Sodium Super Ionic Conductor-Type Hybrid Electrolytes for High Performance Lithium Metal Batteries

Composite solid electrolytes (CSEs), composed of sodium superionic conductor (NASICON)-type Li1+xAlxTi2-x(PO4)3 (LATP), poly (vinylidene fluoride-hexafluoro propylene) (PVDF-HFP), and lithium bis (trifluoromethanesulfonyl)imide (LiTFSI) salt, are designed and fabricated for lithium-metal batteries. The effects of the key design parameters (i.e., LiTFSI/LATP ratio, CSE thickness, and carbon content) on the specific capacity, coulombic efficiency, and cyclic stability were systematically investigated. The optimal CSE configuration, superior specific capacity (~160 mAh g−1), low electrode polarization (~0.12 V), and remarkable cyclic stability (a capacity retention of 86.8%) were achieved during extended cycling (>200 cycles). In addition, with the optimal CSE structure, a high ionic conductivity (~2.83 × 10−4 S cm−1) was demonstrated at an ambient temperature. The CSE configuration demonstrated in this work can be employed for designing highly durable CSEs with enhanced ionic conductivity and significantly reduced interfacial electrolyte/electrode resistance.


Introduction
Reliable energy storage technologies are needed for future-generation electronic devices (e.g., laptop computers and cellphones), electric vehicles (EVs), and electric grids. Depending on the application, the required attributes for energy storage may vary; however, high power density and capacity utilization, along with enhanced stability, safety, and environmental friendliness, are urgently demanded [1,2]. All solid-state energy-storage devices (e.g., batteries and supercapacitors) with compact designs and light weights are very promising for enabling high energy densities, as well as enhanced safety and thermal stability, since the safety hazards and physicochemical limitations associated with the liquid electrolytes can be mitigated via utilizing solid electrolytes [3][4][5][6][7]. An ideal solid electrolyte is expected to possess a high level of ionic conductivity, a wide electrochemical stability window, and compatibility (chemically and electrochemically) with the electrode materials [4].

Fabrication of CSEs
To fabricate the composite solid-state electrolytes, first, lithium bis(trifluoromethanesul fonyl)imide (LiTFSI, Alfa, Bracknell, UK, purity: 98%) and PVDF-HFP (Sigma, Saint Louis, MO, USA, molecular weight: 400,000) were homogeneously mixed using a glove box, and the N-methyl pyrrolidone (NMP, Showa, Tokyo, Japan, purity: 95%) was poured into the mixture. Highly crystalline LATP powders with an average particle size of 5 µm were supplied from Gold Carbon Co., Ltd., Taoyuan City, Taiwan. The LATP powders were synthesized by using an efficient sol-gel method, followed by thermal calcination. The LATP powders were gradually added to the solution, and the entire CSE slurry was uniformly dispersed through a planetary milling process with a rotation speed of 700 rpm for 1 h. The weight ratio of the PVDF-HFP in the CSE films was approximately 25 wt.%. The resulting LATP-containing slurry was then coated onto a polymeric membrane (polyethylene (PE) separator (Hipore AC-0881, Yan Tin Chemical Co., Ltd., Hong Kong, China) with a nominal thickness of approximately 8 µm and an uncompressed porosity of approximately 38%. The nominal weight of the PE membrane was~4.6-4.9 g m −2 . The thicknesses of the CSEs films casted on the PE membranes were tuned using a doctor blade (i.e., 50, 100, and 150 µm). The actual thicknesses were approximately 40, 70, and 100 µm. The above preparation process was carried out in a glove box, after which the CSEs were dehydrated at 140 • C in a vacuum oven overnight.

