Lithium phosphorus sulfide (li7p3s11) solid electrolyte for solid state batteries enabled by metal halide (e.g., zrcl4) doping

Abstract

A solid electrolyte for an all solid state lithium battery includes lithium phosphorus sulfide(LPS), which has been doped during synthesis with a metal halide, such as zirconium tetrachloride. The synthesis includes adding the metal halide to lithium sulfide and phosphorus pentasulfide precursor materials, followed by ball milling and heat treatment.

Description

Lithium Phosphorus Sulfide (Li PsSn) Solid Electrolyte for Solid State Batteries Enabled by Metal Halide (e.g., Z1CI4) DopingCROSS REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to USSN 63 / 738,180, filed December 23, 2024, under relevant portions of 35 USC §119 and 35 USC §120. The herein referred to priority application is incorporated by reference in its entirety.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0002] The following was made with government support under a grant by the National Science Foundation (NSF -2052611). The government has certain rights in the invention.BACKGROUND

[0003] Lithium-ion (Li-ion) battery technology, first commercialized by Sony in 1991, has been a cornerstone of portable energy storage solutions. However, this existing technology faces significant limitations in meeting the growing demand for higher energy density batteries, primarily due to the reliance on graphite anodes, which offer a theoretical capacity of only 372 mAh / g Additionally, the use of flammable liquid electrolytes renders these batteries susceptible to severe safety hazards, particularly in the event of thermal runaway. This latter situation underscores the urgent need for the development of safer and more efficient alternatives. All solid state lithium metal batteries (ASSLBs) have emerged as a promising solution, potentially addressing both energy density limitations and safety concerns through the incorporation of solid electrolytes and lithium metal anodes. By eliminating the risks that are associated with liquid electrolytes and enabling higher capacity anodes, ASSLBs now represent a significant advancement in battery technology, paving the way for next-generation energy storage applications.

[0004] All-solid-state lithium metal batteries (ASSLBs) have the potential to enhance energy density from approximately 270 Wh / kg to 400 Wh / kg, while also addressing the noted safety concerns associated with conventional lithium-ion batteries. To achieve each goal, it is essential to successfully couple lithium metal anodes with solid electrolytes and transition metal oxide 132974769.1cathodes. There are various types of known solid electrolytes that have been identified and capable of conducting lithium ions, which can be broadly classified into two (2) categories: solid polymer electrolytes (SPEs) and inorganic ceramic electrolytes (ICEs).

[0005] SPEs are synthesized by incorporating conductive lithium salts, such as lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) or lithium perchlorate (LiClCL), into a polymer matrix, including poly(ethylene oxide) (PEO), polyacrylonitrile (PAN), or poly(vinylidene fluoride) (PVDF). While SPEs offer excellent mechanical flexibility and ease of fabrication, their performance in ASSLBs is often limited by low room-temperature ionic conductivity, low critical current density, and instability when interfaced with lithium metal anodes.

[0006] ICEs can be subdivided into two (2) primary categories: oxides and sulfides. Oxide electrolytes, such as lithium lanthanum zirconate (LLZO) and lithium titanium oxide (LTO), are known for their relatively high ionic conductivity and broad electrochemical stability windows. However, oxide electrolytes often require high processing temperatures and exhibit brittleness, which can hinder their practical applications in solid-state batteries. On the other hand, sulfide electrolytes, including LivPsSn, LiePSsCl, and LiioGeP Sn, demonstrate significantly higher ionic conductivities as compared to their oxide counterparts, making sulfide electrolytes more suitable for facilitating rapid lithium-ion transport. Furthermore, sulfide electrolytes can be synthesized at comparatively lower temperatures, thereby reducing energy consumption. The superior mechanical properties and higher ionic conductivity of sulfides position them as a more advantageous option for ASSLBs, addressing many of the limitations associated with polymer and oxide electrolytes and paving the way for improved battery performance.

