Plastic Monolithic Mixed‐Conducting Interlayer for Dendrite‐Free Solid‐State Batteries

Abstract Solid‐state electrolytes (SSEs) hold a critical role in enabling high‐energy‐density and safe rechargeable batteries with Li metal anode. Unfortunately, nonuniform lithium deposition and dendrite penetration due to poor interfacial solid–solid contact are hindering their practical applications. Here, solid‐state lithium naphthalenide (Li‐Naph(s)) is introduced as a plastic monolithic mixed‐conducting interlayer (PMMCI) between the garnet electrolyte and the Li anode via a facile cold process. The thin PMMCI shows a well‐ordered layered crystalline structure with excellent mixed‐conducting capability for both Li+ (4.38 × 10–3 S cm–1) and delocalized electrons (1.01 × 10–3 S cm–1). In contrast to previous composite interlayers, this monolithic material enables an intrinsically homogenous electric field and Li+ transport at the Li/garnet interface, thus significantly reducing the interfacial resistance and achieving uniform and dendrite‐free Li anode plating/stripping. As a result, Li symmetric cells with the PMMCI‐modified garnet electrolyte show highly stable cycling for 1200 h at 0.2 mA cm–2 and 500 h at a high current density of 1 mA cm–2. The findings provide a new interface design strategy for solid‐state batteries using monolithic mixed‐conducting interlayers.

Note: To make a better comparison, we dropped DME on the LLZTO (noted as DME-LLZTO) with the same amount as Li-Naph-LLZTO and then vacuum-dried the samples for 1 min, 2 min, 3 min, 4min, 5 min, 10 min, and 15 min, respectively. The samples were then rinsed in DMSO-d6 for NMR analysis. As shown in Figure S2a, the 1 H-NMR signals of DME molecules decreased significantly ( Figure S2a inset) within 1 min under vacuum and almost completely vanished after 5 min due to the volatile nature of DME (boiling point of 85 o C). We also compare the evaporation rate of DME with the presence of naphthalene (without the addition of Li). As shown in Figure S2b, there is barely any DME signals observed after vacuum-drying after 15 min. As a result, free DME solvent can be almost fully eliminated within the first few minutes of a vacuum-drying process during the Li-Naph(s) preparation process. Further removing the coordinated DME molecules in the crystalline structure requires considerable longer time duration under vacuum. As shown in Figure S2c, the ratio between Naph and DME remains 1:1 after 12 h under dynamic vacuum. To separate the electronic and ionic resistance, the Li-Naph solution was cautiously dropped on the stainless steel disc and vacuumed for 15 min to assemble a symmetric SS/Li-Naph/SS cell. For SS/Li-Naph/SS, the thickness of Li-Naph(s) layer is 0.2 mm, and its diameter is 15 mm. The ionic conductivity is from σi=L/S×Ri, and electric conductivity is from σe=L/S×Re, where S and L refer to surface area and thickness of the Li-Naph(s) layer.
According to Huggins's EIS method, mixed conducting materials exhibited two semicircles in the complex plane. In the equivalent circuit ( Figure S2b), R0 represents the contact resistance, Ri represents ionic resistance, and Re refers to electronic resistance, Cgeom is geometric capacitance. At lower frequencies, the equivalent circuit should be interpreted as a parallel combination of electronic resistance with geometric capacitance. At higher frequencies, the impedance due to the relatively large Cint becomes insignificant and is not observed in the measurements. As a result, the equivalent circuit at higher frequencies is a parallel arrangement with three legs: geometric capacitance Cgeom with two resistances (Ri and Re in parallel). There are two intercepts on the horizontal axis of the impedance spectrum ( Figure S2a), noted as R1 and R2. (R2-R0) refers to Re. and (R1-R0) results from the parallel combination of Re and Ri, which meets the following equation: 1/(R1-R0) = l/Ri + 1/Re. Therefore, both electronic and ionic conductivities can be estimated by the Huggins method. To validate the conductivity results from the Huggins method, DC polarization was also carried out to test the electronic conductivity of the Li-Naph(s). As shown in Figure S4, under a DC polarization of 10 mV, the current stabilizes at ~ 1.075 mA. After the long-term DC polarization, the current should be only from electronic conduction with no net flow of Li + ions.
The electronic conductivity of the Li-Naph(s) is calculated to be 1.22 × 10 -3 S cm -1 , which is close to the value of 1.01 × 10 -3 S cm -1 obtained from the Huggins method.   Note: From the integral area of the peak, it can be calculated that the mole ratio of Naph and DME in the Li-Naph(s) is 1:2.

Prepared of Li-Naph-LLZTO-wash:
Li-Naph-LLZTO was washed by DME over 3 times until the dark-green color vanished, followed by vacuum drying at glove-chamber for 12 h, which is noted as Li-Naph-LLZTOwash. Subsequently, Li-Naph-LLZTO-wash was assembled with Li anode to form Li symmetric cell for galvanostatic measurement. Note: The Li anode was rinsed by anhydrous DME three times to remove residual Li-Naph(s) and dried under vacuum. The samples were transferred into the XPS test chamber in a sealed sample holder without exposure to an ambient atmosphere.           Group2 0.2699 g 0.789 g cm -3 0.2297 g 5.297 g cm -3 5.5 g cm -3 96.3% The relative density of LLZTO ceramic pellets was evaluated by the Archimedes' principle, which is shown as the following formula, ρrelative = ρreal/ρtheoretical×100% (1)

ρreal = mLLZTO•ρethanol/(mLLZTOmsubmerged) (2)
Where the ρrelative is the relative density of the LLZTO, the ρreal and ρtheoretical are the tested density and the theoretical density of the LLZTO, mLLZTO is the mass of the LLZTO, msubmerged is the apparent mass of the LLZTO when submerged in the ethanol, and ρethanol is the density of the ethanol. Two groups of relative density data are shown in Table S4.