Si-doped Li6PS5I with enhanced conductivity enables superior performance for all-solid-state lithium batteries
Liang Ming , Dan Liu , Qiyue Luo , Chaochao Wei , Chen Liu , Ziling Jiang , Zhongkai Wu , Lin Li , Long Zhang , Shijie Cheng , Chuang Yu
Lithium-ion batteries (LIBs) are widely used in various energy storage fields including electric vehicles and plenty of mobile appliances for their high energy density and promising chemical stability [1-5]. However, current LIBs are approaching its energy limit and encountering potential safety risks including thermal runaway, flammability. One promising alternative strategy is to replace flammable liquid electrolytes with non-flammable solid electrolytes due to their better safety and higher energy density, which can obtain enhancements with the application of Li metal. Nevertheless, one of the major obstacles hindering the scalable application of solid electrolytes is the poor room-temperature conductivity which typically between 10−5 S/cm and 10−4 S/cm. Moreover, the volume changes that occur in both the electrode materials and the solid electrolytes during cycling directly contribute to structure deterioration and a decline in battery performance: (1) The formation of space charge layer between the cathode particles and SEs; (2) Hinderance of ion transport due to the thick solid electrolyte layer; (3) Uneven distribution of cathode active materials. Therefore, the crucial factor in obtaining excellent electrochemical performance in ASSLBs lies in the development of solid electrolytes with high ionic conductivity and exceptional compatibility with electrodes. Among inorganic solid-state electrolytes, sulfide electrolytes are proved as the promising candidates in the field of solid-state electrolyte research due to the high room temperature ionic conductivity, and wide electrochemical window. Recently, the ionic conductivity of some sulfide-based solid electrolytes, such as Li10GeP2S11.7O0.3 [6], Li10.35Ge1.35P1.65S12 [7], Li10GeP2S12 [8], has demonstrated levels of conductivity comparable to, or even surpassing certain liquid electrolytes (10−2 S/cm).
Li-argyrodite Li6PS5I show considerable electrochemical stability towards lithium metal [9,10]. Due to the restricted Li+ transport path within cage-structure of Li6PS5I, it has a particularly low room temperature ionic conductivity (10−6 S/cm), which is correlated with the high migration barrier and the absence of the site disorder of I−/S2− due to the neatly arranged lattice structure [11]. Significant improvements in the ionic conductivity of Li6PS5I can be achieved through aliovalent substitution: Ge4+ [12], Sn4+ [13], and Si4+ [14] with larger ionic radius and different to replace the P5+. For instance, Zhang et al. [15] reported that the partially P substitution with Ge in the PS43− tetrahedral structure of Li6PS5I, Li6.6P0.4Ge0.6S5I has an ultra-high ionic conductivity of 1.84 × 10−2 S/cm at room temperature. Nevertheless, high cost of Ge and poor electrochemical stability with cathode active materials limit its widespread application in all-solid-state-batteries. By introducing Sn in the structure, Zhao et al. [13] have successfully synthesized Li6.2P0.8Sn0.2S5I with increased ionic conductivity of 3.5 × 10−4 S/cm. However, a solubility limit of approximately 20% Sn4+ is observed in Li6+xP1-xSnxS5I, which is indicative of no site disorder occurs within the structure. Due to the low cost and higher solubility limit, Si was chosen as the substitution on P site. Based on previous study, the series of Si-doped Li6PS5I exhibits a large drop in activation barrier and highly enhanced I−/S2− disorder are obtained when Si4+/P5+ ratio is 0.5 [16].
The ternary material-LiNixCoyMnzO2 (x + y + z = 1) exhibit high specific capacity and super-high stability [16-23]. Although the electrochemical stability of Li6.5P0.5Si0.5S5I obtains enhancement, the severe irreversible side reactions occur between the oxide cathode and sulfide electrolyte lead to an increasing interfacial resistance and decomposition during cycling. To mitigate this issue, an effective way is to establish a protective coating to the NCM secondary particles, such as LiNbO3 or Li2ZrO3, being regarded as the most practical coating materials [24]. Although coatings generally enhance the interfacial stability, the discharge capacity is sacrificed due to interfacial resistance between the electrolyte and the LiNi0.6Co0.2Mn0.2O2 [25-32]. Besides, employing a solid electrolyte that exhibits better compatibility with LiNi0.6Co0.2Mn0.2O2 active material and possesses satisfactory ionic conductivity as ionic conductor in the cathode composite represents an effective strategy to achieve both high energy density and enhanced cyclability simultaneously. Among different kinds of solid electrolytes equipped with above advantages, Li3InCl6 shows superior stability and high room temperature conductivity (10−3 S/cm), suggesting its promising potential to be applied in sulfide-based all-solid-state-lithium batteries [33,34].