[0007] Despite the above noted advantages, the practical integration of sulfide electrolytes into ASSLBs is still challenged by a number of factors that include low critical current density, a narrow electrochemical stability window with electrodes, and high interfacial impedance with lithium metal anodes and transition metal oxide cathodes. These factors contribute to the rapid capacity fading observed in ASSLBs, necessitating further research to optimize the performance and reliability of this promising technology.232974769.1

[0008] I 7P3S 11 (LPS), a superionic conductor first reported by Mizuno et al. in 2006, is a promising sulfide solid electrolyte for the development of high-energy-density all-solid-state lithium metal batteries (ASSLBs). This solid electrolyte can be synthesized by ball milling, subsequently followed by heat treatment. Room temperature ionic conductivities ranging from 10’4to 10’3S.cm’1have been reported for LPS electrolytes. However, the practical integration of LPS electrolytes into ASSLBs is impeded by several factors, including a low critical current density (CCD), a narrow electrochemical stability window, adverse side reactions with lithium metal anodes, and transition metal oxide cathodes. The reported room temperature CCD for LivPsSii are typically less than 1 mA / cm2, which is insufficient for effective practical applications. Doping Li? P3 S 11 with various impurity elements has been investigated as a potential strategy to enhance CCD. Doping can alter the structural and electrochemical properties of solid electrolytes, facilitating improved performance. Several doping agents, including M0S2, InaS?, ZrCL, ZnO5, CaL6, Ce2S3, SnSe2 have been studied; however, their impact on increasing CCD of LPS has, at best, been generally modest.BRIEF DESCRIPTION

[0009] It has been determined that a metal halide, and more specifically inexpensive zirconium tetrachloride (ZrCL), when used as a doping agent to modify LivPsSn (LPS), leads to a significant enhancement in the critical current density (CCD) of this solid electrolyte. More specifically, the introduction of a small quantity of ZrCL during the synthesis process preserves the crystal structure of LPS, while the CCD increases markedly from 0.36 mA / cm2to over 2 mA / cm2at room temperature, without applying any external pressure during cell cycling. In addition to the substantial improvement in CCD, ZrCL doping results in higher ionic conductivity, as well as lower activation energy. Furthermore, the interfacial stability between LPS and a lithium metal anode is enhanced, wherein Li / LFP full cells having doped LPS demonstrate higher specific capacity and stable performance, as compared to full cells having non-doped and pristine LPS as a solid electrolyte. As discussed herein, these performance improvements are described and are also confirmed using various means that include X-ray diffraction (XRD), Raman spectroscopy, scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDS), and X-ray photoelectron spectroscopy (XPS).332974769.1

[0010] According to an example, optimized doping of 2 wt.% ZrCL increases the CCD of a LPS pellet from 0.36 mA / cm2to 2.0 mA / cm2at room temperature, while the ionic conductivity also improves from 9.8x10’4S / cm to 1.4 x 10’3S / cm. Li / Li symmetrical cells with 2 wt.% ZrCL-doped LPS as an electrolyte can exhibit improved cycling stability at 0.1 mA / cm2, as compared to that of a pristine electrolyte. Furthermore, Li / LFP full cells having ZrCL-doped LPS demonstrate stable performance as compared to cells having undoped LPS. Enhanced lithium dendrite suppression and improved electrochemical properties of ZrCL-doped LPS create marked potential for significantly advancing ASSLB technology.

[0011] The herein described concepts using doped sulfide solid state electrolytes enable safer and higher capacity, more stable solid-state lithium-ion batteries for use in electric vehicles, power systems, aerospace, medical and portable electronic applications.

[0012] Doped sulfide solid electrolytes made in accordance with teachings described herein show lower bulk and interfacial impedance, as well as higher critical current density to support fast charging in solid-state lithium-ion batteries. In addition, these electrolytes possess lower activation energy that assists in faster ionic migration and are more stable with lithium metal as compared to control LPS samples, as shown and demonstrated in CV curves, and also as discussed in greater detail herein. Furthermore, the cycling performance of solid state lithium batteries that include the doped sulfide solid electrolytes was improved. The processing / synthesis is novel and battery performance is improved. The herein described invention solves the abovenoted instability problems found in solid state electrolytes.

[0013] These and other features and advantages will be readily apparent from the following Detailed Description, which should be read in conjunction with the accompanying drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0014] FIG. 1(A) is an X-ray diffraction (XRD) spectrum plot of Li2S and LPS precursor after predetermined times (8h and 16 h) of ball milling;432974769.1

[0015] FIG. 1(B) is an X-ray diffraction (XRD) spectra plot of a control LPS sample and x wt % in which x = 1, 2, 3 and 4, respectively) ZrCL doped LPS in accordance with aspects of the present teachings;

[0016] FIG. 1(C) is an X-ray diffraction (XRD) spectra plot of LPS that is heat treated at various processing times;

