Author s Accepted Manuscript

Transcription

1 Author s Accepted Manuscript Sealing ZnO nanorods for deeply rechargeable high-energy aqueous battery anodes Yamin Zhang, Yutong Wu, Haoran Ding, Yu Yan, Zhubo Zhou, Yong Ding, Nian Liu PII: DOI: Reference: To appear in: S (18) NANOEN3026 Nano Energy Received date: 31 July 2018 Revised date: 8 September 2018 Accepted date: 10 September 2018 Cite this article as: Yamin Zhang, Yutong Wu, Haoran Ding, Yu Yan, Zhubo Zhou, Yong Ding and Nian Liu, Sealing ZnO nanorods for deeply rechargeable high-energy aqueous battery anodes, Nano Energy, This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

2 Sealing ZnO nanorods for deeply rechargeable high-energy aqueous battery anodes Yamin Zhang 1, Yutong Wu 1, Haoran Ding 1,2, Yu Yan 1,3, Zhubo Zhou 1, Yong Ding 4, and Nian Liu 1,* 1 School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, USA 2 School of Chemical Engineering and Technology, Tianjin University, Tianjin , China 3 School of Physical Sciences, University of Science and Technology of China, Hefei, Anhui , China 4 School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, USA nian.liu@chbe.gatech.edu website: *Correspondence: Abstract Rechargeable Zn-based batteries are a safe alternative to Li-ion for compatibility with aqueous electrolyte. Also, theoretical volumetric energy density of Zn-based batteries (e.g. Zn-air) is ~85% of lithium-sulfur battery. However, the rechargeability and specific capacity of Zn anodes are limited by passivation and dissolution. Here we report a ZnO@TiN x O y core/shell nanorod structure for deeply rechargeable Zn anodes. The small diameter (<500 nm) of ZnO prevents passivation and allows full utilization of active materials, while the thin and conformal TiN x O y coating mitigates Zn dissolution in alkaline electrolyte, mechanically maintains the nanostructure, and delivers electron to nanorods. As a result, the ZnO@TiN x O y core/shell nanorod anode achieves superior specific capacity and cycle life compared with bulk Zn foil and uncoated ZnO nanorod anodes. The discharge capacity of this anode is twice as large as that of the uncoated ZnO nanorod anode. Remarkably, our ZnO@TiN x O y nanorod anode achieves a much higher specific discharge capacity of 508 mah/g(zn) than that of many previously reported zinc anodes. It can deeply cycle >640 times (64 days) in a beaker cell and deliver excellent long-term electrochemical performance (more than 7,500 cycles) when cycled under start-stop conditions. The nanoscale design principles reported here is an important step towards practical deeply 1

3 rechargeable Zn anodes, and can potentially be applied to overcome intrinsic limitations of other battery materials that involve soluble intermediates or insulating discharge products. Graphical Abstracts The x O y nanorod anode was synthesized by two steps: hydrothermal growth of ZnO nanorods on the carbon paper followed by an atomic layer deposition. The conformal TiN x O y coating mitigates Zn dissolution, mechanically maintains the shape of zinc anode, and delivers electron to ZnO nanorods. This zinc anode shows excellent electrochemical performance in alkaline electrolyte. Keywords Aqueous batteries; Rechargeable; Zinc oxide; Titanium nitride; Nanomaterial 1. Introduction Alternative energy carriers have been sought after due to fossil fuels slow regeneration and environmental concerns such as climate change, air and water pollution[1]. When carrying clean electricity from solar or wind, batteries are promising to alleviate the current energy and environmental problems.[2 4] Lithium-ion batteries[5 10] are widely used energy storage system because of their outstanding energy density and rechargeability. However, safety is always a concern because the use of flammable organic electrolytes. On the other hand, batteries that use aqueous electrolytes have enhanced safety (Supplementary video), ion conductivity, and cost-effectiveness.[11 18] Within the stability window of water, zinc is an attractive anode material because it is the most active metal that is stable with water. By using aqueous electrolyte, zinc-based batteries not only are safer, but also can be manufactured in ambient air rather than 2

4 dry room, and has much higher tolerance to moisture and air during operation. Having two valence electrons and high density, zinc metal has three times the volumetric capacity of lithium metal. Among various zinc-based batteries, Zn-air[19,20] has a theoretical volumetric energy density (4400 Wh/L) that is more than three times of conventional Li-ion batteries (1400 Wh/L), and approaching Li-S batteries (5200 Wh/L). Primary Zn-air batteries have already been the battery of choice for hearing aids, which require extremely high energy density and safety. Finally, zinc is abundant, low-cost, and environmentally benign, rendering them suitable for large scale applications. Although stable cycling of Zn anodes in mild acidic electrolyte[15,16,21] has been demonstrated, an alkaline electrolyte is ideal for zinc-air batteries[22 25], because oxygen cathode has minimum overpotential in alkaline electrolyte. However, a deeply rechargeable (>50% depth of discharge) Zn anode in lean alkaline electrolyte (mass ratio of electrolyte to electrode < 100:1) is still lacking[26 30], due to multiple challenges. In aqueous alkaline electrolyte, zinc anode undergoes two consecutive reactions (equation (1) and equation (2)): complexation (or dissolution) and electroreduction reactions in charging; electrooxidation and precipitation reactions in discharging. Complexation/Precipitation: ZnO + H 2 O + 2OH Zn(OH) 4 2 (1) Electroreduction/Electrooxidation: Zn(OH) e Zn + 4OH (2) Different from conventional intercalation electrodes in lithium ion batteries, this solid-solute-solid (ZnO-Zn(OH) Zn) transformation[31] of zinc electrode inherently represents a series of challenges: (i) discharge product ZnO passivates the surface of Zn, preventing the latter from further discharging; (ii) ZnO is insulating, which can be hardly recharged to metallic Zn; (iii) ZnO precipitation from soluble zincates occurs randomly on the electrode surface, and changes the morphology of electrode over cycling. Zn metal foil is the most commonly used Zn anode in aqueous batteries. However, the passivation problem (nonconductive property of ZnO) of Zn foil limits the utilization (<1%) of Zn foil anode and makes it non-rechargeable (Fig. 1a). Under 10 ma discharge, we can only get 1.7 mah capacity of Zn foil with 0.25 mm thickness and 1 cm diameter in coin cell (Supplementary Fig. S1a). The thickness of the passivation layer (or the critical passivation size) is ~ 2 µm (Supplementary Fig. S1b). Microporous Zn sponges[26,32,33] have been made to enhance the rechargeability, yet its feature size is ~ 10 µm (> 2 µm), so only 40% DOD of them can be achieved. Sub-micron-sized Zn anodes[34 37] including ZnO nanoplates[38] and ZnO nanoparticles[39] are also investigated. However, the anode dissolution problem, resulting from large electrode-electrolyte surface area[40] of these sub-micron-sized Zn anodes, is not well controlled or clearly stated. 3

