With the introduction of China's dual carbon policy, there is a growing demand for electrified vehicles and scalable energy storage. Nevertheless, conventional liquid-state lithium battery systems pose safety hazards, and their energy density has essentially reached its limit. All solid-state lithium batteries (ASSLBs), offering both higher safety and energy density, as well as extended cycle life, hold the potential to fundamentally address these issues. As it is widely acknowledged, raising the upper cut-off charging voltage is a crucial method to drive the advancement of high-energy-density ASSLBs [1-8]. Therefore, to achieve high energy density in ASSLBs, it is crucial to develop solid-state electrolytes (SSEs) that are stable at high voltages [9-12]. These SSEs should possess a wide electrochemical stability window (ESW) with a high oxidation potential exceeding 4.2 V (vs. Li⁺/Li⁰).

In currently reported electrolytes, oxide electrolytes have demonstrated their potential for high oxidation potentials, exceeding 5.0 V (vs. Li⁺/Li⁰), making them compatible with high-voltage cathodes in high-energy-density applications [13,14]. However, their intrinsic rigidity often leads to poor electrode/electrolyte interface contact. Sulfide solid electrolytes garner significant attention owing to their exceptionally high ionic conductivity and ease of mechanical processing. Nevertheless, their limited ESW within the range of 1.5–2.5 V (vs. Li⁺/Li⁰) is also considered a primary obstacle to their utilization in high-voltage ASSLBs. Direct contact between sulfide electrolytes and layered oxide cathodes results in substantial interfacial resistance due to inevitable side reactions [15-19]. Additionally, the interfacial resistance is exacerbated by the solid electrolyte's decomposition under high voltage.

In recent years, halide electrolytes have emerged as newcomers in the field and have shown promising developments. Advanced halide solid electrolytes (e.g., Li₃InCl₆, Li₃ErCl₆, Li₂Sc_2/3Cl₄) have achieved Li-ion conductivity values exceeding 1 mS/cm at room temperature [20-26]. Ma et al. have designed even lower-cost solid electrolytes such as Li_1.75ZrCl_4.75O_0.5 with Li-ion conductivity as high as 2.42 mS/cm [27]. Additionally, halide electrolytes, such as chloride-based electrolytes, exhibit a relatively high theoretical potential of up to 4.2 V (vs. Li⁺/Li⁰). Asano et al. reported a Li₃YCl₆ SSE with a wide ESW of 0.62–4.21 V (vs. Li⁺/Li⁰) [28]. However, these compounds are still inadequate for the application in ASSLBs at higher voltages. The oxidation of lithium chloride electrolyte occurs above 4.3 V, resulting in the formation of lithium-deficient metal chloride products, such as YCl₃, InCl₃, and ErCl₃ [29,30]. These products lack sufficient Li⁺ conduction pathways, leading to continuous consumption of the SSEs and thereby reduces the overall battery performance. According to Mo et al.'s simulation results, fluoride-based SSEs exhibit the highest oxidation stability potential compared to other halide-based SSEs [29]. However, the strong electronegativity of fluorine (F) yields a relatively lower Li-ion conductivity.

In this work, partially substitution of Cl⁻ with F⁻ in the Li₃InCl₆ structure was employed to achieve acceptable Li-ion conductivity and high oxidation stability for the construction of high-voltage ASSLBs. The optimized Li₃InCl_5.5F_0.5 electrolyte demonstrates enhanced voltage stability, exceeding 4.5 V (vs. Li⁺/Li⁰). ASSLBs consisiting of the bare LiNi_0.7Mn_0.2Co_0.1O₂ and Li₃InCl_5.5F_0.5 electrolyte show superior electrochemical performance under upper cut-off voltages of 4.5 and 4.8 V (vs. Li⁺/Li⁰). Multiple characterizations were combined to further elucidate the working mechanism. This work proposes a strategy for designing high-voltage stable solid electrolytes for ASSLBs with high energy density.