[0017] FIG. 1(D ) is a Raman spectrum plot of an undoped LPS sample;

[0018] FIG. 1(E) is a Raman spectrum plot of a doped LPS sample in accordance with an embodiment of the present teachings;

[0019] FIGS. 2(A) and 2(B) are scanning electron microscope (SEM) image of control LPS powder;

[0020] FIGS. 2(C) and 2(D) are scanning electron microscope (SEM) images of 2 wt% LPS powder that has been doped with ZrCL in accordance with aspects of the present teachings;

[0021] FIG. 2(E) is an electrochemical impedance spectroscopy (EIS) Nyquist plot of SS / SS symmetrical cells made in accordance with aspects of the present teachings using control (undoped) LPS and 1, 2, 3 and 4 wt. % zirconium tetrachloride doped LPS pellet as electrolytes for ionic conductivity measurement;

[0022] FIG. 3 is an electrochemical impedance spectroscopy (EIS) plot comparing a control sample of SS / LPS / SS with a sample of SS / 2 wt% LPS doped with ZrCL / SS;

[0023] FIG. 4(A) is an Arrhenius plot of SS / LPS / SS;

[0024] FIG. 4(B) is an Arrhenius plot of SS / 2 wt% LPS doped with ZrCL / SS;532974769.1

[0025] FIG. 5(A) is a cyclic voltammetry (CV) curve of LPS in the potential range from -0.5V to 0.5V (vs. Li / Li+) at a scanning rate of ImV / s at room temperature;

[0026] FIG. 5(B) is a cyclic voltammetry (CV) curve of 2 wt% LPS doped with ZrCL in the potential range from -0.5V to 0.5V (vs. Li / Li+) at a scanning rate of ImV / s at room temperature;

[0027] FIG. 6(A) depicts a galvanostatic charge / discharge profile of Li / LPS / Li at increasing current density;

[0028] FIG. 6(B) depicts a galvanostatic charge / discharge profile of Li / 2 wt% ZrCL doped LPS / Li at increasing current density.

[0029] FIG. 7 depicts a galvanostatic charge / discharge profiles of Li / LPS / Li and Li / 2 wt.% ZrCL4 doped LPS / Li at 0.1 mAcm-2, 0.1 mAhcm’2

[0030] FIG. 8(A) graphically depicts long-term cycling performance of full cells having ZrCL doped LPS at 0.1 C;

[0031] FIG. 8(B) graphically depicts 1stand 50thcycle charge / discharge voltage profiles of an embodiment of a Li / LPS / SPE / LFP full cell at 0.1 C;

[0032] FIG. 8(C) graphically depicts 1stand 50thcycle / discharge voltage profiles of an embodiment of an Li / LPS / SPE / LFP full cell at 0.1C; and

[0033] FIGS. 9(A) - 9(D) are high resolution X-ray photoelectron spectroscopy (XPS) plots of undoped and doped electrolytes in accordance with aspects of the present invention.DETAILED DESCRIPTION

[0034] The following describes structural characterizations for the preparation and synthesis of lithium phosphorus sulfide (LPS), as doped with a metal halide and more specifically, zirconium tetrachloride. For purposes of fabrication, battery grade Li2S (99.98% trace metal basis) was 632974769.1used. For purposes of the preparation of various samples and cells employing prepared samples, P2SS(99%), and ZrCL (> 99.5% trace metal basis), were used.Sample preparation

[0035] For purposes of the following discussion, a number of control LivPsSn (LPS) and x weight % (x = 1, 2, 3, and 4) ZrCL doped LivPsSn samples were synthesized by a high-energy planetary ball milling technique, which was followed by heat treatment. For pristine (control) LPS, Li2S and P2S5 powders were weighed in a 7:3 molar ratio and ground into agate mortar and pestle to mix uniformly inside an argon- filled glovebox. The ground mixture was transferred into stainless steel jars, each containing a plurality (e.g., 10) of stainless steel balls having an appropriate diameter (e.g., about 8 mm). The jars were sealed using Kapton tape and mounted on a planetary ball milling apparatus. Details relating to the latter milling apparatus are well known in the field and do not require further explanation. The milling was then performed for an appropriate number (e.g., 16) of cycles, each milling cycle being defined by a 30 minute forward step, followed by a 5 min pause delay, a 30 minute reverse step, and then a 5-minute pause delay step. The amorphous LPS powder obtained after ball milling was then transferred to glass vials, which were then wrapped with aluminum foil and placed on a metal holder. The powdered LPS samples were heat treated on a hot plate at 240 °C for 1 hour inside an argon filled glovebox to obtain crystalline LPS. A lollipop thermometer was inserted into one of the glass vials to monitor the temperature. Several experiments were performed to determine an optimal ball milling time and an optimal heat treatment time, as discussed in greater detail below. For the synthesis of doped LPS, 1, 2, 3, and 4 wt.% of ZrCL was added into the starting material and the remaining steps of the method were identical to that of the prior control LPS sample synthesis.Sample characterization

