Stable cycling of sodium all-solid-state batteries with high-capacity cathode presodiation

Abstract

Capacity retention and cycling performance in a sodium all-solid-state battery (NaSSB) are enhanced by employing a cathode pre-Na strategy by activating Na2S within a composite of Na3PS4 (NPS) and acetylene black (AB) (Na2S-NPS-C). The NPS serves as an ionic conductor while AB provides electronic conductivity.

Description

[0001] STABLE CYCLING OF SODIUM ALL-SOLID-STATE BATTERIES WITH HIGH-CAPACITY CATHODE PRESODIATION

[0002] RELATED APPLICATIONS

[0003] This application claims the benefit of the priority of U.S. Provisional Application No. 63 / 749,389, filed January 24, 2025, which is incorporated herein by reference in its entirety.

[0004] GOVERNMENT RIGHTS

[0005] This invention was made with government support under FM2134764 awarded by the National Science Foundation. The government has certain rights in the invention.

[0006] FIELD OF THE INVENTION

[0007] The present invention relates to sodium all-solid-state batteries (NaSSBs) and more specifically to methods and compositions for reducing sodium loss during initial cycling, providing significantly enhanced cycling performance with superior capacity retention.

[0008] BACKGROUND

[0009] With the widespread adoption of electronic devices and the increasing use of renewable energies, interest in energy storage devices like batteries has surged. Despite the market dominance, lithium-ion batteries (LIBs) face limitations due to resource availability and sustainability concerns. As an alternative, sodium-ion batteries (SIBs) have gained attention due to earth-abundant sodium (Na) and cost-effective transition metal-based cathodes. However, the flammability of their electrolytes remains a significant safety concern.

[0010] Sodium all-solid-state batteries (NaSSBs) with an alloy-type anode (e.g., Sn and Sb) offer superior capacity and energy density compared to hard carbon anode. However, the irreversible loss of Na+at the alloy anode during the initial cycle results in diminished capacity and stability, impairing full-cell performance.

[0011] NaSSBs have emerged as a safer technology that uses nonflammable and nonvolatile inorganic solid electrolytes (SEs), significantly reducing the risk of thermal runaway. However, the practical application of NaSSBs faces significant obstacles due to the low capacity and energy density resulting from the electrode selection, low initial Coulombic efficiency (ICE), and poor cycling stability. The hard carbon (HC) anode widely used in SIBs exhibits limited capacity, typically around 250-320 mAh g'1, and a low volumetricenergy density attributable to its low tapped density, often less than 1 g em'3. In addition, SEs must be added to the anode composite to form an ionic conducting pathway, reducing its energy density and increasing costs. Accordingly, the need remains for a high-capacity anode to boost the energy densities of NaSSBs.

[0012] Na metal anode is appealing in NaSSBs due to its high capacity and lowest potential, however, it is incompatible with most SEs, leading to detrimental issues such as persistent degradation reactions, dendritic growth, and low critical current density. Na-containing alloys, such as Na-tin (Sn) alloy (e.g., NagS ), are confronted with stability issues when exposed to ambient air, which poses significant challenges during fabrication processes. In contrast, Na-free alloy anodes such as Sn offer a notably higher capacity (847 mAh g'1) and a high volumetric density of 7.3 g em'3, making them attractive for high energy density full cells. Moreover, the Na-free alloy anode demonstrates environmental compatibility and stability under atmospheric conditions, mitigating challenges associated with anode fabrication. Additionally, alloy-type anodes can function without needing SEs and carbon additives due to their inherent electronic conductivity and Na+diffusivity. This property not only enhances cell energy density but also reduces the production costs of NaSSBs, making them more economically viable and sustainable for large-scale manufacturing.

[0013] Despite the advantages above, NaSSB full cells with alloy anodes often experience inevitable Na+loss after the first charging / discharging cycle due to Na+trapping in the anode and forming anode / electrolyte interphases. This results in low ICE and considerably reduced energy density and cycling stability due to an unbalanced negative / positive (N / P) ratio in full cells. In this regard, the presodiation (pre-Na) strategy is crucial to mitigate the initial capacity loss, which can be realized by either anode pre-Na or cathode pre-Na methods. Anode pre-Na materials (e.g., Na metal, biphenyl / tetrahydrofuran) often exhibit a low potential and high chemical reactivity, akin to typical Na-containing alloy anodes. This method introduces significant manufacturing complexities and raises safety concerns, making their large-scale production impractical. Conversely, cathode pre-Na methods involve the addition of Na-rich additives into the cathode composites, offering the advantages of precise regulation of pre-Na capacity and compatibility with the widely used anode / cathode systems. The effectiveness of this method hinges on carefully selecting a high-capacity pre-Na agent to maximize the overall cell energy density. When designing pre-Na agents, three fundamental requirements should be considered. First, the pre-Na agent should exhibit significantly higher capacities than cathode materials, which is crucial forenhancing capacity. Second, it should ensure controlled Na+transport behavior below the charge cutoff voltage and avoid further reduction in the discharge potential range of the cathode. Third, the addition and decomposition of the pre-Na agent should not have detrimental effects such as impeding Na+transportation or releasing gases, which are crucial for maintaining stable and efficient battery performance over multiple charging / discharging cycles.

[0014] Research on sacrificial pre-Na additives has primarily focused on liquid-based systems, with relatively little exploration within NaSSBs. Besides, many of these pre-Na agents exhibit electronic and ionic insulating properties. While this characteristic might not significantly impact liquid battery systems due to the good wettability and high ionic conductivity of liquid organic electrolytes, it is detrimental in NaSSBs. For example, one promising pre-Na candidate is Na?S, which meets all the design requirements. However, its ionic insulating nature (~10'nS cm'1) could impede Na+transport, leading to poor redox kinetics and low Na donation capacity in all-solid-state battery architectures. This situation underscores the need to activate Na?S as a pre-Na agent for NaSSBs.

[0015] SUMMARY

[0016] The inventive cathode presodiation strategy provides a practical solution to enhance the full-cell performance and advance the transformation from half-cell to full-cell applications of NaSSBs. The approach employs a Na-rich material to address the aboveidentified challenges, effectively adding a backup sodium source that will release extra sodium during charging. Leveraging the high theoretical capacity and suitable voltage window, Na?S is chosen as the Na donor, which is activated by creating a mixed electronion conducting network, delivering a high capacity of 511.7 mAh g'1. By adding a small amount (i.e., 3 wt.%) of Na?S to the cathode composite, aNaCrCh || Sn full cell demonstrated capacity improvement from 90.8 mAh g'1to 118.2 mAh g'1(based on cathode mass). The capacity -balanced full cell can thus cycle to more than 300 times with > 90% capacity retention.