In this work, we had successfully introduced Si4+ into the Li6PS5I to substitute the P5+ using ball milling method following by an annealing route to synthesize Li6.5P0.5Si0.5S5I. The refinement spectra of XRD diffraction data verified the structural changes, indicating that Si4+ did cause lattice expansion of argyrodite and increased the disorder of lattice sites. The ionic conductivity was significantly increased after subsequent high-temperature heat treatment (3.6 × 10−3 S/cm), indicating that Si doping formed a favorable diffusion transport path for Li ions. To moderate the side reactions between Li6.5P0.5Si0.5S5I and LiNi0.6Co0.2Mn0.2O2. Li3InCl6 with highly stability is adopted to design a new configuration. This designed configuration delivers an initial high discharge capacity of 150.2 mAh/g and maintains 78.5% of capacity at 0.5 C after 250 cycles at room temperature.
To unravel structural variations before and after Si4+ substitution, XRD Rietveld refinement of both Li6PS5I and Li6.5P0.5Si0.5S5I are performed using the GSAS. The low Rwp and Rp of those refinements indicates the high reliability [28]. As shown in Figs. 1a and b and Fig. S1a (Supporting information), the major diffraction peaks of both samples are well indexed to the referenced Li7PS6 (PDF#34–0688, space group of F-3m) [16]. The outcome of refined pattern with Rwp which was 5.56% indicates that Li6.5P0.5Si0.5S5I can be categorized as Face Center Cubic type. However, the major diffraction peaks of annealed Li6.5P0.5Si0.5S5I show slight shift to the lower angle, suggesting the partially substitution of P5+ with Si4+ has resulted in the lattice expansion of the argyrodite phase. Additionally, the lattice parameters of Li6PS5I and Li6.5P0.5Si0.5S5I: a = 10.2007 Å, and V = 1061.436 Å3; a = 10.2032 Å, and V = 1062.125 Å3. Additionally, the structural data of above information can well demonstrate the P which belongs to PS43− unit substitution by Si induces the increasing S2−/I− site disorder, thereby flattening the migration barrier. Based on the previous NMR result, the poor ionic conductivity of Li6PS5I is due to homogenous distribution of S2− and I− over 4a and 4c sites [7]. The visual structural image of the Li6.5P0.5Si0.5S5I shown in Fig. 1c illustrates that Si has substituted partially P in the PS43− structure, which in further functioning as the modified diffusion path for efficient lithium-ion conduction. Schematically illustrated in Fig. 1b, the formation of (Six/P1-x)Sx(3 + x)− facilities overall lattice expansion and the increased charge carrier density influences the lithium substructure. The impedance of Li6.5P0.5Si0.5S5I was tested in the temperature range of 30–70 ℃ and the resistance of it drops following increasing temperature as depicted in Fig. S1c (Supporting information) [35-38]. Consequently, the potential impacts enhance site disorder and reduce the migration barrier of rate-determining step. The morphology of Li6.5P0.5Si0.5S5I was characterized by SEM and the corresponding results are depicted in Fig. 1e. The prepared particle shows large size of ~ 200 µm with homogenous distribution of Si, I, P, and S elements. Further evidence from the Raman spectroscopy results is presented in Fig. S1b (Supporting information). The spectra of Li6PS5I exhibits a prominent peak at around 424 cm−1 with weaker peaks at 190 and 560 cm−1 [38], which are attributed to the PS43− units present in the structure. While for the obtained Li6.5P0.5Si0.5S5I electrolyte, a strong new peak near the characteristic peak of PS43− at 375 cm−1 is assigned to the SiS44− unit. Moreover, the strongest peak located at 424 cm−1 slightly shifts to the lower wavenumber direction which is related to the appearance of a longer bond and the reduction in the intensity justify the doping impact caused by the substitution, which can enhance the I−/S2− disorder [39-44].
