STABILIZING ZINC ELECTRODEPOSITION IN A BATTERY ANODE BY CONTROLLING CRYSTAL GROWTH A Thesis Presented to the Faculty of the Graduate School of Cornell University In Partial Fulfillment of the Requirements for the Degree of Master of Science by Shuo Jin August 2021 © 2021[Shuo Jin] ABSTRACT Reversible electrodeposition of metals at liquid-solid interfaces is a requirement for long cycle life in rechargeable batteries that utilize metals as anodes. The process has been studied extensively from the perspective of the electrochemical transformations that impact reversibility, however the fundamental challenges associated with maintaining morphological control when a intrinsically crystalline solid metal phase emerges from an electrolyte solution have been less studied, but provide important opportunities for progress. Here we propose a crystal growth stabilization method to reshape the initial growth and orientation of crystalline metal electrodeposits. The method takes advantage of polymer-salt complexes (PEG-Zn2+-aX-) (a=1,2,3) formed spontaneously in aqueous electrolytes containing zinc (Zn2+) and halide (X-) ions to regulate electro-crystallization of Zn. It is shown that when X = I, the complexes facilitate electrodeposition of Zn in a hexagonal closest packed (HCP) morphology with preferential orientation of the (002) plane parallel to the electrode surface. This facilitates exceptional morphological control of Zn electrodeposition at planar substrates and leads to high anode reversibility and unprecedented cycle life. Preliminary studies of the practical benefits of the approach are demonstrated in Zn-I2 full battery cells, designed in both coin cell and single-flow battery cell configurations. In both contexts, we find that control of the Zn crystallography enables batteries with long-term cycling stability at high areal capacity. The crystal growth stabilization method therefore provides an exciting pathway toward low-cost and large-scale storage of electrical energy. i BIOGRAPHICAL SKETCH Shuo Jin was born in China, on May 2, 1996. He received bachelor’s degree in Chemical Engineering from the South China University of Technology, Guangzhou, followed by his Master of Science at Chemical Engineering, Cornell University. At Cornell University, he carried out independent research in the exciting field of developing next-generation aqueous zinc batteries. ii ACKNOWLEDGMENTS I would like to thank my advisor Professor Lynden Archer for helping me so much, providing me useful insights, and motivating me throughout my research. It’s my great advisor who helps me strengthen my confidence in scientific research. I would like to extend my thanks to Dr. Jingxu zheng, Dr. Qing Zhao, and the rest of Archer group for leading me into this exciting research area, and always giving me inspired suggestions. I would also like to thank my friends and also cooperators, Peng-Yu Chen, Yiqi Shao, Xiaosi Gao, and Jasper Qiu, for helping me explore the wonderful research world. Finally, I would like to express my gratitude to my parents for their endless love and support and for the opportunity to carry out my graduate studies at Cornell University. iii TABLE OF CONTENTS Abstract i Biographical Sketch ii Acknowledgements iii Table of Contents iv Chapter 1: Introduction Metal electrodeposition: an overview 1 Crystal growth control by halide ions 3 Zinc electrodeposition system 4 Chapter 2: Zinc crystal growth control process Zinc crystal growth control results 11 The electrochemical properties of different Zn salt electrolytes 14 The bonding effect between Zn ions and PEG300 18 The Zinc crystal growth control process 21 Chapter 3: Zinc anode and full battery The reversible electrodeposition of Zn anode 35 Aqueous Zinc full battery 41 Chapter 4: Conclusion iv CHAPTER 1 INSTRUCTION Metal electrodeposition: an overview Electrodeposition of metals has been studied extensively from an electrochemistry perspective. The fundamental simplicity of the electrode reaction (Mz+ + ze- M) belies the underlying complexity of the ion de-solvation and crystallization processes that ultimately yields a solid metal phase, from electroreduction of solvated ions in a liquid electrolyte medium1. This aspect of electrodeposition is always important but is not emphasized in conventional studies of the process as a scalable method for manufacturing thin coatings of metals2, colloids3 and polymer4 on electronically conductive substrates. The strong revival of interest in rechargeable batteries that employ metals as anodes, including lithium (Li), sodium (Na), potassium (K), magnesium (Mg), calcium (Ca), zinc (Zn), and aluminum (Al)5-8, which are all crystalline solid materials with unique crystal structures and well-known to electrodeposit in morphologies that often have no relationship to the orientation or structure of the substrate, underscores the need for improved understanding and regulation of crystal growth during metal electrodeposition processes. A body of work already exists that shows that regulation of crystal structure can play an important role in the metal electrodeposition morphology. Among these studies, the 1 recent contribution by Zheng et al.,6 stand out because it shows that aligned graphene or Au interfacial layers strongly influence the crystal orientation and coating morphology formed by electrodeposited Zn and Au, respectively. Significantly, in the case of the graphene-Zn system, there is evidence that a textured graphene layer can template Zn crystal growth well into the electrodeposited metal bulk, yielding Zn electrodes with unprecedented, high levels of reversibility in simple ZnSO4 aqueous electrolytes. The present study is also inspired by recent works showing that primarily (110) textured Li electrodeposits are formed in some electrolytes9 (e.g., 1 M LiTFSI + 1% LiNO3 in DOL/ DME and 5M S8 +1M LiTFSI+1% LiNO3 in DOL/DME, etc), but not in others. These findings suggest that unregulated crystal growth processes at metal anodes must be understood and managed in order to achieve the high levels of morphological control needed for electrochemical cells based on metal anodes to achieve levels of long-term performance relevant for practical applications. Consequences of poor control of metal electrodeposit morphology at a battery anode has been reviewed extensively during the last decade,10,11 and it is now known that failure to control morphology at a metal anode during charging leads to a long list of problems, all of which leads to poor reversibility and premature failure of the battery anode. The list includes: (1) concentration of electric field lines at irregular deposit features (e.g. bumps, edges, etc.) to drive nonplanar electrodeposit growth in the form of diffusion-limited, dendritic structures12,13 which may bridge the inter-electrode space and cause short-circuit the battery; (2) uneven metal deposition during the charge cycle leads to uneven electrowinning during the discharge, producing undesirable effects such as orphaning, when the metal deposit breaks away from the 2 substrate and therefore becomes electrochemically inaccessible.14-16 This process has been shown by means of operando optical visualization studies in sodium batteries to drive large decreases in Coulombic efficiency in battery cycling.17 Crystal growth control by halide ions The use of salt and molecular additives for controlling growth of desired crystalline facets and shape of crystalline nanomaterials is a well-developed concept in materials synthesis.18-21 For example, salts, surfactants and polymers are routinely used in hydrothermal synthesis to induce and/or block certain crystal planes to reform the crystal growth structure18. Selective growth of (111) oriented triangular/truncated triangular particles of Au, Pt, Ag, etc. have been shown to be enabled by halide ions in solution.18,19 The ions are believed to