[0017] According to embodiments disclosed herein, the inventive approach employs a cathode pre-Na strategy by activating Na?S within a composite of NasPS4 (NPS) and a carbon source, e.g., acetylene black (AB) (Na2S-NPS-C), in which NPS serves as the ionic conductor while AB contributes the electronic conductivity. The composite's electronic conductivity reaches around ~10'3S cm'1, and the ionic conductivity improves from theorder of -10'11(for Na2S) to -10'5S cirT1(for the composite). An optimal NPS: Na?S mass ratio (1:2) yields the highest first cycle charge capacity of 511.7 mAh g'1, corresponding to 90% of its theoretical capacity. When implemented in a NaCrCh (NCO) || Sn full cell configuration, the capacity (based on cathode mass) can increase from 90.8 to 118.2 mAh g'1by adding only 3 wt.% of the pre-Na agent. After 1stcycle activation, no gas was released, and only ~1.7 wt.% of the inactive material remained in the cathode. Moreover, enhanced cycling performance was confirmed due to the continuous extraction of Na+from pre-Na agent residue in the following cycles. As a result, the NCO||Sn full cell with cathode pre-Na agent shows significantly enhanced cycling performance with capacity retention of 90.4% after 300 cycles, far superior to 80.5% for cells without the pre-Na agent. The efficacy of this pre-Na technique in augmenting battery performance was also demonstrated in other cathode / anode combinations, including NaxNii / sFei / sMm / sCh (NFM) cathode and Sb anode, showcasing its potential for broad applications across various NaSSB chemistries.

[0018] NaCrCh or NaFeMnNiO acts as the active material on the cathode side. However, the pre-Na agent (a Na2S-NPS mixture) is added to the cathode composite to compensate for the sodium loss (low ICE) from the anode side. This is a general problem for anode materials, such as the tin and antimony.

[0019] The improvements provided with the inventive approach can be viewed from two aspects. The first one is the pre-Na agent provided as an additive to the cathode, designed to be applied to different types of all-solid-state batteries with various cathode and anode material systems as a one-size-fits-all solution. By adding only < 5% additive to the cathode composite's original recipe, the cell performance can be highly improved (the material's structure stability, initial discharge capacity, and long cycling performance). This can be compared to existing approaches, for example, US11949092B2, which discloses use the Na2S-NPS composite as the cathode directly, where Na2S is the active material, and NPS is the solid electrolyte. Such practice is directed to a specific type of battery with a fixed cathode material system rather than a solution to the problem that widely exists for all batteries.

[0020] The second is that the synthesis of the pre-Na agent only includes ball milling without requiring any heat treatment. Ball milling is a simple and easily performed method widely used in solid-state synthesis. It is safer, less energy-consuming, and more environmentally friendly. According to the inventive approach, ball milling is essential.The inventive scheme is expected to be commercially applicable to next-generation NaSSBs for renewable energy storage, offering safer and more sustainable alternatives to lithium-ion batteries, for grid-scale energy storage, leveraging the low-cost and abundant sodium resources. The approach provides enhanced energy density and capacity retention rate with improved cell performance compared to existing products.

[0021] BRIEF DESCRIPTION OF THE DRAWINGS FIGs. 1 A-1G illustrate characterizations and mechanism study of Na+loss in NaSSB full cell with Sn anode during the first cycle, where FIG. 1A diagrammatically depicts a scheme of cathode presodiation strategy used to compensate for Na trapping in the Sn anode and avoid Na loss in the cathode; FIG. IB shows charge / discharge GITT voltage curves of Sn||Na9Sn4 half-cell to show phase transition of Sn alloy anode and Na+trapping in the first cycle; FIG. 1C is a plot showing Na+diffusion coefficient in different phases of Na-Sn alloy determined by GITT; FIG. ID provides XRD patterns of Na-Sn alloy after the 1stcycle of sodiation and desodiation; FIG. IE shows the 1stcycle charge / discharge curve of the NCO||Sn full cell and the NCO||Na9Sn4 half-cell to illustrate their differences in ICE and highlight Na+trapping in the full cell; and FIGs. 1F-1G are SEM images and EDX mapping of the Sn anode after the initial sodiation and desodiation process, respectively.

[0022] FIGs. 2A-2H illustrate pre-Na agent design and characterizations, where FIG. 2A shows the theoretical capacity and the decomposition voltage of the potential cathode additives for presodiation; FIG. 2B diagrammatically illustrates an exemplary scheme describing the preparation of the pre-Na agent using ball milling to mix a cathode material (NFS), a pre-sodiation agent (Na2S), and a carbon source (AB); FIG. 2C provides XRD patterns of the pre-Na agents with various NFS and Na?S ratios (NFS: Na?S) after ball milling; FIG. 2D is a diagram showing the calculated theoretical capacity and the trend of the actual capacity of pre-Na agent with different Na?S ratios; FIG. 2E plots ionic conductivity of different pre-Na agent compositions; FIG. 2F is an Arrhenius plot comparing the pre-Na agent as mixed by hand (circle) and ball milling (triangle) to show the conductivity and activation energy; FIG. 2G provides Nyquist plots of the hand-mixed and ball-milled pre-Na agent assembled with NagS as half-cell before charging; and FIG.

[0023] 2H is a diagram of the scheme to show the pristine Na?S failing to conduct electrons and Na+and its activation by ball milled with NFS and AB to form ionic and electronic pathways.FIGs. 3A-3D illustrate characterization of the particle morphology and structure of the pre-Na agent, where FIGs. 3A-3C are SEM images of the Na?S, NPS, and the pre-Na agent, respectively; FIG. 3D is a cryo-TEM image of the pre-Na agent particles after ball milling where the areas within the boxes are partially magnified enlarged to show the NPS (lower box), Na?S particle (upper right box), and AB structure (upper left box).

