High performance separator, process of preparation and use in solid state batteries

Abstract

The application relates to the technical field of solid-state battery separator membranes, and particularly discloses a high-performance separator membrane, a preparation process and application in a solid-state battery. Through the synergistic effect of molecular design of a reactive multifunctional monomer (N, N-bis-substituted acrylamide type phosphonic acid base ferrocene derivative, Fc-AAm-PA), partial lithiation acid-base balance regulation, double covalent immobilization and double-sided asymmetric composite structure, the Fc-AAm-PA is embedded in a PI porous skeleton, a polyether acrylate crosslinking network and a fast ion conductor (LLZTO) composite system, the prepared high-performance composite solid-state separator membrane simultaneously realizes the synergistic effect of multiple performances such as high ion conductivity, high migration number, a wide electrochemical window, inhibition of lithium dendrites, high-voltage positive electrode interface resistance and inhibition of redox shuttling in a single separator membrane system, and provides a key separator membrane material solution for high-energy density, long-life and high-safety lithium metal solid-state batteries.

Description

Technical Field

[0001] This invention relates to the field of solid-state battery separator technology, specifically to a high-performance separator, its preparation process, and its application in solid-state batteries. Background Technology

[0002] Lithium metal batteries are considered an important direction for the development of high-energy-density rechargeable batteries due to their high theoretical specific capacity (3860 mAh / g) and low electrochemical potential (-3.04 V relative to the standard hydrogen electrode). However, traditional liquid electrolyte systems suffer from problems such as flammability, leakage, narrow electrochemical window, severe interfacial side reactions, and lithium dendrite growth, which seriously limit their application in high-safety power batteries and large-scale energy storage batteries.

[0003] Composite solid / quasi-solid membranes (polymer-inorganic ceramic composite electrolyte membranes) can balance safety, mechanical strength and ion conductivity to a certain extent, and are currently a research hotspot. Among them: (1) Polyether polymers (such as PEO and polyether acrylate) contain ether oxygen segments, which can react with Li + Coordination occurs and promotes lithium salt dissociation, but its room temperature ionic conductivity is low and its mechanical strength is limited; (2) Inorganic ceramic fast ionic conductors (such as lithium lanthanum zirconium tantalum oxide LLZTO, Al2O3, etc.) have high intrinsic ionic conductivity, wide electrochemical window and excellent thermal stability, which can significantly improve the mechanical strength, thermal dimensional stability and interface stability of the membrane. However, unmodified ceramic particles usually have problems such as high surface energy, easy agglomeration, poor wettability with polymer matrix and poor interface compatibility, resulting in large ceramic / polymer interface impedance and continuous Li + The transmission channels are difficult to construct, and the ionic conductivity and mechanical properties of the composite system do not achieve synergistic enhancement.

[0004] Furthermore, to improve the stability of the high-voltage cathode interface, ferrocene (Fc) compounds are preferred due to their highly reversible Fc / Fc ratio at approximately 3.5V (relative to the lithium reference electrode). + Single-electron redox pairs have been attempted to be introduced into battery systems as redox buffers, overcharge protection, or interface modulation additives. However, free ferrocene molecules have a high diffusion coefficient in electrolytes or polymer matrices and readily shuttle across membranes, leading to the oxidation of Fc. +Ferrocene cells diffuse from the positive electrode to the negative electrode and are reduced back to Fc by lithium metal, shuttling back and forth. This causes problems such as continuous self-discharge, decreased coulombic efficiency, capacity decay, and the accumulation of side reactions at the SEI (Solid Electrolyte Interphase, formed at the negative electrode / electrolyte interface) and CEI (Cathode Electrolyte Interphase, formed at the positive electrode / electrolyte interface), making it difficult for them to function stably under long-term cycling conditions. Therefore, how to immobilize the ferrocene unit while retaining its redox activity is a key problem that urgently needs to be solved.

[0005] Furthermore, porous polyimide (PI) membranes are often used as the supporting framework for composite membranes due to their excellent heat resistance (≥300℃), mechanical strength, and dimensional stability. However, PI membranes have strong chemical inertness and weak polarity, resulting in poor interfacial bonding and pore wettability with the polymer functional layer. After curing, problems such as interlayer delamination and insufficient pore blockage are prone to occur, which limits the stability of the integrated support / functional layer structure.