Assembly of Li-Metal Batteries
The electrochemical characterization of the CSEs was conducted using a coin cell configuration (type: CR2032) with commercially available LFP (Ubiq Technology Co., Taoyuan, China) as the active material for the cathode electrode and Li metal as the anode material. To fabricate the LFP cathode, the LFP powders (average size of~0.5 µm) were mixed with a binder (PVDF) and a conducting medium (Super-P, Taiwan Maxwave Co., Taipei City, Taiwan) with a 90:7:3 wt.% in an NMP solvent to form the LFP slurry. The mixture was blended using a three-dimensional mixer equipped with Zr balls for 2 h to form a uniform slurry. The resulting slurry was subsequently coated on the Al foil substrate with a doctor blade. The LFP cathode sheets were dried at 110 • C in a vacuum oven overnight. Afterward, the LFP sheets were compressed and then cut into the desired shape for the battery assembly. The surface loading of the LFP was approximately 6-7 mg cm −2 , with an average thickness of 70 µm, and the thickness of Li metal was approximately 10 µm. The as-prepared CSEs (with PE membranes) were placed on the LFP cathode, and the entire composite sheet was tightly compressed under a pressure of approximately 200 kg cm −2 and then baked at 140 • C in a vacuum oven for 24 h to remove any residual organic solvent (e.g., NMP) from the CSEs. After drying, the Li|CSE|LFP batteries were assembled as the CR-2032-type coin cells in a glove box with LFP cathodes, Li-metal anodes, and CSE (i.e., LATP plus LiTFSI plus PVDF-HFP) solid-state electrolytes.

Materials and Electrochemical Characterization
The morphologies of the LFP, LATP, and CSE were characterized using field-emission scanning electrode microscopy (FE-SEM, JEOL JSM-6700F, Tokyo, Japan). The crystalline structure of the LFP, LATP, CSE samples was analyzed through X-ray diffraction (XRD, Brucker D2 diffractometer with Cu target, Billerica, MA, USA). To explore the real-time performance of the as-prepared CSEs, the charge/discharge cycling experiments were performed at different C rates (which varied from 0.1 to 3 C) within the voltage range of 2.8-4.0 V at ambient temperatures. The batteries were first charged using a conventional protocol of constant current-constant voltage (i.e., different C rates to 4.0 V, with a 0.01 mA cut-off current), followed by discharging to 2.8 V at a constant current. Electrochemical impedance spectroscopy (EIS, CH Instruments 608C, Austin, TX, USA) was also conducted to quantify the polarization distribution of the coin cells assembled with the various CSEs. The EIS measurements were carried out at different potentials within the frequency range of 10 mHz to 100 kHz. Figure 1a shows typical XRD patterns of pristine LFP powders, including the characteristic peaks of crystalline LFP. The diffraction peaks within the LFP sample matched quite well with the standard orthorhombic olivine phase of LFP (JCPDS Card No.: 83-2092: a = 10.334 Å, b = 6.010 Å, and c = 4.693 Å) [30]. The XRD pattern confirmed that the LFP powders were highly crystalline orthorhombic olivine [31]. The FE-SEM image of the pristine LFP sample, as shown in Figure 1b, illustrates the homogeneous dispersion of the quasi-spherical powders, with an average particle size of~500 nm.

Results and Discussion
In this study, the composition of the Li 1+x Al x Ti 2-x (PO 4 ) 3 with x = 0.3 (i.e., Li 1.3 Al 0.3 Ti 1.7 (PO 4 ) 3 ) was specifically chosen to demonstrate its potential for high-performance batteries. The LATP powders, which were made using the citric acid-assisted sol-gel synthesis method followed by calcination, possessed highly crystalline structures, as depicted in Figure 2a.
The XRD pattern of the as-synthesized LATP powder can be indexed to the standard NASICON-type structure (i.e., rhombohedral lattice, Card No.: ICDD 00-035-0754). The characteristic peaks at 2θ = 20.8, 24.5, 29.7, 33.3, and 36.5 • were assigned to the crystalline planes of (012), (104), (113), (024), and (116), respectively [32,33]. It is important to note that there is no other diffraction peak that commonly appears in the case of any impurity (e.g., AlPO 4 [34]), revealing the high-phase purity of the synthesized LATP powders. An FE-SEM image of the LATP powders is provided in Figure 2b, showing the rectangular-shaped particles homogeneously dispersed within the structure. In this study, the composition of the Li1+xAlxTi2-x(PO4)3 with x = 0.3 (i.e., Li1.3Al0.3Ti1.7(PO4)3) was specifically chosen to demonstrate its potential for high-performance batteries. The LATP powders, which were made using the citric acid-assisted solgel synthesis method followed by calcination, possessed highly crystalline structures, as depicted in Figure 2a. The XRD pattern of the as-synthesized LATP powder can be indexed to the standard NASICON-type structure (i.e., rhombohedral lattice, Card