[0036] As discussed herein, structural characterizations were performed by taking X-ray diffraction (XRD) and Raman spectra measurements of the prepared control and doped LPS samples. For purposes of these characterization, powder X-ray diffraction (XRD) measurements as further discussed with reference to FIGS. 1(A) and 1(B), were conducted using a Bruker D2 phaser equipped with Cu-Ka radiation (A, = 1.50405 A) within a 20 range from 10° to 60°, with a 0.05° step and a scan speed of 0.5 seconds per step. For XRD measurements, the powdered sample(s) was spread uniformly on a silicon diffraction plate (such as those manufactured and 732974769.1sold by the MTI Corporation), covered with adhesive removed Kapton tape and then placed on the custom-built sample holder. Each of the foregoing steps were performed inside an argon-filled glovebox to avoid air exposure to the powdered sample. For Raman measurements, as discussed in FIGS. 1(C) and 1(D), all control and doped LPS samples were sealed in a transparent glass pipe within the glovebox and the Raman spectra were obtained using a Renishaw In Via Confocal Raman microscope at an excitation of 532 nm. In addition, morphological and elemental characterizations were also performed using a scanning electron microscope (Axia ChemiSEM) equipped with an energy dispersive spectrometry (EDS).

[0037] X-Ray Photoelectron Spectroscopy (XPS) data were taken in a Thermo Fisher ESCALAB QXi X-ray Photoelectron Spectrometer Microprobe.. The C is signal of adventitious carbon (285 eV) was used for charge correction and data fitting was performed using XPS peak fit software. Samples were protected from the contact with air during the transfer from the argon-filled glovebox to the XPS vacuum chamber.Catholyte preparation for full cells

[0038] A solid polymer electrolyte (SPE) membrane was used as a catholyte between LFP and LPS in prepared full cells, as described in a later portion. The SPE was prepared by dissolving an optimized mass ratio of 20% poly (vinylidene fluoride-co-hexafluoropropylene) (PVDF-HFP) as a polymer host, 35% Lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) as a lithium salt, and 45% succinonitrile as a plasticizer in acetone (0.4 gm / ml). The mixture was stirred at 80 °C for 2 hours to make the mixture into a homogenous solution, and then 400 pl of the mixture was cast on stainless steel blocks before placing in a desiccator for 12 hours. Then, the formed membranes were dried in a vacuum oven at 55 °C for 12 hours.Electrode fabrication for full cells

[0039] To make Lithium iron phosphate (LiFePO₄) cathodes for full cells, as discussed in a later portion, a uniform slurry prepared by mixing active material, binder and carbon black in 8: 1: 1 ratio was casted on an aluminum current collector using doctor blading, and then was dried in a vacuum oven for 12 hours at 80 °C. 6 mm circular cathode electrodes were punched to use in CR2032-coin cells. The areal active mass loading for these cathodes was ~ 2 mg cm2.832974769.1Electrochemical characterization:

[0040] For ionic conductivity measurements as discussed in a later portion, circular LPS pellets of diameter 10 mm were obtained by pressing 80 mg of electrolyte powder at 500 psi for 2 minutes. Electrochemical impedance spectroscopy (EIS) of SS / LPS / SS ion blocking coin cells were conducted at frequencies from 1 MHz to 1 kHz with 10 mV excitation potential at room temperature. Li / LPS / Li and Li / 2 wt.% ZrCL doped LPS / Li symmetrical coin cells were assembled for a critical current density (CCD) test and long-term plating / stripping. The CCD test was conducted using a time-controlled protocol with a step size of 0.05 mAcm-2and a step time of 30 minutes. Plating / stripping of symmetrical cells was performed at a current density of 0.1 mAcm-2with charge-discharge of 1 hour each. Li / LPS / SPE / LFP full cells were fabricated in CR2032 coin-cells format and were electrochemically cycled at 0.1 C at room temperature using CCCV cycling protocol in a Landt electrochemical testing system.Structural characterizationX-ray Diffraction (XRD)