regulate growth by selectively adsorbing on some facets20, such as at the (111) plane of Au, Pt and (002) plane of Zn. We hypothesize that similar approaches could be used to regulate metal crystallization and for controlling morphology of metal electrodes in electrochemical cells. Specifically, we propose that integration of inhibitors with fast adsorption kinetics to specific crystal facets can be used to selective regulate crystal growth. If crystal growth on planes parallel to the electrode substrate are selectively slowed, for instance, (110) plane of Li and Cu, (002) plane of Zn and Ti, and (111) plane of Au and Ag, it should be possible to drive anisotropic, planar crystal growth in two dimensions (2D) at a metal electrode.21 On the basis of the findings in reference 6, we hypothesize further that if the approach allows both reshaping of undesired crystal growth structures and 3 regulation of the crystal growth orientation, it will not only lead to better morphological control, but should enable higher levels of reversibility of the plating and stripping processes during charge and discharge of the electrode. Zinc electrodeposition system To evaluate our hypotheses, we use Zn in the aqueous electrolyte as a model metallic anode system for in-depth study. There are a variety of consideration for selecting Zn for the study; the most important include: (i) Zn anodes in neutral aqueous electrolytes do not form complicated solid-electrolyte interphase (SEI) layers.6,12,21,22 The presence of such layers would complicate adsorption processes at the electrode and we believe reduce the fidelity with which polymer, salt and other known crystal growth- regulation agents are able to regulate crystal growth process at a Zn anode. (ii) Our prior success regulating Zn crystallization in the Zn-graphene system6 confirm the importance of consideration (i). We note, however, that because the crystal growth regulation is produced by additives in the electrolyte bulk, unlike the Zn-graphene system which relies on homoepitaxial Zn growth to preserve crystal structures formed at the Zn/graphene interface, the method is simultaneously simpler, easier to scale-up, and is in principle suitable for much higher electrode capacities than possible with any epitaxial regulation strategy. (iii) Aqueous rechargeable Zn-ion batteries (ZIB) in which liquid water is the electrolyte solvent offers a large number of practical advantages in terms of low-cost, high safety, ease of scaling-up results from small- 4 scale coin-cell studies to larger format batteries, and rich Zn resources in regions all over the world. Taken together these benefits position ZIBs as a plausible alternative energy storage platform to Lithium-ion batteries, particularly in situations where very large-scale and low-cost electrical energy storage is needed23-25. (iv) Zinc’s elastic modulus is approximately twenty-times higher than that of Li. It means that the consequences of poor morphological (e.g., non-planar deposition, metal orphaning, and short-circuits) are likely to be substantially greater than the corresponding Li case. This underscores the need for easy proliferation of Zn branch dendrite stresses the importance to find an effective method to stabilize the Zn deposition.26,27 To obtain smooth and compact zinc electrodeposition, we propose a crystal growth stabilization method that inhibits growth of the (002) plane of the hexagonal closest packed (HCP) Zn. This requires a strong inhibitor that shows highly selective preferential absorption at (002) plane, long-lasting activity, and possibility for resupply from the electrolyte bulk. As already noted, this approach offers multiple advantages over the already powerful Zn epitaxial crystal growth strategy reported in our earlier study.6 The most important in the present situation are the versatility in interrogating additives with widely varying adsorption characteristics and the ability to regulate the crystal growth from the electrolyte bulk. Meaning that crystal growth structure and orientation can, at least in principle, be regulated in each layer, which makes the method more easily applicable in practical batteries with high areal capacity of Zn anode. This feature is in reality far more important in ZIBs than in studies of lithium anode batteries because the low discharge voltage of ZIB compared to lithium 5 batteries28, makes low capacity Zn anodes less competitive from a practical point of view. 6 REFERENCES 1 Liang, Y., Dong, H., Aurbach, D. & Yao, Y. Current status and future directions of multivalent metal-ion batteries. Nature Energy, 1-11 (2020). 2 Lawless, K. R. Growth and Structure of Electrodeposited Thin Metal Films. Journal of Vacuum Science and Technology 2, 24-34, doi:10.1116/1.1492395 (1965). 3 Sun, F. et al. Morphology Control and Transferability of Ordered Through- Pore Arrays Based on the Electrodeposition of a Colloidal Monolayer. Advanced Materials 16, 1116-1121, doi:10.1002/adma.200400006 (2004). 4 Gleason, K. K. Nanoscale control by chemically vapour-deposited polymers. Nature Reviews Physics 2, 347-364, doi:10.1038/s42254-020-0192-6 (2020). 5 Lin, D., Liu, Y. & Cui, Y. Reviving the lithium metal anode for high-energy batteries. Nat Nanotechnol 12, 194-206 (2017). 6 Zheng, J. et al. Reversible epitaxial electrodeposition of metals in battery anodes. Science 366, 645-648 (2019). 7 Lin, M. C. et al. An ultrafast rechargeable aluminium-ion battery. Nature 520, 325-328 (2015). 8 Deng, Y. et al. On the Reversibility and Fragility of Sodium Metal Electrodes. Advanced Energy Materials 9, 1901651 (2019). 9 Shi, F. et al. Strong texturing of lithium metal in batteries. Proc Natl Acad Sci U S A 114, 12138-12143 (2017). 7 10 Zheng, J. et al. Regulating electrodeposition morphology of lithium: towards commercially relevant secondary Li metal batteries. Chem Soc Rev 49, 2701- 2750, doi:10.1039/c9cs00883g (2020). 11 Liu, W., Liu, P. & Mitlin, D. Tutorial review on structure - dendrite growth relations in metal battery anode supports. Chem Soc Rev 49, 7284-7300, doi:10.1039/d0cs00867b (2020). 12 Tikekar, M. D., Choudhury, S., Tu, Z. & Archer, L. A. Design principles for electrolytes and interfaces for stable lithium-metal batteries. Nature Energy 1, 1-7 (2016). 13 Sawada, Y., Dougherty, A. & Gollub, J. P. Dendritic and fractal patterns in electrolytic metal deposits. Phys Rev Lett 56, 1260-1263 (1986). 14 Zheng, J. et al. Physical Orphaning versus Chemical Instability: Is Dendritic Electrodeposition of Li Fatal? ACS Energy Letters 4, 1349-1355 (2019). 15 Fang, C. et al. Quantifying inactive lithium in lithium metal batteries. Nature 572, 511-515 (2019). 16 Higashi, S., Lee, S. W., Lee, J. S., Takechi, K. & Cui, Y. Avoiding short circuits from zinc metal dendrites in anode by backside-plating configuration. Nat Commun 7, 1-6 (2016). 17 Deng, Y. et al. On the Reversibility and Fragility of Sodium Metal Electrodes. Advanced Energy Materials 9, doi:10.1002/aenm.201901651 (2019). 18 Shankar, S. S., Bhargava, S. & Sastry, M. Synthesis of gold nanospheres and nanotriangles by the Turkevich approach. J Nanosci Nanotechnol 5, 1721-1727 (2005). 8 19 Wiley, B., Herricks, T., Sun, Y. & Xia, Y. Polyol Synthesis of Silver Nanoparticles: Use of Chloride and Oxygen to Promote the Formation of Single-Crystal, Truncated Cubes and Tetrahedrons. Nano Lett 4, 1733-1739 (2004). 20 Magnussen, O. M., Ocko, B. M., Adzic, R. R. & Wang, J. X. X-ray diffraction studies of ordered chloride and bromide monolayers at the Au(111)-solution interface. Phys Rev B Condens Matter 51, 5510-5513 (1995). 21 Zheng, J. et al. Spontaneous and field-induced crystallographic reorientation of metal electrodeposits at battery anodes. Science Advances 6, eabb1122 (2020). 22 Zheng, J. & Archer, L. A. Controlling electrochemical growth of metallic zinc electrodes: Toward affordable rechargeable energy storage systems. Science Advances 7 (2021). 