Binary anion and cation co-doping enhance sulfide solid electrolyte performance for all-solid-state lithium batteries

INTRODUCTION

With the burgeoning expansion in electronic devices, electric vehicles, and energy storage solutions, the demands placed on lithium-ion batteries have grown significantly^[1-5]. Traditional lithium-ion batteries utilizing liquid electrolytes face safety concerns, and their energy density is approaching a plateau. By transitioning to solid electrolytes, all-solid-state lithium batteries (ASSLBs) have emerged as a promising solution to address these challenges. It shows great potential for these batteries to offer increased safety, durability, higher energy density, and greater flexibility in battery packaging design^[6-10].

Solid electrolytes, a critical component of ASSLBs, have garnered increasing attention. Among various solid electrolytes, sulfide solid electrolytes have gained prominence due to their exceptional characteristics, including high ionic conductivity (> 1 mS cm^-1), favorable mechanical properties, and minimal grain boundary resistance. Nonetheless, solid-state batteries relying on sulfide electrolytes face significant hurdles, including poor air stability, chemical/electrochemical stability, and issues related to interfacial compatibility^[11-15].

One of the most effective strategies for enhancing the stability of sulfide electrolytes is through element doping. Anion doping strategies have witnessed prominence, with partial substitution of sulfur by oxygen being the prevailing approach^[16,17]. This strategy is widely employed because it confers notable benefits. Firstly, by incorporating oxygen atoms, the system forms non-bridging oxygen (O), which is more stable than bridging sulfur (S), thereby improving the overall stability of the electrolyte. Furthermore, the lattice mismatch between oxygen ions and oxide-based positive electrodes is minimal. Partially replacing sulfur with oxygen can effectively prevent oxygen from the positive electrode from infiltrating the sulfide electrolyte, thereby significantly inhibiting undesirable interface side reactions in sulfide-based ASSLBs. However, the incorporation of oxygen impedes the migration of Li⁺. Notably, as the level of oxygen doping increases, the conductivity of the electrolyte may experience a gradual decline. On the other hand, when considering cation doping, the selection of dopants aligns with the Hard and Soft Acid-Base Theory (HSAB). According to this theory, hard acids have a preference for reacting with hard bases, while soft acids tend to react more readily with soft bases. Combining a softer acid, such as Sn⁴⁺, Ge⁴⁺, As⁵⁺, and Sb⁵⁺, with the soft base sulfur (S) can significantly enhance the air stability of the sulfide electrolyte^[18-20].

Oxide doping constitutes a synergistic amalgamation of anion and cation doping methodologies, demonstrating the capability to concurrently enhance the conductivity and stability of sulfide electrolytes^[21-25]. Liu et al. presented a pioneering endeavor by doping of Li₃PS₄ glass-ceramic electrolytes with ZnO, demonstrating the synthesis of a new Li_3+3xP_1-xZn_xS_4-xO_x solid electrolyte^[26]. This advancement featured the incorporation of Zn, replacing a fraction of P, while O replaced a segment of S within the Li₃PS₄ matrix. The resultant electrolyte showcased enhanced air stability alongside improved ionic conductivity. Notably, there was no obvious structure change in the Li_3+3xP_1-xZn_xS_4-xO_x electrolyte after exposure to air. Ahmad et al. synthesized a Li_6.988P_2.994Nb_0.2S_10.934O_0.6 glass-ceramic electrolyte with Nb and O co-doping^[27]. Wherein oxygen partially supplanted the sulfur bridge bond within the P₂S₇^4- group, thereby engendering the formation of P₂OS₆^4- entities, which contributed significantly to augmenting the air stability of sulfide electrolyte. In addition, the increase in the crystallinity of the conductive phase coordinated a reduction in the activation energy associated with the migration of lithium ions, which culminated in an appreciable enhancement of ionic conductivity within the glass-ceramic electrolyte. Xu et al. introduced a congener substitution strategy to optimize the electrochemical performance of Li₇P3S₁₁ (LPS) via chemical bond and structural regulation^[28]. The as-obtained Li₇P_2.9Sb_0.1S_10.75O_0.25 solid electrolyte shows improved solid-solid interfacial compatibility compared to LPS. Such enhancements are conducive to augmenting the reversible capacity, rate capability, coulombic efficiency, and cycling stability of ASSLBs. The oxide doping strategy has emerged as an effective methodology for mitigating the air sensitivity and instability issues associated with sulfide electrolytes without compromising the critical ionic conductivity. A systematic investigation of the effects of doping on the ionic conductivity, electrochemical stability, and interfacial stability is essential to guide the selection of appropriate doping elements and concentrations for optimizing the performance of the solid electrolytes.