To prepare the series of F-doped Li₃InCl₆ electrolytes, the mixture of starting materials (LiCl, InCl₃, and InF₃) was first ball-milled and followed by a heat treatment process to promote the crystallinity and Li-ion conductivities of the obtained materials. The primary diffraction peaks of the synthesized Li₃InCl_6-xF_x materials are well matched to the pure Li₃InCl₆ phase with a monoclinic phase (Fig. 1a). XRD refinements were conducted on Li₃InCl_5.5F_0.5 to elucidate structural distinctions. Fig. 1b displays the measured XRD patterns and their fits, demonstrating a good agreement with the simulated XRD peaks. The corresponding crystal structure is shown in Fig. 1c, and the relevant parameters obtained from the refinements can be found in Table S1 (Supporting information).The total resistances of the Li₃InCl_6-xF_x (x = 0, 0.3, 0.5, 0.7, and 0.9) materials decrease with increasing F⁻ dopant amounts (Fig. 1d), indicating that F⁻ doping strategy reduces Li-ion conductivity of the Li₃InCl₆ electrolyte. The room temperature Li-ion conductivities deduced from the resistance results are 1.40 mS/cm for Li₃InCl₆, 1.17 mS/cm for Li₃InCl_5.7F_0.3, 1.00 mS/cm for Li₃InCl_5.5F_0.5, 0.95 mS/cm for Li₃InCl_5.3F_0.7, and 0.48 mS/cm for Li₃InCl_5.1F_0.9, respectively (Fig. 1e). An obvious drops in Li-ion conductivities is observd as the dopant amount increases from x = 0.7 to x = 0.9, which aligns well with the XRD results that the intensity of the major diffraction peaks of the pure phase of Li₃InCl₆ experiences a sharp decline. Further Li-ion conductivities of these prepared samples under the selected temperatures were also confirmed that ionic conductivities decrease with increasing F⁻ amount in the structure in Fig. 1f. Considering the ionic conductivity of different compositions, Li₃InCl_5.5F_0.5 with an acceptable Li-ion conductivity of 1.00 mS/cm was chosen as the target SSE in this study. Electronic conductivity changes of Li₃InCl₆ electrolyte before and after introducing F⁻ was also investigated. (Fig. 1g) Finally, electrochemical stability of Li₃InCl_5.5F_0.5 electrolytes was also studied using the typical characterization method as reported [24]. As depicted in Fig. 1h, the Li₃InCl_5.5F_0.5 electrolyte exhibits superior voltage stability comapred to the bare Li₃InCl₆ electrolyte (around 4 V vs. Li⁺/Li⁰) with a high stable potential over 4.5 V (vs. Li⁺/Li⁰). To investigate the electrochemical performance of the electrolytes, ASSLBs consiting of the bare LiNi_0.7Mn_0.2Co_0.1O₂ and Li₃InCl₆/Li₃InCl_5.5F_0.5 electrolytes were assembled and cycled within the voltage range of 3.0–4.5 V (vs. Li⁺/Li⁰).

