
==== Front
Proc Natl Acad Sci U S A
Proc Natl Acad Sci U S A
PNAS
Proceedings of the National Academy of Sciences of the United States of America
0027-8424
1091-6490
National Academy of Sciences

38252819
202309981
10.1073/pnas.2309981121
research-articleResearch ArticlechemChemistry410
Physical Sciences
Chemistry
Suppressing metal corrosion through identification of optimal crystallographic plane for Zn batteries
Ren Lingxiao a b 1
Hu Zhenglin b 1
Peng Chengxin c 1 https://orcid.org/0000-0002-7643-4150

Zhang Lan d 1
Wang Nan e 1 https://orcid.org/0000-0002-2450-0689

Wang Fei f g https://orcid.org/0000-0002-2057-5130

Xia Yongyao f g https://orcid.org/0000-0001-6379-9655

Zhang Suojiang sjzhang@ipe.ac.cn
d 2 https://orcid.org/0000-0002-9397-954X

Hu Enyuan enhu@bnl.gov
e 2
Luo Jiayan jyluo@sjtu.edu.cn
a h 2
aState Key Laboratory of Metal Matrix Composites, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
bSchool of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China
cSchool of Materials and Chemistry, University of Shanghai for Science and Technology, Shanghai 200093, China
dInstitute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China
eChemistry Division, Brookhaven National Laboratory, Upton, NY 11973
fDepartment of Chemistry, Fudan University, Shanghai 200433, China
gDepartment of Materials Science, Fudan University, Shanghai 200433, China
hZhangjiang Institute for Advanced Study, Shanghai Jiao Tong University, Shanghai 200240, China
2To whom correspondence may be addressed. Email: sjzhang@ipe.ac.cn, enhu@bnl.gov, or jyluo@sjtu.edu.cn.
Edited by Alexis Bell, University of California, Berkeley, CA; received June 13, 2023; accepted December 1, 2023

1L.R., Z.H., C.P., L.Z., and N.W. contributed equally to this work.

22 1 2024
30 1 2024
22 7 2024
121 5 e230998112113 6 2023
01 12 2023
Copyright © 2024 the Author(s). Published by PNAS.
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND).

Significance

Efficient zinc metal stripping and plating is the key for highly reversible Zn batteries. The current belief is that epitaxy growth with (002) texture is the best possible solution as (002) plane has the smallest surface energy. Here, we examine the conventional theory by making a series of Zn anodes with tuned texture. We found that while monocrystalline (002) faceted Zn electrode indeed promises the highest critical current density, the (100) faceted electrode with vertically aligned deposits is the most important one in suppressing Zn metal corrosion and promising the best reversibility resulting from the lowest electrochemical surface area of (100) faceted electrode. Such fundamental understanding of structure–property relationships may bring breakthrough toward advanced battery materials design.

Direct use of metals as battery anodes could significantly boost the energy density, but suffers from limited cycling. To make the batteries more sustainable, one strategy is mitigating the propensity for metals to form random morphology during plating through orientation regulation, e.g., hexagonal Zn platelets locked horizontally by epitaxial electrodeposition or vertically aligned through Zn/electrolyte interface modulation. Current strategies center around obtaining (002) faceted deposition due to its minimum surface energy. Here, benefiting from the capability of preparing a library of faceted monocrystalline Zn anodes and controlling the orientation of Zn platelet deposits, we challenge this conventional belief. We show that while monocrystalline (002) faceted Zn electrode with horizontal epitaxy indeed promises the highest critical current density, the (100) faceted electrode with vertically aligned deposits is the most important one in suppressing Zn metal corrosion and promising the best reversibility. Such uniqueness results from the lowest electrochemical surface area of (100) faceted electrode, which intrinsically builds upon the surface atom diffusion barrier and the orientation of the pallets. These new findings based on monocrystalline anodes advance the fundamental understanding of electrodeposition process for sustainable metal batteries and provide a paradigm to explore the processing–structure–property relationships of metal electrodes.

Zn battery
single crystal
epitaxy
surface energy
corrosion
DOE | EERE | Office of Sustainable Transportation | Vehicle Technologies Office (VTO) 100011884 DE-SC0012704 Enyuan Hu MOST | National Key Research and Development Program of China (NKPs) 501100012166 2021YFB2500100 Jiayan Luo
==== Body
pmcFor next-generation batteries, metals-including lithium (Li), sodium (Na), potassium (K), magnesium (Mg), calcium (Ca), zinc (Zn), and aluminum (Al)-have been proposed as anode candidates owing to their high energy density (1–10). To be used as rechargeable battery electrodes, the metal anodes have to be plated and stripped over hundreds of cycles reversibly (11, 12). However, the propensity of metal anodes to form random morphology during plating speeds up the capacity fading and internal short-circuit, presenting a fundamental barrier to achieve high reversibility (13). The Zn anode is of great interest for its sustainability, low toxicity, and intrinsic safety in aqueous batteries (14). More importantly, Zn electrodeposits are typically in the form of hexagonal platelets exposing the minimum surface energy Zn(002) facet due to the hexagonal close-packed (HCP) crystallographic anisotropy (SI Appendix, Fig. S1) (15), which can be used as a model system to study the regulation of electrodeposit morphology evolution.