[0041] FIG. 1(A) depicts an X-ray diffraction (XRD) spectrum plot of LPS precursor that was obtained after 8 hours and 16 hours of ball milling, as previously described. As shown, prominent Li2S peaks are present in the 8 hour ball milled sample, indicating that the Li2S has not reacted fully. However, no Li2S peaks are present in the XRD spectrum plot of 16 hour ball milled sample, which indicates that the precursors Li2S and P2S5 have reacted fully to form an amorphous precursor after ball milling for 16 hours.

[0042] In order to determine the optimal heat treatment time, the obtained amorphous LPS precursor was heat treated for 0.5 hour, 1 hour, and 2 hours on a hot plate within an argon filled glovebox and the temperature was monitored by using a lollipop thermometer. FIG. 1 (B) shows the XRD patterns of three (3) different heat treatment times i.e., 0.5 hours, 1 hour and 2 hours, respectively, at 240°C. Each of the samples showed the sharp peaks corresponding to LPS, indicating the formation of a crystalline LPS phase. In addition, a close comparison of the XRD spectrum showed that the peaks of 0.5 hour heat treated sample are wider than those of either the 1 hour or 2 hour heat treated samples. This latter variation indicates that the 0.5 hour heat treated sample contains a certain amount of amorphous phase. There is no obvious difference in the 932974769.1XRD peaks between 1 hour and 2 hour heat treated samples, which indicates that a 1 hour heat treatment time may be optimal in order to synthesize a crystalline LPS. Accordingly, subsequent batches of synthesis were done using 1 hour heat treatment at 240 °C inside the argon- filled glovebox on a hot plate.

[0043] FIG. 1(C) shows the XRD spectrum patterns of pristine LPS and 1, 2, 3, and 4 wt.% ZrCL doped LPS. All of the XRD patterns show the presence of all the major peaks corresponding to LPS. Similar XRD patterns and absence of any additional peaks in doped LPS show that introducing a small amount of ZrCL does not change the original crystal structure of LPS.Ram an rn easuremen t

[0044] Raman spectrum of control and doped LPS are comparatively shown in FIGS. 1(D) and 1(E), respectively. The peaks at 420 cm1, and 405 cm-1in the control LPS sample of FIG. 1(D) belong to the stretching vibration of P-S bonds in the PS43’ (monomer) and P2S74’ (dimer monomer) ions, respectively, showing a major peak that is centered at around 405 cm-1with a shoulder peak at or around 420 cm1. These noted peaks correspond to the local structural unit of P2S74’ and PS43’, respectively, which matches well with that of the reported literature. The Raman spectra of doped LPS in FIG. 1(E) shows an upper shift of both of the peaks corresponding to PS43’ and P2S74’. This latter showing indicates that ZrCL doping influences the lattice dynamics and the chemical environment around each of the P2S74’ and PS43’ units, potentially contributing to improved electrochemical properties, such as enhanced ionic conductivity or stability.Morphology

[0045] FIGS. 2(A)- 2(D) depict scanning electron microscope (SEM) images of LPS and 2 wt.% ZrCL doped LPS samples, respectively, for purposes of comparison. Both of the depicted samples show pseudo-spherical primary particles of less than 2 pm, wherein each primary particle is aggregated and connected with the other particles to form larger secondary particles. Larger secondary particles are observed in the ZrCL doped LPS samples, as compared to the undoped LPS.Electrochemical characterizationIonic Conductivity1032974769.1

[0046] Solid electrolytes having an ionic conductivity of the order > 10“4S cm-1are required to develop a high-energy bulk-type all-solid-state lithium battery. Room temperature ionic conductivities in the range of 10’4to 10’2S cm'1are reported for LPS by different groups. The observed discrepancies in the ionic conductivity may be due to the difference in synthesis methods and ionic conductivity measurement conditions like stack pressure, fabrication pressure, type of blocking electrodes used, etc. In order to reveal the effects of zirconium doping amounts on Li+ionic conductivity, a series of glass ceramic electrolytes Li₇P₃S₁₁-x wt.% doped ZrCL (x = 0, 1, 2, 3, and 4 wt.%) were synthesized by using high-energy planetary ball milling followed by heat treatment. The electrochemical impedance spectroscopy (EIS) plots for ion blocking symmetrical cells using stainless steel as blocking electrodes and LPS, x wt.% ZrCL doped LPS as electrolyte are shown in FIG. 2(E). These plots consist of a half semicircle and a spike, which correspond to contributions from the bulk / interfacial resistance of the SE-electrode and the diffusion impedance, respectively. The Lithium ionic conductivities (o) of the electrolytes were calculated by using the formula (1):in which L is the electrolyte pellet thickness, A is the contact area between the electrodes and LPS pellet, and R is the electrolyte impedance.