23 Zhong, C. et al. Decoupling electrolytes towards stable and high-energy rechargeable aqueous zinc–manganese dioxide batteries. Nature Energy 5, 440-449 (2020). 24 Choi, J. W. & Aurbach, D. Promise and reality of post-lithium-ion batteries with high energy densities. Nature Reviews Materials 1, 1-16 (2016). 25 Wang, F. et al. Highly reversible zinc metal anode for aqueous batteries. Nat Mater 17, 543-549 (2018). 26 Biton, M., Tariq, F., Yufit, V., Chen, Z. & Brandon, N. Integrating multi- length scale high resolution 3D imaging and modelling in the characterisation and identification of mechanical failure sites in electrochemical dendrites. Acta Materialia 141, 39-46 (2017). 9 27 Khor, A. et al. Review of zinc-based hybrid flow batteries: From fundamentals to applications. Materials Today Energy 8, 80-108 (2018). 28 Albertus, P., Babinec, S., Litzelman, S. & Newman, A. Status and challenges in enabling the lithium metal electrode for high-energy and low-cost rechargeable batteries. Nature Energy 3, 16-21 (2018). 10 CHAPTER 2 ZINC CRYSTAL GROWTH CONTROL PROCESS Zinc crystal growth control results Figure 1. Electrochemical crystal growth of Zn in different electrolytes. Scheme illustrating the Zn crystal growth processes hypothesized (A) in ZnSO4 electrolytes and (B) in ZnX2 + oligomeric polyethylene glycol (PEG 300 g/mol PEG300) electrolytes. Here X = Cl, Br, I is a halogen species. (C) (G) Scanning electron microscopy (SEM) is used to reveal hexagonal platelet structures formed by electroreduction of Zn from 2M ZnSO4 and 2M ZnSO4 + 5% PEG300 electrolytes. (D) (E) (F) Sphere-like structure formed in 2M ZnCl2, ZnBr2 and ZnI2 electrolyte, respectively revealed by SEM analysis. (H) Sphere-like structure formed in 2M 11 ZnCl2+5 wt% PEG300 electrolyte. (I) Oval-like structure formed in 2M ZnBr2+5 wt% PEG300 electrolyte revealed by SEM analysis (J) SEM analysis reveals that large hexagonal platelets are formed by electroreduction of Zn from 2M ZnI2+5 wt% PEG300 electrolytes. Figure 2. SEM image of Zn deposits in (A) 2M ZnI2 + 5 wt%PEG300 electrolytes, (B) 2M ZnI2 electrolyte. (6 min, J=4 mA cm-2) Figure 3. SEM image of Zn deposits in (A) 2M ZnI2 + 5 wt%PEG300 electrolyte, (B) 2M ZnI2 electrolyte. (12 min, J=4 mA cm-2) 12 Figure 1 reports the effect of salts and polymer additives on the orientation and morphology of Zn at planar electrodes. The results suggest that halide ions introduce opposite influences in hydrothermal synthesis18 and electrodeposition of Zn (Fig. 1C~F). In hydrothermal synthesis, chemisorption of halide ions has been reported to lead to highly tunable and oriented crystal growth, while they are seen cause rough and unregulated crystal growth in Zn electrodeposition (Fig. 2~3). The scanning electron microscope (SEM) images show that, hexagonal platelet electrodeposits form in ZnSO4 electrolytes (Fig. 1C), whereas a thin scaffold consisting of inner unit sphere-like features form in aqueous ZnX2 (X = Cl, Br, I) electrolytes (Fig. 1D~F). In hydrothermal synthesis chloride ions are reported to be highly effective in inhibiting growth at crystal planed where strong and specific chemisorption occurs29. Based on the Wulff plot for Zn10, we expect this adsorption to be preferentially at the (002) facets for Zn. The poor ability of any of the Zn salts to promote ordered crystal growth raises obvious questions about the strength and longevity of the halide ion adsorption at the potentials where deposition occur. 13 The electrochemical properties of different Zn salt electrolytes Figure 4. Electrochemical characteristics of ZnSO4 and ZnX2 (X=Cl, Br, I) aqueous electrolytes. Electrochemical impedance spectroscopy (EIS) analysis of (A) 2M ZnSO4 and various ZnX2 aqueous electrolytes. (B) Same as (A) except containing a fixed concentration (5 wt%) PEG300. (C)(D) EIS analysis of 2M ZnSO4 and ZnI2 with/without (5 wt%) PEG300. (E) Interfacial resistance (Rint) calculating by fitting the data in (A)(B). (F) Ion conductivity calculating from (A). Linear sweep voltammetry of current-voltage curve of ZnSO4 and various ZnX2 aqueous electrolytes (G)0.05M, (H) 0.5M. The plateau or turning points indicate the value of the limiting current. (I) Exchange current (i0) of 2M ZnSO4 and ZnX2 aqueous electrolytes 14 calculated from Tafel plots. (J) Current-Time curve: 20mV polarization experiments of 2M ZnSO4 and ZnX2 aqueous electrolytes for 5 hours. To diagnose the underlying problems of Zinc halide electrolytes, we performed electrochemical impedance spectroscopy (EIS) analysis to characterize the interfacial properties of these electrolytes on the Zn plate surface. As shown in Fig. 4A, the Nyquist plots of ZnSO4 and ZnCl2 show a single semicircle shape, whereas a mass transport region appears in ZnBr2 and ZnI2 electrolyte. Significantly, compared with the interfacial resistance of ZnSO4 and ZnCl2 electrolytes (158Ω and 161Ω), ZnBr2 and ZnI2 exhibit remarkably small interfacial resistance, 1.1Ω and 1.8Ω, respectively. Considering the minor change in ion conductivity (Fig. 4F), we hypothesize that the lowest interfacial resistance of ZnBr2 and ZnI2 results from different ion-solvation environment and the strong interfacial ion complexes adsorption on the Zn plate surface. According to previous reports, in the three Zn halide electrolytes studied here, different initial complexes dominate the ionic states due to the electronegativity difference. In ZnCl2, stable hexagonal six-coordination complex30 [ZnCl4(H 2-2O)2] predominate the complexes states. In ZnBr 312 and ZnI 322 electrolyte, in addition to the stable complexes [ZnX 2-4] (X=Br, I), unstable complexes [ZnX ]- 3 , [ZnX2] and [ZnX]+ (X=Br, I) are also present in large proportions. In ZnSO4 electrolytes, Zn solvation ions, [Zn(H2O) ]2-6 separately exist with SO 2-4 . Taken together all these works, the minimal interfacial resistance in ZnBr2 and ZnI2 electrolytes are therefore concluded to be consequences of the interfacial specific adsorption of the unstable complexes [ZnX ]-3 , [ZnX2] and [ZnX]+ (X=Br, I). 15 To confirm the interfacial specific ion complexes adsorption process, limiting current, which can well reflect the ion transportation properties, at different concentrations was obtained. At low concentration (0.05M) of these four Zn salts electrolytes(Fig. 4G), the limiting currents are close to each other due to the concentration of the unstable ion complexes is low, meaning the interfacial ion complexes adsorption is not strong enough; at higher concentration electrolytes (0.5M) (Fig. 4H), owning to the concentration of unstable ion complexes highly increases and varies in each electrolyte, ZnI2 electrolyte shows the highest limiting current and an obvious increasing trend was found from ZnSO4 to ZnI2 electrolytes. At high concentration electrolytes (1M), Zn halide salts show higher limiting current than ZnSO4 electrolytes, proving that the interfacial unstable ion-complexes adsorption should be the main reason results in the low interfacial resistance. Exchange current is another powerful parameter to confirm the interfacial unstable ion complexes adsorption. It is shown in Fig. 4I that the exchange current of Zn halide salts (i.e.,100 mA cm-2) is one order of magnitude higher than that of ZnSO4 electrolyte (i.e.,10-1 mA cm-2), indicating that these unstable ion complexes have much lower de-solvation energy than Zn solvation ion [Zn(H 2-2O)6] . Polarization experiments are also in accordance with the exchange current results. 