No.: ICDD 00-035-0754). The characteristic peaks at 2θ = 20.8, 24.5, 29.7, 33.3, and 36.5° were assigned to the crystalline planes of (012), (104), (113), (024), and (116), respectively [32,33]. It is important to note that there is no other diffraction peak that commonly appears in the case of any impurity (e.g., AlPO4 [34]), revealing the high-phase purity of the synthesized LATP powders. An FE-SEM image of the LATP powders is provided in Figure 2b, showing the rectangular-shaped particles homogeneously dispersed within the structure. The typical XRD pattern of CSEs (i.e., composition: LATP plus LiTFSI plus PVDF-HFP) is depicted in Figure 3a. According to Figure 3, the representative crystalline planes of the LATP lattices are present and the crystalline PVDF-HFP morphology can be identified at approximately 19.0, 20.6, and 26.9°, corresponding to the α, β, and γ crystalline phases of a PVDF-HFP structure, respectively [35]. To further explore the interfacial layer, a cross-sectional view FE-SEM image of the interface of the LFP cathode and the CSEs was recorded (see Figure 3b). It is clear that the CSE layer tightly covered the LFP cathode sheets, where the CSE coating formed a dense and solid film with a uniform thickness of approximately 40 μm. There was no obvious porosity, and a cavity appeared at the interface between the CSE and the LFP cathode, indicating that there was good adhesion to the electrode. To ensure the uniformity of the CSEs, the elemental mapping including Ti, P, O, Al, and F was employed to characterize the chemical distribution at the cross-sectional  In this study, the composition of the Li1+xAlxTi2-x(PO4)3 with x = 0.3 (i.e., Li1.3Al0.3Ti1.7(PO4)3) was specifically chosen to demonstrate its potential for high-performance batteries. The LATP powders, which were made using the citric acid-assisted solgel synthesis method followed by calcination, possessed highly crystalline structures, as depicted in Figure 2a. The XRD pattern of the as-synthesized LATP powder can be indexed to the standard NASICON-type structure (i.e., rhombohedral lattice, Card No.: ICDD 00-035-0754). The characteristic peaks at 2θ = 20.8, 24.5, 29.7, 33.3, and 36.5° were assigned to the crystalline planes of (012), (104), (113), (024), and (116), respectively [32,33]. It is important to note that there is no other diffraction peak that commonly appears in the case of any impurity (e.g., AlPO4 [34]), revealing the high-phase purity of the synthesized LATP powders. An FE-SEM image of the LATP powders is provided in Figure 2b, showing the rectangular-shaped particles homogeneously dispersed within the structure. The typical XRD pattern of CSEs (i.e., composition: LATP plus LiTFSI plus PVDF-HFP) is depicted in Figure 3a. According to Figure 3, the representative crystalline planes of the LATP lattices are present and the crystalline PVDF-HFP morphology can be identified at approximately 19.0, 20.6, and 26.9°, corresponding to the α, β, and γ crystalline phases of a PVDF-HFP structure, respectively [35]. To further explore the interfacial layer, a cross-sectional view FE-SEM image of the interface of the LFP cathode and the CSEs was recorded (see Figure 3b). It is clear that the CSE layer tightly covered the LFP cathode sheets, where the CSE coating formed a dense and solid film with a uniform thickness of approximately 40 μm. There was no obvious porosity, and a cavity appeared at the interface between the CSE and the LFP cathode, indicating that there was good adhesion to the electrode. To ensure the uniformity of the CSEs, the elemental mapping including Ti, P, O, Al, and F was employed to characterize the chemical distribution at the cross-sectional The typical XRD pattern of CSEs (i.e., composition: LATP plus LiTFSI plus PVDF-HFP) is depicted in Figure 3a. According to Figure 3, the representative crystalline planes of the LATP lattices are present and the crystalline PVDF-HFP morphology can be identified at approximately 19.0, 20.6, and 26.9 • , corresponding to the α, β, and γ crystalline phases of a PVDF-HFP structure, respectively [35]. To further explore the interfacial layer, a crosssectional view FE-SEM image of the interface of the LFP cathode and the CSEs was recorded (see Figure 3b). It is clear that the CSE layer tightly covered the LFP cathode sheets, where the CSE coating formed a dense and solid film with a uniform thickness of approximately 40 µm. There was no obvious porosity, and a cavity appeared at the interface between the CSE and the LFP cathode, indicating that there was good adhesion to the electrode. To ensure the uniformity of the CSEs, the elemental mapping including Ti, P, O, Al, and F was employed to characterize the chemical distribution at the cross-sectional CSEs, as shown in the Supplementary Materials (see Figure S1). As shown in Figure S1, the colorful dots were well