In this work, the concurrent anion and cation co-doping strategy is proposed for optimizing the structural and electrochemical properties of LPS-based solid electrolytes employing oxides as deponents [Figure 1A]. We systematically investigate the effects of SiO₂, GeO₂, and SnO₂ doping on the structure, ionic conductivity, oxidative stability, and lithium plating/stripping behavior of Li_7+xP_3-xM_xS_11-2xO_2x (M = Si, Ge, Sn) (LPS-xMO₂) electrolytes. The electrochemical performance of the doped LPS-xMO₂ electrolytes is evaluated in ASSLBs. It reveals that this co-doping strategy yields improved ionic conductivity and enhanced electrochemical stability. Specifically, SnO₂-doped LPS-xSnO₂ electrolytes exhibit higher ionic conductivity, while SiO₂-doped LPS-xSiO₂ electrolytes exhibit superior oxidative stability. Consequently, ASSLBs based on NCM811 cathodes and LPS-0.05SnO₂ electrolytes demonstrate higher initial capacity, while those utilizing LPS-0.05SiO₂ electrolytes exhibit superior initial coulombic efficiency, rate capability, and cycling stability.

Figure 1. (A) Schematic illustrating the preparation process of sulfide electrolyte. X-ray diffraction patterns of the doped sulfide electrolytes (B) LPS-xSiO₂, (C) LPS-xGeO₂, (D) LPS-xSnO₂ (x = 0, 0.02, 0.05, 0.08, 0.1) and Raman spectra of the doped sulfide electrolytes (E) LPS-xSiO₂, (F) LPS-xGeO₂, (G) LPS-xSnO₂ (x = 0, 0.02, 0.05, 0.08, 0.1).

EXPERIMENTAL

RESULTS AND DISCUSSION

The crystal structure and structural units of the electrolytes were characterized through XRD and Raman spectroscopy. The XRD results [Figure 1B-D] show that when the doping amount is ≤ 0.05 mol, the diffraction patterns of the doped sample remain in congruence with that of the pristine LPS electrolyte, thus preserving an intact body-centered cubic structure. However, for doping concentration exceeding 0.08 mol, unidentified new diffraction peaks are evident, signifying the limits of dopant concentration. In addition, the lattice constant variation of LPS with an increasing concentration of MO₂ [Supplementary Table 1] and XRD patterns for reference materials (LPS, MO₂) [Supplementary Figure 1] was analyzed. It shows that as the doping concentration of MO₂ increases, the lattice parameters exhibit a gradual increase within the range of 0 ≤ x ≤ 0.08, indicative of the successful incorporation of MO₂ into LPS. Raman spectra [Figure 1E-G] show peaks at approximately 407, 421 and 386 cm^-1 across all spectra. The peaks centered at 407 and 421 cm^-1 correspond to the symmetric stretching vibration of the P-S bonds in the P₂S₇^4- and PS₄^3- units, respectively, indicating the formation of a highly conductive crystalline LPS phase. A minor peak at approximately 386 cm^-1 corresponds to the low-conductivity phase of Li₄P₂S₆^[29-31]. Furthermore, it is noteworthy that the moderate co-doping of anion and cation atoms into LPS exerts a relatively modest impact on the fundamental framework of the LPS-xMO₂ glass-ceramic electrolytes in comparison to the unaltered LPS electrolyte. Supplementary Figure 2 shows the morphology and element distribution of the typical LPS-0.05MO₂ (M = Si, Ge, Sn) electrolytes characterized by SEM. A typical spherical powder morphology is discernible, with diameters of about 1-3 µm. Notably, Si, Ge, Sn, and O elements are evenly distributed in the electrolyte particles.

