The overwhelming majority of lithium-ion batteries currently in use feature a graphite anode, which is approaching its maximum theoretical capacity. Lithium metal batteries are regarded as pivotal enablers for the forthcoming generation of lithium-ion technology, largely due to their superior energy density. Indeed, lithium (Li) metal has the lowest reduction potential (-3.04 V vs. std H), a very low density (0.534 g.cm-3), and one of the highest specific capacities (3860 mAh.g-1) [1-3]. However, in contrast to the graphite anode, which is an intercalation anode, lithium metal is a conversion anode that functions as an active material reservoir during cycling. Accordingly, the energy density of the battery is directly proportional to the thickness of the Li metal anode. To achieve high specific and volumetric energy density, it is necessary to utilize an ultra-thin lithium metal anode, with a thickness of less than 25 micrometers. In addition to the technical challenge of producing such a thin lithium metal anode using the current method of extrusion and lamination, its high chemical and electrochemical reactivity and process complexity represents a significant obstacle to its commercialization [4]. Indeed, lithium metal is susceptible to dendritic growth during electroplating, which can result in continuous reaction with the electrolyte. This can lead to a rapid decrease in the active species reservoir and, subsequently, a decline in battery capacity. The solid electrolyte interphase (SEI) that forms on lithium because of its reaction with the electrolyte is typically heterogeneous and mechanically unstable. Consequently, the lithium-ion flux during plating is heterogeneous, which facilitates the growth of dendrites. It has been demonstrated that during subsequent stripping, it is more advantageous to dissolve lithium from the dendrite root, resulting in the formation of detached Li that undergoes chemical reactions with the electrolyte. This detached Li, referred to as dead Li, is passivated and completely electronically insulated, preventing its participation in further electrochemical reactions.

Despite considerable efforts to enhance the SEI of Li metal through morphological and compositional modifications, employing techniques such as electrolyte formulation, electrolyte additives, or Li metal coating, the native passivation layer on the Li metal surface is frequently disregarded [5-6]. This layer, which is formed during processing, handling, and storage, is typically composed of lithium carbonate (Li2CO3), lithium hydroxide (LiOH), lithium oxide (Li2O), and can include impurities such as Si or Cu particles [7]. It has been demonstrated that the nucleation through the native passivation layer, irrespective of the electrolyte system, is inhomogeneous and exerts a significant influence on the growth morphology. To mitigate the impact of the native passivation layer, a range of techniques are employed in battery research, including brushing the Li metal surface, roll-pressing, chemical reactions, and shortening the storage period. However, these techniques are typically not subjected to rigorous control.

In this study, we propose the use of vacuum thermal evaporation as a method for the production and coating of high-quality, ultra-thin, and high-performance lithium metal anodes. This method has been demonstrated to be a scalable process in other industries, including semiconductors and solar cells and could represent a valuable solution for commercialization of high performance Li-metal anodes. Initially, X-ray photoelectron spectroscopy (XPS) coupled with argon-ion sputtering was employed to demonstrate that the vacuum thermally evaporated Li metal exhibited a native passivation layer that was one order of magnitude thinner than the extruded lithium metal that is commercially available. The bare evaporated lithium metal displays enhanced performance in a symmetrical configuration relative to extruded lithium metal. Electrochemical impedance spectroscopy (EIS) coupled with distribution of relaxation time (DRT) demonstrated that the evaporated Li, due to its minimal passivation layer, exhibited an exceptionally low charge transfer resistance in comparison to the extruded one, even after formation. This study demonstrates that plating and stripping on evaporated Li is more favourable and can suppress or inhibit dendrite formation, thereby prolonging cyclability in full cells. Postmortem analyses demonstrate that evaporated lithium, due to its superior surface quality, impedes the formation of dead lithium and results in a denser cycled lithium metal. Consequently, in both ether-based electrolytes with lithium iron phosphate (LFP) and carbonate-based electrolytes with nickel manganese cobalt (NMC), evaporated lithium exhibits significantly enhanced performance compared to extruded lithium.
It is notable that the cycle life of the system employing evaporated Li with NMC622 and a carbonate electrolyte is three times longer than that of extruded Li (Figure 1b). As vacuum thermal evaporation can be performed on a variety of substrate shapes, custom-made evaporated Li metal anodes were produced for pouch cells. The developed Li-metal anode was also tested with LFP cathode material with a loading of 2 mA/cm2. The cells from several runs and Li-metal sources from different suppliers highlighting the high stability of the process.
The high purity and controllable surface state provide excellent electrochemical properties, allowing for the design of an artificial SEI from a passivation-free interphase. Based on this optimal surface configuration, a protective coating has been developed to further enhance the electrochemical performance.