dispersed at the cross-sectional view of the FE-SEM images, indicating well-prepared CSEs with LATP ceramics and LiTFSI salts. For all test cells, the chargedischarge curves were symmetric at 0.5 C, revealing the reversibility of the Li + intercalation and de-intercalation [38][39][40]. In addition, as illustrated in Figure 4, the discharge capacity as a function of cycle number for the as-prepared CSEs (i.e., LiTFSI/LATP ratio: 2.0) reached as high as approximately 161 mAh g −1 . Importantly, the electrode polarization (i.e., the potential difference between the charging and discharging plateau, ∆E) was vastly influenced by the LiTFSI/LATP ratio, where ∆E demonstrated the following order: LiTFSI/LATP ratio: 2.0 (approximately 0.25 V) < LiTFSI/LATP ratio: 1.8 (approximately 0.28 V) < LiTFSI/LATP ratio: 1.5 (approximately 0.31 V). According to this observation, a higher LiTFSI content tended to facilitate the ionic migration within the composite layer, alleviating the ionic diffusion resistance in the solid phase. CSEs, as shown in the Supplementary Materials (see Figure S1). As shown in Figure S1 the colorful dots were well dispersed at the cross-sectional view of the FE-SEM images indicating well-prepared CSEs with LATP ceramics and LiTFSI salts.  For all test cells, the chargedischarge curves were symmetric at 0.5 C, revealing the reversibility of the Li + intercala tion and de-intercalation [38][39][40]. In addition, as illustrated in Figure 4, the discharge ca pacity as a function of cycle number for the as-prepared CSEs (i.e., LiTFSI/LATP ratio: 2.0 reached as high as approximately 161 mAh g −1 . Importantly, the electrode polarization (i.e., the potential difference between the charging and discharging plateau, ΔE) wa vastly influenced by the LiTFSI/LATP ratio, where ΔE demonstrated the following order LiTFSI/LATP ratio: 2.0 (approximately 0.25 V) < LiTFSI/LATP ratio: 1.8 (approximately 0.28 V) < LiTFSI/LATP ratio: 1.5 (approximately 0.31 V). According to this observation, a higher LiTFSI content tended to facilitate the ionic migration within the composite layer alleviating the ionic diffusion resistance in the solid phase.
To further explore the efficacy of the as-prepared composite solid-state electrolytes the cyclic performance of the test cells equipped with different CSEs were analyzed at 0.5 C (see Figure 4d-f). Capacity retention and coulombic efficiency are usually employed in evaluating the cyclic stability of CSEs upon being cyclically charged/discharged. As illus trated in Figure 4, the coin cell assembled with the CSE (i.e., LiTFSI/LATP ratio: 2.0) ex hibited a high-capacity retention capability (~99.6%), along with excellent coulombic effi ciency (~99.3%), after 100 cycles. In contrast, the other two cell configurations displayed poor cyclic performance (e.g., low-capacity retention of approximately 50-82% after 100 cycles), confirming the critical role of the CSE configuration on the electrochemical per formance of the Li-metal batteries.  Next, the influence of CSE thickness on the electrochemical performance of the lithium-ion batteries with Li-metal anodes was explored, as shown in Figure 5. The thicknesses of the CSE films on the LFP cathodes were controlled at 40, 70, and 100 μm. The Limetal batteries were cyclically charged and discharged for 100 cycles at 0.5 C at an ambient temperature. According to Figure 5, the reduced CSE thickness resulted in the improved cyclic performance of the coin cells. Compared to other configurations, the Li-metal batteries equipped with 40 μm-thick CSE films displayed the best cyclic stability (i.e., the capacity retention was ~96.3%) and a remarkable coulombic efficiency (~92.5%) after 100 cycles. Indeed, reducing the CSE thickness reduced the ohmic polarization. In contrast, thicker CSE layers imparted longer diffusion pathways for the Li + ions. In particular, at higher C-rate operations, an increased ohmic polarization may result in the poor cyclic performance of Li-metal batteries during long-duration cycling.  To further explore the efficacy of the as-prepared composite solid-state electrolytes, the cyclic performance of the test cells equipped with different CSEs were analyzed at 0.5 C (see Figure 4d-f). Capacity retention and coulombic efficiency are usually employed in evaluating the cyclic stability of CSEs upon being cyclically charged/discharged. As illustrated in Figure 4, the coin cell assembled with the CSE (i.e., LiTFSI/LATP ratio: 2.0) exhibited a high-capacity retention capability (~99.6%), along with excellent coulombic efficiency (~99.3%), after 100 cycles. In contrast, the other two cell configurations displayed poor cyclic performance (e.g., low-capacity retention of approximately 50-82% after 100 cycles), confirming the critical role of the CSE configuration on the electrochemical performance of the Li-metal batteries.
