Chinese Chemical Letters  2025, Vol. 36 Issue (8): 110308   PDF    
In situ nanoscale insights into the interfacial degradation of Zn metal anodes
Jiao Wanga,b, Shuang-Yan Langa,b, Zhen-Zhen Shena,b, Gui-Xian Liua,b, Rui Wena,b,*     
a CAS Key Laboratory of Molecular Nanostructure and Nanotechnology, Beijing National Laboratory for Molecular Sciences, CAS Research/Education Center for Excellence in Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China;
b University of Chinese Academy of Sciences, Beijing 100049, China
Abstract: Zn metal batteries are highly attractive because of their high theoretical specific capacity, intrinsic safety and resource availability. However, further development is significantly hindered by low Coulomb efficiency, which is closely linked to reaction processes occurring at electrode/electrolyte interfaces. Herein, we have achieved a real-time visualization and comprehensive analysis of the interfacial evolution of Zn metal anode via in situ AFM in organic and aqueous electrolytes, respectively. The processes of uneven nucleation, dendrite growth, the ZnO formation and the dissolution of Zn substrate are directly probed in aqueous electrolyte, which induces interfacial deterioration and ultimately results in battery failure. In organic electrolyte, the in situ observations show that the homogeneous nuclei form on the Zn surface to induce the dendrite-free deposition, however, exhibiting poor Zn plating/stripping reversibility. This work delves into the dynamic evolution and electrochemical behaviors regulated by solvents, which provides in-depth understanding of structure-reactivity correlations and further interfacial engineering.
Keywords: In situ AFM    Interfacial processes    Degradation mechanism    Organic electrolyte    Zn metal batteries    

Zn metal batteries (ZMBs) represent promising candidates due to the high theoretical specific capacity (820 mAh/g), intrinsic safety and relatively low cost of Zn [1-5]. However, the practical utilization of ZMBs is invariably hindered by the thermodynamic instability of the Zn anode in an aqueous environment, which leads to two unavoidable interfacial issues: (1) Uncontrolled Zn dendrites due to heterogeneous Zn deposition, (2) low Zn plating/stripping Coulomb efficiency (CE) from inevitable side reactions between Zn metal anodes and aqueous electrolytes [6-11]. Therefore, numerous efforts have been dedicated to tackling the above challenging problems by tuning the solvation structure of Zn2+ [12, 13], which include the exploitation of ionic liquids [14-16], electrolyte additives [17-20], and novel solvents [21-23].

Among the aforementioned strategies, organic electrolytes exhibit high reversibility and a wide electrochemical window, making them compatible with high-voltage cathode materials, which can address water-associated parasitic reactions [24-26]. It has been reported that alcohol-based electrolytes can alleviate water corrosion of the Zn anode and suppress hydrogen gas evolution, furthermore, allowing the battery to operate in extreme temperatures at −40 ℃ [27, 28]. Despite the extensive efforts to develop advanced organic electrolytes for ZMBs, the practical application has not yet been realized. This is because Zn ion batteries consisting of organic electrolytes often exhibit poor rate performance and low specific capacity [27, 28]. However, in-depth comprehension of the evolution processes and degradation mechanism of the electrode/electrolyte interface with various solvents remains elusive, which is beneficial for interfacial regulation and electrolyte optimization. In situ atomic force microscopy (AFM) with high spatial resolution and low invasion exhibits unique strengths in exploring the detailed interfacial processes at nanoscale. In addition, considering the extensive research and low cost of MeOH, we selected it as the representative non-aqueous solvent for further research.

In the present work, in situ AFM was employed to investigate the failure mechanisms of Zn metal anodes with different electrolyte solvents. In aqueous electrolyte, the uneven nucleation and dendrite growth on the Zn metal surface increase the uneven distribution of the interfacial electric field. During the stripping process, the substrate undergoes dissolution resulting in cracks or pits that alter electrode shape, while a small amount of deposited Zn dissolves and reacts with water to generate ZnO that passivates the electrode surface. The combination of uneven nucleation, growth, irreversible plating/stripping processes, and the formation of ZnO byproduct collectively lead to battery failure. In MeOH-based electrolyte, ZnF2 and ZnS enriched SEI originating from electrolyte decomposition induces uniform nucleation and dendrite-free deposition. However, the deposited Zn exhibits low plating/stripping reversibility, which decreases the CE. These results revealed the failure mechanism of the Zn metal anodes, providing insights for advanced battery optimization and structure design.