The metal plating can be fundamentally considered as an electro-crystallization process in which both the substrate and solid electrolyte interphase (SEI) are important. Well-designed SEIs have been demonstrated to be an effective way to manipulate the deposition morphology (16). Zn deposits can be densely and vertically aligned through Zn/electrolyte interface modulation (17, 18). By providing a substrate with low lattice mismatch, the Zn platelets can be locked horizontally through electrochemical epitaxy (15, 19). The strategy of deposit orientation regulation has successfully improved Zn plating/stripping efficiency, but the apparent contradictory electrodeposition morphology presents a challenge to further improving the reversibility of metal anodes. Besides, the metal foil electrodes, which are expected to be used by themselves, are less explored in terms of their own structures than the interphases or heterogeneous substrates, although it is known that structure dictates properties.

Monocrystalline metals with grain boundaries-free structures and specific facet indices have always been featuring a lot of interesting properties in materials science. For example, monocrystalline copper exhibits higher electrical conductivity than polycrystalline copper benefitting from the elimination of electron scattering at the grain boundaries (20, 21). Single crystals Mg3(Sb3Bi)2 exhibit higher thermoelectric performance near room temperature than their polycrystalline counterparts (22). Catalytic metals with certain crystal surface have been proved to offer advantageous reaction paths (23). Epitaxial growth of thin films, e.g., boron nitride, demands large-area monocrystalline metal substrates (24). Faceted monocrystalline metal anodes have a great potential to systematically regulate the deposit orientation and to clarify the aforementioned contradiction. However, the synthesis of high-performing monocrystalline metal anodes with desirable facets is challenging, hindering a fundamental linkage between the facets and electrochemical behaviors.

Here, taking advantage of the apparent anisotropy of HCP crystals, we fabricate a library of faceted monocrystalline Zn metal foil anodes via a low-cost recrystallization technique. We find that the orientation of Zn platelet deposits can be well controlled following the Wulff construction prediction. The critical current densities (CCDs) of the monocrystalline Zn anodes are in the order: Zn(002) > Zn(100) > Zn(102) > Zn(101) > Zn(110), whereas the critical areal capacity (CAC) and electrode reversibility are in the order: Zn(100) > Zn(002) > Zn(102) > Zn(101) > Zn(110). We reveal the underlying mechanisms by considering the surface atom diffusion barrier, electrodeposit growth mode, and electrode corrosion. This fundamental understanding of electrodeposition process can be applied to other metal anodes and will advance the development of sustainable batteries.

Results and Discussion

Preparation and Characterization of Monocrystalline Zn Foils.

The as-received Zn foils, typically prepared by roll calendering of metallurgic Zn, are polycrystalline as observed in the X-ray diffraction (XRD) pattern and electron backscatter diffraction (EBSD) (Fig. 1). To prepare monocrystalline Zn foils, the Zn foils are sandwiched between two graphite blocks and heated to 500 ∘C under argon/H2 (5%) flow for remelting (the melting point of Zn is 420 ∘C ). Zn recrystallization was then done by cooling the sample to 400 ∘C . During the solidification process, the cooling rate is critical to the crystal structure. At higher cooling rate, facets with lower surface energy will grow faster than those with higher surface energy. The surface energy disparity of the main Zn facets (Fig. 1A) allows us to prepare monocrystalline Zn foils with different facets. By tuning the cooling rate from 4 to 1 ∘C/min precisely (SI Appendix, Fig. S2), we were able to obtain a library of monocrystalline Zn foils with (002), (100), (101), (102), and (110) facets.

Fig. 1. Preparation and characterization of monocrystalline Zn foils. (A) Surface energy of different Zn facets. (B) Optical image of as-received Zn and monocrystalline Zn foils. The typical size is ~10 × 1 cm2. (C) XRD pattern, (D) EBSD inverse pole figure color map and (E) (001) pole figure, (F) 2D XRD pattern of five faceted monocrystalline and as-received Zn foils.

The clamping and relative low temperature above the melting point ensure the shape of monocrystalline Zn foils largely unchanged (Fig. 1B). The sole characteristic peak identified by the facet indices is observed in the XRD patterns (Fig. 1C). The monocrystalline nature of the five recrystallized Zn foils was further confirmed by the monotony in the corresponding EBSD inverse pole figure color maps and pole figures (Fig. 1 D and E) and two dimensional XRD patterns (Fig. 1F). Inverse pole figures and XRD conducted at interval spots of a foil show that the entire foil has the same crystallographic orientation (SI Appendix, Fig. S3). The size of produced Zn foils was adjustable, merely limited by the measurements of clamping blocks and furnace.