[0024] FIGs. 4A-4E illustrate electrochemical characterization of the NCO cathode composite with pre-Na agent added, where FIG. 4A provides a comparison of charging and discharging voltage curves between NCO cathode and pre-Na agent; FIGs. 4B-4C are charge and discharge curves of the first cycle of NCO||Na9Sn4half-cell and NCO||Sn fullcell, respectively, with different pre-Na agent ratios; FIG. 4D provides a comparison of cathode discharge capacity (left bars) and full cell energy density (right bars) (based on the mass of cathode + anode + pre-Na) after adding the pre-Na agent; and FIG.4E compares the long-term cycling performance of the NCO||Sn full cell with and without pre-Na agent.

[0025] FIGs. 5A-5C illustrate the composition evolution of the cathode composite during the charging process, showing XPS analysis of the C Is, S 2p, and P 2p elements, respectively, to characterize the change of chemical composition at different states of charge (SOC).

[0026] FIG. 6A is a schematic of equivalent circuit used in EIS analysis; FIGs. 6B-6G illustrate evolution of the cathode composite impedance with the pre-Na agent during the 1stcharging process, where FIGs. 6B-6D are Nyquist plots of the cells with pre-Na agent as the cathode at SOC of 0%, 70%, and 100%, respectively; FIGs. 6E-6G show DRT results derived from the Nyquist plots with peaks fitted at different SOC (20%, 70%, and 100%) to show multiple electrochemical processes.

[0027] FIGs. 7A-7C plot voltage profile as a function of specific capacity for the 1st cycle tested from the full cells with NFM-Sn, NFM-Sb, and NCO-Sb, respectively; FIGs. 7D-7F plot the cycling performance of the same cells, with and without presodiation.

[0028] DETAILED DESCRIPTION OF EMBODIMENTS

[0029] The inventive easy-to-implement cathode presodiation strategy employs a Na-rich material to address the challenges of irreversible sodium ion loss at the ally anode during the initial cycle. Leveraging the high theoretical capacity and suitable voltage window, in exemplary embodiments, Na?S is selected as the Na donor, which is activated by creating a mixed electron-ion conducting network, delivering a high capacity of 511.7 mAh g '. Byadding a small amount (i.e., 3 wt.%) of Na?S to the cathode composite, a NaCrCh || Sn full cell demonstrates capacity improvement from 90.8 to 118.2 mAh g1(based on cathode mass). The capacity-balanced full cell can thus cycle to more than 300 times with >90% capacity retention. This approach provides a practical solution to enhance the full-cell performance and advance the transformation from half-cell to full-cell applications of NaSSBs.

[0030] The Na loss in the alloy -type anode

[0031] FIG. 1A diagrammatically illustrates the inventive approach to cathode presodiation to compensate for sodium trapping in the anode while avoiding sodium loss in the cathode. A series of alloying processes of the Sn anode occurs during sodiation, leading to sequential formation of distinct phases. To comprehensively evaluate the Na+transport and kinetic characteristics in the Sn anode, a galvanostatic intermittent titration technique (GITT) test was conducted at 0.05 C, involving a charging / discharging period of 0.5 hours followed by a 2-hour rest interval. FIG. IB illustrates the voltage profile during the first cycle of the Sn anode. The profile reveals a series of alloying processes of the Sn anode during sodiation (discharging step for Sn || NaySm half-cell), marked by the sequential formation of NaiSm, Na9Sn4, and NaisS phases, which are characterized by distinct voltage plateaus. For the subsequent desodiation (charging step for Sn || NaySm half-cell) process, the NaisS alloy transforms back to NagS and NaiSm with significant hysteresis. However, the rest of Na (~0.8 mol of Na per Sn) remains trapped within the anode due to the slow Na+diffusion kinetics of the Na-Sn alloy at low Na content, which was also revealed by the evolution of the Na+diffusion coefficient as shown in FIG. 1C. Initially, the diffusion coefficient of the pristine Sn anode is approximately 2xl0'6cm2s-1. As Na begins to alloy with Sn at a potential of about 0.40 V (vs. NagS , the same below), the NaiSm phase is formed, and the Na+diffusion coefficient drops sharply to -10'9cm2s-1. This drastic decrease reflects the slow kinetic properties of the Na-Sn alloy at low Na content. As sodiation progresses, the formation of the NagS phase at the second plateau at 0.08 V is accompanied by an increase in the diffusion coefficient (>10‘7cm2s-1), indicating enhanced transport properties compared to the NaiSm phase. At the end of the sodiation step, the NaisS phase was formed with the potential decreases to -0.13 V. A moderate Na+diffusion coefficient of -10"7cm2s-1was obtained at this stage. For the desodiation (Na+extraction) process, a similar Na+diffusion coefficient, ranging from high to low Na content, was observed at each stageof the Na-Sn alloy. As Na extraction nears the completion of the NaiSm phase, the diffusion coefficient rapidly declines to ~1O'10cm2s-1, indicating difficulty in removing the remaining Na. Consequently, some Na becomes irreversibly trapped within the anode, leading to permanent Na loss.

[0032] FIG. ID shows the X-ray diffraction (XRD) pattern of the Sn anodes after charging / discharging in the half-cell to identify the composition of the products. The characteristic peaks of NaisS and NagS were detected from the fully sodiated electrodes. After desodiation, the XRD pattern showed a strong signal of the NaiSm phase and lower Na+content of NaS . Joshua et al. calculated the formation enthalpy per atom versus Na concentration in the Na-Sn compound and constructed the convex hull to show the stable structures. The results show that compared with pure Sn, NaiSm has a lower formation energy, about -0.02 eV / atom, which results in a barrier for Na to return to the cathode during discharge. Such alloying components as discharged products have been observed in earlier works.

[0033] The NCO || Na9Sn4 half cells and NCO || Sn full cells were tested to elucidate the Na loss phenomena, as shown in FIG. IE. For the NCO || NagS half-cell, a high capacity of 117.5 mAhg'1and a high ICE of 97.61% was achieved. In contrast, a lower capacity of 94.6 mAhg'1and ICE of 80.6% were achieved for the NCO || Sn full cell. In the NCO || NagS half-cell, the excess Na inventory in the NagS counter electrode facilitates the restoration of Na stoichiometry in the NCO cathode, leading to a high ICE. Conversely, in the NCO || Sn full cell, a significant portion of Na+becomes trapped within the Sn anode during the alloying process, resulting in a much lower ICE.