The schematic illustration of the in situ AFM cell is delineated in Fig. 1a, where the Zn foil was used as the working electrode, and the Zn strip was used as both counter and reference electrodes. The interfacial processes of the Zn anode in aqueous electrolyte (1 mol/L Zn(OTf)2) were first disclosed. Before conducting the in situ experiment, the Zn anode surface was carefully polished to remove the passivation layer. Fig. 1b shows the AFM image of the Zn surface at open circuit potential (OCP). Real-time AFM images of Zn plating/stripping at a current density of 40 µA/cm2 are displayed in Figs. 1c-e and f-k, respectively. The corresponding galvanostatic profile of the plating/stripping processes is shown in Fig. S1 (Supporting information). In addition, the different stages for each frame are accordingly marked in Fig. S1, which is conducive to comprehending the correlation between electrochemistry and interfacial evolution. At 1100 s, several particles occur on the electrode surface (Fig. 1c), which could be ascribed to the uneven Zn nucleation. As plating proceeds, the deposited Zn gradually grows larger and no new Zn nuclei appear on the Zn surface (Figs. 1d and e, Fig. S2 in Supporting information). The corresponding height section profiles (Fig. 1g) reveal that the Zn deposition proceeds from 0.59 µm (Fig. 1c) to 0.93 µm (Fig. 1e) horizontally and from 53.1 nm (Fig. 1c) to 209.5 nm (Fig. 1e) vertically, which illustrates the quicker growth rate along the vertical direction. Due to the uneven distribution of the electric field and rapid ion diffusion, Zn2+ tends to accumulate at the deposited Zn, resulting in the formation of Zn dendrites.

Download:
Fig. 1. (a) Schematic illustration of the in situ electrochemical AFM cell. AFM topography images of the Zn metal anode (b–e) during the plating and (f-k) during the stripping processes with different stages at a current density of 40 µA/cm2 in aqueous electrolyte. (g) The height section of the Zn surface along the green dotted lines in (b–e). (l–o) The corresponding DMT modulus of the Zn surface in (h–k). The white arrows indicate the scanning direction. The scale bars are 1 µm.

During the subsequent stripping process, the dissolution step tends to occur preferentially at the Zn substrate, inducing the formation of cracks marked by yellow arrows in Fig. 1f. In contrast, there is only a slight dissolution on the deposited Zn (Fig. S3 in Supporting information). With stripping prolonged, the cracks gradually become wider and deeper. Remarkably, the amorphous material signed by blue arrows, showing a lower modulus (Figs. 1l-o), appears near the deposited Zn (Figs. 1h and i), possibly due to the side reaction. This material spreads and covers the whole substrate as stripping progresses (Figs. 1j and k). In addition, the volume of deposited Zn increases evidently, further illustrating the parasitic reaction between the Zn anode and electrolyte (Figs. S4e-j in Supporting information). The quantitative analyses show that the roughness of the Zn surface initially decreases slightly due to the dissolution of the deposits, followed by a rapid increase as the Zn substrate also dissolves (Fig. S5 in Supporting information). The on-site formed film exhibits distinct mechanical properties compared to the Zn substrate and deposited Zn (Figs. 1l-o). The corresponding 3D AFM images are shown in Fig. S6 (Supporting information) to further illustrate the dynamic evolution of the Zn anode in aqueous electrolyte. The chemical properties of this film will be discussed later.