Homoepitaxial Deposition on (002) Faceted Monocrystalline Zn.

Heteroepitaxy has been reported for locked Zn planar deposition on aligned graphene substrates prepared by shearing coating of graphene slurry (15). A semi-coherent interface will form between the deposit and substrate when their lattice mismatch ( δ=a-b/b   , where a and b are the lattice constants of the substrate and deposit, respectively) is less than 25%. The critical thickness of epitaxy tc=kQ-lnδ/δ   (where k is constant related to the Poisson’s ratio and lattice property of the deposit, Q bears upon the shear modulus among the substrate, deposit, and interface), however, drops sharply with δ and virtually exhausts when δ exceeds 3% (Fig. 2A) (25). Unlike epitaxial thin film growth which is merely at nanometer scale, the deposition thickness for metal anodes is up to tens of micrometers to obtain a practically useful capacity (e.g., 17.1 and 48.3 µm for 10 mAh/cm2 Zn and Li, respectively). High-quality single-crystal Zn disks as electrode show excellent homoepitaxy that can sustain high capacity deposition (26), but its prohibitive production cost prevents the practical application (19, 26). Here, the (002) faceted monocrystalline Zn prepared by low-cost recrystallization technique can also ensure high-quality homoepitaxy, as the grains are millimeter sized in the monocrystalline electrodes (SI Appendix, Fig. S4), a few orders of magnitude larger than the aligned graphene sheets (15) or micron grains in textured Zn electrodes (27, 28), significantly mitigating the grain boundary strain.

Fig. 2. Homoepitaxial deposition on (002) faceted monocrystalline Zn electrodes. (A) Dependence of epitaxy critical thickness on lattice mismatch, based on van der Merwe’s theory. (B) GIXRD patterns of deposits on (002) faceted monocrystalline Zn electrodes in 2 M ZnSO4 aqueous electrolyte at different capacity. (C) The intensity ratio of (002) peak derived from GIXRD patterns of deposits on (002) monocrystalline Zn electrodes at different capacity in 2 M ZnSO4 aqueous electrolyte, 1 M Zn(OTF)2 + 20 M LiTFSI water-in-salt electrolyte, and 0.25 M Zn(OTF)2 in AN electrolyte. (D) Cross-section TEM images of deposits on (002) faceted monocrystalline Zn electrodes and the corresponding fast fourier transform (FFT) patterns.

We evaluated the homoepitaxy on (002) faceted monocrystalline Zn electrodes using grazing incidence X-ray diffraction (GIXRD, SI Appendix, Fig. S5), a surface crystallographic technique to characterize the deposits. Indeed, the (002) peak dominates in the deposits plated in 2 M ZnSO4 aqueous electrolyte though other peaks gradually rose along with capacity increase (Fig. 2B). The intensity ratio of (002) peak is still up to nearly 80% at the capacity of 4 mAh cm−2. The intensity ratio could be further improved in 1 M Zn(OTF)2 + 20 M LiTFSI water-in-salt electrolyte or 0.25 M Zn(OTF)2 in acetonitrile (AN) electrolyte (Fig. 2C and SI Appendix, Fig. S6). This is due to the suppressed corrosion from aqueous electrolyte, which will be discussed later. The homoepitaxy is further verified by transmission electron microscope (TEM) and the corresponding electron diffraction patterns (Fig. 2D).

Growth Orientation Following Wulff Construction Prediction.

The monocrystals allow us to exclude the complication of other variables and focus on understanding how the facet triggers the Zn plating/stripping behavior change from the microstructure to chemical activity and its implications for the electrochemical properties. We first investigated the morphological variation of Zn deposits. In the most widely used 2 M ZnSO4 aqueous electrolyte for Zn batteries, Zn exhibits a strong tendency to deposit as platelets (13, 15), which are widely reported for polycrystalline Zn electrodes and also observed on all the monocrystalline Zn electrodes (Fig. 3A). The exposed basal plane of platelets is Zn(002) facet (SI Appendix, Fig. S1). This is because (002) is the most stable facet for its lowest surface energy. The platelets, however, are organized in apparently different patterns on the monocrystalline Zn electrodes. For example, the platelets lie down on Zn(002) electrode but stand vertically on Zn(100) and Zn(110) electrodes. On closer examination, the platelets on Zn(110) have larger lateral size than on the Zn(100). Besides, the platelets on Zn(101) and Zn(102) feature different angles with respect to their substrates, with the former being larger. The individual angles between the platelet deposits and the monocrystalline electrodes are independent of the plating factors such as current density or deposition capacity (SI Appendix, Fig. S7).

Fig. 3. Electro-crystallization on monocrystalline Zn. (A) Top-view SEM of 5 mAh cm−2 Zn deposited on monocrystalline Zn electrodes. (B) Wulff construction and (C) atomic structure schematic of HCP Zn showing the angle between (002) facet (the gray hexagonal plane) and (002), (100), (101), (102), and (110) facets.