[0034] To study the morphology evolution and Na distribution, FIGs. IF and 1G show the focused ion beam milling-scanning electron microscopic (FIB-SEM) images and energy dispersive X-ray spectroscopic (EDX) mapping of the Sn anode in the full cell at the charged and discharged states, respectively. After charging (FIG. IF), the Sn anode alloys with Na and experiences apparent volume expansion. The observed striped structure could be attributed to the different phases of Na-Sn alloy due to a nonuniform sodiation reaction. The EDX mapping of Na shown in the cyan areas suggests a higher Na concentration within the charged anode. After discharging, the particles contract when Na+is removed from the alloyed anode, with the thickness reduced from ~16.1 z / m to ~6.8 / / m. While most of the Nais removed, a significant amount remains at the anode as shown in the EDX image, indicating irreversible Na trapping.

[0035] The trapped Na results in a Na deficiency in the cathode, structural changes, and degradation in both the cathode and anode over subsequent cycles, leading to capacity decay in full cells. Introducing Na-rich additive with a high decomposition capacity to the cathode is logically feasible to compensate for the Na inventory loss due to these reasons. Hence, properly controlling the N / P ratio and the presodiation extent is key to ensuring there is sufficient Na in the cathode for stable battery cycling.

[0036] Pre-Na agent design and characterization

[0037] Theoretical capacity is a primary consideration when selecting pre-Na agents. Composites with heavy anions and more complex structures may not be ideal due to their lower capacities. Pre-Na agents with higher capacities (> 500 mAhg'1) are advantageous because they enhance the battery energy density while minimizing the negative impact of introducing additional reaction products. In this regard, simple compounds including sodium nitride (NasN), sodium oxide (Na2O), sodium phosphide (NasP), and sodium peroxide (Na2C>2) exhibit potential, as indicated by shaded window in FIG. 2A. However, for NasN, Na2O, and Na2C>2, the nitrogen or oxygen gas generated in the decomposition process during the first charging cycle will cause volume expansion and deterioration of the interface contact thus increasing impedance and accelerating cell performance degradation. Although NasP provides a high capacity of 804 mAh g'1with no gas release, its low decomposition voltage (<1 V vs. Na / Na+) is incompatible with the electrochemical stability window of the widely used halide and sulfide SEs. This incompatibility triggers reactions with SEs and the formation of insulating products, which significantly decrease the ionic conductivity, escalating the battery resistance and impedance.

[0038] In contrast, Na2S has emerged as a more promising pre-Na agent with a moderate decomposition voltage (~1.8 V) and a high theoretical capacity of 687 mAh g'1. Unlike nitrides and oxides, its decomposition yields only elemental sulfur without gas production, avoiding volume expansion and gas-related degradation. However, pure Na2S suffers from extremely low ionic conductivity (c\a) of 5.4* 10'11S cm'1and is almost electronically insulating, which hinders its electrochemical oxidation and the Na+extraction during charging. To overcome these limitations, it is essential to create ionic and electronic transport pathways that facilitate the electrochemical decomposition of Na2S. As onesolution, NPS is introduced to activate Na?S for pre-Na due to their similar chemical compositions. As a widely used SE in NaSSBs, NPS offers a notably high ionic conductivity (IxlO’4to 2*10'4S cm'1) with an oxidation voltage of approximately 2.5 V. This is higher than the decomposition voltage of Na?S but lower than typical cathode operation voltage (e.g., NCO, NFM), making it an ideal choice for activating Na?S with minimal impact. The higher desodiation voltage of NPS allows Na?S to contribute Na+at a lower voltage, followed by Na+extraction from NPS itself, maximizing its capacity utilization. After the decomposition of NPS, there is a decrease in ionic conductivity for the pre-Na agent, which prevents Na+from returning and ensures the irreversible pre-Na process.

[0039] According to embodiments of the inventive scheme, the composite of a cathode material, pre-sodiation agent, and carbon Na2S-NPS-C pre-Na composites were synthesized by simple ball milling. The effects of ball milling and the varying ratios of Na?S and NPS in these composites were investigated. A scheme describing ball milling to obtain the pre-Na agent is shown in FIG. 2B, where the milling conditions The large particles of Na?S were fragmented into smaller sizes during the ball milling process and combined with cathode material, in this case, NPS particles and a carbon source, e.g., AB, to obtain a higher ionic and electronic conductivity as a composite. As will be apparent to those in the art, in addition to NPS, appropriate CAM (cathode active material) for implementation under the inventive scheme may include, but is not limited to, NCO (e.g., NaCrO?), NFM (e.g., NaFeMnNiO), and other sodium -based layered transition metal oxides (NaxM02, where M = Ni, Co, Mn, Fe, Cr, V, etc.) , which can reversibly host Na+ while exhibiting high redox potentials and theoretical capacities. FIG. 2C shows the XRD patterns of Na2S, NPS, and synthesized pre-Na composites. After ball-milling, pre-Na agents exhibit identical peaks of Na2S with increased peak intensity for higher Na2S content in pre-Na agents. The weak NPS peak indicates that the crystalline NPS was converted into an amorphous phase after the ball-milling process. Meanwhile, the Na2S peak was broadened, indicating the grain size reduction.

[0040] The initial charging capacity of pre-Na with different Na2S and NPS ratios is tested in a cell configuration of pre-Na agent || NPS || Na9Sn4(FIG. 2D). The theoretical capacity of the pre-Na agent was calculated based on the total Na+content in the composition (total mass ofNa2S and NPS) (Eq. 1):

[0041]

[0042] Where Ctis the theoretical capacity of the pre-Na agent;Na2SandNPSare the mass ratio of Na?S and NPS in pre-Na agent, respectively; CNa2Sand CNPSare capacity for Na?S (687 mAh g'1) and NPS (352 mAh g'1), respectively.