Applying MeOH solvent is considered an effective method to tailor the interfacial chemistries as shown in Fig. 2a, thereby inhibiting the dendrite formation and widening the voltage window [29]. The interfacial processes in MeOH-based electrolyte (Zn(OTf)2: MeOH = 1:111 by molar ratio), including the processes of Zn nucleation, growth, and dissolution, were disclosed by in situ AFM. Fig. S7 (Supporting information) shows the typical galvanostatic profile in MeOH-based electrolyte. The Zn anode surface is clean at OCP (Fig. 2b). When the current density of 40 µA/cm2 is applied for 330 s, uniform particles generate on the Zn anode surface (Fig. 2c). Notably, the amount of Zn nuclei is much greater than that in aqueous electrolyte (Fig. 2d), demonstrating the homogeneous Zn nucleation. During the subsequent plating process (Figs. 2e and f), Zn nuclei uniformly grow and entirely cover the Zn surface, exhibiting a lamellar structure. Furthermore, ex situ SEM characterizations of the Zn anode with the same current density have been further carried out. It is revealed that the hexagonal lamellar evenly deposits on the Zn anode surface, indicating dendrite-free morphology (Figs. S8a and b in Supporting information). Even increasing the capacity to 80 mAh/cm2, the deposited layer is also compact and uniform (Figs. S8c and d in Supporting information), which is significantly distinguished from the dendrite growth in aqueous electrolyte (Fig. S9 in Supporting information). Such a structure of compact lamellar is conducive to stabilizing the electrode/electrolyte interface. Subsequently, the same current density and capacity were applied to disclose the Zn dissolution. There are no obvious changes in the Zn anode surface during the whole stripping process (Fig. 2g and Fig. S10 in Supporting information), illustrating poor reversibility in MeOH-based electrolyte. The additional quantitative analysis of the roughness variation is shown in Fig. 2h, directly tracking the rapid burgeon during the plating process and minimal change upon stripping. The corresponding 3D AFM images were exhibited in Fig. S11 (Supporting information).

Download:
Fig. 2. (a) Schematic illustration of the solvation structure in MeOH-based electrolyte. AFM topography images of the Zn metal anode (b–f) during the plating and (g) during the stripping processes with different stages at a current density of 40 µA/cm2 in MeOH-based electrolyte. (h) The roughness variation of the Zn surface during the plating/stripping processes. The white arrows indicate the scanning direction. The scale bars are 1 µm.

In order to investigate the cyclic characteristic of Zn in various electrolytes, the Zn plating/stripping processes during 2nd cycle are further examined in detail. It is indicated that the amorphous film and "dead" Zn still remain at the Zn anode surface after the 1st stripping (Fig. 3a and Fig. S12 in Supporting information). During the 2nd plating process, the undissolved Zn gradually swells and keeps growth (as marked by blue dotted lines in Figs. 3ac) because Zn2+ tends to settle at the tip position, which further increases the inhomogeneity of the interfacial electric field. The volume of the deposition in the blue dotted boxes increases from 0.34 µm3 in Fig. 3a to 1.94 µm3 in Fig. 3c. The detailed volume variation of the electrode surface upon 2nd plating is exhibited in Fig. S13 (Supporting information). During the 2nd stripping process, only partial deposits are dissolved, as shown by blue dotted lines in Figs. 3d-f. There is no obvious change in the deposited Zn indicated by yellow arrows, which further uncovers the accumulation of "dead" Zn during cycling. In addition, the Zn block marked by green arrows becomes swell during the stripping process, which corresponds to the parasitic reaction and aligns with the behavior observed in the 1st cycle. It means that side reactions between the Zn anode and electrolyte occur all the time, constantly depleting the active materials to cause the battery failure.

Download:
Fig. 3. 3D AFM topography images of the Zn anode (a-c) during the 2nd plating and (d-f) during the 2nd stripping processes with different stages at a current density of 40 µA/cm2 in aqueous electrolyte. The scale bars are 1 µm. The white arrows indicate the scanning direction.

In MeOH-based electrolyte, the deposits obtained in 1st plating process entirely remain on the electrode surface before 2nd plating. After plating at a current density of 40 µA/cm2 for 290 s, a large amount of Zn forms on the Zn surface as shown in Fig. S14 (Supporting information). The subsequent process was not captured due to the drastic change of the Zn electrode. Combined with the 1st plating/stripping processes, it can be seen that in MeOH-based electrolyte, irreversible dissolution and the accumulation of deposited layer during long cycling may pierce the separator, resulting in the reduction of the Coulombic efficiency as well as internal short circuit, which is the main cause of the battery failure. Fig. S15 (Supporting information) shows the Coulombic efficiency nearly at 0% in MeOH-based electrolyte, in agreement with the results observed by in situ AFM experiments. In aqueous electrolyte, the Coulombic efficiency gradually increases, but not higher than 50%. In addition, the voltage profile of the Zn-Zn cell in MeOH-based electrolyte shows a gradual increase (Fig. S16 in Supporting information), which results from the irreversible plating/stripping.

In addition to the above in situ observations of the morphological evolution in aqueous and organic electrolytes, the interfacial properties were further disclosed by ex situ characterizations. The samples were obtained after plating-stripping processes in various electrolytes with a density current of 40 µA/cm2 and capacity of 40 µAh/cm2. In the F 1s spectra (Fig. 4a), both ZnF2 (684.8 eV) and organic F (688.8 eV) can be detected on the Zn electrode surface in MeOH-based electrolyte [30, 31]. The ZnF2 fraction increases from 26.5% at 0 s to about 91.7% at 100 s, and maintains stability at 300 s. In addition, the appearance of ZnS [32] in the S 2p spectra further suggests the decomposition of electrolyte (Fig. 4b), which also increases with the Ar+ etching (Fig. 4c). It indicates that OTf anion or its derivatives mainly exist in the outer region, and inorganic phases like ZnF2 and ZnS are mainly distributed in the inner region. ZnS and ZnF2, are considered suitable SEI components with relatively high ionic conductivity, which can effectively lower the diffusion energy barriers of Zn2+ and suppress the dendrite growth [33, 34]. In aqueous electrolyte, there is no ZnF2 or ZnS signal in XPS spectra (Figs. S17 and S18 in Supporting information). In the O 1s spectra (Fig. 4d), a new peak at 530.5 eV can be detected, which is attributed to ZnO [35]. With the increase in the etching depths, the intensity of ZnO gradually decreases. Furthermore, the Raman peak at 433 and 563 cm−1 provides the further chemical composition of ZnO on the Zn anode surface (Fig. 4e) [16, 36]. Fig. 4f shows the TEM and the element mapping of products after cycle in aqueous electrolyte, evidencing the formation of ZnO (Fig. 4f and Fig. S19 in Supporting information). Therefore, it can be revealed that the amorphous film observed from in situ investigation mainly consists of ZnO. Such amorphous ZnO is considered electrochemically inactive, which can passivate the electrode surface and induce uneven nucleation [37].

Download:
Fig. 4. Ex situ characterizations of the Zn anode surface. XPS spectra of (a) F 1s and (b) S 2p at different etching depths on the Zn anode after cycle in MeOH-based electrolyte. (c) The depth profiles of atomic composition on the Zn surface in aqueous electrolyte. XPS spectra of (d) O 1s at different etching depths on the Zn anode after cycle in aqueous electrolyte. (e) Raman spectra of Zn anode surface in aqueous electrolyte. (f) TEM micrograph with elemental mapping images of the product in aqueous electrolyte. The scale bar is 100 nm.

As schematically illustrated in Fig. 5, the interfacial evolution upon Zn plating/stripping processes was proposed. In traditional aqueous electrolyte (1 mol/L Zn(OTf)2), the solvation shell of Zn2+ ions is predominantly occupied by water molecules, forming [Zn(H2O)6]2+ (Fig. 5a1). During the plating process, Zn tends to unevenly nucleate on the electrode surface (Fig. 5a2). Subsequently, Zn2+ prefers to deposit at the nucleation sites to reduce surface energy, meaning the "tip effect", which leads to the formation of Zn dendrites (Fig. 5a3). During the stripping process, only a small amount of the deposited Zn can be dissolved, which induces the accumulation of "dead" Zn during long cycling. In addition, undissolved Zn deposits react with water molecules to form the inactive ZnO, which covers the electrode surface. Besides, significant dissolution of the substrate results in the formation of cracks, altering the electrode shape (Fig. 5a4). Dendrite formation, parasitic reactions, irreversible dissolution and shape change of electrode together induce the battery failure in aqueous electrolyte.