Ionic conductivity of the prepared Li6PS5I and Li6.5P0.5Si0.5S5I were characterized via the AC impedance spectroscopy operating at different temperatures from room temperature to 70 ℃. As shown in Fig. 2a, the modified Li6.5P0.5Si0.5S5I delivers a much lower total resistance than that of the bare Li6PS5I at room temperature based on the Nyquist plots, indicating a higher Li-ion conductivity. Specifically, the Li6PS5I electrolyte demonstrates a room temperature conductivity of 1.36×10−5 S/cm, while the conductivity of Li6.5P0.5Si0.5S5I is 3.6 × 10−3 S/cm, which is threefold higher than the former (Li6PS5I). This suggests that the incorporation of Si in Li6PS5I structure can significantly enhance the ionic conductivity. In addition, temperature-dependent Li-ion conductivities were also plotted in Fig. 2b. The activation energy deduced from the Arrhenius plots are 0.28 eV and 0.17 eV for Li6PS5I and Li6.5P0.5Si0.5S5I electrolytes, respectively. Previous research has found that [31], when the Si4+/P5+ ratio exceeds 0.3, impurities are produced, and when the ratio of Si4+/P5+ further rises to 0.5, the I−/S2− site disorder and (P1-xSix)S4(3 + x)- lattice volume increase. The occurrence of lattice expansion phenomenon in the structure can promote the cage jump of Li+ in the PS43− lattices [43]. In accordance with charge conservation, the amount of Li-ion and vacancies would be increased to compensate the substitution of Si4+ with P5+ in tetragonal structure. This means an increased concentration of carriers, which promotes the overall diffusion between the lattices and therefore reduces the activation energy barrier in the diffusion process, yielding a facilitated Li-ion conductivity. However, the electronic conductivities of both electrolytes were also tested under 0.50 V whose results are shown in Fig. 2c, indicating the room temperature electronic conductivities of the pristine Li6PS5I and Li6.5P0.5Si0.5S5I electrolytes are 8.9 × 10−9 S/cm and 4.32×10−8 S/cm, respectively. Different from the pre-assumption that the overall property of Li6.5P0.5Si0.5S5I is enhanced compared to the pristine Li6PS5I. Conductive decomposition products may be the potential cause for accelerating electrolyte decomposition and severe degradation of the corresponding battery performance during the charge/discharge process. We suppose that the presence of SiS2, Li21Si5, and Li4SiS4 due to the decomposition of Si-doped electrolyte, which exhibits slight electrical conductivity, contributes to the higher electronic conductivity observed in Li6.5P0.5Si0.5S5I [38,39]. Besides, electrochemical stability of the prepared Li6.5P0.5Si0.5S5I electrolyte was also investigated by CV at the voltage range of 0.0–4.0 V (vs. Li-In) using the Li6.5P0.5Si0.5S5I + C/Li6.5P0.5Si0.5S5I/Li-In battery at room temperature. As depicted in Fig. 2d, a reduction peak located at 1.7 V and an oxidation peak at 2.0 V (vs. Li-In), demonstrating that the electrochemical window of the synthesized Li6.5P0.5Si0.5S5I is between 1.7 V and 2.0 V (vs. Li-In), which is good accordance with the reported articles [40-43]. Interestingly, double peaks are detected in the initial cycle and disappear in the subsequent cycles. That is associated with the formation of Li3PS4 during the 1st scan and the decomposition of Li3PS4 produce S and P2S5, which is reflected by the peak observed at 2.3 V (vs. Li-In). For Li6PS5I, its resistance is too high to scan the reduction peak in CV within this potential range
To evaluate the application of the prepared Li6.5P0.5Si0.5S5I solid electrolyte in full cell, the bare LiNi0.6Mn0.2Co0.2O2 and the Li-In alloy were chosen as the cathode and anode materials in combination with the Li6.5P0.5Si0.5S5I electrolyte to assemble all-solid-state lithium batteries. The assembled batteries were cycled between 2.4 V and 3.7 V (vs. Li-In) at different C-rates. As shown in Fig. 3a, the LiNi0.6Mn0.2Co0.2O2/Li6.5P0.5Si0.5S5I/Li-In battery shows slow increase of voltage from 2.4 V to 3.1 V during the initial charging process, reflecting the large space charge effect between the bare LiNi0.6Mn0.2Co0.2O2 active material and the sulfide electrolyte in the cathode mixture. Additionally, the decomposition of Li6.5P0.5Si0.5S5I occurred at the voltage range of 2.4–3.7 V (vs. Li-In) shown in Fig. 2d. Although typical charge/discharge voltage plateau are observed in the curve for the first cycle of the above battery, it shows an initial discharge capacity of 135 mAh/g with an ultralow Coulombic efficiency of 59.8% at 0.1 C. Moreover, this battery suffers fast discharge capacity degradation during the cycling test in Fig. 3b. After 40 cycles, this battery can only sustain a discharge capacity of less than 40.0 mAh/g at room temperature. The poor capacity and cyclability, low Coulombic efficiency are associated with the side reaction and space charge effect between the bare LiNi0.6Mn0.2Co0.2O2 and the prepared Li6.5P0.5Si0.5S5I electrolyte [44,45]. Based on our previous work, it seems that Li3InCl6 electrolyte displays fast Li-ion conductivity and superior stability towards the pristine high voltage cathode, such as LiNixMnyCozO2 (x + y + z = 1) [46-50]. Therefore, Li3InCl6 electrolyte was introduced to address the poor compatibility issues associated with the cathode mixture mentioned above. As is depicted in Fig. 3c, the LiNi0.6Mn0.2Co0.2O2/Li6.5P0.5Si0.5S5I/Li-In exhibits discharge capacities of 129.4 mAh/g at 0.05 C, 112 mAh/g at 0.1 C, 77.5 mAh/g at 0.2 C, 33.3 mAh/g at 0.5 C, and 3.1 mAh/g at 1.0 C, respectively. By comparison, LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/Li6.5P0.5Si0.5S5I/Li-In displays much higher discharge capacities of 181.6 mAh/g at 0.05 C, 148.3 mAh/g at 0.1 C, 134.0 mAh/g at 0.2 C, 107.1 mAh/g at 0.5 C, and 72.6 mAh/g at 1.0 C. Therefore, it is suggested that both the interfacial stability and Li-ion diffusion rate across the interface of active material/SEs also positively affect the rate capability of the corresponding all-solid-state lithium batteries.