20mV polarization voltage was used to polarize the Zn symmetric cell for 5 hours. For ZnSO4 and ZnCl2 electrolytes (Fig. 4J), the current density is at the microampere level, and the transfer number can be obtained, 0.31 and 0.275, respectively. However, for ZnBr2 and ZnI2 electrolytes, Zn electrodeposition occurs at such low polarization voltage, as we clearly observed: (1) the current density 16 is at the milliampere level; (2) as Fig. 4J inset shows, the sand’s time points were observed; (3) for ZnI2 electrolytes, the obvious breakpoint indicates the short circuit of the symmetric cell. These results confirm that the unstable ion complex adsorption highly influence the interfacial properties, further affect the crystal growth structure and electrodeposition morphology. Selective crystal growth mentioned above shows that strong interfacial halide adsorption will lead to the various growth rates of different crystal planes. For the halide ion preferred adsorption plane, the growth rate will be greatly affected and then lead to unexpected crystal structure. Here, SEM images shown in Fig. 1 reveal that the adsorbed ion complex did distort the crystal structure from hexagonal platelet to sphere-like structure and lead to unstable reversible electrodeposition, which should result from the specific unstable ion complexes adsorption to certain crystal plane. However, it is known in the literature that multi-valent halide salts form complexes with ether polymers in aqueous solutions33-36. We hypothesized that by tuning the molecular weight of the polymer and composition of these complexes, it should be possible to produce long-lived adsorption and, via the halide ion component, target the adsorption to the (002) Zn crystal facets. 17 The bonding effect between Zn ions and PEG300 Figure 5. Characteristics of ZnSO4 and ZnX2 (X=Cl, Br, I) aqueous electrolytes containing oligomeric polyethylene glycol (PEG 300 g/mol) as a soluble additive. 1H Nuclear Magnetic Resonance (NMR) analysis of (A) Various 2M ZnX2 aqueous electrolytes containing a fixed concentration (5 wt%) PEG300. (B) Same as (A) except the concentration of ZnI2 in the electrolytes is varied while that of PEG300 is kept fixed at 5 wt%. (C) Ion conductivity of 2M ZnI2 and ZnI2+ PEG300 electrolytes. (D) 18 Cyclic voltammetry (CV) for the first 50 cycles in a 2M ZnI2 and ZnI2+5 wt% PEG300 electrolyte recorded at a fixed scan rate of 10mV/s. To evaluate this concept, a screening framework was developed to evaluate promising materials as follows: (1) the material should bond with unstable ion complexes [ZnX3]-, [ZnX +2] and [ZnX] , enabling fast and preferential adsorption to the (002) plane of Zn37; (2) the adsorbed material should selectively inhibit growth at the (0002) crystal plane by regulating the ion flux38,39; (3) the material should be nonconductive and charge-free; (4) it should also easily dissolve in the aqueous electrolyte and have limited influence on ion transport features and electrochemical stability of the electrolyte. And, finally, considering the ultimate application value, the material should also be (5), inexpensive. Within this screening framework, Polyethylene glycol with a molecular weight of 300 g/mol (PEG300) was singled out as a promising candidate material that meets all the specified criteria 37,40,41. Nuclear Magnetic Resonance (NMR) was used to first understand the initial complexes formed in aqueous electrolytes containing Zn salts with PEG300. It is known that the de-shielding effect results in the right shift of 1H peak in NMR spectra, therefore, the 1H peak in the PEG backbone reflects the complexes stability of PEG300, Zn2+, and halide ions. Compared with the PEG 1H peak in 5%PEG300 + 2M ZnSO4 electrolyte, 5% PEG300 + 2M ZnI2 shows the strongest de-shielding effect (Shift to the far left), meaning that the PEG readily forms complexes PEG300-Zn2+-aI- (a=1,2,3) with the unstable initial complexes [ZnI - 3] , [ZnI2] and [ZnI]+. Additionally, 19 in Fig. 5B, 1H NMR spectra also show that the de-shielding effect is enhanced as the concentration of ZnI2 increases, further supporting the idea that a greater [ZnI -3] enhances formation of the PEG-Zn2+-aI- (a=1,2,3) complex. EIS analysis (Fig. 4C, D, E) shows that the interfacial resistance greatly increases in ZnSO4 electrolyte, while only minor increases in Zn halide electrolyte, indicating that unlike pure oligomer adsorption process in ZnSO4 electrolyte, the ion-oligomer complex adsorption happens in Zn halide + PEG300 electrolytes, and is as strong as pure ion complexes adsorption. The decrease of exchange current with PEG300 additives (Fig. 4I), also revealed that the ion-oligomer complexes are more stable than pure ion complexes but still possess lower de-solvation energy than Zn solvation ion [Zn(H2O) ]2-6 . On this basis, we firstly choose 2M ZnI2+5 wt% PEG300 electrolytes as the most promising complex for in-dept studies. Conductivity measurements (Fig. 5C) indicate that a high concentration of polymer complexes in solution has a negative impact on bulk ion diffusion. Cyclic voltammetry (CV) results also demonstrate that 5 wt% PEG300 has no negative effect on the Zn stripping/plating efficiency and reaction kinetics (Fig. 5D). In addition, this low concentration of short-chain PEG has a negligible influence on the electrolyte viscosity (e.g., the viscosity of the electrolyte with and without 5 wt% PEG300 was measured to be 3.202 mPas and 2.819 mPas using a ViscoQC 300). 20 The Zinc crystal growth control process Figure 6: Electrochemical crystal growth structure and orientation of electro- reduced Zn in 2M ZnI2 and ZnI2+5 wt% PEG300 aqueous electrolytes. (A) X-ray diffraction (XRD) analysis of Zn deposits after 24 min growth under galvanostatic conditions at a rate of 4 mA cm-2. (B) The I002:I101 peak ratio deduced from XRD measurements for Zn deposit after 6 min, 12 min, and 24 min growth, again at a fixed rate of 4 mA cm-2. (C) (D) Two-dimensional grazing incident XRD profiles for Zn deposits formed after 24 min growth in 2M ZnI2 and ZnI2+5 wt% PEG300 electrolyte, respectively, at 4 mA cm-2. Crystal growth structure in ZnI2+5 wt% PEG300 electrolyte for deposition at (E) 40 s, (I) 2 min. The corresponding Energy dispersive spectroscope (EDS) of Zn (F) (J), carbon (G) (K) and Iodine (H) (I). 21 Figure 7. Large-scale high contrast nucleation SEM images of Zn deposits in (A) (C) 2M ZnI2 and (B) (D) ZnI2 + 5 wt.%PEG300 electrolyte. (A) (B)1 min, J=4 mA cm-2, (C)(D) 1.5 min, J=4 mA cm-2. All images are in the same scale. 22 Figure 8. SEM image of Zn deposits in 2M ZnI2 electrolyte. (A) 2 min, J= 4 mA cm-2. The corresponding Energy dispersive spectroscope (EDS) of Iodine (B) without DI water washing (C) after DI water washing for 30s. 23 Figure 9. XPS characterization of the Zn foil after 50 cycles. (A) 2M ZnI2. (B) 2M ZnI2 + 5 wt% PEG300 electrolytes. Figure 10. QCM characterization of PEG300 adsorption performance. (A)PEG300 adsorption/dissipation frequency and three harmonics. (B) PEG adsorption thickness, about 150nm. Quartz Crystal Microbalance with Dissipation Monitoring (QCM-D): QCM-D is an extremely sensitive mass balance that is capable of detecting mass changes in the nanogram range. It utilizes a piezoelectric crystal that oscillates at a fixed frequency on application of voltage. When mass is added/removed from the surface, it effects the frequency and damping of the crystal oscillation (dissipation factor). A decrease in frequency indicates addition of mass to the surface while a high (low) dissipation value indicates a soft (rigid) film. Additionally, these values at different harmonics can be modeled to obtain viscoelastic properties of the adsorbed layer. 