[0006] In the existing technology, strategies such as simply adding ceramic fillers, simply adding free functional small molecules, or physically mixing simple ceramics, small molecules, and polymers are all difficult to simultaneously meet the following requirements: (1) uniform dispersion and interfacial compatibility of ceramic particles in the polymer matrix; (2) continuous and efficient Li + Transmission channel; (3) Stable immobilization and anti-shuttle of functional units such as ferrocene; (4) High thermal dimensional stability, mechanical strength and wide electrochemical stability window of the diaphragm; (5) Simultaneous suppression of dendrites on the lithium anode side and high voltage interface stability on the cathode side.

[0007] To address the aforementioned issues, those skilled in the art have attempted to improve compatibility by modifying the surface of ceramics with functional molecules containing phosphonic acid groups. However, if the phosphonic acid groups exist in the form of completely free acid, they are prone to corroding the ceramic surface; if they exist in the form of completely lithium salt, they will weaken the interfacial anchoring ability with the ceramic. Neither of these is conducive to long-term interfacial stability.

[0008] Therefore, it is necessary to provide a novel reactive multifunctional monomer and its composite solid membrane to achieve efficient anchoring on ceramic surfaces, covalent fixation of polymer networks, precise control of acid-base balance, and synergistic optimization of bifacial interfaces. Summary of the Invention

[0009] This invention designs and synthesizes a reactive multifunctional monomer (N,N-disubstituted acrylamide-type phosphonate-based ferrocene derivative, Fc-AAm-PA) that simultaneously possesses ceramic anchoring groups (phosphonic acid groups), polymer network covalent bonding groups (acrylamide double bonds), and electrochemically active groups (ferrocene units). Through partial lithiation and bifacial asymmetric structure design, Fc-AAm-PA is embedded in a composite system of a porous PI framework, a polyether acrylate crosslinking network, and a fast ion conductor (LLZTO). This achieves efficient anchoring on the ceramic surface, covalent fixation of functional units, acid-base balance regulation, and synergistic optimization of the dual interfaces. The specific details are as follows:

[0010] At the molecular design level: three functional groups—phosphonic acid groups, polymerizable acrylamide double bonds, and ferrocene—are integrated into the same molecular framework. The phosphonic acid groups are chemically anchored to the LLZTO ceramic surface via POM bonds. The acrylamide double bonds are covalently fixed by copolymerizing with polyether acrylate during UV curing to form a cross-linked network. The ferrocene units are unable to migrate freely due to this double covalent locking.

[0011] Acid-base regulation: By using a partial lithiation strategy (22 mol%), the optimal balance is achieved between the interfacial anchoring ability of phosphonic acid (retaining some free -P-OH) and the neutralization of the system pH (partial conversion to -P-OLi is required to avoid corrosion of ceramics).

[0012] Double-sided asymmetric functionalization: Utilizing a PI porous membrane as a high-temperature framework, a functional layer containing Fc-AAm-PA@LLZTO ceramic is coated on the negative electrode side (using the high modulus of the ceramic to suppress dendrites), and a pure polymer functional layer containing partially lithium-ionized Fc-AAm-PA is coated on the positive electrode side (utilizing lithium phosphonate groups and Fc / Fc...). + Redox reactions stabilize CEI, enabling differentiated adaptation of the positive and negative electrode interfaces.

[0013] Integrated process: The process of slurry coating and UV in-situ curing is adopted to copolymerize and crosslink the acrylamide double bonds of Fc-AAm-PA with the acrylate double bonds of mPEGA and PEGDA under UV initiation in the same step. The immobilization of functional units and the diaphragm molding are completed simultaneously, and the process is simple and scalable.