To validate the improved electrochemical stability of F-doped Li₃InCl₆ electrolytes, the bare LiNi_0.7Mn_0.2Co_0.1O₂ was selected as the cathode active material combined with the pristine Li₃InCl₆ and Li₃InCl_5.5F_0.5 electrolytes and Li-In anode to assemble all-solid-state lithium batteries. To aviod interfacail instability between the halide electrolytes and Li-In anode in the batteries, a thin layer of Li_5.5PS_4.5Cl_1.5 electrolyte was added to isolate the direct contact [31]. Therefore, three types of batteries configurations, LiNi_0.7Mn_0.2Co_0.1O₂/Li_5.5PS_4.5Cl_1.5/Li-In, LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆/Li₃InCl₆/Li_5.5PS_4.5Cl_1.5/Li-In, and LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5/Li₃InCl_5.5F_0.5/Li_5.5PS_4.5Cl_1.5/Li-In were constructed in this work. These batteries were cycled with a wide charge/discharge voltage range of 2.4–3.9 V (vs. Li-In, equivelent to 3.0–4.5 V vs. Li⁺/Li⁰) to assess their electrochemical stability. When cycled at 0.1 C, the LiNi_0.7Mn_0.2Co_0.1O₂/Li_5.5PS_4.5Cl_1.5/Li-In battery delivers an initial discharge capacity of 157.9 mAh/g with a coulombic efficiency of 69.19%, while these values for the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆/Li₃InCl₆/Li_5.5PS_4.5Cl_1.5/Li-In battery are 187.6 mAh/g and 77.97%, and for the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5/Li₃InCl_5.5F_0.5/Li_5.5PS_4.5Cl_1.5/Li-In battery are 218.9 mAh/g and 86.68%. The LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆ cathode mixture displays greater initial discharge capacity and coulombic efficiency compared to the LiNi_0.7Mn_0.2Co_0.1O₂-Li_5.5PS_4.5Cl_1.5. This is due to the superior interfacail stability of the Li₃InCl₆ electrolyte than the Li_5.5PS_4.5Cl_1.5 electrolyte. Sulfide electrolytes are known to undergo serious side reactions and space charge effects towards the bare layered cathode materials in a typical charge/discharge [32-34]. Replacing sulfide electrolytes in the cathode mixture with lithium halide electrolytes can significantly increase capacities and coulombic efficiencies during the first few cycles [35]. When zooming in on the charge/discharge voltage window, the side reaction effects become worse due to the degradation of sulfide electrolytes at a higher upper cut-off voltage of 4.5 V (vs. Li⁺/Li⁰) in this work. Based on the LSV test results shown in Fig. 1h, Li₃InCl_5.5F_0.5 exhibits superior voltage stability compared to bare Li₃InCl₆. This indicates that the Li₃InCl_5.5F_0.5 electrolyte may result in higher capacity and coulombic efficiencies in solid-state batteries. This assumption is supported by our battery performance results presented in Fig. 2. The fabricated LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5/Li₃InCl_5.5F_0.5/Li_5.5PS_4.5Cl_1.5/Li-In battery delivers the highest discharge capacity and Coulombic efficiency values among these three battery configurations in Fig. 2a. Moreover, the electrode polarizations of the cathode mixtures decrease in the sequence of LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5, LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆, and LiNi_0.7Mn_0.2Co_0.1O₂-Li_5.5PS_4.5Cl_1.5 mixtures, which are due to the electrochemical stability of different solid electrolyte when cycled with the bare LiNi_0.7Mn_0.2Co_0.1O₂ active material in a high cut-off voltage. During the subsequent cycling tests, the LiNi_0.7Mn_0.2Co_0.1O₂/Li_5.5PS_4.5Cl_1.5/Li-In battery exhibits a rapid decline in discharge capacities and can only retain 27.0% of its initial discharge capacity with a value of 42.7 mAh/g after 100 cycles. In contrast, the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆/Li₃InCl₆/Li_5.5PS_4.5Cl_1.5/Li-In battery demonstrates much higher discharge capacities and maintains 52.5% of its original value, with a discharge capacity of 98.4 mAh/g for the 100^th cycle. While the assembled LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5/Li₃InCl_5.5F_0.5/Li_5.5PS_4.5Cl_1.5/Li-In battery exhibits the best cycling performance among these three batteries. Specifically, it provides a discharge capacity of 175.1 mAh/g after 100 cycles with a capacity retention of 80.0%. The solid-state lithium batteries that contain Li₃InCl_5.5F_0.5 electrolytes in both the cathode mixture layer and solid electrolyte layer demonstrate the highest discharge capacity retention among these batteries. Furthermore, the dQ/dV plots of these different cathodes based on the charge/discharge curves in Fig. 2b were also performed. These curves depict characteristic oxidation–reduction peaks corresponding to various phase transitions during the processes of lithiation and delithiation. As illustrated in Fig. 2c, the oxidation/reduction peaks of the electrode based on LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5 highly overlap in different cycles. For the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆ electrode, at the 50^th cycle, the distinct oxidation peak at 3.65 V and the corresponding reduction peak at 3.56 V disappeared, indicating an irreversible phase transition. This is likely the cause of the sudden and severe capacity decay (Fig. 2d). In addition, based on LiNi_0.7Mn_0.2Co_0.1O₂-Li_5.5PS_4.5Cl_1.5 electrode, the dQ/dV curve at the 100^th cycle exhibits oxidation/reduction peaks that are broader and much lower in intensity compared to the corresponding peaks in the first cycle curve, indicating a deterioration in the reversibility of lithiation (Fig. 2e). The Nyquist plots of the three battery configurations before and after 100 cycles are shown in Figs. 2f-h. The changes in EIS spectra for the three batteries are primarily concentrated in the low-frequency region, as opposed to the high-frequency region. The equivalent circuits were employed to analyze all nyquist plots (Figs. S2 and S3 in Supporting information). In terms of the fitting outcomes outlined in Table S2 (Supporting information), the values of R_ct2 are the highest for all three battery configurations. This indicates that the primary contributor to the increase in impedance is the interfacial resistance between the Ni-rich layered oxide and the electrolyteHG. This aligns with findings reported in other relevant studies [36-39]. Additionally, by comparing the R_ct2 values of the three configurations, the battery based on Li₃InCl_5.5F_0.5 exhibits the smallest R_ct2 value after 100 cycles compared to the other two batteries, implying that Li₃InCl_5.5F_0.5 can enhance interfacial stability and possess better high-voltage tolerance. Furthermore, electrochemical performances of these batteries at a higher C-rate of 0.5 C within the same voltage window were also examined. The LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5 cathode shows higher discharge capacities and smaller electrode polarizations for the selected cycles compared to the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆ and LiNi_0.7Mn_0.2Co_0.1O₂-Li_5.5PS_4.5Cl_1.5 cathodes (Fig. 2i). All batteries exhibit good cycling performances when charged/discharged at 0.5 C (Fig. 2j), while these cathode mixtures contain lithium halide electrolytes (Li₃InCl₆ and Li₃InCl_5.5F_0.5) demonstrate markedly higher discharge capacities compared to the LiNi_0.7Mn_0.2Co_0.1O₂-Li_5.5PS_4.5Cl_1.5 cathode during cycling. The latter cathode exhibits discharge capacities of 95.3 mAh/g and 86.4 mAh/g for the 1^st and 80^th cycles when cycled at 0.5 C, respectively. In comparision, the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆ cathode shows much higher discharge capacities of 169.5 mAh/g for the first cycle and 156.3 mAh/g for the 80^th cycle, and the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5 cathode demonstrates even higher discharge capacities of 180.9 mAh/g at the beginning and 172.1 mAh/g after 80 cycles. The cathodes consisting of lithium halide electrolytes exhibit superior capacities compared to those using sulfide electrolytes due to better voltage stability at higher voltages and excellent interfacial stability with bare high nickel layered active materials, such as LiNi_0.7Mn_0.2Co_0.1O₂. Our previous research found that the interfacail instability between Li₃InCl₆ and Li_5.5PS_4.5Cl_1.5 electrolytes yields poor electrochemical performance for the corresponding batteries [16]. The improved battery performance of the Li₃InCl_5.5F_0.5-based solid-state battery, compared to the Li₃InCl₆-based battery, is attributed to the enhanced interfacial stability between the halide and Li_5.5PS_4.5Cl_1.5 electrolytes resulting from F-doping.