5 Figure 1 Schematic of morphological changes of zinc electrode during electrochemical cycling. a, Zn foil shows very low utilization (<1%) because of ZnO passivation layer. The critical passivation size is ~ 2 µm, as shown above. b, The feature size of ZnO nanorod is smaller than the critical passivation size, however, the large electrode-electrolyte surface area accelerates anode dissolution and promotes electrode shape change. Moreover, due to the relatively insulating property of ZnO, electrons can only be distributed on carbon paper, which leads to fast complexation and electroreduction reactions on the root of nanorods in charging. As a result, the nanorods will detach from carbon paper. c, The shape of ZnO@TiN x O y nanorod anode retains during cycling with the sealed nanorod structure. Herein, we solve Zn anode s dilemma of passivation and dissolution with a sealed structure (ZnO@TiN x O y nanorod anode), with feature size smaller than the critical passivation size, and a thin and conformal coating to prevent dissolution of anode (Fig. 1c). We used the hydrothermal method[41] to grow ZnO nanorods on carbon fiber paper (Fig. 1b). Then, the atomic layer deposition (ALD) technique was used to form a strong and conductive stable TiN x O y coating on the ZnO nanorods. Our structure has a few advantages: (i) the feature size of ZnO nanorod is smaller than critical passivation size; (ii) the carbon paper framework and TiN x O y coating, which encapsulates ZnO nanorod, function as an electrical pathway so that all ZnO nanorods are electrochemically active; (iii) the TiN x O y coating enables fast hydroxide/water diffusion as well 4

6 as blocks large zincates from escaping during electrochemical cycling, thus prevents anode structure fracture. 2. Material and methods 2.1. Synthesis of ZnO nanorods Briefly, carbon paper (Fuel Cell Store) was first heat-treated at 500 C for 1 h in air to increase its wettability. Then, ZnO nanorods were grown on carbon paper by a wet chemical process. carbon paper was soaked in an aqueous solution containing 0.1 M KMnO 4 (Sigma Aldrich) for 1 h to form a seed layer. The seeded carbon paper was then dipped into a glass bottle with a precursor solution containing 50 ml zinc nitrate hexahydrate (30 mm, Alfa Aesar), 50 ml hexamethylenetetramine (30 mm, Sigma Aldrich), and ammonia ( % NH 3 basis, Sigma Aldrich)[41]. The sealed bottle was placed into an oven at 90 C. After that, the white-colored carbon paper ZnO nanorods was obtained by water washing and drying at 80 C for 3 h. Different mass loadings of ZnO nanorods on carbon paper ranging from 0.5 mg/cm 2 to 5.5 mg/cm 2 were synthesized by adjusting the carbon paper area per bottle, NH 3 concentration, hydrothermal time and hydrothermal times, as summarized in Supplementary Table. S Synthesis of ZnO@TiN x O y core/shell nanorods The synthesis of ZnO@TiN x O y core/shell nanorods was conducted in Cambridge FIJI Plasma ALD system. First of all, the TiN was deposited onto the ZnO nanorods. The precursors of TiN were Tetrakis(dimethylamido)Titanium(IV) (TDMAT, Sigma Aldrich) and N 2. During TiN ALD process, the recipe was run 100 or 200 cycles at 250 C. The TiN ALD recipe is shown in Supplementary Fig. S2. Then, when ZnO@TiN nanorods were exposed to air, the TiN was partially oxidized to TiN x O y (evidenced by XPS spectra in Fig. 2k). Thus, for accuracy purpose, we named our anode as ZnO@TiN x O y in this manuscript Electrochemistry To investigate the Zn anode, we make full batteries using Ni(OH) 2 as the rechargeable cathode. The Ni(OH) 2 cathodes were harvested from commercial Ni-Zn AA batteries from PowerGenix. Coin cell: Coin-type batteries were assembled using CR2032 cases (MTI Corporation), our zinc anodes (round disk, 1 cm diameter) and Ni(OH) 2 cathodes with excess capacity, as shown in Supplementary Fig. S8. Coin cell has small volume of electrolyte, which is required for practical application. The aqueous electrolyte consists of 4 M KOH (Sigma Aldrich, 99.99%), 2 M KF (Alfa Aesar, 99.99%) and 2 M K 2 CO 3 (Alfa Aesar, %).[42] Glass fiber (GE Healthcare, Whatman TM ) was used as the separator. For start-stop operation, ZnO@TiN x O y nanorod anode and Zn foil were pre-activated. ZnO@TiN x O y nanorod anode was pre-cycled three times at 0.5C between 1.4 and 2 V. Zn foil (0.02% DOD) was firstly discharged 5