LiNi0.6Mn0.2Co0.2O2, while the other two weak peaks located at 3.41 V and 3.55 V reflect the phase transitions from the monoclinic M to the hexagonal H2 phase and the hexagonal H2 phase to M1 phase [51-54]. The diminished reduction peaks in Fig. 3d suggest the lower discharge capacity of the LiNi0.6Mn0.2Co0.2O2/Li6.5P0.5Si0.5S5I/Li-In battery. As shown in Fig. 3e, the LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/Li6.5P0.5Si0.5S5I/Li-In exhibits higher level of overlapped oxidation/reduction peaks than that of the pristine cathode, indicating the highly energy conversion efficiency and structural stability of double electrolytes layer. Less M phase is observed during the charging process in Fig. 3f, due to the suppression of corresponding phase transition was observed. With the charging process goes on at high states of charge, the complete removal of Li+ from the lithium layers leads to the formation of the H3 phase at approximately 3.60 V [55]. Moreover, as depicted in Fig. 3f, the LiNi0.6Co0.2Mn0.2O2@Li3InCl6 cathode mixture shows smaller polarization voltage (61.1 mV) than that of the LiNi0.6Mn0.2Co0.2O2 (72.9 mV) cathode during the first cycle and more stable cycling performances in the subsequent cycles, which agrees well with the battery performance depicted in Fig. 3a.
Additionally, electrochemical performances of both batteries operated at a higher charge/discharge C-rate was also investigated. As displayed in Fig. 3g, the LiNi0.6Mn0.2Co0.2O2/Li6.5Si0.5P0.5S5I/Li-In battery delivers an initial discharge capacity of 97.2 mAh/g at 0.5 C with a Coulombic efficiency of 51.6% and maintains a discharge capacity of 4.2 mAh/g after 250 cycles. In contrast, the LiNi0.6Co0.2Mn0.2O2@Li3InCl6/Li3InCl6/Li6.5P0.5Si0.5S5I/Li-In battery delivers much higher discharge capacity (152.0 mAh/g vs. 97.2 mAh/g) at 0.5 C and Coulombic efficiency (85.0% vs. 51.6%) for the first cycle. Moreover, it remains a discharge capacity of 118 mAh/g with a capacity retention of 78.0% after 250 cycles. The superior electrochemical performances are due to the high Li-ion conductivity of Li6.5P0.5Si0.5S5I electrolyte and the excellent solid/solid interfacial and voltage stability towards the bare LiNi0.6Co0.2Mn0.2O2 materials provided by the Li3InCl6 electrolyte.
To investigate the side reactions and interfacial durability between the active materials LiNi0.6Mn0.2Co0.2O2 and sulfide solid electrolytes before and after cycling, ex-situ XPS was carried out on the cathode mixtures of LiNi0.6Mn0.2Co0.2O2/Li6.5Si0.5P0.5S5I/Li-In and LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/Li6.5Si0.5P0.5S5I/Li-In all-solid-state lithium batteries. As shown in the Fig. S2a (Supporting information), two typical doublet at 161.7 eV and 160.7 eV are observed in the S 2p spectra of the pristine Li6.5Si0.5P0.5S5I, which are ascribed to the S in the (Six/P1-x)S4(3 + x)- tetrahedra. A minor peak at 164.5 eV (orange) is also observed, which is attributed to the sulfite species caused by the oxidation when contacting with the rarefied oxygen. In contrast, when the cathode was cycled after 50 cycles, the intensities of the above two XPS signals due to the S 2p clear decrease (Fig. S2c in Supporting information). Moreover, XPS signals at 164.1 eV and 165.3 eV are also detected in the spectra, representing the S information in (Si/P)Sx, which are attributed to the decomposition of Li6.5Si0.5P0.5S5I electrolyte and the severe degradation occurring at the interfacial section between the Li6.5Si0.5P0.5S5I and bare LiNi0.6Mn0.2Co0.2O2 active material. These side reaction products have a negative effect on the capacity and cyclability properties of the LiNi0.6Mn0.2Co0.2O2 /Li6.5Si0.5P0.5S5I/Li-In battery in Fig. 3. Our previous research have found that Li3InCl6 electrolyte presents notable chemical/electrochemical compatibility towards the bare layered NCM ternary cathode material compared to sulfide electrolytes [29,36,55]. In this study, Li6.5Si0.5P0.5S5I has been replaced by Li3InCl6 electrolyte in the cathode mixture to enhance the interface stability. Moreover, a thin layer of Li3InCl6 solid electrolyte layer was introduced to act as an isolation layer to avoid the direct contact between the cathode mixture layer and the Li6.5Si0.5P0.5S5I electrolyte layer. Furthermore, two XPS signals located at 445.0 eV and 453.0 eV attributed to the In 3d3/2 and In 3d5/2 are observed in the In spectra (Fig. S2b in Supporting information), indicating the presence of Li3InCl6 electrolyte. However, minor variations can be detected in the In 3d spectra after 50 cycles (Fig. S2d in Supporting information), suggesting the ultrastable between the bare LiNi0.6Mn0.2Co0.2O2 and the Li3InCl6 electrolyte. This enables superior battery performances in Fig. 3g.