24 Fig. 6 reports the main result of the present study. The Zn crystal growth morphology in ZnI2 aqueous electrolyte, with/without PEG300 is investigated using Scanning electron microscopy (SEM), and planar stainless steel is the deposition substrate. As already noted, in pure ZnI2 electrolyte, the Zn crystal growth shows the sphere-like structure, which grows larger as deposition time increases, ultimately producing unregulated morphologies loosely termed Zn dendrites16 , which are easily orphaned upon cycling 14, as shown in Fig.2~3. In stark contrast to the Zn deposited in the ZnI2 electrolyte, 2D hexagonal platelet structures are observed 5wt% PEG300 in ZnI2 system (Fig. 1J). These results are thought to reflect the adsorption control provided by PEG300, which is consistent with our expected crystal growth pattern (Fig. 1B). To develop a more in-depth understanding of the crystal growth process, crystal structures at different nucleation time are shown in Fig. 6E, I. The results indicate that the initial crystal growth is consistent with the hypothesis (Fig. 1B) and that preferential growth is favored along the 2D substrate plane direction and preferentially expose the (002) plane of Zn, which is confirmed by the nucleation density result (Fig.7). To better evaluate changes in the nucleation density induced by the electrolyte additives, we select large scale (1.2mm) high-contrast images (Fig.7) at different nucleation time points: 1 minute and 1.5 minutes, J=4 mA cm-2. The much higher nucleation density observed in the ZnI2+ 5 wt% PEG300 electrolyte system indicates that the nucleation rate along the 2D direction is much faster than in ZnI2 electrolyte, which is again consistent with the idea that Zn crystals prefer to expose the (002) plane because of the slow growth rate of the (002) plane42. 25 Combining the results of XRD and EDS analysis show that PEG300 is preferentially adsorbed to the (002) plane due to the preferential chemisorption of iodide ion (Fig. 6H, L). At different nucleation time points, the elements of iodine and carbon (carbon represents the PEG300) are observed to be uniformly distributed at high density on the exposed plane parallel to the substrate (Fig. 6G, K). We note that all samples used for these studies were thoroughly washed by Deionized (DI) water for 30 seconds to remove the residual salt; the fact that the iodine and carbon signal are still obvious, signifying that the preferential adsorption process is very stable. This can be contrasted with the case where the PEG300 is not used (see Fig.8), where washing lowers the iodine count by 50%. X-ray photoelectron spectroscopy (XPS) and Quarts Crystal Microbalance (QCM) were used to provide more insight into the adsorption kinetics and to assess the role of PEG300. Zn symmetric cell cycles in 2 M ZnI2 and ZnI2 + 5 wt% PEG300 electrolytes for 50 cycles (25 mA cm-2, 3.2 mAh cm-2). After cycling and washing with DI water, XPS analysis of the Zn foil shows obvious Zn 2p and I 3d peaks in the ZnI2 electrolyte (Fig.9), whereas there are no Zn 2p and I 3d peak in the ZnI2+5 wt% PEG300 electrolyte (Fig.9), revealing that the adsorbed PEG300 covers the Zn foil and likely exceeds the sensitive thickness (10nm) of XPS. This latter conclusion is supported by QCM adsorption results. Specifically, the QCM results (Fig.10A) show that on flowing ZnI2+5 wt% PEG300 electrolyte, both the frequency and dissipation values manifest significant shifts (Fig.10A). Furthermore, we note that the shift in different harmonics do not overlap and the dissipation values are high (>1x10-6), indicating that the polymer layer adsorbed on the surface is 26 viscoelastic in nature and cannot be assumed to be a rigid layer. We employed the Voigt viscoelastic model to analyze the 3rd, 5th and 7th harmonics, revealing that the adsorbed layer has a thickness of approximately 150 nm (Fig.10B). The QCM results also reveal that the adsorption is reversible, both the frequency and dissipation values plummet to zero when the flowing fluid is switched to buffer solution (2M ZnI2), not containing polymer. To more thoroughly investigate the influence of the electrolyte additives on late-stage Zn deposition morphology, we increase the deposition time. As shown in Fig.2~3, the 2D crystal growth pattern shows a much smoother, compacter deposition morphology in ZnI2+5 wt% PEG300 electrolyte compared to the uneven deposition morphology in ZnI2 electrolyte, which can further be explained by x-ray diffraction (XRD). The peak intensity ratio between I002 and I101 of the XRD pattern reflects the crystal texturing behavior21. A larger I002: I101 means deposits are more (002) texturing, which is parallel to the substrate, meaning smooth and compact deposition. As shown in Fig. 6A~D, the I002: I101 is larger and more stable in ZnI2+5 wt% PEG300 electrolyte than in ZnI2 electrolyte at different deposition time points, meaning the Zn electrodeposition morphology are highly (002) plane orientation under the influence of adsorption PEG300. Taken together our observations therefore suggest that we can successfully reshape the crystal structure and further regulate the crystal electrodeposition orientation. The crystal structure has a decisive influence in the deposition morphology, which will greatly affect the cycle life of Zn anode. 27 REFERENCES 1 Liang, Y., Dong, H., Aurbach, D. & Yao, Y. Current status and future directions of multivalent metal-ion batteries. Nature Energy, 1-11 (2020). 2 Lawless, K. R. Growth and Structure of Electrodeposited Thin Metal Films. Journal of Vacuum Science and Technology 2, 24-34, doi:10.1116/1.1492395 (1965). 3 Sun, F. et al. Morphology Control and Transferability of Ordered Through- Pore Arrays Based on the Electrodeposition of a Colloidal Monolayer. Advanced Materials 16, 1116-1121, doi:10.1002/adma.200400006 (2004). 4 Gleason, K. K. Nanoscale control by chemically vapour-deposited polymers. Nature Reviews Physics 2, 347-364, doi:10.1038/s42254-020-0192-6 (2020). 5 Lin, D., Liu, Y. & Cui, Y. Reviving the lithium metal anode for high-energy batteries. Nat Nanotechnol 12, 194-206 (2017). 6 Zheng, J. et al. Reversible epitaxial electrodeposition of metals in battery anodes. Science 366, 645-648 (2019). 7 Lin, M. C. et al. An ultrafast rechargeable aluminium-ion battery. Nature 520, 325-328 (2015). 8 Deng, Y. et al. On the Reversibility and Fragility of Sodium Metal Electrodes. Advanced Energy Materials 9, 1901651 (2019). 9 Shi, F. et al. Strong texturing of lithium metal in batteries. Proc Natl Acad Sci U S A 114, 12138-12143 (2017). 28 10 Zheng, J. et al. Regulating electrodeposition morphology of lithium: towards commercially relevant secondary Li metal batteries. Chem Soc Rev 49, 2701- 2750, doi:10.1039/c9cs00883g (2020). 11 Liu, W., Liu, P. & Mitlin, D. Tutorial review on structure - dendrite growth relations in metal battery anode supports. Chem Soc Rev 49, 7284-7300, doi:10.1039/d0cs00867b (2020). 12 Tikekar, M. D., Choudhury, S., Tu, Z. & Archer, L. A. Design principles for electrolytes and interfaces for stable lithium-metal batteries. Nature Energy 1, 1-7 (2016). 13 Sawada, Y., Dougherty, A. & Gollub, J. P. Dendritic and fractal patterns in electrolytic metal deposits. Phys Rev Lett 56, 1260-1263 (1986). 14 Zheng, J. et al. Physical Orphaning versus Chemical Instability: Is Dendritic Electrodeposition of Li Fatal? ACS Energy Letters 4, 1349-1355 (2019). 15 Fang, C. et al. Quantifying inactive lithium in lithium metal batteries. Nature 572, 511-515 (2019). 16 Higashi, S., Lee, S. W., Lee, J. S., Takechi, K. & Cui, Y. Avoiding short circuits from zinc metal dendrites in anode by backside-plating configuration. Nat Commun 7, 1-6 (2016). 17 Deng, Y. et al. On the Reversibility and Fragility of Sodium Metal Electrodes. Advanced Energy Materials 9, doi:10.1002/aenm.201901651 (2019). 18 Shankar, S. S., Bhargava, S. & Sastry, M. Synthesis of gold nanospheres and nanotriangles by the Turkevich approach. J Nanosci Nanotechnol 5, 1721-1727 (2005). 29 19 Wiley, B., Herricks, T., Sun, Y. & Xia, Y. Polyol Synthesis of Silver Nanoparticles: Use of Chloride and Oxygen to Promote the Formation of Single-Crystal, Truncated Cubes and Tetrahedrons. Nano Lett 4, 1733-1739 (2004). 20 Magnussen, O. M., Ocko, B. M., Adzic, R. R. & Wang, J. X. X-ray diffraction studies of ordered chloride and bromide monolayers at the Au(111)-solution interface. Phys Rev B Condens Matter 51, 5510-5513 (1995). 