A stable solid electrolyte interface (SEI) is the pre-requisite to reach long cycling in Li-metal batteries. As already mentioned, this layer has a very complex chemical and elemental structure and in case of Li-metal anodes is very unstable. The applications of artificial SEI (a-SEI) is considered as a valid approach to improve the stability of Li-metal anodes. The processing routes for the fabrication of a-SEI on the Li-metal anode are very limited because of the tighter restrictions compared to other anode or cathode materials. Here we are proposing the use thermal evaporation method as for the Li-metal deposition, for the deposition and engineering of the a-SEI. This unique approach consists in depositing the film in-situ directly after Li metal deposition (fabrication of the Li-metal anode) without breaking the vacuum. This has the advantages of preventing the formation of unwanted passivation layer on the Li metal as well as improving the electrochemical performance and reducing the reactivity of lithium, allowing easier handling in dry room environment. Moreover, the PVD process is highly versatile and flexible allowing to tune the elemental composition in the a-SEI and precisely control its thickness. Our approach consists depositing composite coatings and tune their chemical properties by thermally evaporating at the same time from different sources. This leads to the deposition of multi-elemental composite coatings with uniform chemical distribution as shown in the energy dispersive X-ray spectroscopy.

After chemical engineering and thickness series variation of the composite coating we found an optimum layer design. The electrochemical performance of bare Li-metal and coated Li-metal by PVD in monolayer pouch cells with LFP cathode have excellent electrochemical performance. The Li metals cells outperform of more than 200 cycles the refence cell (uncoated) highlighting the great potential of our approach to form stable SEI. As next step we are now testing the coating with pouch cells utilizing NMC811 and ultra-thin evaporated Li metal (25 µm) have demonstrated the ability to reach over 350 cycles, substantiating the viability of this production method as a promising avenue for the practical deployment of Li metal anodes.

Evaporated lithium metal has been also identified as an optimal starting point for enhancing the performance of batteries when utilising solid polymer electrolytes. In this study, we demonstrate that our in-house developed solid polymer electrolyte, in combination with the ultrathin evaporated lithium metal, can achieve more than 350 cycles using NMC622 (2 mAh.cm-2) and a limited Li reservoir (25 um). This enables a projected gravimetric energy density of 325 Wh.kg-1 and a volumetric energy density of 651 Wh.L-1 at the pack level. The high purity surface of evaporated lithium facilitates the formation of a stable SEI, thereby enabling rapid Li-ion diffusion, thus permitting the solid polymer electrolyte to cycle at high current densities, reaching 10 mA.cm-2 (5C) with a capacity of 2 mAh.cm-2. Subsequent research will encompass the implementation of this technology utilizing even higher loading, with the objective of attaining potentially remarkably elevated energy densities.

In conclusion, this work presents a novel approach to processing and developing Li metal anodes, which can facilitate the transition from laboratory-based research to practical applications of Li metal batteries

[1] M. Fichtner et al., Adv. Energy Mater. 12, 2102904 (2022).
[2] P.U. Nzereogu, Appl. Surf. Sci., Adv. 9, 100233 (2022).
[3] M.A. Azam, Energy Storage 33, 102096 (2021).
[4] R. Wang, J. Energy Chem. 48, 145–159 (2020).
[5] G. M. Hobold et al., Nat. Energy, vol. 6, pp. 951-960, (2021).
[6] M. Srout et al., Small, vol. 19, pp. 2206252, (2023).
[7] N. Rospars et al., Comm. Mat., 5, Article number: 179 (2024)