Finally, in-situ EIS was performed on both batteries when they are charged/discharge at 0.2 C under room temperature to reveal the resistances variations of different sections. Figs. 4a–d display the recorded electrochemical impedance spectra at different cut-off voltages during the charging/discharging processes. The Nyquist plots exhibit five distinct components that provide insights into the underlying electrochemical processes. During the initial cycling, the presence of a single semicircle in the spectra indicates the interfacial resistance between the bare LiNi0.6Mn0.2Co0.2O2 and Li6.5Si0.5P0.5S5I [46]. As depicted in Figs. 4g–j, besides the smaller typical semicircle of CEI, an elongated tail detected in the low frequencies range reflects interfacial resistance between the Li6.5Si0.5P0.5S5I and Li3InCl6. LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/Li6.5Si0.5P0.5S5I/Li-In exhibits smaller interfacial resistances than that of the bare LiNi0.6Mn0.2Co0.2O2 with the Li6.5Si0.5P0.5S5I electrolyte [56,57]. The combination of bare LiNi0.6Mn0.2Co0.2O2 active materials with Li3InCl6 electrolytes as the Li-ion additives can enhance both the solid/solid interface stability and Li-ion conductivity of the cathode mixture, inducing higher columbic efficiencies. Furthermore, time constants variations of different charge/discharge of states were also investigated to unravel the Li-ion kinetics of those two all-solid-state batteries. The bulk resistance of the solid electrolyte layer and the grain boundary resistance between solid electrolyte particles exhibit very small relaxation times, typically ranging from 10−7–10−6 s and 10−5–10−4 s. Generally, peaks located at 10−3–10−2 s and 10−2–10−1 s are attributed to the Rcathode/SE and Ranode/SE, respectively. Finally, Li-ion diffusion in the cathode mixture delivers the smallest rate, with time constants typically exceeding 10 s [36,53,57]. As shown in Figs. 4b and e, the DRT peaks of this battery are mainly distributed in the range of 10−3–10−2 s and 10−2–10−1 s, indicating that the major contributions to the total resistance of this battery are RSE/cathode and RSE/anode and exhibit small changes during charging. During the discharge process, the boundary resistance between the anode and SEs displays a slight increase. This increase in resistance means that the side reactions and ion transfer processes occurring at the anode-electrolyte interface are relatively slow, leading to a slight impedance rise. By contrast, the LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/Li6.5Si0.5P0.5S5I/Li-In battery (Figs. 4h and k) delivers smaller time constants and lower interfacial resistances, suggesting that introducing Li3InCl6 electrolyte layer in the newly established battery configuration leads to smaller total battery resistances during cycling than the LiNi0.6Mn0.2Co0.2O2. Furthermore, one characteristic peak located at the range of 1–10 s indicate the diffusion resistance of LiNi0.6Mn0.2Co0.2O2 particles in the cathode mixture is one major part of the total resistance and the RSE/cathode, RSE/anode dropped significantly with the introduction of Li3InCl6 electrolyte layer. To exhibit the resistance variations visually, the 2D intensity distribution plots derived from the DRT (distribution of relaxation time) results, show similar changes. The intensity distribution plot of LiNi0.6Mn0.2Co0.2O2 /Li6.5Si0.5P0.5S5I/Li-In during charging visualizes a prominent diffraction peak between the time range of 10−3–10−2 s as depicted in Fig. 4c. Additionally, reduced intensity of LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/Li6.5Si0.5P0.5S5I/Li-In is observed in Fig. 4i [55,56]. As shown in Figs. 4f and l, both battery configurations present a similar trend during discharging process. The disappearing intensity in Fig. 4l of LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/Li6.5Si0.5P0.5S5I/Li-In suggests a lower interfacial impedance, suggesting improved battery performance [58]. By way of conclusion, the measurements provide further evidence that Li3InCl6 shows favorable compatibility with LiNi0.6Mn0.2Co0.2O2 and efficient ion transport at CEI compare to Li6.5Si0.5P0.5S5I. Regarding previous XPS analysis, electrochemical performance, the application of Li3InCl6 both as cathode composite and interfacial barrier between the cathode active materials and Li6.5Si0.5P0.5S5I electrolytes can significantly enhance battery efficiency and mitigate the side reactions.