21 Zheng, J. et al. Spontaneous and field-induced crystallographic reorientation of metal electrodeposits at battery anodes. Science Advances 6, eabb1122 (2020). 22 Zheng, J. & Archer, L. A. Controlling electrochemical growth of metallic zinc electrodes: Toward affordable rechargeable energy storage systems. Science Advances 7 (2021). 23 Zhong, C. et al. Decoupling electrolytes towards stable and high-energy rechargeable aqueous zinc–manganese dioxide batteries. Nature Energy 5, 440-449 (2020). 24 Choi, J. W. & Aurbach, D. Promise and reality of post-lithium-ion batteries with high energy densities. Nature Reviews Materials 1, 1-16 (2016). 25 Wang, F. et al. Highly reversible zinc metal anode for aqueous batteries. Nat Mater 17, 543-549 (2018). 26 Biton, M., Tariq, F., Yufit, V., Chen, Z. & Brandon, N. Integrating multi- length scale high resolution 3D imaging and modelling in the characterisation and identification of mechanical failure sites in electrochemical dendrites. Acta Materialia 141, 39-46 (2017). 30 27 Khor, A. et al. Review of zinc-based hybrid flow batteries: From fundamentals to applications. Materials Today Energy 8, 80-108 (2018). 28 Albertus, P., Babinec, S., Litzelman, S. & Newman, A. Status and challenges in enabling the lithium metal electrode for high-energy and low-cost rechargeable batteries. Nature Energy 3, 16-21 (2018). 29 Filankembo, A. & Pileni, M. P. Is the template of self-colloidal assemblies the only factor that controls nanocrystal shapes? The Journal of Physical Chemistry B 104, 5865-5868 (2000). 30 Irish, D. E., McCarroll, B. & Young, T. F. Raman Study of Zinc Chloride Solutions. The Journal of Chemical Physics 39, 3436-3444, doi:10.1063/1.1734212 (1963). 31 Kálmán, E., I. Serke, G. Pálinkás, G. Johansson, G. Kabisch, M. Maeda, and H. Ohtaki. Complex formation in an aqueous ZnBr2 solution based on electron diffraction, X-ray scattering and Raman spectra. Zeitschrift für Naturforschung A 38, 225-230 (1983). 32 Wakita, H., Georg Johansson, Magnus Sandström, Peter L. Goggin, and Hitoshi Ohtaki. Structure determination of zinc iodide complexes formed in aqueous solution. Journal of solution chemistry 20, 643-668 (1991). 33 Bulgariu, L. & Bulgariu, D. The Partition Behavior of Zn(II) using Halide Ions Extractants in Aqueous Peg‐Based Two‐Phase Systems. Separation Science and Technology 42, 1093-1106, doi:10.1080/01496390601174307 (2007). 34 Bulgariu, L., Bulgariu, D., Sârghie, I. & Măluṭan, T. Cd(II) extraction in PEG- based two-phase aqueous systems in the presence of iodide ions. Analysis of 31 PEG-rich solid phases. Open Chemistry 5, 291-302, doi:10.2478/s11532-006- 0048-7 (2007). 35 Bulgariu, L. & Bulgariu, D. Extraction of gold(III) from chloride media in aqueous polyethylene glycol-based two-phase system. Separation and Purification Technology 80, 620-625, doi:10.1016/j.seppur.2011.06.018 (2011). 36 Banik, S. J. & Akolkar, R. Suppressing Dendrite Growth during Zinc Electrodeposition by PEG-200 Additive. Journal of The Electrochemical Society 160, D519-D523, doi:10.1149/2.040311jes (2013). 37 Feng, Z. V., Li, X. & Gewirth, A. A. Inhibition due to the interaction of polyethylene glycol, chloride, and copper in plating baths: a surface-enhanced Raman study. The Journal of Physical Chemistry B 107, 9415-9423 (2003). 38 T. P. Moffat, a. et al. Superconformal Electrodeposition of Copper in 500–90 nm Features. Journal of The Electrochemical Society 147, 4524-4535 (2000). 39 Hebert, K. R. Role of Chloride Ions in Suppression of Copper Electrodeposition by Polyethylene Glycol. Journal of The Electrochemical Society 152, C283 (2005). 40 Hasegawa, M., Okinak, Y., Shacham-Diamand, Y. & Osaka, T. Void-free trench-filling by electroless copper deposition using the combination of accelerating and inhibiting additives. Electrochemical and Solid State Letters 9, C138 (2006). 32 41 Wilms, M., Broekmann, P., M. Kruft, Z. P., Stuhlmann, C. & Wandelt, K. STM investigation of specific anion adsorption on Cu (111) in sulfuric acid electrolyte. Surface science 402, 83-86 (1998). 42 Biswal, P., Stalin, S., Kludze, A., Choudhury, S. & Archer, L. A. Nucleation and Early Stage Growth of Li Electrodeposits. Nano Lett 19, 8191-8200 (2019). 43 Shen, C. et al. Graphene-Boosted, High-Performance Aqueous Zn-Ion Battery. ACS Appl Mater Interfaces 10, 25446-25453 (2018). 44 Zhao, K. et al. Ultrathin Surface Coating Enables Stabilized Zinc Metal Anode. Advanced Materials Interfaces 5, 1800848 (2018). 45 Zhang, Q. et al. The Three-Dimensional Dendrite-Free Zinc Anode on a Copper Mesh with a Zinc-Oriented Polyacrylamide Electrolyte Additive. Angew Chem Int Ed Engl 58, 15841-15847 (2019). 46 Liang, P. et al. Highly Reversible Zn Anode Enabled by Controllable Formation of Nucleation Sites for Zn‐Based Batteries. Advanced Functional Materials 30, 1908528 (2020). 47 Zeng, Y. et al. Dendrite-Free Zinc Deposition Induced by Multifunctional CNT Frameworks for Stable Flexible Zn-Ion Batteries. Adv Mater 31, e1903675 (2019). 48 Yuksel, R., Buyukcakir, O., Seong, W. K. & Ruoff, R. S. Metal‐Organic Framework Integrated Anodes for Aqueous Zinc‐Ion Batteries. Advanced Energy Materials 10, 1904215 (2020). 33 49 Chen, A., Zhichao Shi, Dan Bizzotto, Jacek Lipkowski, Bruno Pettinger, and Christoph Bilger. Iodide adsorption at the Au (111) electrode surface. Journal of Electroanalytical Chemistry 467, 342-353 (1999). 34 CHAPTER 3 ZINC ANODE AND FULL BATTERY The reversible electrodeposition of Zn anode 35 Figure 1: Galvanostatic cycling behavior of Zn anodes in 2M ZnI2 and ZnI2 + 5 wt% PEG300 aqueous electrolytes. Cycling performance of Zn||Zn symmetric cells at (A) 25 mA cm-2, 3.2 mAh cm-2. (B) 15 mA cm-2, 15 mAh cm-2. (C) 10 mA cm-2, 20 mAh cm-2. (D)Coulombic efficiency measured in Zn plate-strip experiments in Zn||carbon matrix half-cells in a 2M ZnI2+5 wt% PEG300 electrolyte. The inset is the voltage-capacity curve for the corresponding Zn||carbon matrix half-cell in which 2M ZnI2 is used as the electrolyte for current densities of 25 mA cm-2 and 3.2 mAh cm-2. (E) The Voltage-areal capacity plot for Zn||carbon matrix half-cell in 2M ZnI2+5 wt% PEG300 electrolyte at 25 mA cm-2 and 3.2 mAh cm-2. 36 Figure 2. SEM images of Zn foil after cycling in (A) (C) 2M ZnI2 + 5 wt% PEG300 (B) (D) 2M ZnI2. (A) and (B) are in the same scale. Figure 3. Cycle performance of Zn symmetric cell at 25 mA cm-2, 3.2 mAh cm-2. To evaluate the reversibility of Zn deposited at high areal capacity, we measured the electrochemical cycling performance of Zn anodes in Zn symmetric cell and plating/stripping Coulombic efficiency (CE) of Zn deposits in Zn /carbon matrix half- cell. Fig. 1A~C show the galvanostatic cycle curves of Zn/Zn symmetric battery at various current densities and areal capacity. Compared with ZnI2 electrolyte, ZnI2+5wt% PEG300 electrolyte shows a better cycle performance due to the 2D 37 hexagonal crystal growth structure and high (002) plane orientation deposition. It is worth noting that the highest area capacity of 15 mAh cm-2 and a cycle time of 630 hours were achieved for the first time43-48 on a Zn anode with a current density of 15 mA cm-2 (Fig. 1B). Furthermore, at a current density of 25 mA cm-2, the cycle time with the areal capacity 3.2 mAh cm-2 is over 4000 cycles and 1200 hours, which is the longest cycle time at such high current density and areal capacity till now (Fig. 1A). Even under the unprecedented high area capacity of 20mAh/ cm2 in coin cell, the Zn anode can still work stably over 200 hours (Fig. 1C). In ZnI2+5 wt% PEG300 electrolyte, the Zn foil after 1000 cycles (25 mA cm-2 and 3.2 mAh cm-2) shows the uniform plating/stripping morphology compared to the dendritic surface of Zn foil in the ZnI2 electrolyte after short circuit (50cycles) (Fig. 2). On the other hand, the CE, which reveals the ratio between the plating and stripping metal in each cycle, also indicates the better