The optimal doping concentration was determined by measuring the ionic conductivity of LPS-xMO₂(M = Si, Ge, Sn, x=0, 0.02, 0.05, 0.08, 0.1) electrolytes through EIS employing a symmetrical SS/SSE/SS cell configuration [Figure 2A-C]. The ionic conductivity of pristine LPS was determined to be 1.07 × 10^-3 S cm^-1[Supplementary Figure 3]. As the doping level increased, the ionic conductivities exhibited a gradual increment followed by a decline. Remarkably, LPS-0.05SiO₂, LPS-0.08GeO₂, and LPS-0.05SnO₂ electrolytes reached a peak ionic conductivity of 1.32 × 10^-3, 2.06 × 10^-3, and 2.53 × 10^-3 S cm^-1, respectively [Supplementary Figure 4]. The LPS-0.05SnO₂ electrolyte exhibits a 2.5 times improvement in ionic conductivity compared to pristine LPS. However, with further increasing doping levels, the EIS curves displayed a distinctive semicircle in the high-frequency region due to the presence of impurities in the electrolytes, consistent with XRD results [Supplementary Figure 5]. Moreover, the activation energies of the electrolytes were derived from Arrhenius plots based on the Arrhenius equation:

where σ, T, k, and A represent the Li⁺ conductivity, the absolute temperature, the Boltzmann constant, and the pre-exponential factor, respectively. The Arrhenius plots of activation energy are presented in Figure 2D-F. Notably, at higher doping concentrations, the electrolytes exhibited elevated activation energies, indicative of more pronounced barriers to lithium-ion migration.

Figure 2. Nyquist plots for the solid electrolytes measured at room temperature. (A) LPS-xSiO₂ (x = 0.02, 0.05, 0.08, 0.1). (B) LPS-xGeO₂ (x = 0.02, 0.05, 0.08, 0.1). (C) LPS-xSnO₂ (x = 0.02, 0.05, 0.08, 0.1). Corresponding Arrhenius plots of ionic conductivity of solid electrolytes. (D) LPS-xSiO₂ (x = 0, 0.02, 0.05, 0.08, 0.1). (E) LPS-xGeO₂ (x = 0, 0.02, 0.05, 0.08, 0.1). (F) LPS-xSnO₂ (x = 0, 0.02, 0.05, 0.08, 0.1).

It was found that the ionic conductivity of glass-ceramic electrolytes is correlated to the proportion of the high conductive crystalline LPS phase characterized by the typical local structural units of P₂S₇^4- and PS₄^3-[32]. When cations displace P⁵⁺ ions and bond with S^2-, two possible scenarios emerge: the replacement of either P₂S₇^4- or PS₄^3-. Importantly, the energy required for substituting P₂S₇^4- is lower than that for PS₄^3-, thus favoring cation doping to favorably target the P₂S₇^4-[33,34]. Additionally, the ionic radii of Si⁴⁺, Ge⁴⁺, and Sn⁴⁺ are larger than that of P⁵⁺ ions, which means that cation-doping will broaden the lithium-ion conducting channels and generate lithium defects, thereby facilitating lithium-ion transport.