The rate capabilities of the above battery configurations were further studied to reveal the effect of F-doping on electrochemcial performances at higher cycling current densities under room temperature. All batteries were charged/discharged at these chosen C-rates of 0.1, 0.2, 0.5, 1.0, and 2.0 C between 2.4 V and 3.9 V (vs. Li-In). As illustrated in Figs. 3a-c, the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5 cathode shows higher discharge capacities and smaller electrode polarizations compared to the other two cathodes, LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆ and LiNi_0.7Mn_0.2Co_0.1O₂-Li_5.5PS_4.5Cl_1.5, when cycled at these chosen C-rates. When the charge/discharge C-rates increase step-by-step from 0.1 C to 2.0 C, and then recover back to 0.1 C, the LiNi_0.7Mn_0.2Co_0.1O₂-Li_5.5PS_4.5Cl_1.5 cathode delivers dsicharge capacities of 171.8, 143.9, 92.1, 38.1, 2.2, and 169.1 mAh/g, respectively. In contrast, the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆ cathode shows much higher corresponding discharge capacity values of 198.0, 179.3, 162.6, 139.1, 99.4, and 181.9 mAh/g, respectively. Whereas the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5 cathode diplays the highest discharge capacities at the same C-rates among these batteries, 222.3, 204.8, 182.3, 157.0, 120.2, and 209.7 mAh/g, respectively. Moreover, cycling performance results in Fig. 3d also confirm the conclusion that the solid-state battery utilizing Li₃InCl_5.5F_0.5 electrolyte exhibits the best rate capability. To further verify the improved voltage stability of Li₃InCl₆ electrolyte after F-doping, the Li₃InCl₆-based and Li₃InCl_5.5F_0.5-based ASSLBs were also cycled at 0.5 C with an even higher upper cut-off voltage of 4.2 V (vs. Li-In, corresponding to 4.8 V vs. Li⁺/Li⁰) at RT. The Li₃InCl₆-based battery demonstrates similar chagre/discharge profiles with comparable voltage plateau differences at the beginning when compared to the Li₃InCl_5.5F_0.5-based battery (Fig. 3e). With cycling goes on, the latter delivers much higher discharge capacities and smaller electrode polarizations for the chosen cycles (20^th, 40^th, 80^th). The corresponding dQ/dV curves obtained from the charge/discharge curves in Figs. 3f and g exhibit the same conclusion. These dQ/dV peaks observed for the Li₃InCl_5.5F_0.5-based battery are highly overlaped each other for different cycles (Fig. 3f), whereas these peaks for the Li₃InCl₆-based battery exhibit clear intensity decay in Fig. 3g, indicating a fast degradation of discharge capacities during cycling. After 80 cycles within the range of 3.0–4.8 V (vs. Li⁺/Li⁰), compared with the LiN_i0.7Mn_0.2Co_0.1O₂-Li₃InCl₆ electrode, the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5 electrode exhibits a much smaller voltage gap between the oxidation and reduction peaks (0.14 V vs. 0.61 V). This agrees well with the following cycling performance of these batteries in Fig. 3h. Compared to the Li₃InCl_5.5F_0.5-based battery, the Li₃InCl₆-based battery shows slightly lower discharge capacities during the first 40 cycles, and then delivers much lower discharge capacities in the subsequent 40 cycles. Specicallly, the Li₃InCl₆-based battery delivers a dischagre capacity of 177.8 mAh/g for the 3^rd cycle, and suffers a fast capacity degradation in the following 77 cycles and sustains a dischagre capacity of 44.8 mAh/g for the 80^th cycle with a capacity retention of 25.2%. For comparison, the Li₃InCl_5.5F_0.5-based battery shows discharge capacities of 189.4 mAh/g and 145.9 mAh/g for the corresponding 3^rd and 80^th cycles with a much higher capacity retention of 77.0%. The higher capacities and superior cyclability of the Li₃InCl_5.5F_0.5-based battery compared to the Li₃InCl₆-based battery is attributed to the enhanced electrochemical stability after introducing F⁻ in the structure of Li₃InCl₆ electrolyte when cycled at an ultrahigh upper cut-off voltage (4.8 V vs. Li⁺/Li⁰). This result is consistent with previous battery performance depicted in Fig. 2b. EIS was further performed on both battery configurations before and after 80 cycles to reveal the resistance variations. As illustrated in Figs. 3i-k, the Li₃InCl₆-based battery experiences a notable rise in overall resistance, peaking at 6480 Ω after 80 cycles. In contrast, the Li₃InCl_5.5F_0.5-based battery shows a smaller total resistance during the same cycling process, remaining at 584.9 Ω. Table S3 (Supporting information) lists the specific numerical values of the fitted impedances for each component. Due to the decreased electrochemical stability of Li₃InCl₆ electrolyte at 4.8 V (vs. Li⁺/Li⁰), these side reaction products caused by Li₃InCl₆ and LiNi_0.7Mn_0.2Co_0.1O₂ in the cathode mixture result in a large interfacail resistance, as detected in Fig. 3k. In contrast, Li₃InCl_5.5F_0.5 electrolyte with higher electrochemical stability shows excellent interface stability with the bare LiNi_0.7Mn_0.2Co_0.1O₂ when cycled at 4.8 V (vs. Li⁺/Li⁰), leading to a much smaller resistances for the corresponding battery.