To understand the facet-dictated plating behavior and microstructure variation of the deposits, we built Wulff construction for HCP Zn structure. The angles between (002) facet and the other facets (Fig. 3 B and C) are highly consistent with those observed between the platelet deposits and the monocrystalline substrates. This suggests that the metal plating on the monocrystalline Zn electrodes is an electro-crystallization process. The seeds embedded in monocrystalline substrates determine the deposit crystal preferential growth orientation. In comparison, the deposits are randomly oriented and unevenly distributed when the as-received polycrystalline Zn is used as the electrode (SI Appendix, Fig. S8).

Electrochemical Characterization.

The different faceted monocrystalline Zn metal anodes are expected to exhibit distinct electrochemical behaviors. Their CCDs are first investigated. Here, we define the CCD as the current density that triggers the sudden voltage drop as the sign of short-circuiting during symmetrical cells cycling under step-increased current density (29). With horizontally locked deposits, the monocrystalline Zn(002) anode exhibits the highest CCD of 80 mA cm−2 in 2 M ZnSO4 aqueous electrolyte with plating/stripping capacity of 0.5 mAh cm−2 (Fig. 4A). This value is high enough for fast battery charging/discharging. Interestingly, the CCD of the Zn(100) anode is as high as 60 mA cm−2, which is unexpected considering its deposits are vertically aligned. In contrast, the CCD of Zn(110) whose deposits are also vertically aligned is as low as 2 mA cm−2, being even lower than that of the as-received polycrystalline Zn anode (12 mA cm−2). The CCDs of Zn(101) and Zn(102) with tilted deposits are 18 and 25 mA cm−2, respectively. We further measured the CAC—the maximum areal capacity that can be deposited before short-circuiting of Zn symmetrical cells under a given current density (Fig. 4B) (30). Surprisingly, Zn(100) had the largest CAC of 13.1 mAh cm−2 at 1 mA cm−2 current density, 62% higher than Zn(002) (8.1 mAh cm−2). The CACs of Zn(101), Zn(102), Zn(110), and polycrystalline Zn are 3.6, 5.7, 2.0, and 2.9 mAh cm−2, respectively.

Fig. 4. Electrochemical characterization. (A) CCD of monocrystalline and as-received polycrystalline Zn electrodes in symmetrical cells cycling under step-increased current density. (B) CAC of monocrystalline and as-received polycrystalline Zn electrodes at 1 mA cm−2 current density. (C) Capacity retention of 10 μm monocrystalline and as-received polycrystalline Zn electrodes cycled in symmetrical cells at 20% DOD under the current density of 2.4 mA cm−2. (D) Cycling performance of Zn||MnO2 full cells with controlled N/P ratios. The thickness of monocrystalline Zn anodes for N/P = 5 is 10 μm and for N/P = 50 is 100 μm.

To evaluate the reversibility, Zn||Zn symmetrical cells using 10 μm Zn (~6 mAh cm−2) electrodes were cycled at 20% depth of discharge (DOD) under the current density of 2.4 mA cm−2, and the remaining available capacity was measured by full stripping the cycled electrodes to 0.5 V (Fig. 4C and SI Appendix, Fig. S9). The capacity maintenance of the monocrystalline Zn anodes is largely in the same order of their CACs. Zn(100) has the best reversibility, retaining 91.2% of its initial capacity. In addition, its monocrystalline property of the electrode is proved to be well maintained even after 100 deposition/dissolution cycles (SI Appendix, Figs. S10–S12). Then, 85.4%, 67.7%, and 58.7% capacity retentions are observed, respectively, for Zn(002), Zn(102), and Zn(101), but all are higher than polycrystalline as-received Zn (39.1%). Zn(110) cannot last for 100 cycles owning to its lowest CAC. This means that Zn(100) contributes to best performance in large capacity and long-cycle working conditions. We also evaluate the practicability of Zn(100) in full cell by pairing Zn anodes with the most widely used cathode α-MnO2. MnO2 was well mixed with carbon nanotubes (CNTs) (SI Appendix, Fig. S13) and 0.1 M MnSO4 was added to the aqueous electrolyte to prevent Mn dissolution. With mass loading of MnO2 (∼6.8 mg cm−2) and a small N:P ratio (5:1), the cell using a Zn(100) anode provides a high capacity retention of 85.7% over 1,000 cycles (Fig. 4D), in comparison to 62.6% using a Zn(002) anode. As expected, the cell performance with as-received Zn anode decays quickly at low N:P. Higher N:P ratio (50:1) helps to alleviate the downtrend but still inferior to the monocrystalline anodes. Charge/discharge curves of Zn/MnO2 cells are shown in SI Appendix, Fig. S14.

Electrodeposition Mode and Degradation Mechanism.