[0043] After ball milling with AB, pure NPS exhibits a capacity of 148.8 mAh g'1after being charged to 4V, achieving only 42.3% of the theoretical value. This insufficient decomposition is likely due to the poor kinetics from the continuous decrease in \ along with the process of Na+extraction. The charging curve for pure NPS is a continuous slope, with no plateau at any specific voltage. As the Na?S content increases, the Na+donation capacity rises dramatically due to the substantial release of Na+from Na?S. The highest charge capacity of 511.7 mAh g'1was achieved at the mass ratio of NPS: Na?S at 1:2, corresponding to an approximate molar ratio of 1:6. The theoretical capacity for this composition is estimated to be around 575.3 mAh g'1, indicating that near 90% of the Na+are utilized during first charging. However, a capacity decrease is observed when the Na?S ratio increases beyond this point. Only 417.5 mAh g'1was obtained at the ratio of NPS: Na?S at 1:3. This reduction in capacity can be attributed to the decreased volume ratio of NPS electrolytes leading to poor percolation network for Na+conduction. The differential capacity (dQ / dV) curve reveals the reaction plateaus of each composite ratio, with two characteristic peaks appearing at around 2.0V and 2.3 V, corresponding to the two stages of the phase transformation of the Na2S during the decomposition. At the end of the first stage at 2.1V, the 1 :2 ratio sample achieves a capacity of 259.5 mAh g'1, about 60% of the Na2S theoretical capacity. This indicates an intermediate product of polysulfide Na2Sxwas generated. In the second stage, Na2Sxcontinuously loses Na, contributing to the remaining charge capacity. However, without mixing with NPS electrolyte, the peak intensity in the 2ndstage of the Na2S-AB mixture is not apparent. Due to the low r\ of Na2S, it is difficult to fully activate Na2S by further extracting Na+, leading to a much lower capacity. Therefore, sufficient NPS is critical to maintain ion transport pathways within particles to fully extract the Na+capacity of the pre-Na agents.

[0044] To reveal the electrochemical properties of the pre-Na composites with various ratios, their \ and electronic conductivities (ue-) were measured and compared to quantitatively dissect the role of NPS and AB. The r\ of Na2S-NPS-C composites were measured by the DC polarization method using a symmetrically-layered cell of NagS || NPS || Na2S-NPS-C || NPS || NagS . The NPS layers function as electronic block layers,and their resistance was subtracted when calculating \aof the pre-Na agent. The NagS alloy added on each side was a Na source and current collector. The r\a of the pre-Na agent with different NPS: Na?S ratios were determined from the obtained value after applying a 0.1V bias voltage for 300 s, which shows a much-improved c\athan pure Na?S and an increasing trend with a higher NPS ratio. Specifically, the uNa+ was 3.9><10'5, 1.8* 10’5, and 1.2* 10'5S- cm'1for composites with NPS: Na?S mass ratio of 1:1, 1:2 and 1:3, respectively, as shown in FIG.2E. The ue- of pure Na?S is extremely low, which behaves as an electronic insulator. However, after ball milling with AB, the ue- was improved to 1.3 * 10’2, 1.8* 10‘2, and 2.8* 10'2S cm'1for the respective NPS: Na?S mass ratios. Compared with the r\aof the pre-Na agents, their oe- were nearly three orders of magnitude higher, indicating that ionic conductivity is the main factor limiting charge transfer and capacity utilization. The NPS: Na2S=l:2 mixture with the highest specific Na+capacity was selected for further analysis.

[0045] To understand the impact of the ball-milling process, the ionic conductivities of samples prepared by ball-milling and hand-mixing were measured at different temperatures to construct an Arrhenius curve. As shown in FIG. 2F, the ball-milling samples exhibited higher ionic conductivity (about one magnitude) across all temperatures compared to the hand-mixing samples, with a lower activation energy (0.35 eV vs. 0.44 eV). Both samples were assembled into half-cells with NagS for electrochemical impedance spectroscopy (EIS) measurement. FIG. 2G indicates that the hand-mixing sample exhibits no electrochemical activity. In contrast, the ball-milling sample showed an RPof approximately 107 , with the characteristic peak frequency of 104Hz. This difference suggests that ball milling is essential to activate Na?S by forming a conductive pathway for ions and electrons between the particles, which is represented by the darker “chains” shown in the right panel of FIG. 2H

[0046] SEM and cryogenic transmission electron microscopy (cryo-TEM) with EDX were utilized to characterize NPS and carbon distribution within the pre-Na agent composites. The pristine Na?S and NPS particles exhibit sizes around 3.5 pm and 1.5 pm, respectively shown in FIGs. 3A-3B. FIG. 3C provides the corresponding SEM image for the pre-Na agent. After the ball milling process, the large Na?S particles were reduced to smaller particles at around 1.5 pm, and the softer sulfide NPS particles were ground and dispersed around the Na?S particles. SEM-EDX mapping of the pre-Na agent demonstrated a uniform elemental distribution of S, P, Na, and C. Cryo-TEM shows lattice fringes with a spacing of0.494 nm, corresponding to the (101) lattice of crystalline NPS (FIG.3D, lower inset panel. Distances of 0.325 nm and 0.233 nm are directed towards the (200) and (220) directions of Na?S, respectively, with these characteristic fringes filling most of the field of view as the primary phase (FIG. 3D, upper right inset panel). Cluster-like fringe structures can be observed, corresponding to typical carbon structures from the AB in the pre-Na agent (FIG.

[0047] 3D, upper left inset panel). EDX mapping of a single particle of pre-Na composite reveals that NPS (characterized by the P signal) and Na?S are well-mixed in particle scale after ball milling. This reveals the structural characteristics of the pre-Na agent and elucidates the mechanism of using NPS and AB to improve ion / electron conductivity.

[0048] Electrochemical performance

[0049] FIG. 4A provides the charge and discharge profile of the NCO cathode and the pre-Na agent in the voltage window of 2.25 V to 3.5V. The NCO cathode exhibited a capacity of 117.5 mAh g'1, while the pre-Na agent exhibited a significantly higher charging capacity of 511.7 mAh g'1with a negligible discharge capacity in the same voltage range. This property meets the requirement that after the pre-Na agent releases its stored Na+during the cathode charging process, while not taking in Na+during discharging. Further investigation was conducted on NCO cathode composites containing 60 wt.% of active material and 1 wt.%, 2 wt.%, and 3 wt.% of pre-Na agent (FIG. 4B). After aligning them to the same capacity, it was observed that Na+is extracted from the pre-Na agent before reaching the NCO charging plateau. Adding 3wt.% of pre-Na agent to the composite increased the charging capacity from 119.0 mAh g'1(w / o pre-Na agent) to 150.2 mAh g'1. All cells displayed similar discharge capacities, indicating that the extra Na+transferred from the cathode is accumulated at the anode side.