electrodeposition performance of ZnI2+5 wt% PEG300 electrolyte. Considering that the limitation of planar stainless steel at high areal capacity, we choose 3D interwoven carbon fibers (carbon matrix) as a deposition substrate which can supply a high deposition area. In stark contrast to the rapid battery failure observed in ZnI2 electrolyte (Fig. 1D), ZnI2+5 wt% PEG300 electrolyte manifests a high level of reversibility (99.1%~99.8%) at high areal capacity 3.2 mAh cm-2 (Fig. 1D). The voltage profile confirms that the plating/stripping reaction is stable (Fig. 1E). The final important consideration is that since the successful crystal growth control with PEG300 in ZnI2 electrolyte, will it work in ZnBr2 and ZnCl2 electrolyte? We found that PEG300 can partially control the crystal growth in ZnBr2+5 wt% PEG300 38 electrolyte. As shown in previous crystal growth structure, compared with the sphere- like structure, complete large oval structures with a planar plane facing us can be observed. However, the crystal growth pattern is more complicate in ZnCl2+5 wt% PEG300 electrolyte. At the initial deposition period, we observe two kinds of structure: sphere-like and hexagonal platelet structure. As deposition time increases, only sphere-like structures are observed. Hence, we think that PEG300 fails to reshape the crystal growth structure in ZnCl2+5 wt% PEG300 electrolyte. Cycling performance of the Zn symmetric cell is consistent with the control result of crystal growth structure (Fig. 3). The best crystal growth control, the hexagonal platelet structure, in ZnI2+5 wt% PEG300 electrolyte shows the longest cycle time, whereas the failure of crystal control in ZnCl2+5 wt% PEG300 electrolyte shows the shortest cycle time. Two considerations are taken into account to the greatly different crystal growth structures in ZnX2(X=Cl, Br, I) +5 wt% PEG300 electrolytes: (1) as 1H NMR spectra analysis shows, due to the high concentration of unstable complexes [ZnI3]- in ZnI2 electrolyte, the energy barrier of forming PEG300-Zn2+-aI- (a=1,2,3) complex is lower, which has an indispensable influence in the adsorption process as we think it’s halide’s preferential adsorption that effects the PEG300 adsorption39. (2) Under high current density, compared with chloride and bromide, iodide has a strong chemisorption ability to the gold (111) electrode surface due to the covalent bond character between adsorbed iodide and metal surface49. Correspondingly, we believe that iodide also has the strongest preferential chemisorption ability to the (002) crystal plane of Zn, which highly enhance the control ability of crystal growth process in ZnI2+PEG electrolyte compared with other two Zn halide electrolytes. Furthermore, 39 the poor cycling performance of Zn symmetric cells in ZnCl2+5 wt% PEG300 and ZnBr2+5 wt% PEG300 electrolytes indicate the importance of hexagonal platelet crystal growth structure and highly oriented deposition morphology to the Zn anode. On this basis, we conclude that regulating metal electrodeposition morphology by controlling crystal growth structure and orientation is a rational and available method. 40 Aqueous Zinc full battery Figure 4: Full 4Zn||I2 battery cell performance in 2M ZnI2 and ZnI2 + 5 wt% 4 PEG300 aqueo4us electrolytes. (A) Cycling performance of Zn||I2 coin cells at 2 mA 3 cm-2. (B) Charge/discharge curve of Zn||I coin cell at a current density of 2 mA cm-24 2 . (C) Coulombic efficiency of Zn||I2 electrochemical flow battery. (D) The Voltage- 41 areal capacity profile in the first few cycles of a 10 mAh cm-2 Zn||I2 flow battery with PEG300 additive at 40 mA cm-2. (E) Photograph of our single-flow Zn||I2 flow cell. (F) Deposition morphology of Zn on the carbon felt anode current collector for the Zn||I2 flow cell after 10 cycles, with/without PEG300. Figure 5. Charge/discharge curve of Zn/I2 battery in coin cell. Voltage-Capacity curve at different current density. Benefiting from the high areal capacity and long cycle time of the Zn anode, we further investigate the performance of Zn/I2 full batteries in coin cell and flow battery, respectively. In Zn/I2 coin cell, due to strong adsorption ability to the I -3 , activated carbon (AC) can be used as cathode materials, which can well reduce the shuttle effect of I -3 in the aqueous electrolyte. The Zn║AC full battery shows the stable 42 charge/discharge plateau and different areal capacity at different current densities (Fig. 5). The reaction during charge/discharge should be: Anode: Zn2+ + 2e- ↔ Zn Cathode: 3I- ↔ I - -3 + 2e Total reaction: Zn2+ + 3I- ↔ Zn + I - 3 Fig. 4A, B reports that the Zn/I2 coin cell has a high areal capacity of 1 mAh cm-2 at a current density of 2 mA cm-2, achieving 1000 cycles without fading. To demonstrate the advantages of the high area capacity Zn anode system and push it to the height of application, the flow battery is a good model to prove our system. We choose the previously reported electrolyte50: 6M KI and 3M ZnBr2 and adding 5 wt% PEG300 to the anolyte side. Considering that the iodine deposition will block the flowing tube, we use single-flow battery here50. As Fig. 4C~D show, the Zn/I2 flow battery has a high areal capacity of 10 mAh cm-2 at a current density of 40 mA cm-2, stability running 120 cycles with high Coulombic efficiency (CE) 99.5%. After 120 cycles, the CE shows a little drop. However, after we replace the catholyte, the CE recovers to 99.5%, meaning the iodine deposition side is the main problem for our system. We replace the catholyte three times and the flow battery system can work over 200 cycles. On the contrary, the CE keeps dropping after we replace the catholyte in the electrolyte without PEG300. The deposition morphology (Fig. 4F) well reflects the smooth deposition in PEG300 additive electrolyte. Owing to the non-dendrite Zn anode with PEG300 additives, theoretically, we can keep the flow battery working for a long time by replacing the catholyte. 43 REFERENCES 1 Liang, Y., Dong, H., Aurbach, D. & Yao, Y. Current status and future directions of multivalent metal-ion batteries. Nature Energy, 1-11 (2020). 2 Lawless, K. R. Growth and Structure of Electrodeposited Thin Metal Films. Journal of Vacuum Science and Technology 2, 24-34, doi:10.1116/1.1492395 (1965). 3 Sun, F. et al. Morphology Control and Transferability of Ordered Through- Pore Arrays Based on the Electrodeposition of a Colloidal Monolayer. Advanced Materials 16, 1116-1121, doi:10.1002/adma.200400006 (2004). 4 Gleason, K. K. Nanoscale control by chemically vapour-deposited polymers. Nature Reviews Physics 2, 347-364, doi:10.1038/s42254-020-0192-6 (2020). 5 Lin, D., Liu, Y. & Cui, Y. Reviving the lithium metal anode for high-energy batteries. Nat Nanotechnol 12, 194-206 (2017). 6 Zheng, J. et al. Reversible epitaxial electrodeposition of metals in battery anodes. Science 366, 645-648 (2019). 7 Lin, M. C. et al. An ultrafast rechargeable aluminium-ion battery. Nature 520, 325-328 (2015). 8 Deng, Y. et al. On the Reversibility and Fragility of Sodium Metal Electrodes. Advanced Energy Materials 9, 1901651 (2019). 9 Shi, F. et al. Strong texturing of lithium metal in batteries. Proc Natl Acad Sci U S A 114, 12138-12143 (2017). 44 10 Zheng, J. et al. Regulating electrodeposition morphology of lithium: towards commercially relevant secondary Li metal batteries. Chem Soc Rev 49, 2701- 2750, doi:10.1039/c9cs00883g (2020). 11 Liu, W., Liu, P. & Mitlin, D. Tutorial review on structure - dendrite growth relations in metal battery anode supports. Chem Soc Rev 49, 7284-7300, doi:10.1039/d0cs00867b (2020). 12 Tikekar, M. D., Choudhury, S., Tu, Z. & Archer, L. A. Design principles for electrolytes and interfaces for stable lithium-metal batteries. Nature Energy 1, 1-7 (2016). 13 Sawada, Y., Dougherty, A. & Gollub, J. P. Dendritic and fractal patterns in electrolytic metal deposits. Phys Rev Lett 56, 1260-1263 (1986). 