Next, the influence of CSE thickness on the electrochemical performance of the lithiumion batteries with Li-metal anodes was explored, as shown in Figure 5. The thicknesses of the CSE films on the LFP cathodes were controlled at 40, 70, and 100 µm. The Li-metal batteries were cyclically charged and discharged for 100 cycles at 0.5 C at an ambient temperature. According to Figure 5, the reduced CSE thickness resulted in the improved cyclic performance of the coin cells. Compared to other configurations, the Li-metal batteries equipped with 40 µm-thick CSE films displayed the best cyclic stability (i.e., the capacity retention was~96.3%) and a remarkable coulombic efficiency (~92.5%) after 100 cycles. Indeed, reducing the CSE thickness reduced the ohmic polarization. In contrast, thicker CSE layers imparted longer diffusion pathways for the Li + ions. In particular, at higher C-rate operations, an increased ohmic polarization may result in the poor cyclic performance of Li-metal batteries during long-duration cycling. Next, the influence of CSE thickness on the electrochemical performance of the lithium-ion batteries with Li-metal anodes was explored, as shown in Figure 5. The thicknesses of the CSE films on the LFP cathodes were controlled at 40, 70, and 100 μm. The Limetal batteries were cyclically charged and discharged for 100 cycles at 0.5 C at an ambient temperature. According to Figure 5, the reduced CSE thickness resulted in the improved cyclic performance of the coin cells. Compared to other configurations, the Li-metal batteries equipped with 40 μm-thick CSE films displayed the best cyclic stability (i.e., the capacity retention was ~96.3%) and a remarkable coulombic efficiency (~92.5%) after 100 cycles. Indeed, reducing the CSE thickness reduced the ohmic polarization. In contrast, thicker CSE layers imparted longer diffusion pathways for the Li + ions. In particular, at higher C-rate operations, an increased ohmic polarization may result in the poor cyclic performance of Li-metal batteries during long-duration cycling. The interfacial layer between the CSEs and LFP cathodes played a critical role, affecting the rate capability and cyclic stability of the coin cells. It is generally recognized that conductive carbon particles (Super-P in this case), due to their intrinsic hydrophobicity The interfacial layer between the CSEs and LFP cathodes played a critical role, affecting the rate capability and cyclic stability of the coin cells. It is generally recognized that conductive carbon particles (Super-P in this case), due to their intrinsic hydrophobicity [41,42], significantly influence the wetting property of a CSE layer. The wetting characteristic of a CSE is strongly related to the surface tension of an LFP cathode toward gel-phase composite electrolytes. To inspect the influence of surface hydrophobicity, the CSEs were coated onto three types of LFP cathodes (with three different contents of the carbon medium (5, 7, and 15 wt.%)), and the corresponding electrochemical performances were analyzed. The charge-discharge curves at 0.5 C, as well as the cyclic performances of the different Li-metal batteries, are shown in Figure 6. As clearly demonstrated in Figure 6, the LFP cathode containing 7 wt.% Super-P conductive carbon delivered the lowest polarization (approximately 0.25 V), excellent coulombic efficiency (approximately 97.5%), and higher-capacity retention (approximately 90.1%) after 100 cycles compared to other samples. This finding reflects that an unoptimized C content within the cathode sheets can easily result in the unwanted peeling of CSEs from the LFP cathode, and it can even ease the formation of the Li dendrites that lead to poor cyclic stability.