[0050] NCO cathodes were tested with and without the pre-Na agent in an NCO| | Sn full cell with an N / P ratio of 1.1 and a voltage window of 1.75 V to 3.6V (FIG. 4C). These cells exhibit a similar ICE of -80%, while their discharge capacity, plotted as the left bars in FIG.

[0051] 4D, increases with more pre-Na being added in the cathode composite. With 3wt.% of pre-Na agent, the discharge capacity of the NCO||Sn full cell increased from 90.8 mAh g'1to approximately 118.2 mAh g'1, similar to the NCO||Na9Sn4 half-cell, which corresponding to a 30.2% increase in capacity. This translates to a 21% increase in energy density (right bars in FIG. 4D) based on the total weight of the cathode, anode, and pre-Na agent. The full cell with the pre-Na additive also demonstrated stable cycling and significantly enhancedcapacity retention (FIG. 4E). After 300 cycles at 0.5C, the capacity retention of the full cell without pre-Na was 80.5%, while the cell with 3% pre-Na agent achieved a 90.4% capacity retention. This suggests that even though most of Na+has been extracted, the residual Na+in the pre-Na agent acts as a “Na+reservoir” that may continuously compensate for the Na+loss during the cycling, thereby enhancing cycling stability.

[0052] The pre-Na mechanism

[0053] Using SEM imaging of the post-charging cathode under backscattering mode, dark spots could be observed randomly dispersed within the cathode composite. EDX revealed that these areas contain sulfur (S) and phosphorus (P)-rich species, suggesting they are residues of pre-Na. X-ray photoelectron spectroscopy (XPS) was used to determine the composition of these species. In the pristine pre-Na agent, the C-C bond seen in the Cis spectrum (FIG. 5A) indicates the presence of carbon (AB). The S 2p spectrum (FIG. 5B) reveals distinctive PS43' peaks relating to NPS at 161.6 eV and Na?S at 159.6 eV. The P 2p spectrum (FIG. 5C) also displays a PS43' peak at 131.4 eV. At the charging step, the decomposition of Na?S is proved by the reduction of its peak intensity and the appearance of new peaks at 163.8 eV. This indicates the formation of elemental S as the final product. Concurrently, the peak intensity of PS43' decreases in both the S 2p and P 2p spectra, further confirming the decomposition of NPS. By the end of the charging process (100% SOC), the peaks for Na?S and PS43' are no longer detectable. A significant increase in the S peak and a new peak at 162.4 eV are observed, reflecting the formation of Na-S bonds and the polysulfide compound (Na2Sn) from Na?S decomposition. Simultaneously, the P 2p spectrum shows new peaks at 133.2 eV and 134.2 eV, corresponding to the oxidized product P2S5.

[0054] EIS and distribution of relaxing time (DRT) were used to analyze the decomposition process of the pre-Na agent at different SOCs. The cell was charged at 0.1C and paused for an EIS test per hour. Over the entire charging process, the cell Ohmic resistance (Ro) was maintained at about 480 Q, indicating no electrolyte-related degradation. FIG.6A illustrates the equivalent circuit used in EIS analysis of evolution of the pre-NA agent during the first charging. Referring to the EIS curves in FIGs. 6B-6D, a consistent intercept with the real axis, represented by a resistor R1 with a value of approximately 450 Q, was observed in the high-frequency range of the spectrum. This value remains relatively stable across various charging depths, which matches the resistance of the NPS solid electrolyte, suggesting itsassociation with the SE layer. However, there is a noticeable variation in the intermediate frequency semicircle corresponding to the state of charge (SOC). As the charging progresses, the radius of this semicircle gradually increases, revealing increased charge transfer resistance at the cathode side, implying a hindered charge transfer process. During charging, the decomposition of Na?S into element sulfur brings in non-conductive products and leads to a volume change in the active material. This alteration affects the stress / strain conditions at the cathode and subsequently modifies its conductive contact with SE. As a result, there is an evident rise in the charge transfer resistance (R2). Towards the end of the charging cycle, a third semicircle (R3 / CPE3) becomes distinct in the impedance spectrum. This new feature suggests the formation of an additional interface at the cathode side, correlating with a decrease in conductivity, likely due to the decomposition of NPS in the pre-Na agent at the late stages of charging. After most Na+are extracted from Na?S, NPS will perform as a Na source and release Na+until the c\ais insufficient to sustain the continuous decomposition of Na?S. As the SOC increases from 0 to 70% (FIGs. 6B-6C), the polarization resistance R2 of the battery in the equivalent circuit rose from 61.4 Q to 181.9 Q, which relates to the charge-transfer reactions with the characteristic frequency at about 105to 103Hz. This corresponds to the continuous extraction of Na+from Na?S. When SOC is higher than 70%, a new semicircle with a characteristic frequency at 103to 101appears (R3 / CPE3), indicating that NPS starts to decompose and leads to slower Na+transportation kinetics inside pre-Na particles. As the Na+was further extracted from the pre-Na agent, R3 increased dramatically (FIG. 6D), resulting from the NPS decomposition and the depletion of Na+from pre-Na.

[0055] The DRT curves corresponding to different SOCs are shown in FIG. 6E-6G, which were deconvoluted to six different peaks (P1-P6), corresponding to the dominant electrochemical processes, further elucidating the decomposition process of the pre-Na agent. The Pl peak, with a time constant of IO’6s, remains unchanged during the charging process, corresponding to the resistance of Na+ions transferred across the SE grain boundaries. Peak P2-P4, with time constants ranging from 10'5s to 10'3s, are related to the charge transfer processes between the electrode and SE, including both the anode and cathode sides. These peaks gradually increase as the SOC increases from 0% to 70%, as shown in FIG. 6, panel g. Given that abundant NagS alloy is used on the anode side, the increased impedance primarily contributes to the cathode side of the pre-Na agent As Na+is gradually extracted from Na?S, the intermediate product of polysulfide Na2Sxhas lowerelectrochemical activity, leading to increased charge transfer resistance. When the pre-Na is further charged to SOC>70%, peaks P5 and P6 with a time constant from 10'3s to 10'1s surge dramatically, with the peak shift to the higher time constant or lower characteristic frequency. These processes are related to diffusion within the pre-Na agent. At this stage, final products of elemental S and P2S5 are generated, which impedes Na+transfer from the pre-Na. Concurrently, the decomposition of NPS begins to affect the Na+diffusion of the pre-Na agent significantly. The DRT analysis supports the EIS discussion well and is consistent with our XPS results.