XPS was employed to assess the effects of doping on the content of P₂S₇^4- and PS₄^3- structural units. Taking SnO₂ doping as an example, Figure 3A shows that in bare LPS, the ratio of P₂S₇^4- (131.2 eV) to PS₄^3-(132.1 eV) is approximately 1.93:1, closely resembling the theoretical ratio of 2:1^[35,36]. With increasing doping levels, a portion of P₂S₇^4- is progressively replaced, resulting in diminishing ratios of P₂S₇^4- to PS₄^3- of 1.87:1, 1.76:1, 1.4:1, and 1.23:1 for LPS-0.02SnO₂, LPS-0.05SnO₂, LPS-0.08SnO₂, and LPS-0.1SnO₂, respectively. Assuming that the LPS structure remains unaltered, the theoretical ratios of P₂S₇^4- and PS₄^3- in LPS-0.08SnO₂ and LPS-0.1SnO₂ should be 1.84:1 and 1.8:1, respectively. However, the substantial decrease in the proportion of P₂S₇^4- groups in LPS doped with 0.08 mol and 0.1 mol SnO₂ indicates that the optimal doping amount to maintain the highly conductive LPS phase structure is approximately 0.05 mol. Similarly, the peak intensity of P-S-P in the P₂S₇^4- unit [Figure 3B] decreases, while the Sn⁴⁺ peak in the Sn 3d spectra and O^2- peak in the O 1s spectra [Figure 3C and D] increase significantly with increasing the doping contents, indicating an elevated oxide doping level in the electrolytes. Similar trends are also observed in the XPS spectra for samples doped with SiO₂ and GeO₂ [Supplementary Figures 6 and 7]. Moreover, the XPS peak positions of Si, Ge, and Sn in the doped electrolytes align with the peaks of Si-S, Ge-S, and Sn-S bonds of Li₁₀SiP₂S₁₂, Li₁₀GeP₂S₁₂, and Li₁₀SnP₂S₁₂^[37-39], further confirming the successful incorporation of SiO₂, GeO₂, and SnO₂ into the LPS electrolyte.

Figure 3. XPS spectra of (A) P 2p, (B) S 2p, (C) Sn 3d and (D) O 1s for LPS, LPS-0.02SnO₂, LPS-0.05SnO₂, LPS-0.08SnO₂ and LPS-0.1SnO₂.

The oxidative decomposition of solid electrolytes has adverse effects on cathode impedance and overall cell performance. To probe the impact of oxide doping on the oxidative decomposition of LPS-based electrolytes, we employed a half-cell setup, with Li-In serving as the anode and a composite of carbon black (AB) (40 wt.%) mixed with LPS-xMO₂ (60 wt.%) on the cathode, composite cathode structures greatly increase the contact area and accelerating the kinetic process of electrolyte decomposition, while close contact greatly reduces the interfacial charge transfer resistance and effectively suppresses polarization. Compared to semi-blocking electrodes, mixed electrodes more accurately reflect the conditions in all-solid-state batteries^[40]. CV was conducted on the Li-In|LPS-xMO₂|LPS-xMO₂-AB cells within a potential range between 2.0 to 4.5 V (vs. Li/Li⁺) at a scan rate of 1 mVs^-1 [Figure 4]. In the case of the pristine LPS electrolyte, the initial positive sweep from open circuit voltage (OCV) revealed a substantial oxidative peak between 3.0 and 4.5 V, with no discernible reductive peak during the subsequent negative sweep from 4.5 to 2.0 V. During the second cycle, the oxidative peak shifted to a higher voltage, and the intensity of the peak significantly diminished, indicating the passivation of the carbon/LPS-xMO₂ electrolyte interphase after oxidation. All the doped LPS-xMO₂ electrolytes exhibited a higher onset oxidative potential, coupled with a lower anodic peak current density during the first cycle compared to the LPS electrolyte. Furthermore, the onset oxidative potential exhibited an increasing trend with rising doping content, concomitant with a noticeable reduction in oxidative peak current density [Supplementary Figure 8]. Similar trends were also observed in the subsequent second cycle. These indicate that the incorporation of SiO₂, GeO₂, and SnO₂ consistently enhanced the oxidative stability of the LPS electrolyte. Among these, LPS-SiO₂ demonstrated the lowest oxidative peak current density due to the non-metallic properties of Si^[32]. In order to analyze the impact of oxide doping on the interface, EIS and XPS were tested on the (LPS + AB/LPS/Li) cells at different voltages. From Supplementary Figures 9-12, it can be evident that upon charging to 4.5 V, the interfacial impedance experiences a significant increase. Subsequent discharge to 2.0 V results in a decrease in the interfacial impedance. Notably, compared to pristine LPS, the cells with doped electrolytes exhibit a substantial reduction in interfacial impedance. Moreover, XPS spectra reveal that electrolytes are oxidized at high voltage, forming oxidation products such as elemental sulfur and P₂S₅. In the doped electrolyte, sulfate ions are detected. During the subsequent discharge to 2.0 V, the oxidation products decrease. It is worth noting that a small amount of oxidation products is still detected in the XPS spectra of a pristine electrolyte after the charge/discharge process, indicating the irreversible oxidative decomposition of the electrolyte at high voltage. These findings align with the CV results, demonstrating that the incorporation of SiO₂, GeO₂, and SnO₂ can effectively suppress the oxidative decomposition of the LPS-xMO₂ electrolytes during cycling to high voltage. As illustrated in Supplementary Figure 13 and Supplementary Table 2, the electronic conductivity of the LPS-xMO₂ electrolytes was assessed through direct current (DC) polarization, employing an ion-blocking cell configuration comprising two blocking electrodes at a bias voltage of 0.5 V. Both the pristine LPS and doped LPS-xMO₂ electrolytes show a low electronic conductivities of about 1.1~1.2 × 10^-9 S cm^-1, which are far lower than the ionic conductivities. The similar values of the pristine and doped LPS-xMO₂ electrolytes suggest that the electronic conductivity does not serve as the decisive factor influencing the electrochemical properties in the doped electrolytes^[41-43].