7 for 2h and re-charged for 2h at the constant current of 1.35 ma. Then it was discharged twice and charged once at the same time interval of 1 h at 1.35 ma. Zn foil (1% DOD) was pre-cycled twice at the constant current of 1 ma. When Zn foil was used as the anode, the cathode harvested from commercial Ni-Zn AA batteries was electrochemically oxidized to 0.6 V vs HgO/Hg reference electrode in a beaker cell with 2 M KOH as electrolyte. Pouch cell: Pouch-type batteries (Supplementary Fig. S11) were assembled using Ampac s SealPAK, our zinc anodes (round disk, 1 cm diameter) and Ni(OH) 2 cathodes with excess capacity. The aqueous electrolyte consists of 4 M KOH (Sigma Aldrich, 99.99%), 2 M KF (Alfa Aesar, 99.99%) and 2 M K 2 CO 3 (Alfa Aesar, %).[42] Celgard 3501 (close to anode) and Freudenberg 700/28K (close to cathode)[26] were used as the separators. Beaker cell: Beaker-type batteries (Supplementary Fig. S8) were assembled using beakers, our zinc anodes (round disk, 1 cm diameter) and Ni(OH) 2 cathodes with excess capacity. 10 ml ZnO saturated 4M KOH (Sigma Aldrich) was used as the electrolyte. Electrochemical impedance spectroscope (EIS) measurements were performed on a Bio-Logic instrument. The frequency range was between 100 KHz and 10 mhz. The amplitude of AC signal was 10 mv. Coin-type batteries, assembled using our zinc anodes and Ni(OH) 2 cathodes were used to measure EIS. 100 μl 4 M KOH, 2 M KF and 2 M K 2 CO 3 electrolyte was added to glass fiber separator. The charge capacity was limited to the theoretical capacity of ZnO (658 mah/g). The theoretical specific capacity (charge capacity) was calculated by C T (mah g 1 ) = 1 MW nf 3.6 where MW is the molar weight of active material, n is the number of electrons transferred in the relevant reaction, and F is the Faraday s constant. The specific discharge capacity of the electrode was calculated by C = It/m where I is the discharge current, t is the discharge time per cycle, and m is active materials mass. 3. Results and Discussion ZnO nanorods are synthesized on carbon paper with mass loading ranging from 0.5 to 5.5 mg/cm 2 (Fig. 2a-c) by adjusting the area of carbon paper placed in hydrothermal reactor, NH 3 concentration, hydrothermal time, etc. (Supplementary Table. S1). In ALD process (Supplementary Fig. S2), the recipe was run 100 or 200 cycles. The TiN x O y mass loadings of 100 cycles and 200 cycles are about mg/cm 2 and 0.19 mg/cm 2, which are only 0.6 wt% and 6

8 1.9 wt% of x O y nanorod anode, respectively (for 3 mg/cm 2 ZnO nanorods). The nanorod morphology does not change after TiN x O y coating (Fig. 2d and Supplementary Fig. S3). High-resolution transmission electron microscopy (HRTEM) images show uniform TiN x O y coating with a thickness of 6.1 nm for ZnO@TiN x O y nanorod with 100 cycles ALD (Fig. 2e, f). ZnO nanorod is hexagonal and TiN x O y coating is amorphous, which are evident from TEM (Fig. 2g, h) and X-ray diffraction (XRD) results (Fig. 3f). In addition to TEM, X-ray photoelectron spectroscopy (XPS) results also indicate complete coverage of TiN x O y on ZnO (Fig. 2i, j and Supplementary Fig. S4), which is crucial for encapsulating zincate during cycling. Besides, nitrogen peak in the XPS survey spectra (Fig. 2i) and three Ti 2p peaks in the high-resolution XPS spectra (Fig. 2k), which belong to TiO 2,[43] indicate that ALD TiN is partially oxidized to TiN x O y. The TiN x O y coating, although only a few nanometers thick, firmly supports the ZnO nanorod, blocks zincates, and enables OH - /H 2 O to pass through. As shown in Fig. 3a, we soaked a ZnO@TiN x O y anode and an uncoated ZnO anode (1 cm diameter disk) into two tubes with 2 ml 4M KOH solution, respectively. Then we measured the dissolved Zn concentration in both solutions using inductively coupled plasma atomic emission spectroscopy (ICP-AES). The dissolved Zn of the ZnO@TiN x O y anode is much lower than that of the uncoated ZnO anode, which means that TiN x O y coating effectively blocks zincates. Furthermore, we assembled coin cells with these two anodes and Ni(OH) 2 cathodes to investigate the influence of the TiN x O y coating during electrochemical cycling. After 2h charge (1h constant current at 1C rate and 1h constant voltage at 1.93 V), uncoated ZnO nanorod anode shows severely morphological degradation, almost all ZnO nanorods detach from carbon paper because of dissolution (Fig. 3b). On the other hand, there is no obvious shape change for ZnO@TiN x O y nanorod anode with 100 cycles ALD (Fig. 3c). ZnO@TiN x O y nanorod anode with 200 cycles ALD maintained its morphology after charging as well (Supplementary Fig. S5). Additional SEM image, elemental mapping (Fig. 3d), TEM image (Fig. 3e), and Zn peaks in XRD results (Fig. 3f) all support that TiN x O y coating confined zinc active material inside, while still allowing it to participate in electrochemical reaction. The mass loading of ZnO nanorods shown in Fig. 3b-e is 0.5 mg/cm 2. For longer nanorods with higher mass loading (~1.7 mg/cm 2 ), there is also no apparent shape change of ZnO@TiN x O y nanorod anode after the first charge and discharge (Supplementary Fig. S6), while ZnO nanorod anode shows severe structure degradation after the first charge (Supplementary Fig. S7). Electrochemical impedance spectroscopy (EIS) was also employed to investigate the electrochemical influence of conductive TiN x O y coating. As shown in the Nyquist plot (Fig. 3g) and equivalent circuit (Randles-Ershler impedance)[44], the charge-transfer resistance (R ct ) of ZnO@TiN x O y nanorod anode (0.8 Ω) is much lower than that of the uncoated ZnO anode (1.8 Ω). This can be attributed to the good conductivity of TiN x O y [45] and the high zincate concentration inside the TiN x O y coating. 7

9 Figure 2 Fabrication and characterization of anodes. a, Schematic of the fabrication process for ZnO@TiN x O y core/shell nanorod anode. b, Low magnification SEM image of ZnO nanorod anode. c, High magnification SEM images of three ZnO nanorod anodes with different mass loadings. d, SEM image of ZnO@TiN x O y nanorod anode. e, TEM image of a ZnO@TiN x O y nanorod. f, HRTEM image of ZnO@100TiN x O y nanorod, showing thickness (6.1 nm) of TiN x O y coating. g, HRTEM image of ZnO@TiN x O y nanorod, showing lattice of ZnO, [002], d= 0.26 nm. h, Electron diffraction pattern of ZnO@TiN x O y nanorod, showing diffraction pattern of ZnO. A [002]; B [200]; C [202]. i, XPS survey of ZnO nanorod and ZnO@TiN x O y nanorod anodes. j, High-resolution XPS spectra of Zn 2p peaks. k, High-resolution XPS spectra of Ti 2p peaks. The samples shown in Fig. 2d-2k are all deposited by ALD for 100 cycles. 8