[0056] Broad application of pre-Na in NaSSBs

[0057] Due to the favorable decomposition voltage of Na?S, this pre-Na agent can be utilized in various NaSSB systems. To demonstrate its broad applicability, further testing was performed in full cells with NaNii / sFei / sMm / sCh (NFM) cathode and / or an antimony (Sb) anode. FIGs. 7A-7C illustrate the first charge and discharge cycles for three different full cell battery configurations: NFM / Sn, NFM / Sb, and NCO / Sb, respectively. The areal capacities were set at approximately 1.2 mAh- cm'2, and the N / P ratio was 1.1. For the Sn anode, the cells using NFM as the cathode demonstrated an ICE of around 82%, comparable to that of NCO. With a 3% pre-Na agent in NFM cathode composite, the charging capacity increased from 124.91 mAh g-1 to 159.66 mAh g'1, with an additional capacity of about 35 mAh g'1released from the OCV to 2.6 V range in the charge curve, as an indicative of the supplementary of the Nat Due to the higher voltage of Na-Sb alloy, the Sb incorporated entire cell exhibits a lower working voltage for the cells with the Sb anode. For both NCO / Sb and NFM / Sb full cells, the ICE was only about 70%, indicating a more significant sodium loss in the Sb anode system than Sn. Additional sodium supplement content is required to obtain a commensurate discharge capacity for various anode materials with lower ICE. To compensate for the Sb anode's higher sodium loss, the pre-Na agent's content in the cathode was adjusted from 3% to 5% for full cell testing. For the NFM / Sb full cells (FIG. 7B), with presodiation, the first-cycle discharge capacity increased from 87.56 mAh g'1to 112.40 mAh g'1. Similarly, for the NCO / Sb full cells (FIG. 7C), in increment from 82.84 mAh g^to 105.90 mAh g'1. By adding 5% pre-Na agent, an extra 35 mAh g'1capacity was supplied during the charging process, revealing the pre-Na agent's feasibility in the Sb anode system. For the full cell with different cathode and anode combinations, thesodium supplementation can take effect in different charge-discharge windows, which benefits from the suitable electrochemical decomposition voltage range of Na?S.

[0058] Observing the long-term cycle stability, all the batteries with pre-Na agents showed a higher capacity retention rate. For NFM / Sn full cells (FIG.7D), the capacity retention rate after 300 cycles is as high as 96.50%. The capacity retention rate of NFM / Sb full cells after 300 cycles is 93.76% (FIG. 7E), and around 78% forNCO / Sb full cells (FIG. 7F).

[0059] The following non-limiting examples provide procedures, results, and discussion to further illustrate the inventive approach:

[0060] Example 1: Material synthesis

[0061] NasPS4 electrolyte (NFS): Two precursors, Na?S (Sigma Aldrich 98%) and P2S5 (Sigma Aldrich 99%), were mixed with a pestle in the mortar. Then, the mixture was transferred into ball milling jars with a total mass of 1 g, sealed with Y-ZrCh grinding media under an argon atmosphere. Ball milling was carried out using a high-energy planetary ball milling machine (TMAX XQM, China). Commercially-available high-energy ball milling machines suitable for this step typically provide rotational speeds ranging from around 400 RPM to 2,000 RPM. For the present example, the rotational speed used was about 700 RPM, for a period of about two hours to about eight hours. For this test, six hours was used. NasPS4 electrolyte powder was collected after ball milling without further treatment.

[0062] Presodiation agent: First, synthesized NPS, Na?S (Sigma Aldrich 98%), and acetylene black (Thermo Scientific 99.9%) with the desired mass ratio were manually mixed in the mortar. The mixture was then transferred into ball milling jars and sealed with Y-ZrCh grinding media under an argon atmosphere. Ball milling was carried out by a ball miller machine (TMAX XQM) proceeding at 600 RPM for 5 hours.

[0063] Example 2: Materials characterization

[0064] X-ray diffraction (XRD) samples were loaded into glass capillary tubes (Charles Supper) of 0.5 mm diameter inside a glovebox, sealed with clay and wrapped with parafilm. Then, the samples were removed from the glovebox and flame-sealed using a butane torch. These samples were evaluated on a Bruker Kappa goniometer coupled with a Bruker APEX II Ultra diffractometer using Mo Ka (k = 0.7093 A) radiation at 40 kV and 40 mA within a Debye-Scherrer geometry.X-ray photoelectron spectroscopy (XPS) data was gathered using the AXIS Supra XPS from Kratos Analytical. The S 2p, P 2p, C is, and F Is spectra were compiled and analyzed using CasaXPS software.

[0065] Focused ion beam (FIB-SEM) was used to characterize the surface and cross-section morphology of the NPS films. The exposed and heat-treated films were mounted onto an SEM sample stage (Ted Pella) and transferred into the FEI Scios DualBeam FIB-SEM using the air-sensitive holder to avoid ambient air exposure. The exposed samples were prepared in a glovebox and sealed before being transferred to the focused ion beam scanning electron microscopy (FIB-SEM) instrument, effectively preventing contamination. EDS mapping was collected using a 10 kV electron beam with a current of 0.1 nA.

[0066] Raman spectroscopy was performed using a Renishaw inVia Raman Microscope and a 532 nm laser source. Due to the samples' air sensitivity, all samples were sealed in two thin glass slides in the glove box and directly exposed to the laser beam during testing.

[0067] TEM images were captured using Titan Cubed Themis Z 60-300, ThermoFisher Scientific, with aberration correction, an X-FEG source, and a Bruker Super-X EDX detector. The test was operated at 300 kV, with a 30 pA beam current, a 21.5 mrad convergence semi-angle, and an 80-379 mrad collection semi-angle. Atomistic models were generated using Crystalmaker software and Fast Fourier transform (FFT) patterns. The inverse analyses and radial intensity profiles were conducted using DigitalMicrograph (Gatan) software.

[0068] Example 3 : Electrochemical characterization

[0069] The ionic conductivity of the pre-Na agent was measured by the DC polarization method using a symmetrical cell (NagS || NPS || pre-Na agent || NPS || NagS ). All materials were added in a 10 mm Poly etheretherketone (PEEK) die and pressed at 70 MPa with titanium plungers. The NPS layers between the Na-Sn alloy, performing as the sodium source, and the pre-Na agent serve as the electron block layer to avoid short circuits from carbon. The o\a of the pre-Na agent with different Na2S-NPS ratios was determined from the obtained value after applying a 0. IV bias voltage for 300s, using Bio-Logic VSP-300.

[0070] Electrochemical impedance spectroscopy (EIS) was measured with a frequency range from 7 MHz to 1 Hz and an AC amplitude of 10 mV using Bio-Logic VSP-300. To test the EIS revolution of the pre-Na agent, the cell configuration as pre-Na || NPS || NagS was assembled and charged at 0.1C for lOh. To avoid the reaction between the NPS andNagSru anode, which complicates the analysis and makes it challenging to identify the reaction on the cathode side, NBH was added between the SE layer and anode as an isolation layer. A 6 mg pre-Na agent was weighed and placed on the NPS-NBH electrolyte layer as the cathode.

[0071] DC polarization was collected by directly testing the pre-Na agent between the Ti plungers, while a potential bias of IV was exerted to stimulate the electrochemical reaction. Solid-state batteries were assembled in a 10 mm PEEK die with two Ti plungers. For pellet cell assembly, 100 mg of commercial halide solid electrolyte (99.9%, Unigrid, USA) was pressed under 110 MPa to form a rigid pellet without cracks as the support layer with a thickness of about 400 pm. To relieve the interfacial reaction at the SE-anode interface, lOmg NCBH (99.9%, Unigrid, USA) was added to form an isolated layer. For the presodiated cells, a 3% or 5% mass ratio of the pre-Na agent was mixed with NCO or NFM cathode composites. The NCO cathode composite was manually prepared by mixing NaCrO? active materials, halide SE, and vapor-grown carbon fibers (VGCF) in a 50:48:2 weight ratio and added to one side of the SE pellet. Since the NFM cathode has a lower conductivity than NCO, an extra ball milling step was required to help improve the cell performance. 0.4 g tin or antimony powder was weighed and mixed with 0.2g CNT for anodes. Then, the mixture was distributed in 0.3 g DI water in a THINKY Mixer at 1800 rpm for 20 min.

[0072] The configuration of full cells is shown as NCO / NFM || halide SE || Sn / Sb. The assembled cells with all three parts were pressed again at 270 MPa before electrochemical testing. The cells were cycled using a Neware Battery cycler at room temperature and under pressure at around 75 MPa.

[0073] The inventive approach employs a universal cathode pre-Na strategy to address the first cycle irreversible capacity loss in NaSSBs. Na?S was used as the Na donor due to its high theoretical capacity, no gas generation, and suitable Na extraction voltage window. The activation of Na?S was achieved by creating a mixed electron-ion conducting network using NPS and AB. The ball milling process is essential for activating Na?S, as it significantly reduces the particle size of Na?S and increases contact with NPS and AB, ensuring a uniform percolation network. The pre-Na agent with the optimal ratio NPS: Na2S=l:2 exhibits enhanced uNa+of 1.8* 10’5S cm'1and os- of 1.8* 10’2S cm'1, achieving a high capacity of 511 mAh g'1. By incorporating a small amount of Na2S-NPS-C pre-Na agent into the cathode, the released Na+compensates for anode trapping, allowing the full cell to achievethe theoretical capacity, thereby maximizing energy density and advancing the development of NaSSBs. Additionally, cycling stability is improved due to the gradual release of residue Na+from the pre-Na agent, offering significant potential for next-generation high-energy-density batteries.

Claims

CLAIMS:

1. A method for limiting capacity decay in a sodium all-solid-state battery (NaSSB) comprising:adding a Na-rich additive to a sodium-based cathode material to form a mixture, wherein the Na-rich additive has a high decomposition capacity relative to the cathode material and has minimal gas release during charging; andball milling the mixture to form a powder and to activate the Na-rich additive.

2. The method of claim 1, wherein the Na-rich additive has a decomposition voltage within a range of 1.8V to 3.5V.

3. The method of claim 1, wherein the Na-rich additive comprises Na?S.

4. The method of claim 1, wherein ball milling is performed in an inert atmosphere.

5. The method of claim 1, wherein ball milling is performed within a range of 400 RPM to 800 RPM for a period in a range of two hours to eight hours.

6. The method of claim 1, wherein the cathode material comprises one or more of NasPS4 (NPS), NaCrCh (NCO) NaFeMnNiO (NFM), and other sodium-based layered transition metal oxides (NaxMCh, where M = Ni, Co, Mn, Fe, Cr, V).

7. The method of claim 1, further comprising adding a carbon source to the mixture.

8. The method of claim 7, wherein the carbon source is one or more of acetylene black (AB), graphite, graphene, carbon nanotubes (CNT), and carbon nanofiber.

9. The method of claim 1, wherein the mixture has a mass ratio of cathode material to Na-rich additive of from 1:1 to 1:6.

10. The method of claim 9, wherein the mass ratio is 1 :2.

11. A method for enhancing capacity retention and cycling performance in a sodium all-solid-state battery (NaSSB) comprising:forming a sodium-based electrolyte by ball milling sources of sodium, phosphorous and sulfur to produce a NPS powder;mixing the NPS powder with a presodiation agent at a desired mass ratio to form a mixed powder;ball milling the mixed powder to form a composite powder; andadding the composite powder to a cathode material to fabricate the NaSSB.

12. The method of claim 11, wherein the mixed powder further comprises a carbon source.

13. The method of claim 12, wherein the carbon source is one or more of acetylene black (AB), graphite, graphene, carbon nanotubes (CNT), and carbon nanofiber.

14. The method of claim 11, wherein the sodium-based electrolyte is NasPS4 (NPS).

15. The method of claim 11, wherein the presodiation agent comprises Na?S.

16. The method of claim 10, wherein the mixture has a mass ratio of NPS powder to presodiation agent of from 1:1 to 1:6.

17. The method of claim 16, wherein the mass ratio is 1 :2.

18. The method of claim 11, wherein the cathode material comprises wherein the cathode material comprises one or more of NasPS4 (NPS), NaCrCh (NCO) NaFeMnNiO (NFM), and other sodium-based layered transition metal oxides (NaxMCh, where M = Ni, Co, Mn, Fe, Cr, V).

Images