[0047] With reference to FIG. 3, the ionic conductivity of 2 wt.% ZrCL doped LPS is a maximum, reaching the value 1.4x1 O’3Scrn’1which is significantly higher than that of undoped control LPS. The Zr4+ions are much larger and more highly charged than Li+. When Zr4+substitutes for Li+in the LPS structure, more lithium vacancies are formed to maintain the charge neutrality. These vacancies create pathways that enhance Li+mobility thereby increasing the ionic conductivity. In addition, the presence of highly electronegative Cl decreases the electron cloud density on the surface of P2S?4“ and PS43’ units, reduces their binding to Li+and thus increases the ionic conductivity. The ionic conductivity decreased upon increasing the doping percentage further. This latter decrease in ionic conductivity may be due to the formation of less ionic conductive phases, such as Li3PS4, and Li₄P₂S₇ in the electrolyte. Due to the highest ionic conductivity of 2 wt.% ZrCL doped LPS, the remaining the tests and characterizations discussed herein compare the control (undoped) and 2 wt.% ZrCl₄ doped LPS electrolyte performance.1132974769.1Activation Energy

[0048] An evaluation of activation energy for Li+diffusion was further made by measuring the temperature dependent ionic conductivity at different temperatures ranging from 30 °C to 120 °C. FIGS. 4(A) and 4(B) depict Arrhenius plots of Stainless steel (SS) / LPS / (SS) and SS / 2 wt.% ZrCh doped LPS / SS, respectively, It can be seen from these plots that the activation energy decreases from 0.26 eV for LPS to 0.23 eV for 2 wt.% ZrCU doped LPS. This decrease suggests that ZrCU doping can effectively reduce the activation energy and enhance the lithium-ion conductivity of solid electrolyte. When ZrCU is introduced into the LPS, many defects are created, and the transmission channels of lithium ions are broadened simultaneously causing the less barrier to the transmission of Li ions.

[0049] FIGS. 5(A) and 5(B) show cyclic voltammetry (CV) curves of asymmetric Li / SE / SS cells with pristine (control) LPS and 2 wt.% ZrCU doped LPS electrolytes, respectively, in the potential range from -0.5 V to 5V (vs. Li / Li+) at a scan rate of 1 mVs'1at room temperature. Only redox peaks corresponding to dissolution (Li— > Li++e“) and lithium deposition (Li1Te’) are observed in the cyclic voltammogram. No other obvious peaks are observed in the voltage range from -0.5 to 5V vs. Li / Li+. This latter result indicates that both electrolytes exhibit a wide electrochemical window up to 5 V vs. Li / Li+. A fluctuation seen in the CV curves of control LPS between 4 V and 5 V may be due to the slight instability between Li and LPS due to side reactions, whereas no such fluctuation occurred in case of ZrCU doped LPS. This latter observation indicates that ZrCU doped LPS is also compatible with Li metal at higher voltages. Critical Current Density (CCD)

[0050] Critical current density is the minimum current density as the lithium dendrites penetrate the electrolyte, causing a short circuit in the cell. The minimum current density is further considered as a measure of the lithium dendrite suppression ability of an electrolyte. To investigate the change on dendrite-suppressing ability of LPS due to ZrCU doping, time controlled critical current density (CCD) tests were conducted on symmetric Li / LPS / Li and Li / 2 wt.% ZrCU doped LPS / Li cells, respectively. Galvanostatic cycling tests were conducted with step-increased current densities at room temperature. The initial current density was 0.05 mA cm'2and current was increased by 0.05 mA cm-2in each successive step. With increasing the current density to a certain value, the voltage rapidly drops near to zero due to the penetration of Li 1232974769.1dendrites inside the electrolyte. The corresponding current density at the short circuit point is named as critical current density.