Considering the effects of the various design parameters on the performance of the coin cells, an optimal CSE configuration (i.e., LiTFSI/LATP ratio, thickness, and carbon content) was engineered (i.e., PVDF-HFP/LATP: 2.0, thickness: 40 µm, C content: 7 wt.%, and heat-treated temperature: 140 • C) for the fabrication of the Li-metal batteries. The typical charge-discharge curves at various rates, as well as the variations in the specific capacity with C rate, for the coin cells equipped with this optimal CSE configuration were recorded (see Figure 7). As shown in Figure 7a, the Li-metal battery was galvanostatically charged and discharged between 2.5 and 4.0 V (vs. Li/Li + ) at different rates (i.e., 0.1, 0.2, 0.5, 1, 2, and 3 C). The performance curves shown in Figure 7 contain the typical charge/discharge plateau occurring at approximately 3.3-3.5 V at 0.1-0.5 C, indicating the presence of a two-phase Fe 3+ /Fe 2+ redox reaction via a first-order transition between the FePO 4 and LFP [31,39,43]. With the optimized CSE configuration, the ∆E value at 0.5 C was dramatically reduced to 0.12 V, significantly reducing the electrode polarization. In addition, according to Figure 7, the Li-metal batteries were robustly charged and discharged at 1-3 C. The discharge capacity as a decreasing function of the C rate is depicted in Figure 7b. Considering the discharge capacity at 0.1 C as the basis for comparison, the capacity retentions were maintained at 98.1% (0.2 C), 96.8% (0.5 C), 87.5% (1 C), 56.3% (2 C), and 25.1% (3 C). Notably, all the coin cells demonstrated 100% capacity retention at 0.1 C after completing the cycling test, indicating the superior Li + reversibility through the CSE layer, even after high-rate cycling experiments.
Membranes 2023, 13, x FOR PEER REVIEW 7 of 1 [41,42], significantly influence the wetting property of a CSE layer. The wetting character istic of a CSE is strongly related to the surface tension of an LFP cathode toward gel-phase composite electrolytes. To inspect the influence of surface hydrophobicity, the CSEs were coated onto three types of LFP cathodes (with three different contents of the carbon me dium (5, 7, and 15 wt.%)), and the corresponding electrochemical performances were an alyzed. The charge-discharge curves at 0.5 C, as well as the cyclic performances of the different Li-metal batteries, are shown in Figure 6. As clearly demonstrated in Figure 6 the LFP cathode containing 7 wt.% Super-P conductive carbon delivered the lowest po larization (approximately 0.25 V), excellent coulombic efficiency (approximately 97.5%) and higher-capacity retention (approximately 90.1%) after 100 cycles compared to othe samples. This finding reflects that an unoptimized C content within the cathode sheet can easily result in the unwanted peeling of CSEs from the LFP cathode, and it can even ease the formation of the Li dendrites that lead to poor cyclic stability. Considering the effects of the various design parameters on the performance of the coin cells, an optimal CSE configuration (i.e., LiTFSI/LATP ratio, thickness, and carbon content) was engineered (i.e., PVDF-HFP/LATP: 2.0, thickness: 40 μm, C content: 7 wt.% and heat-treated temperature: 140 °C) for the fabrication of the Li-metal batteries. The typical charge-discharge curves at various rates, as well as the variations in the specifi capacity with C rate, for the coin cells equipped with this optimal CSE configuration were recorded (see Figure 7). As shown in Figure 7a, the Li-metal battery was galvanostatically charged and discharged between 2.5 and 4.0 V (vs. Li/Li + ) at different rates (i.e., 0.1, 0.2 0.5, 1, 2, and 3 C). The performance curves shown in Figure 7 contain the typica charge/discharge plateau occurring at approximately 3.3-3.5 V at 0.1-0.5 C, indicating the presence of a two-phase Fe 3+ /Fe 2+ redox reaction via a first-order transition between the FePO4 and LFP [31,39,43]. With the optimized CSE configuration, the ΔE value at 0.5 C was dramatically reduced to 0.12 V, significantly reducing the electrode polarization. In addition, according to Figure 7, the Li-metal batteries were robustly charged and dis charged at 1-3 C. The discharge capacity as a decreasing function of the C rate is depicted in Figure 7b. Considering the discharge capacity at 0.1 C as the basis for comparison, the capacity retentions were maintained at 98.1% (0.2 C), 96.8% (0.5 C), 87.5% (1 C), 56.3% (2 C), and 25.1% (3 C). Notably, all the coin cells demonstrated 100% capacity retention at 0.1 C after completing the cycling test, indicating the superior Li + reversibility through the CSE layer, even after high-rate cycling experiments.  