Figure 4. CV curves of the Li/SSEs/SSEs + AB cells within the voltage range from 2 to 4.5 V (vs. Li/Li⁺), (A) the first and second cycles curves of LPS, and the second cycle curves of (B) LPS-xSiO₂ (x = 0.02, 0.05, 0.08, 0.1), (C) LPS-xGeO₂ (x = 0.02, 0.05, 0.08, 0.1), (D) LPS-xSnO₂ (x = 0.02, 0.05, 0.08, 0.1)

Considering both ionic conductivity and electrochemical stability, LPS-0.05MO₂ electrolytes were selected for further electrochemical characterizations. To assess the dendrite-suppressing capability of LPS and LPS-xMO₂ electrolytes, symmetric Li/LPS-xMO₂/Li cells were constructed to determine the critical current density^[44]. As depicted in Figure 5, the critical current density of the pristine LPS electrolyte is 0.64 mA cm^-2 at room temperature. In contrast, the doped LPS-xMO₂ electrolytes exhibit higher critical current densities, with values of 1.14 mA cm^-2 for LPS-0.05SiO₂, 1.02 mA cm^-2 for LPS-0.05GeO₂, and 0.78 mA cm^-2 for LPS-0.05SnO₂, respectively. Notably, LPS-0.05SiO₂ displays the highest critical current density, while LPS-0.05SnO₂ exhibits a lower polarization voltage. The impact of doping on the stripping/platting behaviors of lithium metal was further assessed using symmetric Li/SSE/Li cells. Supplementary Figure 14 illustrates the time-dependent galvanostatic voltage profiles of the Li/SSE/Li cells operating at 0.1 mA cm^-2. The Li/LPS/Li cell exhibited the highest overpotential of ~0.03 V. After 40 h of cycling, the voltage curve exhibited a sudden drop, which was attributed to a short circuit induced by the formation of lithium dendrites within the LPS electrolyte. In contrast, the stripping/plating behavior of the symmetric cell utilizing doped LPS-xMO₂ electrolytes remained stable for 200 h cycling and displayed lower overpotential. This enhanced performance can be attributed to reduced interfacial resistance and improved interfacial stability, thus suppressing lithium dendrite formation. Importantly, among the three doped samples, the SSE doped with SiO₂ exhibited the most robust cycling stability, whereas the SSE doped with SnO₂ displayed lower polarization voltage.