Vertically aligned and horizontally lain Zn electrodeposits have both been obtained previously and are proved highly effective in improving the reversibility. However, they are produced through different regulation strategies, making it difficult to compare the results and investigate further. Here, we are able to systematically tune the deposit orientation by electro-crystallization on a library of monocrystalline Zn electrodes. Zn(002) with horizontally laid deposits exhibits the highest CCD. Zn(100) has the largest CAC and the best reversibility via vertically aligned plating, with slightly lower CCD. Zn(110) has the lowest CCD, smallest CAC and worst reversibility although its deposits are also aligned vertically as Zn(100).

To explain the crystallographic plane-induced electrochemical behavior disparity, we first use density functional theory (DFT) to calculate the surface atom self-diffusion barrier on different Zn planes, which is considered as a descriptor for the occurrence of dendrite growth (31–33). The lowest diffusion barrier of 16 meV is found on (002) plane. The barrier increases to 115, 313, 709, and 1,396 meV for (100), (102), (101), and (110) plane, respectively (Fig. 5A), which is consistent with previous theoretical results that surface with highly mobile atom is less prone to form dendrite.

Fig. 5. Electrodeposition mode and degradation mechanism. (A) DFT calculation of the surface atom self-diffusion barrier on different Zn planes. (B) Schematic of electrodeposit growth model on different faceted monocrystalline Zn electrodes. (C) AFM images and statistics in lateral size and aspects of the collected deposits. (Scale bars are 2 μm in images.) (D) ECSA reflected by the double-layer capacitance of the cycled monocrystalline Zn electrodes at the plated stage after 50 cycles in symmetrical cells with 20% DOD. (E) Linear polarization curves and (F) XRD patterns of the electrodes after 50 cycles with 20% DOD. The thickness of Zn electrodes in (D–F) is 10 µm.

Combining the diffusion barriers with the electro-crystallization orientations as shown in Fig. 3, we speculate the varied growth modes of electrodeposits on different faceted monocrystalline Zn. It is known that HCP crystallography of Zn plays a central role in Zn electrodeposit growth by facilitating the formation of hexagonal Zn platelets exposing (002) facet with the lowest surface energy. On Zn(002), the deposits are locked in horizontal direction and grow quickly on the lateral dimension, forming thin plates with the largest lateral size/thickness aspect ratio (Fig. 5B). This is beneficial for shortening electron transport pathway and improving rate performance of Zn(002). However, the electrochemical epitaxy regulation decays with distance to the substrates (Fig. 2B) and is limited to moderate capacities (14). On Zn(100), along with the thermodynamically dominant (002) plane growth, (100) plane is also prone to develop due to its second lowest diffusion barrier, producing thick vertical aligned plates with the smallest aspect ratio and ordered (100) plane facing the separator, which are not prone to cause short-circuiting. The relatively high diffusion barriers on (102), (101), and especially (110) planes inhibit dimension expansion along direction parallel to substrate. As a result, the deposits on Zn(102), Zn(101), and Zn(110) are relatively thin plates, and the deposits’ tilt angles with the substrates increase from Zn(102) to Zn(101) and to verticality on Zn(110), with the propensity to cause short-circuiting in the same order. We collected the deposits by washing them from the faceted monocrystalline Zn electrodes and imaged them with atomic force microscopy (AFM). The lateral size and aspect ratio of the platelets validated the hypothesis of the growth modes (Fig. 5C). The different growth modes elucidate the order of CAC for the monocrystalline Zn anodes: Zn(100) > Zn(002) > Zn(102) > Zn(101) > Zn(110), corresponding to a gradual decrease in the packing density.

For cycling stability, corrosion of Zn in aqueous electrolyte should be taken into account. Zn is intrinsically chemically unstable in mild acidic environment due to its reducibility (34, 35). When the as-received Zn foil is soaked in 2 M ZnSO4 aqueous electrolytes, its surface is covered by debris of corrosion products and the corresponding XRD patterns exhibited strong peaks belonging to Zn4SO4(OH)6·5H2O (JCPDS: 39-0688). Even monocrystalline Zn(002) with the lowest surface energy and inappreciable grain boundary started to get corroded after soaking for 10 d (SI Appendix, Fig. S15). The plate deposits, which are more accessible by electrolyte than the bulk substrates, can get corroded more easily. To quantitatively compare the electrolyte accessibility, we measured the electrochemical active surface area (ECSA) of the cycled electrodes at the plated stage reflected by the double-layer capacitance based on cyclic voltammetry (CV) (Fig. 5D and SI Appendix, Fig. S16) (36). The ECSA of the cycled electrodes were in the order of Zn(100) < Zn(002) < Zn(102) < Zn(101) < Zn(110), implying the deposits on Zn(100) are the most-densely packed to get rid of side reactions between electrode and electrolyte. This is further confirmed by linear polarization measurements that the corrosion potential for cycled electrodes with lower ECSA shows a positive shift and meanwhile, the corrosion current density decreases, which reveals that lower ECSA electrodes are more advantageous to reduce the degree of hydrogen evolution (Fig. 5E). We also characterize the corrosion byproduct Zn4SO4(OH)6·5H2O on cycled electrodes. XRD data in Fig. 5F verify that Zn electrode degradation rate is proportional to the corrosion degree.

