The instability of solid electrolytes (SE) to Li anode limits the cyclic performance of all-solid-state lithium batteries (ASSLBs). Here, we demonstrate a research path including a series of studies on the rational design of the SE | Li interface. From high-rate Li-alloy anode, lithiophobic LiF-rich layer, porous lithiophobic LiF-Li₃N layer, lithiophobic mix-conductive LiF-Li₃N-Cu layer, to lithiophobic-lithiophilic bifunctional interlayer, we aim to present the developing understanding on Li dendrite suppression for ASSLBs.

• A unique co-doped method was used to enhance the Li self-diffusivity over 10 times compared to Li-Mg alloy by refining the Li grain sizes. With contact loss and void formation suppressed, a critical loading of 4.1 mAh∙cm^-2 at 1.0 mA∙cm^-2 was achieved at room temperature using Ta-doped LLZTO (Li₅La₃Zr_1.5Ta_0.5O₁₂) as a model solid-state electrolyte, which outperformed all previous reports. Both electrochemical data and simulation models were used to discuss the relationship between the dendrite-caused failure and the contact-loss-caused failure.
• A LiF-rich solid electrolyte interphase was in-situ formed between the SE and the Li metal, which suppresses the penetration of Li dendrites. Its highly lithiophobic and electronic insulating nature, and intrinsic electrochemical stability also block side reactions between the SE and Li. The LiF-rich layer enhances the room temperature CCD (critical current density) of Li₃PS₄ to a record-high value of >2 mA cm⁻². Moreover, the Li plating/stripping Coulombic efficiency was escalated from 88% of pristine Li₃PS₄ to more than 98% for LiF-coated Li3PS4.
• A cold-pressed highly lithiophobic, ionic conducting, and electronic insulating Li₃N–LiF composite was used to validate the Li-dendrite-free design criteria, where the highly ionic conductive Li₃N reduces the Li plating/ stripping overpotential, and LiF with high interface energy suppresses dendrites by enhancing the nucleation energy and suppressing the Li penetration into the SE. The Li₃N–LiF layer coating on Li₃PS₄ achieves a record-high CCD of >6 mA cm⁻² even at a high capacity of 6.0 mAh cm⁻². The Coulombic efficiency also reaches a record 99% in 150 cycles.
• A mixed electronic/ionic conductive and lithiophobic layer was in-situ generated by mixing a small amount (0.32 wt%) of CuF₂-LiNO₃ (CL) into Li₆PS₅Cl electrolyte. The CCD of Li₆PS₅Cl-CuF₂-LiNO₃ increases to 1.4 mA cm^–2/1.4 mAh cm^–2 at room temperature, which is much higher than that of pristine Li₆PS₅Cl (0.4 mA cm^–2/0.4 mAh cm^–2) even though mixing 32 wt% CL into Li₆PS₅Cl slightly reduces the ionic conductivity from 2.9 × 10^–3 to 1.5 × 10^–3 S cm^–1. The Cl@S-NMC811-Li₆PS₅Cl-CLA|Li₆PS₅Cl-CL|Li cells with areal capacity of 2.55 mAh cm^-2 achieve a capacity retention of 69.4% after 100 cycles at 1C (1C = 200 mAh g^-1).
• A bifunctional lithiophilic/lithiophobic interlayer was established to achieve both dendrite suppression and resistance reduction for LLZTO SE, which is confirmed by comprehensive material characterizations, simulations and electrochemical evaluations. The optimized Li-Sr | LLZTO | Li-Sr symmetrical cells have a critical current density of 1.3 mA cm^-2 and can be cycled for 1,000 cycles under 0.5 mA cm^-2 at room temperature.

Herein, the concatenation of these studies will provide a comprehensive understanding of the interface design for SE and help conquer the challenges at the SE-lithium interface towards the commercialization of solid-state lithium metal batteries for applications in cellphones, laptops, and EVs.