10 Figure 3 Investigation on the influence of TiN x O y coating on the electrochemical performance of zinc anodes. a, ICP results and image (inset) showing dissolved Zn concentration after soaking the ZnO@TiN x O y and uncoated ZnO anodes in 4M KOH solution. b,c, SEM images of uncoated ZnO nanorod anode (b) and ZnO@TiN x O y nanorod anode (c) before and after 2h charge with 25 μl electrolyte. d, SEM image and elemental mapping of ZnO@TiN x O y nanorod anode after 2h charge. e, TEM image of ZnO@TiN x O y nanorod anode after 2h charge. Fig. 3c-e are from the same anode sample with 100 cycles ALD. f, XRD results of ZnO nanorod and ZnO@TiN x O y nanorod anode before and after charge. The weak ZnO peaks of ZnO@TiN x O y nanorod anode with 200 cycles ALD after charge is from residual unreacted ZnO. Tin foils were used as anode current collectors.[46] g, EIS result and equivalent circuit of uncoated ZnO anode and ZnO@TiN x O y nanorod anode. Z w : Warburg impedance; R ct : charge-transfer resistance; CPE: double layer capacity; R e : total ohmic resistance. Zinc anodes were tested in coin-type cells with lean zinc-free electrolyte here to evaluate their 9

11 real performance (Supplementary Fig. S8), which is different from most of previous investigations[34,36,38,47 53] using beaker cells with a large amount of ZnO saturated electrolyte. When testing the performance of zinc anodes in ZnO saturated electrolyte, all the zincates in the electrolyte could also participate in electrochemical reactions and contribute capacity in addition to the Zn in the anode. However, most previous works reported their specific capacity without counting Zn in the electrolyte, which gives pseudo performance. In an extreme case, even if there is no active material in the anode, only a pure current collector can cycle in ZnO saturated electrolyte. Here we tested two pure carbon fiber paper substrates (1 cm diameter disk) in a beaker cell with 10 ml ZnO saturated 4M KOH electrolyte and a coin cell with 100 ul ZnO saturated 4M KOH electrolyte (Fig. 4a), respectively. They were both cycled at 1 ma with a limiting charge capacity of 1 mah and cut-off voltage of 1.4/2V. The theoretical specific capacity of ZnO is 658 mah/g. As shown in Fig. 4b, if zinc in the electrolyte is not counted, the pure current collector without any active material in the beaker cell can show excellent cycling performance with a pseudo specific discharge capacity of 600 mah/g (Assume there is 1.5 mg ZnO on the anode, which has 1 mah theoretical capacity). However, when zinc in the electrolyte (see Supplementary Table. S2 for equivalent ZnO mass) is counted, the real specific discharge capacity in the beaker cell is only 4 mah/g, while that in the coin cell is ~150 mah/g. These results show that testing zinc anodes in beaker cells with a large amount of ZnO saturated electrolyte will dramatically decrease the overall specific energy. Moreover, it s hard to evaluate the true performance of anodes with a lot of capacity contributed from electrolyte. Coin-type cells use minimum amount of electrolyte and have a higher volumetric capacity compared with beaker cells, which is closer to practical operating condition. Thus, even though the coin cell with lean electrolyte is an extreme harsh testing environment (~25 cycles), we still chose it to test our materials. To evaluate the true performance of anodes, we used ZnO-free electrolytes because capacity contributed by electrolytes could cause an unrealistically high capacity of anodes. We chose 4M KOH, 2M KF and 2M K 2 CO 3 as electrolyte if not otherwise stated because it has lower Zn(OH) 4 2 solubility and thus less shape change than 4 M KOH as evidenced by ICP results (Supplementary Table. S3). As shown in Fig. 4c, the ZnO@TiN x O y nanorod anode affords remarkable battery performance in lean electrolyte configuration even with a high mass loading of active materials (~2.1 mg/cm 2 ). The coin cells are galvanostatically deep-cycled to 100% state of charge (SOC) with 1.5 and 1.9 V as cut-off voltages. The charge capacity was limited to the theoretical capacity of ZnO (658 mah/g). The reversible discharge capacity of ZnO@TiN x O y nanorod anode at tenth galvanostatic cycle is 279 mah g -1, which is twice as large as that of uncoated ZnO nanorod anode (148 mah g -1 ) at a rate of C/2. The discharge capacity of ZnO@TiN x O y nanorod anode decays to below 150 mah g -1 after 30 cycles, versus only 9 cycles for uncoated ZnO nanorod anode. For the ZnO@TiN x O y anode, the TiN x O y coating did not change the overpotential of ZnO anode with almost the same charge profile as the uncoated ZnO (Supplementary Fig. S9). Cycled at a lower rate (C/4) with 50% SOC, the capacity difference between coated and uncoated ZnO nanorod anodes in initial cycles is small, yet ZnO@TiN x O y nanorod anode shows better capacity retention 10