Conclusion

We have demonstrated here the crystallographic engineering tool to pursue and exploit metal anodes capable of reversible plating/stripping that would otherwise be overlooked for batteries. While this work demonstrates the feasibility and the success in zinc battery system which is crucial for sustainable energy, the approach established can be universally extended to explore various metal anodes (SI Appendix, Fig. S17), offering a pathway to address the fundamental structure–property relationships and performance challenges of next-generation battery materials.

Materials and Methods

Preparation of Monocrystalline Metal Foils.

The as-received metal foil was placed on a tungsten substrate and loaded into an atmospheric-pressure chemical vapor deposition system. The system was rapidly heated to a recrystallization temperature, maintaining a few minutes, and was then cooled to a certain temperature for large crystal particle growth. The whole process occurred in an atmosphere of flowing H2 and Ar. For Zn, we heated the foil to 773 K and cooling to 673 K for grain growth. For Al, the corresponding temperatures were 1,093 K and 873 K, respectively.

Synthesis of α-MnO2/CNTs Cathode.

The cathode material, the α-MnO2/CNT composite, was synthesized by a hydrothermal and co-precipitation method. In a typical synthesis process, 0.15 g of purified multi-walled CNTs (Nanjing XFNano Material Tech Co., Ltd.) were refluxed in nitric acid (AR, Aladdin) at 80 ∘C for 6 h. After being washed several times with deionized water, the acidified CNTs were dispersed in 150 mL deionized water. Then, 2.94 g Mn(CH3COO)2·4H2O was added into the as-prepared solution under stirring for 30 min. Subsequently, the as-prepared solution was added dropwise into 80 mL aqueous solution containing 1.27 g KMnO4 (AR, Aladdin) and under stirring for 30 min. The mixed solution was ultrasonically dispersed for 10 min and then transferred to a Teflon-lined autoclave and maintained at 120 ∘C for 12 h. After natural cooling, the dark brown product obtained by filtration was washed with deionized water several times and then dried in a vacuum oven at 80 ∘C for 12 h to obtain MnO2/CNTs.

To prepare the MnO2 cathode, α-MnO2/CNTs, Super P, and polyvinylidene fluoride binder were mixed at a mass ratio of 8:1:1 and dissolved in N-methy-2-pyrrolidone (NMP). The as-prepared homogeneous slurry was then blade-coated onto Ti foil and dried in a vacuum oven at 80 ∘C for 12 h. The mass load of MnO2 is about 6.8 to 7.0 mg cm−2.

Materials Characterization.

We obtained XRD results using a Rigaku SmartLab powder X-ray diffractometer with a Cu-Kα radiation (60 kV and 220 mA), X-ray source, and a 0D scintillation detector. We used a Apreo S LoVac (FEI company) with an EBSD accessory system (Hikari from Ametek) to obtain scanning electron microscopy (SEM) and EBSD results. AFM was performed on Dimension icon (Bruker company). The microstructure of MnO2/CNT was revealed by JEM-2100F field-emission TEM. TEM sample for cross-section view was prepared through resin embedding method.

Electrochemical Measurements.

The electrochemical performance of the single-crystal Zn anode was systemically evaluated using Zn||Zn symmetric cells, Zn||Ti asymmetric cells, and Zn||α-MnO2 full cells. All were assembled to CR2032-type coin cells and tested on LAND-CT2001A battery tester at room temperature. The area of electrodes in this study is 1.13 cm2. Without specific notes, aqueous electrolyte refers to 2 M ZnSO4 solution, “water-in-salt” electrolyte refers to 1 M Zn(CF3SO3)2 and 20 M LiTFSI, and non-aqueous electrolyte refers to 0.25 M Zn(CF3SO3)2 in AN. Electrodes are separated by filter paper/glass fiber in aqueous/non-aqueous electrolyte, respectively. For the Zn||α-MnO2 full cell, the electrolyte was 2 M ZnSO4 + 0.2 M MnSO4. Then, a 10-μm Zn anode was used for measuring capacity retention under N:P ratio of 5:1 and 100 μm Zn was for 500:1. CV and Tafel corrosion curves were conducted on Solartron Metrology, based on a conventional three-electrode configuration with Zn as working electrode, Ti as counter electrode, and saturated calomel electrode (SCE) as reference electrode. The corrosion potential and corrosion current were calculated from Tafel fit system. Symmetric Zn cells were constructed and subjected to single plating and stripping at increasing current densities to explore CCD of an electrode with different facets. Transient voltage responses to galvanostatic electrodeposition at 1 mA cm−2 were measured, and the areal capacities at the first sudden voltage drop were taken to be the critical areal capacities (CACs).