Finally, structural changes of the LiNi_0.7Mn_0.2Co_0.1O₂ particles after cycling with the Li₃InCl₆ and Li₃InCl_5.5F_0.5 electrolytes with a charge/discharge voltage window of 3.0 and 4.5 V (vs. Li⁺/Li⁰) were also explored via ex-situ TEM. As shown in Fig. 4a, LiNi_0.7Mn_0.2Co_0.1O₂ particles maintained their particle integrity after cycling with the F-modified Li₃InCl₆ electrolyte. Additionally, well-defined lattice fringes with values of 0.45 and 0.60 nm which belonging to the LiInF₄ phase are also detected in the image, suggesting the generation of LiInF₄ during cycling in the cathode mixture (Figs. 4b and c). The uniform distribution of materials containing F and In at the edges of the selected particles is affirmed by the EDX mapping results (Fig. 4d). Previous reseach reported that LiInF₄ possesses a wide stable volatge window with a high upper cut-off voltage exceed 6.0 V (vs. Li⁺/Li⁰) [40]. The formation of a small amount of LiInF₄ phase can enhance the electrochemical stability and increase the intrinsic thermodynamic stability limitation of Li₃InCl_5.5F_0.5 electrolyte. This leads to a zoomed charge/discharge voltage window and mitigates the side reaction between the bare LiNi_0.7Mn_0.2Co_0.1O₂ and Li₃InCl_5.5F_0.5 under higher upper cut-off voltages, such as 4.5 and 4.8 V (vs. Li⁺/Li⁰). This analysis is consistent with previous calculation results [40,41]. In comparison, the partcile integrity of LiNi_0.7Mn_0.2Co_0.1O₂ is also remained after cycled with the bare Li₃InCl₆ electrolyte when cycled at 4.5 (vs. Li⁺/Li⁰), whereas the lattice fring of 0.58 nm assigned to the InCl₃ is observed in Fig. 4e, indicating the formation of InCl₃ phase in the LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆ cathode mixture. However, the InCl₃ phase exhibits poor stability under high voltage as reported [42]. Therefore, intense side reactions occur between the bare LiNi_0.7Mn_0.2Co_0.1O₂ and Li₃InCl₆ under 4.5 and 4.8 V (vs. Li⁺/Li⁰) in the cathode mixture when the correponding battery worked at these high upper cut-off voltage, leading to poor battery performances (Figs. 2b and 3h).