[0051] From FIG. 6A, it can be seen see that the overpotential of Li / LPS / Li cells shows a sudden drop in overpotential at a current density of 0.36 mAcm-2. This sudden drop in the overpotential is the indication of lithium dendrites penetration into the LPS electrolyte. Therefore, the CCD of synthesized LPS electrolyte is 0.36 mAcm-2. However, the symmetrical cell with 2 wt.% ZrCL doped LPS electrolyte showed no decrease in overpotential on increasing the current density up to 2 mAcm-2as shown in FIG. 6B. The foregoing shows that lithium dendrites did not penetrate the 2 wt.% ZrCl₄ doped LPS up to a high current density of 2 mA cm “2. The failure of cells with doped LPS was not due to the short circuiting but was due to the high overpotential polarization. Therefore, the doped LPS must have a significantly higher critical current density than that of the control LPS. The increase in overpotential can be the consequence of the SEI growth and contact loss. Since no external pressure was applied to the cells during cycling, there is a chance of contact loss due to the volume expansion of Li metal electrodes. EIS (electrochemical impedance spectroscopy) measurements taken after CCD showed short circuiting of the control cell, but the cell with doped LPS was not short circuited. Hence, the introduction of a small quantity of ZrCl₄ into the LPS during synthesis can increase the critical current density of LPS significantly, which is beneficial for fast charging application of batteries.Li / Li symmetrical cell performance

[0052] Long-term stability of the synthesized electrolytes with Li metal electrolyte was investigated by electrochemical cycling of Li / LPS / Li and Li / 2 wt.% ZrCl₄ doped LPS / Li symmetrical cells at a current density of 0.1 mAcm-2with a charge- dis charge time of 1 h each. The results are shown in FIG. 7. 2 wt.% ZrCl₄ doped LPS, has the lower onset potential compared to LPS, which provides additional proof that the ionic conductivity of doped LPS is greater than the ionic conductivity of the control LPS. Both types of cells showed a gradual increase in overpotential on cycling, which is due to the increased interfacial resistance of the interphase formed between the Li metal and solid electrolyte pellets by continuous reaction between them. However, the overpotential increase rate in case of doped LPS is lower than that of control LPS, thereby proving that the doped LPS is more stable than LPS against Li metal upon electrochemical cycling. Additionally, the symmetrical cell with control LPS cycled only 1332974769.1up to 110 hours before short circuiting, whereas the cells with ZrCL doped LPS did not short circuit until 200 hours. These latter results indicate the ZrCL doping increased the long-term stability of electrolyte with Li-metal by minimizing the detrimental side reactions between Li and LPS.

[0053] High-resolution S 2p and P 2p X-ray photoelectron spectroscopy (XPS) spectra of electrolyte powders were collected using a scan step size of 0.1 eV, as shown in FIGS. 9(A-D). The S 2p spectrum includes spin-orbit-split S 2p3 / 2 and S 2p1 / 2components. Deconvolution of the S 2p spectra resolves sulfur species associated with P-S-Li, P=S, and P-S-P bonding environments. For the undoped electrolyte, S 2p3 / 2 peaks are observed at approximately 160.8 eV, 161.4 eV, and 162.4 eV as shown in FIG. 9(A), respectively, while corresponding peaks in the doped electrolyte appear at approximately 160.93 eV, 161.32 eV, and 162.36 eV FIG. 9(B), indicating a modified bonding environment. The P 2p spectra of both compositions, FIG. 9(C) and FIG. 9(D) exhibit components corresponding to PS43’ and P2S74’ species at approximately 132.5 eV and 131.3 eV, respectively, each comprising spin-orbit-split P 2p3 / 2 and P 2p1 / 2peaks.Electrochemical performance of ASSLBs

[0054] To evaluate the impact of ZrCL doping on the electrochemical performance of all-solidstate lithium batteries (ASSLBs), full cells in CR2032 coin cell formats were assembled with Li as the anode, LiFePO₄ (LFP) as the cathode, and a solid polymer electrolyte (SPE) interlayer between the LFP cathode and the electrolyte pellet. Galvanostatic measurements were conducted from 2.4 V to 4.2 V at a rate of C / 10 at room temperature. No external pressure was applied during this cycling.

[0055] The initial discharge capacities of the cells with pristine LPS and ZrCL-doped LPS were 97.6 mAh g-1and 93.4 mAh g-1, respectively (FIG. 8). After 50 cycles, the discharge capacity of the cell using pristine LPS decreased by 62%, to 36 mAh g-1, while the cell with ZrCL-doped LPS exhibited a significantly lower capacity fade of only 18%, retaining 81 mAh g-1.