Figure 5. Lithium plating/stripping voltage profiles of (A) Li/LPS/Li, (B) Li/LPS-0.05SiO₂/Li, (C) Li/LPS-0.05GeO₂/Li, and (D) Li/LPS-0.05SnO₂/Li cells at different current densities.

To investigate the interfacial reactions between the electrolyte and lithium metal, XPS was employed for characterization^[45]. The sample doped with SnO₂ is used as an example; analysis of XPS spectra [Figure 6] reveals insightful information. In the O 1s spectrum of pristine LPS-0.05SnO₂, two distinct peaks are discerned, attributed to P-O-P and Li-O-P interactions. Upon cycling, the O spectrum collected at the Li/LPS-0.05SnO₂ interface additionally shows a Li₂O peak at 526.6 eV. The presence of Li₂O in the interphase would modulate interfacial lithium deposition behaviors and maintain the stability of the SSE/Li interphase, thus inhibiting lithium dendrite formation within the SSE^[46,47]. The initial S spectrum of LPS-0.05SnO₂ is ascribed to non-bridging sulfur Li-S-P and bridging sulfur P-S-P. However, after cycling, the S spectrum at the LPS-0.05SnO₂/Li interface exhibits an additional peak at 161.0 eV, attributed to Li₂S. Similarly, the P spectrum [Figure 6C] after cycling displays a Li₃P peak (129.3 eV), which was absent in the original sample. As for the pristine LPS electrolyte, the XPS spectra of S 2p and P 2p show similar variations to that of LPS-0.05SnO₂. The initial S spectrum is ascribed to non-bridging sulfur Li-S-P and bridging sulfur P-S-P. An additional peak at 161.0 eV, attributed to Li₂S, is observed after cycling [Supplementary Figure 15A]. A Li₃P peak (129.3 eV) absent in the original sample can be observed in the P spectrum [Supplementary Figure 15B] after cycling. In pristine LPS-0.05SnO₂ electrolyte [Figure 6D], Sn is in the +4 valence state, with binding energies of the Sn 3d5/2 peak at 485.4 eV. After cycling, the collected Sn 3d spectrum at the LPS-0.05SnO₂/Li interphase demonstrates an elemental Sn peak at binding energies of 483.7 eV, signifying the reduction of +4 valence Sn at the interphase to elemental Sn by lithium metal^[48]. This reduction process results in the formation of a lithium alloy phase at the interphase. A similar phenomenon is observed in the XPS data of solid electrolytes doped with SiO₂ and GeO₂[Supplementary Figure 16]. The formation of Li₂O, Li₂S, and Li₃P species facilitates the passivation of the Li/electrolyte interphase, which effectively inhibits the continuous reductive degradation of the solid electrolyte.

Figure 6. XPS spectra of (A) O 1s, (B) S 2p, (C) P 2p, and (D) Sn 3d for LPS-0.05SnO₂ sulfide solid electrolytes before and after five cycles.

