch and low affinity to lock the crystal orientation for the uniform Zn electrodeposition 5, 9 ; (3) increasing the driving force for the nucleation of Zn deposits to induce the uniform distribution of Zn-metal nuclei 10 . However, most of the above approaches rely on the modification of the Zn electrodes or current collectors, which decrease the overall energy density of the cells. Besides, the long-term cycling of Zn electrodes under conditions of high depth of discharge (DOD) and/or high areal capacity of Zn remains challenging. It is highly desired to explore new solutions to enable high effective Zn deposition without sacrificing the energy density.
The morphology of the deposited Zn directly affects the reversibility of Zn electrodes and the lifespan of Zn-based batteries [11][12][13] . It is acknowledged that the morphology of the deposited Zn is related to multiple factors, including the intrinsic crystal anisotropy, electrolytes, substrate chemistry and geometry, and so on 5,10,[14][15][16] . The process of Zn electrodeposition includes the desolvation and reduction of the Zn 2+ ions, and the following formation and growth of the nucleus on conductive substrates. The final morphology of deposited metal is related to both the structure of as-formed grain crystals, which are always observed as irregular hexagonal flakes for Zn, and the interactions between them 17 . Based on the Derjaguin-Landau-Verwey-Overbeek (DLVO) theory, the interactions between Zn deposits in aqueous electrolytes are mainly related to the van der Waals (VDW) attractive force and the electrostatic repulsion due to the electric double layer (EDL) of counterions ( Figure S1 in Supporting Information) [18][19][20] . In ZnSO4 electrolyte, the electrodeposited Zn typically shows a loose and separate structure, indicating a repulsive-force-governed Zn deposition process. To induce dense and compact Zn coherent electrodeposition, we need to regulate the interactions between the Zn deposits from repulsion to attraction. As the VDW force, which depends mainly on the distance between the particles, could be considered as fixed for the Zn deposits, thus the solution is to weaken the EDL repulsion force between the Zn deposits.
Based on the Poisson-Boltzmann (PB) model, the EDL repulsive force between the negatively-charged surfaces of Zn deposits is mainly influenced by the thickness of the EDL, which is known as the Debye length (1/) 17 . Theoretically, by reducing the Debye length, the EDL repulsion force between two charged particles could be reduced. With this theory foundation, in this work, we introduced La 3+ ions, which serve as high-valence competitive ions to decrease the Debye length [21][22][23] , to the aqueous ZnSO4 electrolytes. The electrochemical and morphology characterizations confirmed the presence of the insert La 3+ ions weakens EDL repulsive force between the Zn deposits, changes the preferred orientation of Zn deposits, and results in dense and compact Zn coherent electrodeposition. With La 3+ -modified electrolyte, the corrosion rate of Zn electrodes is significantly relieved with the corrosion current decreased from 421.6 to 6.3 A cm −2 , enabling a high average Coulombic efficiency of > 99.9% for 2100 plating/stripping cycles. Even under an exacting condition with both a high current density (10 mA cm −2 ) and a limited Zn supply (DODZn = 80%), the Zn||Zn cell with the modified electrolyte exhibited a stable Zn deposition for ~ 160 h, whereas the control cell failed to work. The as-proposed strategy demonstrates the importance of the thickness of EDL on the electrodeposition behaviors of Zn 2+ ions and might also be applicable for other metal anodes. To figure out how the ZSL electrolyte affects the Zn electrodeposition, we disassembled the cycled Zn||Zn cells (100 cycles with a current density 1 mA cm −2 of and an areal capacity of 1 mAh cm −2 ) and analyzed the morphology of the cycled Zn electrodes by a scanning electron microscopy (SEM). As the results shown in Figures 1b, the cycled Zn electrode in ZS electrolyte shows a highly porous surface that is stacked by thin Zn flakes with sharp edges, which may pierce the separator and fail the cell 5 . In addition, the loose structure also leads to more side reactions between the Zn electrode and electrolyte due to the larger exposed surface, and thus more by-products accumulation. In comparison, after being cycled under a same test condition, the Zn electrode with ZSL electrolyte displays a dense surface with the deposited Zn flakes closely connected with each other (Figure 1c  The morphology analysis of Zn deposits. It is known that the Zn deposits in ZS electrolyte tend to exhibit a hexagonal-platelet morphology due to the lower thermodynamic free energy of the exposed (002) plane 5   The structure of the Zn deposits obtained from ZS and ZSL electrolytes was further characterized by grazing incidence X-ray diffraction (GIXRD), a powerful tool to investigate the texturing and orientation anisotropy of thin film. As the results shown in Figure 2j, the Zn deposits obtained in ZSL electrolyte display a relatively weaker (002) peak. Quantitatively, the relative intensity ratio of peak (002) (I(002)) to that of peak (100) (I(100)) decreases from 1.4 to 0.92, indicating that the reduced (002) planes for the Zn deposits in ZSL electrolyte.

High
Considering the hcp structure of Zn ( Figure 2k) and reduced (002) plane of the Zn deposits obtained from ZSL electrolyte, it is safe to conclude that the as-obtained Zn deposits are piled up along the c axis, which could be regarded as a coherent deposition 1, 28 . soon after the overpotential was applied, implying the immediate activation of all the nucleation sites and an instantaneous nucleation dominated process. In contrast, the current density for the cell with ZSL electrolyte is characterized by prolonged activation time, indicating that the number of nuclei increases gradually with time and the progressive nucleation is governing during Zn deposition in ZSL electrolyte 29,31 . It is also noticed that the steady current density in ZS electrolyte (~ −26 mA cm −2 ) is higher than that in ZSL electrolyte (~ −22 mA cm −2 ). The differences in the nucleation mechanisms and steady current densities could be ascribed to the absorption of the inert La 3+ ions on the surface of the Zn electrodes, which decrease the number of the active nucleation sites and slow down the formation of the nuclei in ZSL electrolyte.
The absorption of metal ions on an electrode is typically regarded as a monolayeradsorption process 33   charges than that in ZS electrolyte. In this context, the surface of Zn deposits presents a higher potential  ′ than  . The EDL repulsion decreases due to the fewer net charges, and the Stern layer gets thinner in a ZSL electrolyte than in a ZS electrolyte 35 . As illustrated in Figure   4c, in ZS electrolyte, the electrodeposited Zn tends to grow into separate hexagonal plates due to the EDL repulsion dominated interactions between the Zn deposits 36 . While when ZSL electrolyte is used, the competitive absorption of the inert La 3+ ions on the surface of the Zn electrodes reduces the EDL repulsion between the Zn deposits and leads to the coherent electrodeposition of the Zn deposits along (002) plane (Figure 4d).  Characterizations. The morphologies and structures of the samples were characterized by field emission scanning electron microscopy (SEM, Nova NanoSEM 450) equipped with energy-dispersive X-ray spectroscopy (EDS). X-ray diffraction (XRD) patterns were recorded with a Bruker-AXS microdiffractometer (D-8 ADVANCE) using Cu-K  radiation (λ = 1.5406 Å) from 10º to 90º. Grazing incidence X-ray diffraction (GIXRD) patterns were collected from 35 ~ 47º on a Rigaku SmartLab X-ray diffractometer with a Cu-K  radiation with a step size of 0.0001º. Bio-Logic). The Zeta potential was collected on a Zeta potential analyzer (Malvern Zetasizer Nano ZS90).