In summary, Si-substitution strategy was applied to enhance the Li-ion conductivity of the Li6PS5I electrolyte, yielding a high conductivity of up to 3.6 × 10−3 S/cm at room temperature for Li6.5P0.5Si0.5S5I electrolyte. Multiple characterization methods were combined to confirm that Si has been successfully doped in the Li6PS5I structure. The replace of P site with Si leads to larger lattice parameters due to the lattice expansion. Moreover, the obtained Li6.5P0.5Si0.5S5I electrolyte delivers a larger I−/S2− disorder than that of Li6PS5I because of the introduction of Si in the structure, resulting in a higher Li-ion conductivity and a lower activation energy (0.15 eV). Due to the poor interfacial compatibility between the LiNi0.6Mn0.2Co0.2O2 and Li6.5P0.5Si0.5S5I solid electrolyte, the assembled solid-state battery undergoes lower discharge capacities and rapid discharge degradation during cycling. XPS and EIS results confirm that the incompatibility can result in increased interface resistance and side reactions, further contributing to the degradation of the battery's performance over time. To mitigate this effect, Li3InCl6 electrolyte was introduced to replace the Li6.5P0.5Si0.5S5I electrolyte in the cathode mixture layer and act as the buffer layer to isolate the direct contact between the bare LiNi0.6Mn0.2Co0.2O2 and Li6.5P0.5Si0.5S5I. The corresponding all-solid-state battery configuration delivers higher discharge capacities and superior cycling performances. It delivers a high discharge capacity of 150.2 mAh/g for the 1st cycle at 0.5 C and maintained 78.5% of the initial value after 250 cycles. This work provides a promising strategy to design Li6PS5I-based argyrodite electrolytes with higher ionic conductivities enabling superior electrochemical performances with bare layered high voltage cathodes for all-solid-state lithium batteries.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
This work was supported by the National Key Research and Development Program (No. 2021YFB2400300). This work is also supported by the National Natural Science Foundation of China (No. 52177214) and the National Key Research and Development Program (No. 2021YFB2500200) and China Fujian Energy Devices Science and Technology Innovation Laboratory Open Fund (No. 21C-OP202211). We gratefully acknowledge the Analytical and Testing Center of HUST for us to use the facilities.
Supplementary material associated with this article can be found, in the online version, at doi:
H.M. Liang, Z.X. Wang, H.J. Guo, J.X. Wang, J. Leng, Appl. Surf. Sci. 423 (2017) 1045–1053. doi: 10.1016/j.apsusc.2017.06.283
C. Liao, C. Yu, L. Peng, et al., Solid State Ionics 377 (2022) 115871. doi: 10.1016/j.ssi.2022.115871
Y. Qian, L. Zhu, Y. Wang, et al., Renew. Sustain. Energy Rev. 41 (2015) 633–646. doi: 10.1016/j.rser.2014.08.085
J.M. Tarascon, M. Armand, Nature 414 (2001) 359–367. doi: 10.1038/35104644
Y. Lin, S. Huang, L. Zhong, et al., Energy Storage Mater. 34 (2021) 128–147. doi: 10.1016/j.ensm.2020.09.009
Y. Sun, K. Suzuki, K. Hara, et al., J. Power Sources 324 (2016) 798–803. doi: 10.1016/j.jpowsour.2016.05.100
O. Kwon, M. Hirayama, K. Suzuki, et al., J. Mater. Chem. A 3 (2015) 438–446. doi: 10.1039/C4TA05231E
S. Wang, X. Zhang, S. Liu, et al., J. Materiomics 6 (2020) 70–76. doi: 10.1016/j.jmat.2019.12.010
M.A. Kraft, S. Ohno, T. Zinkevich, et al., J. Am. Chem. Soc. 140 (2018) 16330–16339. doi: 10.1021/jacs.8b10282
P.R. Rayavarapu, N. Sharma, V.K. Peterson, S. Adams, J. Solid State Electrochem. 16 (2012) 1807–1813. doi: 10.1007/s10008-011-1572-8