12 than uncoated ZnO nanorod anode (Supplementary Fig. S10). The discharge capacities of uncoated and sealed ZnO nanorod anodes decay to 50% after 31 and 53 cycles, respectively. To probe the behavior of ZnO over cycling, we imaged the electrodes after three galvanostatic cycles at 0.33C. As can be seen in Supplementary Fig. S11, x O y nanorod anode keeps its original morphology after cycling, whereas almost no nanorods can be found on the carbon paper for the uncoated ZnO nanorod anode. The superior performance of ZnO@TiN x O y nanorod anode can be attributed to the small feature size of ZnO and conformal TiN x O y coating. Below the critical passivation thickness, anode passivation problem is eliminated. And the TiN x O y coating serves as an electrical pathway, confines large zincate molecules yet allows OH - and water to pass. As a mechanical backbone, TiN x O y coating protects ZnO nanorods from detaching from carbon paper substrate and thus provides a short zincate mass transfer path for the reaction. Without TiN x O y coating, ZnO nanorod will detach from substrate upon charging (Fig. 3a). This on one hand leads to a much slower mass transport for electrically disconnected ZnO to dissolve in electrolyte to form zincate and then diffuse to current collector. On the other hand, detached ZnO or dissolved zincate may migrate far from anode and never participate in further cycling. Both mechanisms will cause capacity decay over cycling. The utilization of Zn could be potentially modeled by quantitative comparison of electroreduction rate and mass transfer rate of Zn(OH) 4 2. The ZnO@TiN x O y nanorod anode was also tested in pouch cell with ZnO-free electrolyte (Supplementary Fig. S12). It achieves a specific discharge capacity of 408 mah/g (based on ZnO if not otherwise stated), which is 508 mah/g(zn). As shown in Fig. 4d, our ZnO@TiN x O y nanorod anode demonstrates a much higher specific capacity than that of many previously reported zinc anodes (see Supplementary Table. S4 for calculation).[26,34,54,35,38,39,47 51] We also tested our ZnO@TiN x O y nanorod anode in beaker cell (Fig. 4e) and pouch cell (Supplementary Fig. S12) with ZnO saturated electrolyte, which achieves a discharge capacity of ~550 mah/g for >640 cycles (64 days) and 50 cycles, respectively. Fig. 4f shows the cyclic voltammogram (CV) of the ZnO@TiN x O y nanorod anode in coin cell with ZnO free electrolyte. CV of the ZnO@TiN x O y nanorod anode in pouch and beaker cells is shown in Supplementary Fig. S13. In addition, our ZnO@TiN x O y nanorod anode also has excellent performance under start-stop operations, demonstrating potential to replace lead acid batteries in micro-hybrid vehicles[55]. Engine restart, rest and pulse discharge are involved in the start-stop operation.[26] The procedure of test is showed in Fig. 4g. The capacity of ZnO@TiN x O y nanorod anode was kept at 1% depth of discharge (DOD) per duty cycle. The ZnO@TiN x O y nanorod anode maintained 100% discharge capacity for more than 7,500 cycles (Fig. 4h) at 1% DOD. Voltage profile of the ZnO@TiN x O y nanorod anode was shown in Supplementary Fig. S14. Under the same current density (Fig. 4g), Zn foil died after 3,400 cycles, which is less than half of cycle number of ZnO@TiN x O y nanorod anode, even though the DOD of Zn foil is only 0.02% (1/50 of that of ZnO@TiN x O y nanorod anode). This Zn foil cell died with a sudden voltage drop to <0 V because Zn is completely passivated by ZnO. And the cell was severely swelled, possibly due to 11

13 accumulation of hydrogen evolved on the anode (Supplementary Fig. S14). Severe hydrogen evolution occurs after the passivation of Zn anode. The cell with x O y nanorod anode only slightly swelled, which indicated less side reaction and higher utilization of zinc. Less swelling further indicates our ZnO@TiN x O y nanorod anode does not passivate and retains its activity over thousands of cycles. Besides, we also tested Zn foil start stop performance with 1% DOD at the same time interval as shown in Fig. 4g, which showed dramatically discharge capacity decay (Fig. 4h). This result indicates the high stability of ZnO@TiN x O y nanorod anode. ZnO@TiN x O y nanorod anode also demonstrated stable cycling at different cycling rates from 0.25C to 4C (Supplementary Fig. S15). Figure 4 Electrochemical performance of zinc anodes. a, Schematic diagram of beaker cell and coin cell. b, Cycling performance of pure current collector in the beaker cell and coin cell. Real: count all the zinc in the electrolyte. Pseudo: calculation without counting zinc in the 12

14 electrolyte. One dot every four data points. c, Discharge capacity for the first 32 galvanostatic cycles of uncoated ZnO nanorod and x O y nanorod anodes with ~2.1 mg/cm 2 at 0.5C rate in coin cell with ZnO-free electrolyte. 50 μl electrolyte was dropped onto separator and 10 μl electrolyte was dropped onto cathode. Inset: optical image of a coin cell. d, Comparison of specific discharge capacity between our anode and previously reported anodes. Zinc in the electrode and electrolyte are both counted. e, Cycling performance of our ZnO@TiN x O y nanorod anode (2 mg/cm 2 ) with 200 cycles ALD at 0.5C charge and 2C discharge rates in beaker cell with 10 ml ZnO saturated 4M KOH electrolyte. The cut-off voltages are 1.4/2V. One dot every five data points. f, Cyclic voltammogram for ZnO@TiN x O y anode in coin cell at 0.1 mv s -1 scan rate. The CV was done using two electrodes with ZnO@TiN x O y anode and Ni(OH) 2 cathode in ZnO free electrolyte. g, Current density profile as cycled under start-stop conditions in coin cells. h, Long-term discharge capacity retention of Zn foil and ZnO@TiN x O y nanorod anode as cycled under start-stop conditions with 100 μl electrolyte. Zn foil with 0.02% DOD and ZnO@TiN x O y nanorod anode with 1% DOD were cycled at the same current density, which is shown in Fig. 4g. Tin foils were used as anode current collectors. Cells were cycled between 0 and 2 V. 4. Conclusions In summary, our ZnO@TiN x O y nanorod anode achieves very high specific discharge capacity and superior reversibility when testing in a coin cell with lean ZnO-free electrolyte. In commercial PowerGenix AA batteries, which consists Zn metal anode and NiOOH cathode, the discharge capacity decayed to 50% of its initial capacity after only 9 cycles (0.5C, 20 C, charged to 105% theoretical capacity).[46] NiOOH cathodes are very reversible[56], and the Zn anode is the main cause of the poor reversibility. Our ZnO@TiN x O y core/shell nanorod anode structure reported here successfully solves the problems of ZnO passivation and zincate dissolution simultaneously, and significantly improved the cycle life of Zn anode. Because of electrolyte consumption and bubble accumulation resulted from hydrogen evolution side reaction, anodes degraded ultimately when cycled in coin cells with lean electrolyte. We believe this work can be further improved by coating hydrogen evolution suppressive materials. In addition, the mechanistic understanding and design principles could provide guidance to future design of zinc and other metal anodes (e.g. Li, Na, Mg, Ca). Our work prepares the path towards rechargeable Zn-air aqueous batteries, and other rechargeable, high-energy and safe batteries. Additional information Supplementary information is available in Supplementary Information file. Supplementary video 13