The ECSA are evaluated by the double-layer capacitance (Cdl) through the CV measurements at different scan rates in the non-Faradaic region. Half cells with cycled Zn and Pt as the working and counter electrode respectively, were assembled for CV measurement. The electrolyte is 2 M ZnSO4 aqueous solution. The scan rates range from 5 to 30 mV s−1 with an interval of 5 mV s−1 at the potential window of −1.15 to −1.05 V (vs. reversible hydrogen electrode). The square root of the scan rate increased with increasing oxidation peak current. Furthermore, the Randles–Sevcik equation was used to calculate the surface area, as expressed byIp=2.69×105n3/2AD1/2v1/2C0,

where Ip is the oxidation current, D represents the diffusion kinetics of electrolyte, n refers to the number of electrons transferred per molecule (2.0 for Zn ions), C0 is the bulk concentration of Zn ions, and v denotes the scanning rate, respectively. Therefore, the slope of the Ip−v1/2 curve can represent the electroactive surface area (A).

Theoretical Calculations.

The first-principles calculations were performed with the Vienna ab initio simulation package based on the DFT. The general gradient approximation with the Perdew–Burke–Ernzerhof (PBE) formulation for exchange-correlation functions was utilized in the present calculations. The convergence tolerance for geometry optimization was set as 1.0 × 10−5 V for energy, and all the forces on each atom were smaller than 0.02 eV Å−1. The cutoff energy was set 400 eV for the plane-wave basis set. Five or six layers of atoms were considered as the Zn surface in calculation. The vacuum layer on the Zn surface was set as 15 Å. The surface energies of Zn(002)/Zn(100)/Zn(101)/Zn(102)/Zn(110) were calculated by the following formula:γ=Esurf-N×Ebulk2S,

where Esurf is the energy of the slab, Ebulk is the energy of the Zn crystal bulk energy per atom, N is the number of atoms in the slab, and S is the surface area of the slab.

The adsorption energy Eads of Zn ion on the Zn slab was calculated as:Eads=Etotal-Esurf-EZn,

where Etotal is the total energy of the Zn slab bound with one Zn ion, Esurf is the intrinsic energy of the Zn(002)/Zn(100)/Zn(101)/Zn(102)/Zn(110) surface without the adsorbed Zn ion, and EZn is the energy of a single adsorbed Zn ion, respectively. In order to obtain the energetic favorite adsorbed site, the top, bridge, and hollow site were considered for all Zn surfaces. To calculate the diffusion paths and barrier of Zn ion on the Zn surface, the climbing image nudged elastic band method with typically five images along the reaction path was used. The adsorbed Zn ion diffused from the most favorable adsorption position to the next one via the neighboring bridge site. Thus, the activation energy for diffusion can be determined by the difference between the most favorable adsorption energy and the adsorption energy at the bridge position.

Supplementary Material

Appendix 01 (PDF)

Click here for additional data file.

This work was partially supported by the National Natural Science Foundation of China (Grant No. 22179085) and Start-up fund at Shanghai Jiao Tong University. The work at BNL is supported by the Assistant Secretary for Energy Efficiency and Renewable Energy, Vehicle Technology Office of the US Department of Energy through the Advanced Battery Materials Research Program under contract no. DE-SC0012704.

Author contributions

L.R., Z.H., and J.L. designed research; L.R., Z.H., C.P., L.Z., and N.W. performed research; S.Z. and E.H. revised the paper; J.L. wrote the paper; and all authors analyzed the data.

Competing interests

The authors declare no competing interest.

Data, Materials, and Software Availability

All study data are included in the article and/or SI Appendix.