N.J.J. de Klerk, T. Roslon, M. Wagemaker, Chem. Mater. 28 (2016) 7955–7963. doi: 10.1021/acs.chemmater.6b03630
Z. Zhang, J. Yao, C. Yu, et al., J. Mater. Chem. A 10 (2022) 22155–22165. doi: 10.1039/d2ta03168j
F. Zhao, J. Liang, C. Yu, et al., Adv. Energy Mater. 10 (2020) 1903422. doi: 10.1002/aenm.201903422
X. Bai, Y. Duan, W. Zhuang, et al., J. Mater. Chem. A 8 (2020) 25663–25686. doi: 10.1039/d0ta08472g
N. Minafra, S.P. Culver, T. Krauskopf, et al., J. Mater. Chem. A 6 (2018) 645–651. doi: 10.1039/C7TA08581H
J. Zhang, L. Li, C. Zheng, et al., ACS Appl. Mater. Interfaces 12 (2020) 41538–41545. doi: 10.1021/acsami.0c11683
S. Ito, S. Fujiki, T. Yamada, et al., J. Power Sources 248 (2014) 943–950. doi: 10.1016/j.jpowsour.2013.10.005
X. Li, Q. Sun, Z. Wang, et al., J. Power Sources 456 (2020) 227997. doi: 10.1016/j.jpowsour.2020.227997
Y. Zhai, W. Yang, D. Ning, et al., J. Mater. Chem. A 8 (2020) 5234–5245. doi: 10.1039/c9ta13014d
Y. Jin, Q. He, G. Liu, et al., Adv. Mater. 35 (2023) 2211047. doi: 10.1002/adma.202211047
X. Zhao, P. Xiang, J. Wu, et al., Nano Lett. 23 (2023) 227–234. doi: 10.1021/acs.nanolett.2c04140
Z. Zhang, J. Wang, Y. Jin, et al., Energy Storage Mater. 54 (2023) 845–853. doi: 10.1016/j.ensm.2022.11.023
J. Wu, S. Liu, F. Han, et al., Adv. Mater. 33 (2021) 2000751. doi: 10.1002/adma.202000751
K.R. Prakasha, M. Sathish, P. Bera, A.S. Prakash, ACS Omega 2 (2017) 2308–2316. doi: 10.1021/acsomega.7b00381
X. Yi, X. Wang, B. Ju, et al., Electrochim. Acta 134 (2014) 143–149. doi: 10.1016/j.electacta.2014.04.148
X. Zhan, S. Gao, Y.T. Cheng, Electrochim. Acta 300 (2019) 36–44. doi: 10.3390/proteomes7040036
Y.J. Kim, R. Rajagopal, S. Kang, et al., Chem. Eng. J. 386 (2020) 123975. doi: 10.1016/j.cej.2019.123975
J. Li, Y. Liu, W. Yao, et al., Solid State Ionics 349 (2020) 115292. doi: 10.1016/j.ssi.2020.115292
S. Chen, C. Yu, C. Wei, et al., Chin. Chem. Lett. 34 (2023) 107544. doi: 10.1016/j.cclet.2022.05.058
J.W. Lee, Y.J. Park, J. Electrochem. Sci. Technol. 9 (2018) 176–183. doi: 10.33961/JECST.2018.9.3.176
T. Sattar, S.J. Sim, B.S. Jin, et al., Curr. Appl. Phys. 36 (2022) 176–182. doi: 10.1016/j.cap.2022.01.004
S. Ohno, B. Helm, T. Fuchs, et al., Chem. Mater. 31 (2019) 4936–4944. doi: 10.1021/acs.chemmater.9b01857
Z. Jiang, C. Yu, S. Chen, et al., Scr. Mater 227 (2023) 115303. doi: 10.1016/j.scriptamat.2023.115303
Y. Gu, K. Yang, H. Yao, et al., Chin. Chem. Lett. 34 (2023) 108047. doi: 10.1016/j.cclet.2022.108047
Q. Luo, C. Yu, C. Wei, et al., Ceram. Int. 49 (2023) 11485–11493. doi: 10.1016/j.ceramint.2022.11.347
C. Wei, X. Liu, C. Yu, et al., Chin. Chem. Lett. 34 (2023) 107859. doi: 10.1016/j.cclet.2022.107859
F. Strauss, J.H. Teo, J. Maibach, et al., ACS Appl. Mater. Interfaces 12 (2020) 57146–57154. doi: 10.1021/acsami.0c18590
S. Chen, C. Yu, S. Chen, et al., Chin. Chem. Lett. 33 (2022) 4635–4639. doi: 10.1016/j.cclet.2021.12.048
Y. Subramanian, R. Rajagopal, B. Senthilkumar, et al., Electrochim. Acta 400 (2021) 139431. doi: 10.1016/j.electacta.2021.139431
M. Tenhover, M.A. Hazle, R.K. Grasselli, Phys. Rev. B: Condens. Matter. 29 (1984) 6732–6735. doi: 10.1103/PhysRevB.29.6732
H. Triki, A. Benlalli, Q. Zhou, et al., Results Phys. 26 (2021) 104381. doi: 10.1016/j.rinp.2021.104381
F. Han, T. Gao, Y. Zhu, et al., Adv. Mater. 27 (2015) 3473–3483. doi: 10.1002/adma.201500180
I. Hanghofer, M. Brinek, S.L. Eisbacher, et al., Phys. Chem. Chem. Phys. 21 (2019) 8489–8507. doi: 10.1039/c9cp00664h
Z. Zhang, Y. Sun, X. Duan, et al., J. Mater. Chem. A 7 (2019) 2717–2722. doi: 10.1039/c8ta10790d
