Diverse methods for cleaning the Li7La3Zr2O12 (LLZO) surface and utilization of an interfacial layer (IL) between garnet-type LLZO solid-state electrolyte and lithium metal anode are widely recognized as critical strategies to enhance the electrochemical performance of Li-garnet solid-state batteries1,2,3. Despite the growing interest in developing efficient cleaning methodologies and efficient ILs for the Li/LLZO interface, a literature review reveals a lack of consensus within the research community regarding their actual function. While it is generally agreed that both approaches often improve lithium wettability on the LLZO surface, thereby reducing interfacial resistance and voltage polarization, significant debates persist regarding the specific mechanisms occurring during Li plating and stripping at cleaned/IL-functionalized LLZO surfaces during cycling. Furthermore, understanding how the interface evolves during Li plating and stripping cycles remains challenging, primarily due to the complexity of examining the Li/LLZO interface compared to characterizing the Li surface in liquid Li-ion batteries.

Given the importance of unveiling the impact of LLZO surface cleaning and the application of ILs, this commentary summarizes their primary functions by considering experimentally and computationally validated phenomena occurring during Li plating and stripping at the Li/LLZO interface. Numerous review and perspective articles addressing ILs for the Li/LLZO interface4,5,6,7,8,9,10 present a broad overview of various ILs and their reported electrochemical performance, but still lack to offer a broader comprehension of the fundamental processes occurring during Li plating and stripping. This commentary seeks to address this gap by providing a concise analysis of the key interfacial phenomena at the Li/LLZO interface.

Analysis

Let us begin our analysis by discussing the configuration of the LLZO/Li interface without specific LLZO surface cleaning or adding an artificial IL, followed by examining the processes occurring during Li plating and stripping (Fig. 1a). It is important to note that the LLZO surface rarely remains pure during storage and handling11. As revealed by numerous X-ray photoelectron spectroscopy measurements12,13,14,15, LLZO reacts with CO2 and H2O, forming a contamination layer primarily composed of Li2CO3 and LiOH. This issue not only increases the interfacial resistance of the Li/LLZO interface, thereby causing voltage polarization but also significantly enhances the lithiophobicity of the LLZO surface. Lithiophobicity of LLZO has been experimentally demonstrated in multiple studies examining the contact angle of molten Li on the LLZO surface16,17,18. The implications of a lithiophobic contaminated LLZO surface extend beyond just the challenge of establishing full contact with the Li metal anode. Considering reports on liquid-state Li metal batteries suggesting that the lithiophobicity/lipophilicity of current collectors can alter the morphology of deposited Li19,20,21, similar behavior, although not yet confirmed, can also be foreseen in solid-state systems. This correlation can be explained by the fact that dendritic Li tends to form on lithiophobic surfaces as plated Li seeks to minimize contact with the surface. Conversely, Li plating on a lithiophilic surface tends to maximize contact, promoting more uniform, non-dendritic structures. Moreover, the lithiophobicity of the contaminated LLZO surface plays a detrimental role during Li stripping, particularly in promoting the formation of voids at the Li/LLZO interface -a common problem during Li stripping when the rate of Li removal exceeds the rate of Li diffusion within the metal1,22. This occurs because Li tends to minimize contact with the lithiophobic LLZO surface, which amplifies void formation. The formation of voids can significantly decrease cycling stability, leading to increased current densities during subsequent plating, which may ultimately result in dendrite formation.

Fig. 1: Schematics of stripping and plating of Li at the Li/LLZO interface.
figure 1

Drawing of the Li/LLZO interface based on a contaminated (a), cleaned (b), and the IL-functionalized (c) LLZO surface for as-prepared interface and after plating and stripping of Li. The figure illustrates key factors affecting Li-ion transport at the Li/LLZO interface, including the presence of voids and cavities (due to insufficient pressing of Li onto the rough LLZO surface), the Li2CO3/LiOH contamination layer (shown in yellow), and the interfacial layer (shown in blue) between metallic Li and LLZO.

Full size image

To address the issues associated with contaminated LLZO surfaces, Sun et al.23, Kravchyk et al.13 and others12,15,24 demonstrated that cleaning the LLZO surface can significantly reduce Li/LLZO interfacial resistance while also playing a crucial role in mitigating the formation of Li dendrites and voids (Fig. 1b). Contact angle measurements performed by Sakamoto et al.16 revealed that a contaminant-free LLZO surface is lithiophilic, with a contact angle of ca. 95° with molten metallic Li. These experimental observations align with computational results conducted by Liu et al.25, which showed negative surface energy values for the Li/LLZO interface compared to the more positive values computed for the Li/Li2CO3 interface. This suggests that during Li plating on a lithiophilic LLZO surface, the plated Li tends to maximize contact with the surface, promoting more uniform and less dendritic structures. Additionally, during Li stripping, the lithiophilicity of the clean LLZO surface plays a favorable role by mitigating void formation at the Li/LLZO interface. This can be explained by the tendency of Li to maintain maximum contact with the lithiophilic surface, which helps reduce void formation by driving them into the Li metal rather than allowing them to persist at the Li/LLZO interface. Moreover, it should be noted that, in addition to cleaning the LLZO surface, various lithiophilic materials have been tested as interfacial layers between the LLZO surface and Li metal. A comprehensive list of studied interfacial layers can be found in the following reviews5,7,26. Notably, the deposition of interfacial layers at the Li/LLZO interface is performed only after the LLZO surface has been thoroughly cleaned, typically using methods such as thermal treatment13,14,25,27, wet chemical processes16,23,28,29,30,31, or polishing15,27,32. The application of ILs can further enhance lithium wettability on the LLZO surface, thereby reducing interfacial resistance and voltage polarization17,18,33,34,35. For instance, as demonstrated by Hu et al.18, the use of Al as IL resulted in a decrease in interfacial resistance from 950 Ohm·cm² measured for pristine LLZO/Li interface to 75 Ohm·cm². The high lithiophilicity of this IL was also confirmed by complementary contact angle measurements with molten Li18, which revealed significant decrease of Li/LLZO contact angle to 47° compared to 95° observed for a merely cleaned LLZO surface16. These findings suggest that using such ILs could further improve the mitigation of void and Li dendrite formation, similar to the effects observed with clean LLZO surfaces (Fig. 1c). Lastly, it should be noted that both cleaning the LLZO surface and the employment of ILs might significantly improve the surface roughness of the LLZO surface. This factor can be as critical as lithiophilicity since, according to Ceder et al.36, the Li plating at an rough LLZO surface leads to a higher lithium deposition rate in the tip-like regions, causing the formation of the dendrites.

Outlook

In summarizing our analysis of the effectiveness of LLZO surface cleaning and the application of various ILs, it is important to acknowledge that the intrinsic problem of void formation during Li stripping may still persist. This issue is particularly challenging at high current densities and areal capacities, such as those exceeding 1 mA cm-² and 1 mAh cm-², where alternative strategies may be required to address void formation effectively. Below, we outline three potential strategies for mitigating this issue (see Fig. 2 for details):

Fig. 2: Schematic representation of additional strategies to mitigate void formation at the Li/LLZO interface.
figure 2

Illustration of the Li/LLZO interface with a Li-metal alloy anode, LLZO scaffold, and applied external pressure, highlighting various void-mitigation strategies, including (i) the use of pressure, (ii) enhanced Li diffusion within the anode, and (iii) increased Li/LLZO contact area.

Full size image

- The use of pressure: The formation of voids can be mitigated by applying pressure at the Li/LLZO interface, which helps mechanically replenish the voids. Studies by Sakamoto37, Janek38, Chen39, Srinivasan40 and Kovalenko41 have demonstrated that maintaining a critical stack pressure at a given current density can prevent void formation. However, higher current densities necessitate higher pressures, which could potentially induce cracks in the LLZO, leading to dendrite formation due to structural failures. Additionally, the use of bulky, heavy constructions to apply such pressures presents engineering challenges that could significantly reduce the volumetric and gravimetric energy densities of LLZO solid-state batteries.

- Enhancement of Li diffusion in Li metal38,42,43,44,45: While ILs play a crucial role, they are insufficient on their own to fully resolve void formation issues. Modifying the diffusion properties of Li metal is also essential. For instance, research by Janek et al.45 has shown that introducing 10 at% Mg into the lithium metal anode significantly enhances Li diffusion compared to bare Li metal. However, experimental validation of improved cycling stability at the Li-Mg alloy/LLZO interface due to void mitigation is still pending.

- Increasing the Li/LLZO contact area46,47,48,49,50,51,52,53,54: This can be achieved by fabricating porous, scaffold-like LLZO structures, which allow Li to be plated and stripped within the pores, thereby increasing the Li/LLZO contact area and significantly reducing the local stripping current density. Computational studies performed by Kravchyk et al.55 have shown that using 50% porous LLZO membranes with 5 µm pore sizes can reduce the stripping current density by ca. 90%, effectively mitigating void formation at the Li/LLZO interface.

In conclusion, despite significant advancements in this field, the optimal LLZO/Li interface engineering remains a complex challenge. The most effective solution will likely involve a combination of approaches, including the employment of ILs, additives in Li metal, and the use of a porous LLZO surface. It is anticipated that within the next few years, these combined strategies will lead to superior cycling stability of Li metal with LLZO solid-state electrolytes at commercially relevant current densities and areal capacities.