14 Zheng, J. et al. Physical Orphaning versus Chemical Instability: Is Dendritic Electrodeposition of Li Fatal? ACS Energy Letters 4, 1349-1355 (2019). 15 Fang, C. et al. Quantifying inactive lithium in lithium metal batteries. Nature 572, 511-515 (2019). 16 Higashi, S., Lee, S. W., Lee, J. S., Takechi, K. & Cui, Y. Avoiding short circuits from zinc metal dendrites in anode by backside-plating configuration. Nat Commun 7, 1-6 (2016). 17 Deng, Y. et al. On the Reversibility and Fragility of Sodium Metal Electrodes. Advanced Energy Materials 9, doi:10.1002/aenm.201901651 (2019). 18 Shankar, S. S., Bhargava, S. & Sastry, M. Synthesis of gold nanospheres and nanotriangles by the Turkevich approach. J Nanosci Nanotechnol 5, 1721-1727 (2005). 45 19 Wiley, B., Herricks, T., Sun, Y. & Xia, Y. Polyol Synthesis of Silver Nanoparticles: Use of Chloride and Oxygen to Promote the Formation of Single-Crystal, Truncated Cubes and Tetrahedrons. Nano Lett 4, 1733-1739 (2004). 20 Magnussen, O. M., Ocko, B. M., Adzic, R. R. & Wang, J. X. X-ray diffraction studies of ordered chloride and bromide monolayers at the Au(111)-solution interface. Phys Rev B Condens Matter 51, 5510-5513 (1995). 21 Zheng, J. et al. Spontaneous and field-induced crystallographic reorientation of metal electrodeposits at battery anodes. Science Advances 6, eabb1122 (2020). 22 Zheng, J. & Archer, L. A. Controlling electrochemical growth of metallic zinc electrodes: Toward affordable rechargeable energy storage systems. Science Advances 7 (2021). 23 Zhong, C. et al. Decoupling electrolytes towards stable and high-energy rechargeable aqueous zinc–manganese dioxide batteries. Nature Energy 5, 440-449 (2020). 24 Choi, J. W. & Aurbach, D. Promise and reality of post-lithium-ion batteries with high energy densities. Nature Reviews Materials 1, 1-16 (2016). 25 Wang, F. et al. Highly reversible zinc metal anode for aqueous batteries. Nat Mater 17, 543-549 (2018). 26 Biton, M., Tariq, F., Yufit, V., Chen, Z. & Brandon, N. Integrating multi- length scale high resolution 3D imaging and modelling in the characterisation and identification of mechanical failure sites in electrochemical dendrites. Acta Materialia 141, 39-46 (2017). 46 27 Khor, A. et al. Review of zinc-based hybrid flow batteries: From fundamentals to applications. Materials Today Energy 8, 80-108 (2018). 28 Albertus, P., Babinec, S., Litzelman, S. & Newman, A. Status and challenges in enabling the lithium metal electrode for high-energy and low-cost rechargeable batteries. Nature Energy 3, 16-21 (2018). 29 Filankembo, A. & Pileni, M. P. Is the template of self-colloidal assemblies the only factor that controls nanocrystal shapes? The Journal of Physical Chemistry B 104, 5865-5868 (2000). 30 Irish, D. E., McCarroll, B. & Young, T. F. Raman Study of Zinc Chloride Solutions. The Journal of Chemical Physics 39, 3436-3444, doi:10.1063/1.1734212 (1963). 31 Kálmán, E., I. Serke, G. Pálinkás, G. Johansson, G. Kabisch, M. Maeda, and H. Ohtaki. Complex formation in an aqueous ZnBr2 solution based on electron diffraction, X-ray scattering and Raman spectra. Zeitschrift für Naturforschung A 38, 225-230 (1983). 32 Wakita, H., Georg Johansson, Magnus Sandström, Peter L. Goggin, and Hitoshi Ohtaki. Structure determination of zinc iodide complexes formed in aqueous solution. Journal of solution chemistry 20, 643-668 (1991). 33 Bulgariu, L. & Bulgariu, D. The Partition Behavior of Zn(II) using Halide Ions Extractants in Aqueous Peg‐Based Two‐Phase Systems. Separation Science and Technology 42, 1093-1106, doi:10.1080/01496390601174307 (2007). 34 Bulgariu, L., Bulgariu, D., Sârghie, I. & Măluṭan, T. Cd(II) extraction in PEG- based two-phase aqueous systems in the presence of iodide ions. Analysis of 47 PEG-rich solid phases. Open Chemistry 5, 291-302, doi:10.2478/s11532-006- 0048-7 (2007). 35 Bulgariu, L. & Bulgariu, D. Extraction of gold(III) from chloride media in aqueous polyethylene glycol-based two-phase system. Separation and Purification Technology 80, 620-625, doi:10.1016/j.seppur.2011.06.018 (2011). 36 Banik, S. J. & Akolkar, R. Suppressing Dendrite Growth during Zinc Electrodeposition by PEG-200 Additive. Journal of The Electrochemical Society 160, D519-D523, doi:10.1149/2.040311jes (2013). 37 Feng, Z. V., Li, X. & Gewirth, A. A. Inhibition due to the interaction of polyethylene glycol, chloride, and copper in plating baths: a surface-enhanced Raman study. The Journal of Physical Chemistry B 107, 9415-9423 (2003). 38 T. P. Moffat, a. et al. Superconformal Electrodeposition of Copper in 500–90 nm Features. Journal of The Electrochemical Society 147, 4524-4535 (2000). 39 Hebert, K. R. Role of Chloride Ions in Suppression of Copper Electrodeposition by Polyethylene Glycol. Journal of The Electrochemical Society 152, C283 (2005). 40 Hasegawa, M., Okinak, Y., Shacham-Diamand, Y. & Osaka, T. Void-free trench-filling by electroless copper deposition using the combination of accelerating and inhibiting additives. Electrochemical and Solid State Letters 9, C138 (2006). 48 41 Wilms, M., Broekmann, P., M. Kruft, Z. P., Stuhlmann, C. & Wandelt, K. STM investigation of specific anion adsorption on Cu (111) in sulfuric acid electrolyte. Surface science 402, 83-86 (1998). 42 Biswal, P., Stalin, S., Kludze, A., Choudhury, S. & Archer, L. A. Nucleation and Early Stage Growth of Li Electrodeposits. Nano Lett 19, 8191-8200 (2019). 43 Shen, C. et al. Graphene-Boosted, High-Performance Aqueous Zn-Ion Battery. ACS Appl Mater Interfaces 10, 25446-25453 (2018). 44 Zhao, K. et al. Ultrathin Surface Coating Enables Stabilized Zinc Metal Anode. Advanced Materials Interfaces 5, 1800848 (2018). 45 Zhang, Q. et al. The Three-Dimensional Dendrite-Free Zinc Anode on a Copper Mesh with a Zinc-Oriented Polyacrylamide Electrolyte Additive. Angew Chem Int Ed Engl 58, 15841-15847 (2019). 46 Liang, P. et al. Highly Reversible Zn Anode Enabled by Controllable Formation of Nucleation Sites for Zn‐Based Batteries. Advanced Functional Materials 30, 1908528 (2020). 47 Zeng, Y. et al. Dendrite-Free Zinc Deposition Induced by Multifunctional CNT Frameworks for Stable Flexible Zn-Ion Batteries. Adv Mater 31, e1903675 (2019). 48 Yuksel, R., Buyukcakir, O., Seong, W. K. & Ruoff, R. S. Metal‐Organic Framework Integrated Anodes for Aqueous Zinc‐Ion Batteries. Advanced Energy Materials 10, 1904215 (2020). 49 49 Chen, A., Zhichao Shi, Dan Bizzotto, Jacek Lipkowski, Bruno Pettinger, and Christoph Bilger. Iodide adsorption at the Au (111) electrode surface. Journal of Electroanalytical Chemistry 467, 342-353 (1999). 50 Xie, C., Liu, Y., Lu, W., Zhang, H. & Li, X. Highly stable zinc–iodine single flow batteries with super high energy density for stationary energy storage. Energy & Environmental Science 12, 1834-1839 (2019). 50 CHAPTER 4 CONCLUSION In summary, we report that crystal growth stabilization method, including reshaping the crystal growth structure and regulating the crystal deposition orientation, is a good way to control the metal electrodeposition morphology and further has a great impact on the battery anode cycle performance. Taking Zn anode as a good example, the Zn anode achieves unprecedented cycle performance at the high current density and areal capacity using carefully selected PEG-Zn2+-aI- (1=1,2,3) complexes system to control the crystal growth process. On the contrary, the failure of controlling crystal growth structure indicates the poor cycle performance of Zn anode. We further demonstrate the application value of this electrolyte system in coin cell and a single flow battery, both of which show excellent performance at a high areal capacity. Our findings point out the importance to control crystal growth in the metal anode system. To pursue the high areal capacity metal anodes, which can meet the application requirements, crystal growth structure and orientation are the key steps that must be noticed. 51