Download:
Fig. 5. Schematic illustration of the reaction mechanism and structural evolution. Interphasial formation and Zn plating/stripping at the Zn surface in (a1–4) aqueous and (b1–4) organic electrolytes, respectively. (a1) Interface of Zn anode under OCP. (a2) Uneven Zn nucleation. (a3) Growth of dendrites. (a4) The formation of ZnO and "dead" Zn after stripping. (b1) Interface of Zn anode under OCP. (b2) Uniform Zn nucleation. (b3) Dendrite-free growth. (b4) Irreversible dissolution.

When using MeOH as the electrolyte solvent, MeOH molecules and OTf anions contribute to the solvation shell of Zn2+, altering the coordination environment of Zn2+ and the reduction potential of anions, thereby promoting the formation of ZnF2 and ZnS enriched SEI (Fig. 5b1) [29]. During the plating process, Zn uniformly deposits on the electrode surface with more nucleation sites (Fig. 5b2) and gradually grows without forming dendrites (Fig. 5b3). However, during the stripping process, the deposited Zn shows minimal changes, exhibiting poor plating/stripping reversibility (Fig. 5b4). Accumulation of deposited layer during multi-cycling may penetrate the separator and cause a short circuit of the battery. The experiments reveal the failure mechanisms of Zn metal anode in different electrolyte solvents, highlighting the importance of optimizing electrolytes and modifying electrodes to enhance the performance of ZMBs.

In summary, we provide compelling and direct evidence demonstrating the impact of solvents on the morphology evolution and reaction micromechanism of the Zn anode at nanoscale. In aqueous electrolyte, the formation process of Zn dendrite, byproduct ZnO and cracks on the Zn anode surface are directly probed, providing the visualized evidence of the interfacial failure. In organic electrolyte, uniform nucleation can promote dendrite-free deposition, whereas exhibiting poor reversibility, which results in the reduction of Coulomb efficiency and the accumulation of Zn. Comprehending the underlying dynamic evolution and degradation mechanism of Zn anodes in various electrolytes yields tremendous opportunities for battery designs and interfacial engineering in ZMBs.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Jiao Wang: Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft. Shuang-Yan Lang: Data curation, Formal analysis, Supervision, Writing – review & editing. Zhen-Zhen Shen: Data curation, Funding acquisition, Investigation. Gui-Xian Liu: Data curation, Investigation. Rui Wen: Conceptualization, Formal analysis, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – review & editing.

Acknowledgments

This work was financially supported by the CAS Project for Young Scientists in Basic Research (No. YSBR-058), the National Key R & D Program of China (No. 2021YFB2500300), the National Natural Science Foundation of China (Nos. 92372125, 22205241) and the National Postdoctoral Program for Innovative Talents (No. BX20220306) of the Chinese Postdoctoral Science Foundation.

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110308.