Supporting Information

This article is a PNAS Direct Submission.
==== Refs
1 J. F. Parker , Rechargeable nickel-3D zinc batteries: An energy-dense, safer alternative to lithium-ion. Science 356 , 414–417 (2017).
2 B. J. Hopkins, Y. Shao-Horn, D. P. Hart, Suppressing corrosion in primary aluminum-air batteries via oil displacement. Science 362 , 658–661 (2018).30409879
3 S. Hou , Solvation sheath reorganization enables divalent metal batteries with fast interfacial charge transfer kinetics. Science 374 , 172–178 (2021).34618574
4 W. Sun , A rechargeable zinc-air battery based on zinc peroxide chemistry. Science 371 , 46–51 (2021).33384369
5 Y. Chen , Li metal deposition and stripping in a solid-state battery via Coble creep. Nature 578 , 251–255 (2020).32015545
6 S. Li , Codoped porous carbon nanofibres as a potassium metal host for nonaqueous K-ion batteries. Nat. Commun. 13 , 4911 (2022).35987982
7 Z. Zhao-Karger , Calcium-tin alloys as anodes for rechargeable non-aqueous calcium-ion batteries at room temperature. Nat. Commun. 13 , 5264 (2022).36071057
8 Y. Li , Interfacial engineering to achieve an energy density of over 200 Wh kg−1 in sodium batteries. Nat. Energy 7 , 511–519 (2022).
9 S. S. Shinde , Ampere-hour-scale zinc–air pouch cells. Nat. Energy 6 , 592–604 (2021).
10 X. Wang , Ultra-stable all-solid-state sodium metal batteries enabled by perfluoropolyether-based electrolytes. Nat. Mater. 21 , 1057–1065 (2022).35788569
11 J. Muldoon, C. B. Bucur, T. Gregory, Quest for nonaqueous multivalent secondary batteries: Magnesium and beyond. Chem. Rev. 114 , 11683–11720 (2014).25343313
12 J. X. Zheng , Regulating electrodeposition morphology of lithium: Towards commercially relevant secondary Li metal batteries. Chem. Soc. Rev. 49 , 2701–2750 (2020).32232259
13 J. Zheng, L. A. Archer, Controlling electrochemical growth of metallic zinc electrodes: Toward affordable rechargeable energy storage systems. Sci. Adv. 7 , eabe0219 (2021).33523975
14 C. Li, S. Jin, L. A. Archer, L. F. Nazar, Toward practical aqueous zinc-ion batteries for electrochemical energy storage. Joule 6 , 1733–1738 (2022).
15 J. Zheng , Reversible epitaxial electrodeposition of metals in battery anodes. Science 366 , 645–648 (2019).31672899
16 L. Cao , Fluorinated interphase enables reversible aqueous zinc battery chemistries. Nat. Nanotechnol. 16 , 902–910 (2021).33972758
17 R. Zhao , Lanthanum nitrate as aqueous electrolyte additive for favourable zinc metal electrodeposition. Nat. Commun. 13 , 3252 (2022).35668132
18 W. Zhang , Dynamic interphase-mediated assembly for deep cycling metal batteries. Sci. Adv. 7 , eabl3752 (2021).34851670
19 J. X. Zheng, L. A. Archer, Crystallographically textured electrodes for rechargeable batteries: Symmetry, fabrication, and characterization. Chem. Rev. 122 , 14440–14470 (2022).35950898
20 M. Wu , Seeded growth of large single-crystal copper foils with high-index facets. Nature 581 , 406–410 (2020).32461648
21 S. Jin , Colossal grain growth yields single-crystal metal foils by contact-free annealing. Science 362 , 1021–1025 (2018).30337454
22 K. Imasato , Metallic n-type Mg3Sb2 single crystals demonstrate the absence of ionized impurity scattering and enhanced thermoelectric performance. Adv. Mater. 32 , 1908218 (2020).
23 J.-C. Dong , In situ Raman spectroscopic evidence for oxygen reduction reaction intermediates at platinum single-crystal surfaces. Nat. Energy 4 , 60–67 (2019).
24 L. Wang , Epitaxial growth of a 100-square-centimetre single-crystal hexagonal boron nitride monolayer on copper. Nature 570 , 91–95 (2019).31118514
25 J. H. van der Merwe, “Epitaxial Interfaces” in Treatise on Materials Science and Technology, H. Herman, Ed. (Academic, ed. 1, 1973), p. 1.
26 S. D. Pu , Achieving ultrahigh-rate planar and dendrite-free zinc electroplating for aqueous zinc battery anodes. Adv. Mater. 34 , 2202552 (2022).
27 J. Zheng , Textured electrodes: Manipulating built-in crystallographic heterogeneity of metal electrodes via severe plastic deformation. Adv. Mater. 34 , 2106867 (2022).
28 M. Zhou , Surface-preferred crystal plane for a stable and reversible zinc anode. Adv. Mater. 33 , 2100187 (2021).
29 J. Kasemchainan , Critical stripping current leads to dendrite formation on plating in lithium anode solid electrolyte cells. Nat. Mater. 18 , 1105–1111 (2019).31358941
30 P. Bai , Interactions between lithium growths and nanoporous ceramic separators. Joule 2 , 2434–2449 (2018).
31 M. Jäckle, K. Helmbrecht, M. Smits, D. Stottmeister, A. Groß, Self-diffusion barriers: Possible descriptors for dendrite growth in batteries? Energy Environ. Sci. 11 , 3400–3407 (2018).
32 M. Jackle, A. Gross, Microscopic properties of lithium, sodium, and magnesium battery anode materials related to possible dendrite growth. J. Chem. Phys. 141 , 174710 (2014).25381540
33 M. E. Stournara, R. Kumar, Y. Qi, B. W. Sheldon, Ab initio diffuse-interface model for lithiated electrode interface evolution. Phys. Rev. E 94 , 012802 (2016).27575197
34 J. Hao , An in-depth study of Zn metal surface chemistry for advanced aqueous Zn-ion batteries. Adv. Mater. 32 , 2003021 (2020).
35 F. Yang , Understanding H2 evolution electrochemistry to minimize solvated water impact on zinc-anode performance. Adv. Mater. 34 , 202206754 (2022).
36 S. Fleischmann , Continuous transition from double-layer to Faradaic charge storage in confined electrolytes. Nat. Energy 7 , 222–228 (2022).