15 Acknowledgements N. L. acknowledges support from faculty startup funds from the Georgia Institute of Technology. The authors thank S. Zhuo (from KAUST) for discussions, R. Lively, S. Nair, E. Woods, T. Walters, and C. Yang for providing experiment and characterization help. Dexmet Corporation is acknowledged for providing Zn and Ni MicroGrid Metals. Material characterization was performed in part at the Georgia Tech Institute for Electronics and Nanotechnology, a member of the National Nanotechnology Coordinated Infrastructure, which is supported by the National Science Foundation (Grant ECCS ). Author contributions Y.Z. and N.L. conceived the concept and experiments. Y.Z. carried out the material synthesis, characterization and electrochemical measurements. Y.W. conducted the ALD technique. H.D., Y.Y. and Z.Z. participated in part of the synthesis. Y.D. participated in TEM characterization. Y.Z. and N.L. co-wrote the paper. Competing interests The authors declare no competing interests. References and Notes [1] B. Obama, The irreversible momentum of clean energy, Science (80-. ). 355 (2017) doi: /science.aam6284. [2] S. Chu, Y. Cui, N. Liu, The path towards sustainable energy, Nat. Mater. 16 (2017) doi: /nmat4834. [3] Y. Zhang, N. Liu, Nanostructured Electrode Materials for High-Energy Rechargeable Li, Na and Zn Batteries, Chem. Mater. 29 (2017) doi: /acs.chemmater.7b [4] Y. Wu, N. Liu, Visualizing Battery Reactions and Processes by Using In Situ and In Operando Microscopies, Chem. 4 (2018) doi: /j.chempr [5] N. Liu, Z. Lu, J. Zhao, M.T. McDowell, H.-W. Lee, W. Zhao, Y. Cui, A pomegranate-inspired nanoscale design for large-volume-change lithium battery anodes, Nat. Nanotechnol. 9 (2014) doi: /nnano [6] Y. Shi, Y. Zhang, L. Liu, Z. Zhang, J. Wang, S. Chou, J. Gao, H.D. Abruña, H. Li, H. Liu, D. Wexler, J. Wang, Y. Wu, Rapid hydrothermal synthesis of Li3VO4with different favored facets, J. Solid State Electrochem. 21 (2017) doi: /s

16 [7] Y. Jin, B. Zhu, Z. Lu, N. Liu, J. Zhu, Challenges and Recent Progress in the Development of Si Anodes for Lithium-Ion Battery, Adv. Energy Mater. 7 (2017) doi: /aenm [8] Y. Zhang, Y. Jiang, Y. Li, B. Li, Z. Li, C. Niu, Preparation of nanographite sheets supported Si nanoparticles by in situ reduction of fumed SiOinf2/inf with magnesium for lithium ion battery, J. Power Sources. 281 (2015) doi: /j.jpowsour [9] Y. Zhang, Y. Zhu, L. Fu, J. Meng, N. Yu, J. Wang, Y. Wu, Si/C Composites as Negative Electrode for High Energy Lithium Ion Batteries, Chinese J. Chem. 35 (2017) doi: /cjoc [10] L. Dingchang, Y.P. Yan, L. Yayuan, L. Wei, L. Nian, D.R. H., C. Yi, A Silica-Aerogel-Reinforced Composite Polymer Electrolyte with High Ionic Conductivity and High Modulus, Adv. Mater. 0 (2018) doi: /adma [11] Y. Liang, Y. Jing, S. Gheytani, K.Y. Lee, P. Liu, A. Facchetti, Y. Yao, Universal quinone electrodes for long cycle life aqueous rechargeable batteries, Nat. Mater. 16 (2017) doi: /nmat4919. [12] H. Kim, J. Hong, K.Y. Park, H. Kim, S.W. Kim, K. Kang, Aqueous rechargeable Li and Na ion batteries, Chem. Rev. 114 (2014) doi: /cr500232y. [13] D. Kundu, B.D. Adams, V. Duffort, S.H. Vajargah, L.F. Nazar, A high-capacity and long-life aqueous rechargeable zinc battery using a metal oxide intercalation cathode, Nat. Energy. 1 (2016). doi: /nenergy [14] P. Liu, R. Ross, A. Newman, Long-range, low-cost electric vehicles enabled by robust energy storage, MRS Energy Sustain. 2 (2015) E12. doi: /mre [15] H. Pan, Y. Shao, P. Yan, Y. Cheng, K.S. Han, Z. Nie, C. Wang, J. Yang, X. Li, P. Bhattacharya, K.T. Mueller, J. Liu, Reversible aqueous zinc/manganese oxide energy storage from conversion reactions, Nat. Energy. 1 (2016) doi: /nenergy [16] N. Zhang, F. Cheng, J. Liu, L. Wang, X. Long, X. Liu, F. Li, J. Chen, Rechargeable aqueous zinc-manganese dioxide batteries with high energy and power densities, Nat. Commun. 8 (2017) 1 9. doi: /s x. [17] J.-Y. Luo, W.-J. Cui, P. He, Y.-Y. Xia, Raising the cycling stability of aqueous lithium-ion batteries by eliminating oxygen in the electrolyte, Nat. Chem. 2 (2010) doi: /nchem.763. [18] M. Pasta, C.D. Wessells, N. Liu, J. Nelson, M.T. McDowell, R.A. Huggins, M.F. Toney, Y. Cui, Full open-framework batteries for stationary energy storage, Nat. Commun. 5 (2014) doi: /ncomms4007. [19] J. Fu, Z.P. Cano, M.G. Park, A. Yu, M. Fowler, Z. Chen, Electrically Rechargeable Zinc Air Batteries: Progress, Challenges, and Perspectives, Adv. Mater. 29 (2017). doi: /adma [20] Y. Li, H. Dai, Recent advances in zinc air batteries, Chem. Soc. Rev. 43 (2014) doi: /c4cs00015c. 15