References
[1]
W. Zhang, J. Zhang, N. Wang, et al., Nat. Sustain. 7 (2024) 463-473. DOI:10.1038/s41893-024-01300-2
[2]
W. Sun, F. Wang, B. Zhang, et al., Science 371 (2021) 46-51. DOI:10.1126/science.abb9554
[3]
S. Chen, J. Chen, X. Liao, et al., ACS Energy Lett. 7 (2022) 4028-4035. DOI:10.1021/acsenergylett.2c02042
[4]
M. Li, F. Zeng, S. Li, et al., Energy Mater. Devices 1 (2023) 9370007. DOI:10.26599/emd.2023.9370007
[5]
C. Xu, B. Li, H. Du, F. Kang, Angew. Chem. Int. Ed. 51 (2012) 933-935. DOI:10.1002/anie.201106307
[6]
Y. Li, Y. Wang, Y. Xu, et al., Small 18 (2022) 2202214.
[7]
B. Raza, Y. Zhang, J. Chen, et al., Angew. Chem. Int. Ed. 62 (2023) e202302174.
[8]
X. Zhou, B. Wen, Y. Cai, et al., Angew. Chem. Int. Ed. 63 (2024) e202402342.
[9]
S. Xu, J. Huang, G. Wang, et al., Small Methods 8 (2024) 2300268.
[10]
Y. Zhu, G. Liang, X. Cui, et al., Energy Environ. Sci. 17 (2024) 369-385. DOI:10.1039/d3ee03584k
[11]
C. Nie, G. Wang, D. Wang, et al., Adv. Energy Mater. 13 (2023) 2300606.
[12]
Y. Li, J. Wang, S. Wang, et al., Energy Stor. Mater. 66 (2024) 103245.
[13]
J. Wang, Y. Yang, Y. Wang, et al., ACS Nano 16 (2022) 15770-15778. DOI:10.1021/acsnano.2c04114
[14]
Z. Zhao, J. Lai, D.T. Ho, et al., ACS Energy Lett. 8 (2023) 608-618.
[15]
M. Zhao, J. Rong, F. Huo, et al., Adv. Mater. 34 (2022) 2203153.
[16]
H. Qiu, X. Du, J. Zhao, et al., Nat. Commun. 10 (2019) 5374.
[17]
Y.X. Song, J. Wang, X.B. Zhong, et al., Energy Stor. Mater. 58 (2023) 85-93.
[18]
Y. Geng, L. Pan, Z. Peng, et al., Energy Stor. Mater. 51 (2022) 733-755.
[19]
Q. Wu, J. Huang, J. Zhang, et al., Angew. Chem. Int. Ed. 63 (2024) e202319051.
[20]
Y. Zhang, M. Li, S. Liang, G. Fang, Energy Mater. Devices 1 (2023) 9370023. DOI:10.26599/emd.2023.9370023
[21]
K. Zhou, G. Liu, X. Yu, Z. Li, Y. Wang, J. Am. Chem. Soc. 146 (2024) 9455-9464. DOI:10.1021/jacs.4c02150
[22]
A. Naveed, H. Yang, J. Yang, Y. Nuli, J. Wang, Angew. Chem. Int. Ed. 58 (2019) 2760-2764. DOI:10.1002/anie.201813223
[23]
M. Han, T.C. Li, X. Chen, H.Y. Yang, Small 20 (2024) 2304901.
[24]
A. Naveed, H. Yang, Y. Shao, et al., Adv. Mater. 31 (2019) 1900668.
[25]
D. Han, C. Cui, K. Zhang, et al., Nat. Sustain. 5 (2022) 205-213.
[26]
C. Cui, D. Han, H. Lu, et al., Adv. Energy Mater. 13 (2023) 2301466.
[27]
D. Kundu, S. Hosseini Vajargah, et al., Energy Environ. Sci. 11 (2018) 881-892. DOI:10.1039/c8ee00378e
[28]
W. Xu, Y. Wang, Nano-Micro Lett. 11 (2019) 90.
[29]
L. Ma, J. Vatamanu, N.T. Hahn, et al., Proc. Nat. Acad. Sci. 119 (2022) e2121138119.
[30]
L. Cao, D. Li, T. Pollard, Nat. Nanotechnol. 16 (2021) 902-910. DOI:10.1038/s41565-021-00905-4
[31]
W. Xu, J. Li, X. Liao, et al., J. Am. Chem. Soc. 145 (2023) 22456-22465. DOI:10.1021/jacs.3c06523
[32]
L. Cao, D. Li, E. Hu, et al., J. Am. Chem. Soc. 142 (2020) 21404-21409. DOI:10.1021/jacs.0c09794
[33]
X. Song, L. Bai, C. Wang, et al., ACS Nano 17 (2023) 15113-15124. DOI:10.1021/acsnano.3c04343
[34]
Z. Zhao, R. Wang, C. Peng, et al., Nat. Commun. 12 (2021) 6606.
[35]
J. Wang, J.X. Tian, G.X. Liu, Z.Z. Shen, R. Wen, Small Methods 7 (2023) 2300392.
[36]
V.V. Volkov, D.J. Oliver, C.C. Perry, Nat. Commun. 11 (2020) 4908.
[37]
B.E. Hawkins, D.E. Turney, R.J. Messinger, et al., Adv. Energy Mater. 12 (2022) 2103294.