To further assess the electrochemical performance of the LPS-xMO₂ electrolyte, ASSLBs were constructed utilizing NCM811 as the cathode material, LPS-xMO₂ as solid electrolytes, and Li-In as the anode material, respectively. Figure 7A shows the initial charge-discharge voltage profiles of these cells employing various electrolytes. Compared to the bare LPS, the cells with doped LPS-xMO₂ electrolytes exhibit higher initial coulombic efficiency and improved discharge capacity. This can be attributed to the improved ionic conductivity and enhanced anodic oxidative stability of the doped LPS-xMO₂ electrolytes. Notably, among the doped LPS-xMO₂ electrolytes, the cells with LPS-0.05SiO₂ electrolytes show the highest initial coulombic efficiency, whereas the cell with LPS-0.05SnO₂ electrolytes demonstrates the highest initial discharge capacity. In addition, Figure 7B illustrates the rate capability of the cells at varying current rates. All the cells employing doped LPS-xMO₂ electrolytes exhibit improved rate capability compared to the cells with the pristine LPS electrolyte. Although the cells with LPS-0.05SnO₂ electrolyte delivers the highest discharge capacity at 0.05 C, the cell with LPS-0.05SiO₂ electrolytes shows the best rate capability with a high reversible capacity of 75 mAh/g at 3 C. Furthermore, as depicted in Figure 7C, the cells employing doped LPS-xMO₂ electrolytes also exhibit enhanced cycling stability compared to the cells with the pristine LPS electrolyte. Following the formation cycles at 0.05C, the bare LPS exhibits an initial discharge capacity of 154.9 mAh·g^-1 with a capacity retention of 74.63% after 200 cycles at 0.1 C. In comparison, LPS-0.05SiO₂ showcases an initial discharge capacity of 160.5 mAh·g^-1 with a capacity retention of 86%, LPS-0.05GeO₂ exhibits an initial discharge capacity of 174.6 mAh·g^-1 with a capacity retention of 81.8%, and LPS-0.05SnO₂ displays an initial discharge capacity of 184.9 mAh·g^-1 with a capacity retention of 81.1%. These results indicate that while the cell with LPS-0.05SnO₂ electrolyte demonstrates higher initial capacity, LPS-0.05SiO₂ electrolytes offer superior initial coulombic efficiency, rate capability, and cycling stability. This observation aligns with the previously conducted assessments of ionic conductivity and oxidative stability, wherein the LPS-0.05SiO₂ electrolyte demonstrated the highest oxidative stability, while the LPS-0.05SnO₂ electrolyte exhibited the highest ionic conductivity.

Figure 7. Electrochemical performance of the NCM811/SSEs/Li-In ASSLBs, (A) charge-discharge voltage profiles of the first cycle, (B) rate performance at 0.05C, 0.1C, 0.3C, 0.5C, 1C, and 3C, (C) cycling stability, and Coulombic efficiency.

CONCLUSIONS

In summary, an anion and cation co-doping strategy is proposed to functionally enhance both the ionic conductivity and electrochemical stability of the LPS-based sulfide electrolyte. A series of SiO₂-, GeO₂-, and SnO₂-doped Li_7+xP_3-xM_xS_11-2xO_2x (M = Si, Ge, Sn, x = 0, 0.02, 0.05, 0.08, 0.1) sulfide solid electrolytes are prepared via high-energy ball milling and sintering. The effects of doping on the structural and electrochemical properties were systematically screened. XPS characterization indicates that the cations demonstrate a preference for substituting the P⁵⁺ of the P₂S₇^4- unit within the LPS matrix, which expands the Li⁺ transport channels and generates lithium defects to facilitate ion conduction. This significantly enhances the ionic conductivity of the doped LPS-xMO₂ electrolytes. The LPS-0.05SnO₂ electrolyte exhibits a high ionic conductivity of 2.53 × 10^-3 S cm^-1, which is 2.5 times higher than the pristine LPS electrolyte. Further improving the doping levels hinders lithium-ion migration, resulting in elevated activation energy and decreased ionic conductivity. The oxide doping also significantly improves the oxidative stability of the LPS-xMO₂ electrolytes, which is crucial for maintaining stable electrode/electrolyte interphases. While the LPS-0.05SnO₂ electrolyte delivers higher ionic conductivity, the LPS-0.05SiO₂ electrolyte demonstrates superior oxidative stability and higher critical current density. The NCM811-based ASSLBs utilizing the doped LPS-0.05MO₂ electrolytes exhibit higher coulombic efficiency, superior discharge capacity, rate capability, and cycling stability compared to the pristine LPS electrolyte. The cell utilizing the LPS-0.05SnO₂ electrolyte exhibits a higher initial capacity, whereas the LPS-0.05SiO₂ electrolyte demonstrates superior initial coulombic efficiency, rate capability, and cycling stability. This is consistent with the characterization results of ionic conductivity and electrochemical stability, where the LPS-0.05SiO₂ electrolyte shows higher oxidative stability and the LPS-0.05SnO₂ electrolyte exhibits superior ionic conductivity. These findings shed light on doping strategies for optimizing solid electrolyte materials toward the development of high-performance ASSLBs.