Figure 7c illustrates the variations in the charge-discharge curves of the coin cel 0.5 C during extended cycling (i.e., 1-200 cycles). Based on Figure 7c, the ΔE value g ually increased with an increased cycle number, which was primarily due to the grad aging of the CSE layer. However, the capacity retention remained high (i.e., 86.8%), e after 200 cycles. The variation in the capacity retention and coulombic efficiency of the metal battery fabricated with the optimal CSEs is shown in Figure 7d. As clearly il trated in Figure 7d, the Li-metal batteries were cyclically charged/discharge at 0.5 C 200 cycles. The capacity retention was maintained at >85%, with a stable coulombic ciency after 200 cycles. Indeed, such a robust performance was likely due to: (i) the lac dissolution of the iron from the LFP cathode [44,45], (ii) the uniform formation of the s electrolyte interphase layer, and (iii) the lack of a substantial formation of Li dendr [46,47] during the long-duration cycling, confirming that this was a stable design of CSE layer for the Li-metal batteries. However, the gradual aging of the optimal CSEs quires an in-depth investigation. This may presumably be due to a slight electrode po ization during the charge/discharge process, causing the accumulation of Li metal at interface between the Li anode/CSE/LFP cathode. Accordingly, work regarding noscaled LATP, a thinner CSE layer, and a hot compression process is in progress. To further explore the electrochemical performance of the optimized CSE layers, was employed, and the impedance behaviors of the symmetric cells fabricated with as-prepared CSEs were assessed accordingly. Figure 8 shows typical Nyquist plots of test cells at different potentials, where all the curves intersected the x-axis at high quency regions, followed by a depressed semicircle. According to Figure 8, the cha  Figure 7c, the ∆E value gradually increased with an increased cycle number, which was primarily due to the gradual aging of the CSE layer. However, the capacity retention remained high (i.e., 86.8%), even after 200 cycles. The variation in the capacity retention and coulombic efficiency of the Li-metal battery fabricated with the optimal CSEs is shown in Figure 7d. As clearly illustrated in Figure 7d, the Li-metal batteries were cyclically charged/discharge at 0.5 C for 200 cycles. The capacity retention was maintained at >85%, with a stable coulombic efficiency after 200 cycles. Indeed, such a robust performance was likely due to: (i) the lack of dissolution of the iron from the LFP cathode [44,45], (ii) the uniform formation of the solid electrolyte interphase layer, and (iii) the lack of a substantial formation of Li dendrites [46,47] during the long-duration cycling, confirming that this was a stable design of the CSE layer for the Li-metal batteries. However, the gradual aging of the optimal CSEs requires an in-depth investigation. This may presumably be due to a slight electrode polarization during the charge/discharge process, causing the accumulation of Li metal at the interface between the Li anode/CSE/LFP cathode. Accordingly, work regarding nanoscaled LATP, a thinner CSE layer, and a hot compression process is in progress.
To further explore the electrochemical performance of the optimized CSE layers, EIS was employed, and the impedance behaviors of the symmetric cells fabricated with the as-prepared CSEs were assessed accordingly. Figure 8 shows typical Nyquist plots of the test cells at different potentials, where all the curves intersected the x-axis at high frequency regions, followed by a depressed semicircle. According to Figure 8, the charge transfer resistance associated with the CSE film (i.e., the depressed semicircle within the Nyquist plots) demonstrated a decreasing function of the applied potential. One equivalent circuit, as shown in the inset of Figure 8, was proposed to describe the EIS behavior and to quantify the ionic conductivity (σ) of the CSE layer. The equivalent circuit model contained the following components: R e , R inf , Q C , and Z w , representing the resistance of the bulk electrolyte, the interfacial charge transfer resistance, the constant-phase element, and the Warburg impedance associated with the Li + diffusion within the electrode, respectively [48]. The Z-view software package was employed to analyze the impedance spectra of the CSE layer, where the deviation between the experimentally recorded impedance spectra and the model predictions was less than 10%. The ionic conductivity was subsequently calculated using the formula σ = l/R e A [49,50], where the thickness of the CSE layer (l) and the projected active area (A) were used within the formulation.