[0056] The superior initial discharge capacity and the stable cycling performance with reduced capacity fading over 50 cycles highlight the benefits of using ZrCL-doped LPS as an electrolyte. The absence of external pressure during cycling may have contributed to the somewhat lower 1432974769.1discharge capacities compared to previously reported values. Nevertheless, the enhanced cycling stability of ZrCL-doped LPS demonstrates its potential to improve the performance of ASSLBs.

[0057] In summary, LPS solid electrolytes doped with x wt% zirconium tetrachloride (x = 1, 2, 3 and 4) were successfully synthesized by using a planetary ball milling method followed by subsequent heat treatment. Various characterizations (XRD, Raman and SEM-EDS) of control and doped samples clearly indicate that the zirconium tetrachloride can be incorporated into the LPS matrix without disturbing the crystal structure. Electrochemical impedance spectroscopy (EIS) results further demonstrated that the ionic conductivity increased with 1 and 2 wt% zirconium chloride doping and then decreased upon increasing the doping weight percentage to 3 and 4. The noted increase in ionic conductivity is due to the broadening of Li+transmission channels in the LPS structure. The resulting descrease in ionic conductivity is due to the formation of less ionically conductive phases Li3PS4 and Li₄P₂S₇. Critical current density (CCD) tests indicated that the CCD of 2 wt.% zirconium chloride doped LPS reached a high value of 2 mAcm-2, which is signficantly greater than that of the control LPS samples. This increase in CCD is attributable to the decrease in electronic conductivity, increase in ionic conductivity and the formation of a favorable interphase possibly containing highly electron insulating LiCl. Li / Li symmetrical coin cells with doped LPS showed less overpotential and longer cycling, as compared to those with control LPS, proving the overall usefulness of zirconium chloride doping. In addition, the Li / LFP coin cells showed higher capacity retention and specific gravity, when compared to the doped LPS.

[0058] Although described and illustrated with respect to exemplary embodiments thereof, it should be understood to those skilled in the art that the foregoing and various other changes, omissions and additions may be made therein and thereto, without parting from the intended spirit and scope of the present invention.1532974769.1

Claims

CLAIMS1. A solid electrolyte for an all solid-state lithium battery (ASSLB), the solid electrolyte comprising:synthesized lithium phosphorus sulfide (LPS), which has been doped with a metal halide.

2. The solid electrolyte according to claim 1, wherein the metal halide is zirconium tetrachloride.

3. The solid electrolyte according to claim 1, wherein the LPS solid electrolyte is synthesized by ball milling followed by heat treatment.

4. The solid electrolyte according to claim 2, in which the weight percent of doped zirconium tetrachloride is 1% to 4%.

5. The solid electrolyte according to claim 2, in which the weight percent of doped zirconium tetrachloride is 1% to 2%.

6. The solid electrolyte according to claim 3, wherein the doping is performed during synthesis.

7. The solid electrolyte according to claim 6, in which the ball milling is performed between 8 hours and 16 hours.

8. The solid electrolyte according to claim 6, wherein the heat treatment is performed between 0.5 hours and 2 hours.

9. The solid electrolyte according to claim 8, wherein the heat treatment is performed at 240 °C for approximately 2 hours.1632974769.

110. A process for preparing a solid electrolyte for an all solid state lithium battery (ASSLB), the method comprising:providing lithium sulfide (Li2S) and phosphorus pentasulfide (P₂S₅) as starting precursor materials;adding a metal halide as a doping agent to the precursor mixture ’ ball milling the precursor mixture and added metal halide dopant; and heat treating the ball milled mixture11. The process according to claim 10, wherein the metal halide is zirconium tetrachloride.

12. The method according to claim 11, wherein the weight percent of zirconium tetrachloride is between 1 to 4 percent.

13. The method according to claim 11, wherein the weight percent of zirconium tetrachloride is between 1 to 2 percent.

14. The method according to claim 11. in which the ball milling is performed between 8 hours and 16 hours.

15. The solid electrolyte according to claim 11, wherein the heat treatment is between 0.5 hours and 2 hours.

16. The solid electrolyte according to claim 11, wherein the heat treatment is performed at 240 °C for about 2 hours.32974769.1

Images