17 [21] F. Wang, O. Borodin, T. Gao, X. Fan, W. Sun, F. Han, A. Faraone, J.A. Dura, K. Xu, C. Wang, Highly reversible zinc metal anode for aqueous batteries, Nat. Mater. 17 (2018) doi: /s z. [22] K. Wan, Z. Yu, X. Li, M. Liu, G. Yang, J. Piao, Z. Liang, ph Effect on Electrochemistry of Nitrogen-Doped Carbon Catalyst for Oxygen Reduction Reaction, ACS Catal. 5 (2015) doi: /acscatal.5b [23] S. Rojas-Carbonell, K. Artyushkova, A. Serov, C. Santoro, I. Matanovic, P. Atanassov, Effect of ph on the Activity of Platinum Group Metal-Free Catalysts in Oxygen Reduction Reaction, ACS Catal. 8 (2018) doi: /acscatal.7b [24] G. Jürmann, D.J. Schiffrin, K. Tammeveski, The ph-dependence of oxygen reduction on quinone-modified glassy carbon electrodes, Electrochim. Acta. 53 (2007) doi: /j.electacta [25] Y. Li, M. Gong, Y. Liang, J. Feng, J.E. Kim, H. Wang, G. Hong, B. Zhang, H. Dai, Advanced zinc-air batteries based on high-performance hybrid electrocatalysts, Nat. Commun. 4 (2013) doi: /ncomms2812. [26] J.F. Parker, C.N. Chervin, I.R. Pala, M. Machler, M.F. Burz, J.W. Long, D.R. Rolison, Rechargeable nickel 3D zinc batteries: An energy-dense, safer alternative to lithium-ion, Science (80-. ). 356 (2017) doi: /science.aak9991. [27] S.H. Lee, C.W. Yi, K. Kim, Characteristics and electrochemical performance of the TiO 2-coated ZnO anode for Ni-Zn secondary batteries, J. Phys. Chem. C. 115 (2011) doi: /jp110308b. [28] Y.F. Yuan, L.Q. Yu, H.M. Wu, J.L. Yang, Y.B. Chen, S.Y. Guo, J.P. Tu, Electrochemical performances of Bi based compound film-coated ZnO as anodic materials of Ni-Zn secondary batteries, Electrochim. Acta. 56 (2011) doi: /j.electacta [29] S. Higashi, S.W. Lee, J.S. Lee, K. Takechi, Y. Cui, Avoiding short circuits from zinc metal dendrites in anode by backside-plating configuration, Nat. Commun. 7 (2016) 1 6. doi: /ncomms [30] N.D. Ingale, J.W. Gallaway, M. Nyce, A. Couzis, S. Banerjee, Rechargeability and economic aspects of alkaline zinc-manganese dioxide cells for electrical storage and load leveling, J. Power Sources. 276 (2015) doi: /j.jpowsour [31] Z. Zhou, Y. Zhang, P. Chen, Y. Wu, H. Yang, H. Ding, Y. Zhang, Z. Wang, X. Du, N. Liu, Graphene oxide-modified zinc anode for rechargeable aqueous batteries, Chem. Eng. Sci. (2018). doi: /j.ces [32] J.F. Parker, E.S. Nelson, M.D. Wattendorf, C.N. Chervin, J.W. Long, D.R. Rolison, Retaining the 3D framework of zinc sponge anodes upon deep discharge in Zn-Air cells, ACS Appl. Mater. Interfaces. 6 (2014) doi: /am505266c. [33] J.F. Parker, C.N. Chervin, E.S. Nelson, D.R. Rolison, J.W. Long, Wiring zinc in three dimensions re-writes battery performance dendrite-free cycling, Energy Environ. Sci. 7 (2014) doi: /c3ee43754j. [34] J. Long, Z. Yang, Z. Zhang, J. Huang, Sheet-Like Carbon-Coated Zn-Al-Bi Layered 16

18 Double Oxides Nanocomposites Enabling High Performance for Rechargeable Alkaline Batteries, J. Electrochem. Soc. 164 (2017) A3068 A3074. doi: / jes. [35] L. Sun, Z. Yi, J. Lin, F. Liang, Y. Wu, Z. Cao, L. Wang, Fast and Energy Efficient Synthesis of and its Application in Ni-Zn Secondary Battery, J. Phys. Chem. C. 120 (2016) doi: /acs.jpcc.6b [36] X. Zeng, Z. Yang, F. Liu, J. Long, Z. Feng, M. Fan, An in situ recovery method to prepare carbon-coated Zn Al hydrotalcite as the anode material for nickel zinc secondary batteries, RSC Adv. 7 (2017) doi: /c7ra08622a. [37] S. Lee, C. Yi, K. Kim, Characteristics and Electrochemical Performance of the TiO2-Coated ZnO Anode for Ni Zn Secondary Batteries, J. Phys. Chem. C. 115 (2011) doi: /jp110308b. [38] M. Ma, J.P. Tu, Y.F. Yuan, X.L. Wang, K.F. Li, F. Mao, Z.Y. Zeng, Electrochemical performance of ZnO nanoplates as anode materials for Ni/Zn secondary batteries, J. Power Sources. 179 (2008) doi: /j.jpowsour [39] J. Liu, C. Guan, C. Zhou, Z. Fan, Q. Ke, G. Zhang, C. Liu, J. Wang, A Flexible Quasi-Solid-State Nickel-Zinc Battery with High Energy and Power Densities Based on 3D Electrode Design, Adv. Mater. 28 (2016) doi: /adma [40] S.-W. Bian, I.A. Mudunkotuwa, T. Rupasinghe, V.H. Grassian, Aggregation and Dissolution of 4 nm ZnO Nanoparticles in Aqueous Environments: Influence of ph, Ionic Strength, Size, and Adsorption of Humic Acid, Langmuir. 27 (2011) doi: /la200570n. [41] J. Wang, H. Zhong, Z. Wang, F. Meng, X. Zhang, Integrated Three-Dimensional Carbon Paper/Carbon Tubes/Cobalt-Sulfide Sheets as an Efficient Electrode for Overall Water Splitting, ACS Nano. 10 (2016) doi: /acsnano.5b [42] J.F. Parker, I.R. Pala, C.N. Chervin, J.W. Long, D.R. Rolison, Minimizing Shape Change at Zn Sponge Anodes in Rechargeable Ni Zn Cells: Impact of Electrolyte Formulation, J. Electrochem. Soc. 163 (2016) A351 A355. doi: / jes. [43] Thermo Fisher Scientific Inc., Supplemental material, (n.d.). [44] J. Long, Z. Yang, J. Huang, X. Zeng, Self-assembly of exfoliated layered double hydroxide and graphene nanosheets for electrochemical energy storage in zinc/nickel secondary batteries, J. Power Sources. 359 (2017) doi: /j.jpowsour [45] J. Zhang, Y. Li, Y. Zhang, X. Qian, R. Niu, R. Hu, X. Zhu, X. Wang, J. Zhu, The enhanced adhesion between overlong TiNxOy/MnO2nanoarrays and Ti substrate: Towards flexible supercapacitors with high energy density and long service life, Nano Energy. 43 (2018) doi: /j.nanoen [46] P. Bonnick, J.R. Dahn, A Simple Coin Cell Design for Testing Rechargeable Zinc-Air or Alkaline Battery Systems, J. Electrochem. Soc. 159 (2012) A981 A989. doi: / jes. [47] D. Zeng, Z. Yang, S. Wang, X. Ni, D. Ai, Q. Zhang, Preparation and electrochemical 17