Membranes 2023, 13, x FOR PEER REVIEW 9 of 12 contained the following components: Re, Rinf, QC, and Zw, representing the resistance of the bulk electrolyte, the interfacial charge transfer resistance, the constant-phase element, and the Warburg impedance associated with the Li + diffusion within the electrode, respectively [48]. The Z-view software package was employed to analyze the impedance spectra of the CSE layer, where the deviation between the experimentally recorded impedance spectra and the model predictions was less than 10%. The ionic conductivity was subsequently calculated using the formula σ = l/Re A [49,50], where the thickness of the CSE layer (l) and the projected active area (A) were used within the formulation. The ionic conductivity of the CSE sample remained relatively unchanged (2.83 × 10 −4 , 2.45 × 10 −4 , and 2.33 × 10 −4 S cm −1 at 0, 0.5, and 1.0 V vs. Li/Li + , respectively), regardless of the applied potential, confirming a highly stable CSE structure. This finding confirms that the σ value of the as-prepared and highly optimized CSE layer was significantly enhanced via engineering the CSE configuration compared to the LATP-based electrolytes reported Figure 8. Nyquist plots of the symmetric cell fabricated with the optimal CSEs at different potentials, where the inset shows a proposed equivalent circuit. The elements, R e , R inf , Q C , and Z w , represent the resistance of the bulk electrolyte, the interfacial charge transfer resistance, the constant-phase element, and the Warburg impedance associated with the Li+ diffusion within the electrode, respectively.
The ionic conductivity of the CSE sample remained relatively unchanged (2.83 × 10 −4 , 2.45 × 10 −4 , and 2.33 × 10 −4 S cm −1 at 0, 0.5, and 1.0 V vs. Li/Li + , respectively), regardless of the applied potential, confirming a highly stable CSE structure. This finding confirms that the σ value of the as-prepared and highly optimized CSE layer was significantly enhanced via engineering the CSE configuration compared to the LATP-based electrolytes reported in the literature (i.e., those of approximately 1.11 × 10 −4 S cm −1 [13]). For comparison, Tables S1 and S2 in Supplementary Materials show the calculated ionic conductivities of various CSEs in the Li-metal batteries, based on the Nyquist plots, incorporated with the proposed equivalent circuit. We observed that both a higher LiTFSI/LATP ratio and a thinner CSE thickness displayed a positive effect on the ionic conductivity, as demonstrated in the proceeding sections. This reveals that there was an optimal parameter setting in the robust design of the CSEs. Indeed, such an enhanced ionic conductivity at room temperature is largely due to the lower crystallinity of the polymeric electrolytes (i.e., PVDF-HFP) in the presence of the well-dispersed LATP particles within the hybrid heterogeneous structure, which had a critical role in boosting the ionic conductivity. Accordingly, the optimal parameters for the CSE layer demonstrated in this work can be adopted to create a three-dimensional conductive network for facilitating ionic conductivity within hybrid electrolytes while alleviating the interfacial electrolyte/electrode resistance.

Conclusions
In this work, an efficient technique was developed for fabricating high-performance CSEs containing NASICON-type LATP particles, PVDF-HFP, and LiTFSI salt for Li-metal batteries. The key design parameters (i.e., LiTFSI/LATP, CSE thickness, and carbon content) were engineered to enhance the batteries' specific capacity, coulombic efficiency, and cyclic stability. The coin cells equipped with the highly optimized CSEs and assembled with Li-metal anodes and the LFP cathodes demonstrated high specific capacity (~160 mAh g −1 ), reduced electrode polarization (~0.12 V), and superior cyclic stability (capacity retention of 86.8%) after 200 cycles. The ionic conductivity of the optimized CSE layer reached as high as 2.83 × 10 −4 S cm −1 at an ambient temperature. The remarkable cycling performance of the CSE layer demonstrated in this work was primarily due to: (i) the lack of dissolution of iron from the LFP cathode, (ii) the uniform formation of the solid electrolyte interphase layer, and (iii) the lack of substantial Li dendrite growth during the extended cycling. The framework established in this study for designing high-performance solid-state electrolytes with NASICON-type ceramic particles can be adopted to create a conductive ionic pathway for facilitating Li + ionic transport within an electrolyte while alleviating the interfacial electrolyte/electrode resistance in Li-metal batteries. Accordingly, the CSE structures containing LATP powders prepared in this work can be applied for substantially boosting ionic conductivity, specific capacity, and cycle life while mitigating the interfacial resistance of the electrolyte/electrode layer, Li dendrite formation, and ionic diffusion resistance during long-duration cycling for lithium-ion batteries using Li metal as the anode electrode.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

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