M. Brinek, C. Hiebl, K. Hogrefe, et al., J. Phys. Chem. C 124 (2020) 22934–22940. doi: 10.1021/acs.jpcc.0c06090
F. Han, Y. Zhu, X. He, et al., Adv. Energy Mater. 6 (2016) 1501590. doi: 10.1002/aenm.201501590
S.P. Ong, Y. Mo, W.D. Richards, et al., Energy Environ. Sci. 1 (2013) C2EE23355J. doi: 10.1039/C2EE23355J
K. Takada, T. Inada, A. Kajiyama, et al., Solid State Ionics 158 (2003) 269–274. doi: 10.1016/S0167-2738(02)00823-8
J. Wu, L. Shen, Z. Zhang, et al., Electrochem. Energy Rev. 4 (2021) 101–135. doi: 10.1007/s41918-020-00081-4
S. Yang, Z. Zhang, J. Lin, et al., Front. Energy Res. 10 (2022) 945003. doi: 10.3389/fenrg.2022.945003
W.D. Richards, L.J. Miara, et al., Chem. Mater. 28 (2016) 266–273. doi: 10.1021/acs.chemmater.5b04082
Z. Cheng, M. Liu, S. Ganapathy, et al., Joule 4 (2020) 1311–1323. doi: 10.1016/j.joule.2020.04.002
C. Wei, C. Yu, S. Chen, et al., Electrochim. Acta 438 (2023) 141545. doi: 10.1016/j.electacta.2022.141545
H. Li, M. Cormier, N. Zhang, et al., J Electrochem. Soc. 166 (2019) A429–A439. doi: 10.1149/2.1381902jes
Z. Liang, Y. Xiang, K. Wang, et al., Nat. Commun. 14 (2023) S41467-023-36848-8.
Y. Lu, C.Z. Zhao, J.Q. Huang, Q. Zhang, Joule 6 (2022) 1172–1198. doi: 10.1016/j.joule.2022.05.005
C. Wei, S. Chen, C. Yu, et al., Appl. Mater. Today 31 (2023) 101770. doi: 10.1016/j.apmt.2023.101770
Y. Zhao, C. Wu, G. Peng, et al., J. Power Sources 301 (2016) 47–53. doi: 10.1021/acs.energyfuels.5b01775
Figure 1 The corresponding XRD Rietveld refinement patterns of (a) Li6PS5I and (b) Li6.5P0.5Si0.5S5I. Crystal structure of (c) Li6.5P0.5Si0.5S5I. (d) Doping effects to intercage jump of lithium transport and supposed mechanism for enhancing ionic conductivity. (e) SEM image and EDS mapping of Si, I, S, and P elements of the Li6.5P0.5Si0.5S5I electrolyte.
Figure 2 (a) Nyquist plots of the prepared Li6PS5I and Li6.5P0.5Si0.5S5I measured at room temperature. (b) Temperature dependent Li-ion conductivities and (c) electronic conductivities of the obtained Li6PS5I and Li6.5P0.5Si0.5S5I at room temperature. (d) CV curves of the Li6PS5I+C/Li6PS5I/In and Li6.5P0.5Si0.5S5I +C/Li6.5P0.5Si0.5S5I/In measured at room temperature.
Figure 3 (a) The initial charge/discharge curves of the assembled LiNi0.6Mn0.2Co0.2O2/Li6.5P0.5Si0.5S5I/Li-In, LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/Li6.5P0.5Si0.5S5I/Li-In all-solid-state lithium batteries at 0.1 C between 2.4 V and 3.7 V (vs. Li-In). (c) The rate performance test results of both batteries at room temperature. The corresponding dQ/dV curves of (d) LiNi0.6Mn0.2Co0.2O2/Li6.5P0.5Si0.5S5I/Li-In and (e) LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/Li6.5P0.5Si0.5S5I/Li-In. (f) The comparison of dQ/dV curves based on the initial cycle of the above two batteries. The cycling performances of those two all-solid-state batteries at (b) 0.1 C and (g) 0.5 C between 2.4 V and 3.7 V (vs. Li-In) under room temperature.
Figure 4 Electrochemical impedance spectra of the LiNi0.6Mn0.2Co0.2O2/Li6.5P0.5Si0.5S5I/Li-In and LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/Li6.5P0.5Si0.5S5I/Li-In all-solid-state lithium batteries cycled under different (a, g) charge (SOC) and (d, j) discharge states (SOD) during the initial cycle at room temperature. Corresponding DRT calculated from EIS measurements at different states of the initial cycle and 2D intensity color map of the DRT curves of DRT results of (a-f) LiNi0.6Mn0.2Co0.2O2/Li6.5P0.5Si0.5S5I/Li-In and (g-l) LiNi0.6Mn0.2Co0.2O2@Li3InCl6/Li3InCl6/ Li6.5P0.5Si0.5S5I/Li-In.
扫一扫看文章
扫一扫关注我们
DownLoad:
下载:
下载: