Li+ Interstitial Flow in Composite State Electrolyte for Solid-State Lithium Batteries with High-Loading Cathode

Solid-state lithium metal batteries (SSLMBs) hold high energy density and are safe and reliable. However, the polymer-based solid electrolyte possesses low ionic conductivity at ambient conditions and is incompatible with the lithium anode, which seriously hinders their practical application. Solid composite electrolyte (SCE) was prepared by fixing a fast-ion conductor of Li6.4La3Zr1.4Ta0.6O12 (LLZTO) within the P (VDF-ctfe) skeleton. The “Lewis base” effect of LLZTO can be used to construct a fast ion transport layer, realize the fast coordination/decoupling of lithium ions, form a fast ion jump, and improve the density of lithium carriers. The doped SCEs hold a high ionic conductivity of 1.2×10−3 S/cm at ambient temperature and a high migration number of 0.82. In addition, SCE-3% shows intimate compatibility with the lithium anode and LiFePO4 cathode. The assembled Li//SCE-3%/Li battery can be stably tested for 1200h at 0.1 mA/cm2, and the LiFePO4//SCE-3%//Li can charge and discharge for 70 times at 0.5 C, corresponding to 70.1% capacity retention rate, showing excellent electrochemical performance. This work gives a strategy for the design philosophy of SCE with high ion conductivity in SSLMBs.


Introduction
Capacity limitations and safety threats of commonly lithium batteries hinder the development of electric vehicles and the storage of electrical energy in smart grids.SSLMBs have shown enormous potential in breaking through the above limitations, because solid-state electrolytes (SSEs) can simultaneously use lithium anode and high-voltage cathodes, thereby achieving the object of high energy density.However, inhibition of dendrite formation and growth remains a critical challenge during repeated lithium plating and stripping [1].
As an important part of solid-state batteries, SSEs mainly include inorganic ceramics and polymer electrolytes.Ceramic-based inorganic SSEs with high modulus and excellent thermal stability are considered promising candidates for preventing Li dendrite growth.However, the preparation of extremely thin ceramic membranes with sufficient mechanical stability often has the problems of cumbersome processing and high energy consumption, which is not suitable for large-scale production.More importantly, ceramic-based thin films are often hard and fragile, with rigid interfaces, and poor contact with electrodes, resulting in difficulty in ion conduction at the interface [2].In addition, Han et al. have demonstrated that lithium ions easily nucleate at grain boundaries, thereby forming lithium dendrites, including the most promising sulfide electrolytes and garnet oxide electrolytes [3].Polymer solid electrolyte has good flexibility and processability, and they can form a tight interface with cathode and anode, which is a promising SSEs.Lithium conduction in the polymer is based on the hopping and coupling of polymer bulk cations.This transfer mode is inefficient, difficult to operate at room temperature, and cannot be directly used in lithium batteries [4,5].In addition, the incompatibility of polymer solid electrolytes with lithium metal anodes, low chemical stability, and narrow electrochemical window are still key challenges [6].For instance, polyethylene oxide (PEO) is a polymer with high Li-ion conductivity, but it is limited by the problem of a narrow electrochemical window [7,8].Although polyacrylonitrile (PAN) can achieve high ionic conductivity (0.1 mS/cm) and high operating voltage (~5 V) at ambient temperature, it has the problem of cyanide groups reacting with lithium metal, passivating lithium ion conduction [9,10].Recently, the exploration of polymer-ceramic composite solid electrolytes has received extensive attention.Among them, the composite electrolyte composed of flexible polymer and rigid ceramic combines the advantages of constructing fast ion channels at the interface of bulk polymer chains and ceramic particles, which is expected to achieve high electronic insulation and high-flux lithium conduction [11,12].Polyvinylidene fluoride-chlorotrifluoroethylene P(VDF-ctfe) is a polymer material in which the fluorine atoms on the main chain are replaced by chlorine atoms.Due to the asymmetry of its molecular structure and the orientation arrangement of fluorine and chlorine atoms in the arrangement, it presents a relatively high dipole moment as a whole, about 6.4×10 -30 C•m, which can boost the dissociation of lithium salts, thereby providing a high concentration of lithium carriers [13,14].In addition, compared with other polymer materials, P(VDF-ctfe) may be a potential polymer solid electrolyte material due to its high flexibility, sensitive electromechanical response, and exceptional thermal and chemical stability.Although it has been reported that the high electronic conductivity of cubic garnet ceramics can promote Li dendrite growth, the highly dielectric P(VDF-ctfe) polymer can serve as an effective electronic barrier to protect LLZTO from electronic attack [14,15].However, it is still a challenge to prepare polymer-ceramic composite solid-state electrolytes with high voltage resistance and fast ion transport, and the related lithium ion transport mechanism is still unclear.
Herein, a P(VDF-ctfe)-based composite solid-state electrolytes for high cathode-loading SSLMBs is designed.Cubic phase garnet ceramic Li 6.4 La 3 Zr 1.4 Ta 0.6 O 12 is integrated into the P (VDF-ctfe) matrix to optimize Li-ion conductivity and improve its interfacial compatibility.In addition, a low content of succinonitrile was introduced as a plasticizer to lessen the crystallinity of P (VDF-ctfe).Using the Lewis acid-base interaction mechanism between P(VDF-ctfe) and garnet ceramics to achieve fast lithium conduction and good compatibility with lithium anode.The composite electrolyte is matched with the lithium anode and LiFePO 4 to achieve a stable cycle and the preparation of a high-energy-density battery.The mechanisms underlying the interfacial compatibility of composite electrolytes with Li metal anodes are deeply understood.

Material preparation
Preparation of ceramics: Tantalum-doped Li 6.4 La 3 Zr 1.4 Ta 0.6 O 12 (LLZTO) powder was prepared by ball milling and high-temperature sintering.Stoichiometric ratios of lithium hydroxide, tantalum oxide, lanthanum oxide, and zirconyl nitrate were planetarily ball-milled in isopropanol, dried, and annealed at 950°C.Finally, the ball-milled LLZTO powder was pressed into a green body and then annealed in air at 1150°C for 6 hours by the mother powder burying method, and finally, a high-purity cubic phase LLZTO ceramic was obtained.It is assumed that a crucible made of magnesium oxide materials is done.Preparation of composite solid electrolyte: both P(VDF-ctfe) and lithium salt (LITFSI) were dried in an oven for a night to remove moisture.We add 0.5 g of P(VDF-ctfe), 0.05 g of succinonitrile and 0.167 g of LITFSI into DMF solution, and stir at 70°C for 6 h to obtain a transparent solution.We add the nano LLZTO powder into the transparent mixture and fiercely stir to obtain a dark brown gel slurry.Then, coating the brown slurry onto a clean substrate, and transferred to a vacuum oven to dry a night.The composite film with a ceramic mass fraction of 3wt% is marked as SCE-3%.Electrolyte samples without any ceramic additions were recorded as SCE-0%.

Battery assembly
Cathode slurry was prepared in a laboratory planetary mixer, which mix the LiFePO 4 cathode powder, Super-P, and PVDF binder with a proportion of 18:1:1 inappropriate n-methylpyrrolidone (NMP).The coating of cathode slurry is performed on the automatic coating machine in a drying room.A cathode disc with a diameter of ~14 mm was cut into before being assembled into a button cell.The mass loading of the cathode material is greater than 12 mg/cm 2 .

Structure, composition, and performance characterization
The phase structure of LLZTO powder was characterized by X-ray diffraction (XRD, Bruker-D8, CuKa λ=1.5406A˚).The morphology of SCE was surveyed by field emission scanning electron microscopy (FESEM, Hitachi S-4800).X-ray photoelectron spectroscopy (XPS, Thermo Fisher K-Alpha) was conducted to analyze surface interactions.The ionic conductivity of SCE was tested on an electrochemical workstation of Shanghai Chenhua CH660e).The lithium ion migration number test is performed on an asymmetric battery Li//PCSE//SS by combining DC polarization and electrochemical impedance.The cycling stability of the CSE was tested in a Li//SCE//Li battery.The cycle performance was tested on the Wuhan CT3002A at the voltage window of 2.0 -3.8 V.

3.
Results and discussion Figure 1 gives the XRD pattern of LLZTO ceramic powders using a combination of ball milling and high temperature annealing.Diffraction data show that LLZTO has a high-purity garnet cubic phase structure (PDF#45-0109) without secondary phase formation.SEM was tested to characterize the surface morphology of SCE. Figure 2(a-d) exhibits the SEM of the front side of the SCE, and it can be seen that the surfaces of the as-prepared SCE-0% and SCE-3% electrolyte films present two distinct microstructures.In Figure 2a, the surface morphology of SCE-0% is rough and has poor planarity.This may be caused by the strong crystallinity of the composite electrolyte and the uneven volatilization of the DMF solvent.In contrast, SCE-3% showed a relatively flat and smooth surface, and the surface showed a particle aggregate-like morphology.Figure 2d shows this surface difference more clearly.This shows that LLZTO doping has a significant promotion effect on its surface reconstruction, which may be resulted from the Lewis acid-base interaction between the highly active LLZTO ceramics and the polymer bulk phase.In addition, because the smooth surface is rich in the amorphous structure of the polymer, it will enhance the interfacial contact with electrodes, reduce the interfacial resistance, and promote lithium deposition, and further suppress lithium dendrites.Figure 2e-i gives the element distribution of tantalum, zirconium, chloride and lanthanum.The four elements are uniformly dispersed in SCE-3%, which shows that the ceramics are well dispersed and no agglomeration occurs.As shown in Figure 3a, the ionic conductivity of LLZTO-doped CSE is significantly improved at ambient temperature, which is significantly higher than that of the undoped SCE-0% sample (0.26×10 −3 S/cm).Among them, SCE with a doping amount of 3% hold the highest ion conductivity of 1.2×10 -3 S cm -1 , which is beyond the level of most reported PVDF-based composite electrolytes.There may be three main reasons for the high ionic conductivity of SCE doped with a small amount of ceramics.First of all, LLZTO can be uniformly dispersed in the bulk polymer, which greatly reduces the local domain of the PVDF, and significantly improves the lithium ion conductivity.Secondly, due to the strong alkaline effect of LLZTO infiltrated into the polymer, the dehydrofluorination reaction of the PVDF molecular chain is promoted, and the molecular structure of the carbon-carbon double bond is further formed, which enhances the charge delocalization of the PVDF polymer and promotes the dissociation of lithium salt, increasing the concentration of lithium carriers [16,17].Finally, the formed PVDF-LLZTO cross-linked structure realizes the ion percolation layer, and the fast transport channels are constructed, so the lithium ion conduction level is significantly improved.The migration number of Li + is the evaluation of the migration ability of lithium ions in the system, and it is generally expressed by the ratio of the Li ion to the total number of all migrating ions in the system.In liquid lithium batteries, in addition to lithium ion transport, anions also participate in the transport, so the transport level of lithium ions is low, usually between 0.2 and 0.4.However, anion transport will cause more complex surface coupling-decoupling reactions, increasing the ion transport resistance.Therefore, it is very important to reduce the anion conduction and improve the lithium ion transport capacity [18].Figures 3b,c show the potentiostatic polarization curves and electrochemical impedance results based on Li//SCE//Li symmetric cells.The interfacial contact resistance of SCE-3% was 109.3Ω and 127.5 Ω before and after potentiostatic polarization, showing a smaller lithium metal contact resistance, indicating its good interfacial compatibility with metallic lithium.However, as shown in Figure 3c, the interfacial resistance between SCE-0% and Li anode is relatively large, which are 201.1 Ω and 208.1 Ω before and after polarization, respectively.In addition, the polarization current of SCE-3% is significantly greater than that of SCE-0%, indicating that the ion transport level of SCE-3% composite electrolyte is higher.After calculation, the Li migration number of SCE-3% is 0.82 (Figure 3d), which is significantly greater than that of the SCE-0% (0.55).This result further demonstrates that the doping of LLZTO ceramics facilitated the ion transport.

NESP-2023 Journal of Physics: Conference Series 2592 (2023) 012007
To investigate the interface compatibility between the SCE and lithium anode, the galvanostatic cycling experiments for Li ion deposition/stripping were performed.In the voltage-time curve of Figure 4a, although the Li/SCE-0%/Li battery has a stable voltage distribution in the first 30 h, the polarization voltage value is about 75.3 mV.However, the polarization voltage fluctuates greatly around 35 h, which is caused by the instability with lithium metal.This unstable interface prompts the battery to have a large interface side reaction in a short period of time, so the battery fails after dozens of hours of cycling.On the contrary, after a trace amount of LLZTO doping, the interfacial compatibility of SCE-3% with Li metal was significantly improved, and the polarization voltage of the first cycle test was significantly smaller at 21.7 mV.Li plating and stripping reached 1200 h with a low polarization voltage of 31.6 mV (Figure 4b) was further suggest it very stable interface.and SCE-3%.
To further demonstrate the stability of this composite electrolyte to metallic lithium, the surface morphology of its lithium anode after cycling at 0.1 m/cm 2 was analyzed.For Li/SCE-0%/Li in Figure 5a, b, as the testing proceeds, microcracks begin to appear on the surface of the lithium anode.After further magnifying the observation magnification, it was found that this was due to the inhomogeneous product of lithium.As the number of cycles increases, more isolated dead Li accumulates and Li dendrites grow.The battery failed after a short cycle due to the lithium dendrites penetration.This result matches the results of cycling tests on lithium.However, the Li deposition is uniformly for the Li/SCE-3%/Li cells, presenting a smooth and dense surface in Figure 5c, d, even after 1200h cycle, the Li deposition and stripping process always stable, without the existence of lithium dendrites and defects.This further confirms that a small amount of LLZTO ceramic doping can enhance the lithium affinity of the SCE.After confirming the superior compatibility with Li metal, a high-loaded (>12 mg/cm 2 ) LiFePO 4 (LFP) solid-state Li metal battery was assembled to investigate its electrochemical performance.Approximately 5 μL/cm 2 liquid electrolyte was used to reduce the interface resistance.Based on the nominal capacity of 170 mA/g of the LFP, five-stage galvanostatic tests were performed in 2.5-3.8V. Figures 6a, b display the rate performance of LFP//SCE-3%//Li and LFP//SCE-0%//Li cells, respectively.The discharge specific capacities of the LFP//SCEs-3%//Li at various rates of 0.1, 0.2, 0.5, and 1C were 159.4,148.7, 123.5, and 63.8 mAh/g, respectively.When the current is set to 0.1 C again, the specific discharge capacity can recover to a high level of 156.5 mAh/g, indicating that the battery has good reversibility.Benefiting from the fast Li-ion conduction of SCEs-3%, the SCE-3% membrane battery exhibits a better rate performance than that of SCE-0%.However, the rate performance based on SCE-0% almost lost the ability of energy storage at 1C, indicating poor Li-ion conductivity.Figures 6c, and d show the cycle performance of the cells at 0.5C.After the initial stage of activation at 0.1C, their specific capacities are very close.However, the capacity of the LFP//SCE-0%//Li battery rapidly decays after less than 40 cycles, corresponding to battery failure.On the contrary, the LFP//SCE-3%//Li battery was stably cycled 70 times and still maintained a specific discharge capacity of 101.7 mAh/g, corresponding to a capacity retention rate of 70.1%.This further verifies that the trace amount of LLZTO doping plays a major role in interfacial ion conduction.

Conclusion
By preparing lithium-ion conductive garnet LLZTO powder with cubic structure and adding it to the composite solid electrolyte of P(VDF-ctfe)-LITFSI, the effect of ceramic doping amount on lithium ion transport and Effects on interfacial stability in lithium metal batteries.Doping LLZTO ceramics can significantly enhance the ionic conductivity of the SCE.Among them, the SCE with 3% doping possesses the best ion transport performance.Lithium plating/stripping measurements confirmed that SCE has intimate contact with lithium anode and can be cycled stably for up to 1200 h at 0.1 mA/cm 2 , demonstrating excellent cycle stability.In addition, solid-state lithium batteries assembled with highly active LiFeO 4 -loaded cathodes exhibit acceptable rate capability and stable cycle performance.