19 performance of In-doped ZnO as anode material for Ni-Zn secondary cells, Electrochim. Acta. 56 (2011) doi: /j.electacta [48] J. Yu, H. Yang, X. Ai, X. Zhu, A study of calcium zincate as negative electrode materials for secondary batteries, J. Power Sources. 103 (2001) doi: /s (01) [49] L. Wang, Y. Liu, X. Chen, H. Qin, Z. Yang, Zinc Aluminum Antimony Hydrotalcite as Anode Materials for Ni-Zn Secondary Batteries, J. Electrochem. Soc. 164 (2017) A3692 A3698. doi: / jes. [50] B. Yang, Z. Yang, R. Wang, Facile synthesis of novel two-dimensional silver-coated layered double hydroxide nanosheets as advanced anode material for Ni-Zn secondary batteries, J. Power Sources. 251 (2014) doi: /j.jpowsour [51] Y. Liu, Z. Yang, X. Xie, J. Huang, X. Wen, Layered Double Oxides Nano-flakes Derived From Layered Double Hydroxides: Preparation, Properties and Application in Zinc/Nickel Secondary Batteries, Electrochim. Acta. 185 (2015) doi: /j.electacta [52] B. Yang, Z. Yang, R. Wang, Z. Feng, Silver nanoparticle deposited layered double hydroxide nanosheets as a novel and high-performing anode material for enhanced Ni Zn secondary batteries, J. Mater. Chem. A. 2 (2014) doi: /c3ta14237j. [53] B. Yang, Z. Yang, Structure and improved electrochemical performance of a nanostructured layered double hydroxide carbon nanotube composite as a novel anode material for Ni Zn secondary batteries, RSC Adv. 3 (2013) doi: /c3ra41069b. [54] S. Zheng, P. Han, Z. Han, P. Li, H. Zhang, J. Yang, Nano-copper-assisted immobilization of sulfur in high-surface-area mesoporous carbon cathodes for room temperature Na-S batteries, Adv. Energy Mater. 4 (2014) 1 7. doi: /aenm [55] J. Albers, E. Meissner, S. Shirazi, Lead-acid batteries in micro-hybrid vehicles, J. Power Sources. 196 (2011) doi: /j.jpowsour [56] X.-Z. Fu, Q.-C. Xu, R.-Z. Hu, B.-X. Pan, J.-D. Lin, D.-W. Liao, β-coooh coated spherical β-niooh as the positive electrode material for alkaline Zn-NiOOH battery, J. Power Sources. 164 (2007) doi: /j.jpowsour

20 Yamin Zhang received her bachelor's degree in Chemical Engineering and Technology from Tianjin University (Tianjin, China) in Now she is a Ph.D. candidate under the supervision of Prof. Nian Liu in the School of Chemical & Biomolecular Engineering, Georgia Institute of Technology. Her research interest focuses on the development of rechargeable high-energy zinc aqueous batteries and investigation of CO 2 electrochemical reduction. Yutong Wu is currently a Ph.D. student in Dr. Nian Liu s lab at Georgia Institute of Technology, focusing on Zn-based aqueous batteries and in operando imaging of electrochemical reactions. He received his B.S. in 2016 from chemical and biological engineering, Iowa State University. 19

21 Haoran Ding received his Bachelor s Degree in Applied Chemistry from Tianjin University in 2018 and is now pursuing his Ph.D. degree in Chemical Engineering in University of Delaware. He was a summer research student at Dr. Nian Liu s lab in Department of Chemical and Biomolecular Engineering, Georgia Institute of Technology. His research interests include electrochemistry, electrochemical energy storage materials and nanotechnology. Yu Yan will receive his B.S. degree in 2019 from Department of Physics, University of Science and Technology of China. He was a student intern under the supervision of Prof. Nian Liu in School of Chemical & Biomolecular Engineering, Georgia Institute of Technology. He is currently an undergraduate student under the supervision of Prof. Jie Zeng in National Laboratory for Physical Sciences at the Microscale, University of Science and Technology of China. His research focuses on the development of nanomaterials for the catalyst and energy storage. 20

22 Zhubo Zhou is currently a third-year student at Georgia Tech majoring in Mechanical Engineering. Her research interests include the development of clean, renewable energy and the design of energy systems. Dr. Yong Ding is a senior research engineer at Georgia Tech. He received his Ph. D degree in physics from Nanjing University in His research interests mainly focus on the applications of transmission electron microscopy in materials science. 21

23 Nian Liu is an Assistant Professor at Georgia Institute of Technology, School of Chemical and Biomolecular Engineering since January He received his B.S. in 2009 from Fudan University (China), and Ph.D. in 2014 from Stanford University, where he worked with Prof. Yi Cui on the structure design for Si anodes for high-energy Li-ion batteries. In , he worked with Prof. Steven Chu at Stanford University as a postdoc, where he developed in situ optical microscopy to probe beam-sensitive battery reactions. Dr. Liu s lab at Georgia Tech is broadly interested in the combination of nanomaterials, electrochemistry, and light microscopy for understanding and addressing the global energy challenges. Dr. Liu is the recipient of the Electrochemical Society (ECS) Daniel Cubicciotti Award (2014) and American Chemical Society (ACS) Division of Inorganic Chemistry Young Investigator Award (2015). Supplementary video: Battery penetration test to investigate the safety property of organic electrolyte and aqueous electrolye 22

24 Highlights The x O y nanorod anode was uniformly prepared by two steps: hydrothermal growth of ZnO nanorods and an atomic layer deposition. This approach overcomes intrinsic passivation and dissolution problems of zinc anodes. The sealing nanorod anode shows superior specific capacity and electrochemical cycling performance. These nanoscale design principles can be applied to overcome intrinsic limitations of other battery materials that involve soluble intermediates or insulating discharge products. 23