In summary, we have successfully designed F-doped Li₃InCl₆ electrolyte with comparable conductivity and enhanced electrochemical stability. By tuning the F⁻ dopant in the structure, the optimal electrolyte with a composition of Li₃InCl_5.5F_0.5 exhibits a RT Li-ion conductivity of 1.00 mS/cm and improved electrochemcial compatability towards bare LiNi_0.7Mn_0.2Co_0.1O₂ active materials under 4.5 and 4.8 V (vs. Li⁺/Li⁰). Therefore, the Li₃InCl_5.5F_0.5-based solid-state battery demonstrates superior charge/discharge capacities and coulombic efficiencies, rate capability, and cycling performance in comparison to the Li_5.5PS_4.5Cl_1.5-based and Li₃InCl₆-based batteries under the same testing conditions. The Li_5.5PS_4.5Cl_1.5-based battery exhibits rapid capacity decay due to intense side ractions between the sulfide electrolyte and high-nickel active materials. The battery initially provides a discharge capacity of 157.9 mAh/g and retains 27.0% of its capacity after 100 cycles at 0.1 C between 3.0 V and 4.5 V (vs. Li⁺/Li⁰). In comparision, these values are 187.6 mAh/g and 52.5% for the Li₃InCl₆-based battery, and 218.9 mAh/g and 80.0% for the Li₃InCl_5.5F_0.5-based battery, respectively. EIS results confirm that the differences in battery performance are closely related to the interfacial resistances of these different battery configurations. The Li₃InCl_5.5F_0.5-based battery displays the smallest interfacial resistances, while the Li_5.5PS_4.5Cl_1.5-based battery suffers the largest during cycling tests. Ex-situ TEM results verify that severe side reactions occur in the LiNi_0.7Mn_0.2Co_0.1O₂-Li_5.5PS_4.5Cl_1.5 cathode mixture during cycling, while good high voltage stable LiInF₄ phase and poor electrochemical stability InCl₃ phase are generated during cycling at higher upper cut-off voltages for LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl_5.5F_0.5 and LiNi_0.7Mn_0.2Co_0.1O₂-Li₃InCl₆ cathodes, respectively. Additionally, the Li₃InCl_5.5F_0.5-based battery shows superior electrochemical performance when compared to the Li₃InCl₆-based battery at higher charge/discharge C-rates and even at a higher upper cut-off voltage (4.8 V vs. Li⁺/Li⁰). This study sheds a light on designing solid electrolytes with a wide voltage window to achieve high energy densities in ASSLBs when empolying high-voltage cathode materials.

The authors declare no competing financial interest.

This work was supported by the National Key Research and Development Program of China (No. 2021YFB2400300). We also thank the National Natural Science Foundation of China (Nos. 52177214, 22205153) for supporting our work. We gratefully acknowledge the Analytical and Testing Center of HUST and Soochow University for the technical support.

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.109741.