[0014] To achieve the above objectives, the present invention adopts the following technical solution:

[0015] A high-performance membrane includes a porous support layer and a functional composite layer coated on at least one side of the porous support layer;

[0016] The functional composite layer comprises modified ceramic filler, polyether acrylate monomers, crosslinking agent, film-forming polymer, lithium salt, and photoinitiator. The content of each component, based on the total solid content of the functional composite layer by mass, is as follows:

[0017] Modified ceramic filler 25-45 wt%;

[0018] Polyether acrylate monomers 25-40 wt%;

[0019] Crosslinking agent 3-10wt%;

[0020] Film-forming polymer 5-15 wt%;

[0021] Lithium salt 15-25 wt%;

[0022] Photoinitiator 0.5-2wt%;

[0023] The modified ceramic filler is obtained by anchoring N,N-disubstituted acrylamide-type phosphonic acid-based ferrocene derivatives onto the surface of lithium-ion conductor ceramic particles via phosphonic acid groups; the apparent loading of the derivative in the modified ceramic filler is 1.5-3.0 wt%.

[0024] In the modified ceramic filler, the acrylamide double bonds of the derivative are covalently copolymerized with the polyether acrylate crosslinking network, so that the ferrocene units are simultaneously anchored to the ceramic surface by phosphonic acid groups and connected to the polymer network by carbon-carbon covalent bonds.

[0025] Furthermore, the porous support layer is a polyimide porous membrane with a thickness of 10-15 μm, a porosity of 50-65%, and an average pore size of 0.8-1.5 μm;

[0026] The lithium-ion conductor ceramic particles are selected from at least one of garnet-type lithium lanthanum zirconium oxide ceramics, NASICON-type lithium-ion conductor ceramics, and perovskite-type lithium-ion conductor ceramics.

[0027] The polyether acrylate monomer is polyethylene glycol monomethyl ether acrylate with a number average molecular weight of 480-20000.

[0028] The crosslinking agent is polyethylene glycol diacrylate, with a number average molecular weight of 200-2000;

[0029] The film-forming polymer is a vinylidene fluoride-hexafluoropropylene copolymer, wherein the HFP content is 15-25 mol% and the number-average molecular weight Mn is 4 × 10⁻⁶. 5 -8×10 5 ;

[0030] The lithium salt is selected from at least one of lithium bis(trifluoromethanesulfonyl)imide, lithium bis(fluorosulfonyl)imide, and lithium difluorooxalateborate.

[0031] Furthermore, the functional composite layer also contains partially lithium-ionized N,N-disubstituted acrylamide-type phosphonic ferrocene derivatives as reactive comonomers.

[0032] Partial lithiation refers to the replacement of 15-35 mol% of the acidic protons in the phosphonate group of N,N-disubstituted acrylamide-type phosphonate-based ferrocene derivatives with lithium ions to form a lithium phosphonate structure.

[0033] The content of partially lithium-ionized N,N-disubstituted acrylamide-type phosphonic ferrocene derivatives is 1-5 wt% based on the total mass of solid content in the functional composite layer.

[0034] Furthermore, the diaphragm has a double-sided asymmetric structure, comprising:

[0035] A first functional layer coated on the porous support surface facing the negative electrode side, the first functional layer comprising a modified ceramic filler, a polyether acrylate crosslinking network and a lithium salt;

[0036] And a second functional layer coated on the porous support layer facing the positive electrode side, the second functional layer comprising a partially lithium-ionized N,N-disubstituted acrylamide-type phosphonic ferrocene derivative, a polyether acrylate crosslinking network and a lithium salt, and the second functional layer does not contain inorganic ceramic fillers.

[0037] A method for preparing a high-performance separator includes the following steps:

[0038] Step S1: Dissolve the film-forming polymer in an organic solvent to prepare a pre-solution;

[0039] Step S2: Under an inert atmosphere and light-protected conditions, the presol is mixed with polyether acrylate monomers, crosslinking agents, lithium salts, photoinitiators, and modified ceramic fillers, and then dispersed and degassed to obtain a polymerization precursor slurry.

[0040] Step S3: The polymerization precursor slurry is coated on at least one side of the porous support layer and pre-dried.

[0041] Step S4: In-situ UV curing is performed under an inert atmosphere to allow the unsaturated double bonds on the N,N-disubstituted acrylamide-type phosphonic acid ferrocene derivatives on the surface of polyether acrylate monomers, crosslinking agents, and modified ceramic fillers to undergo free radical copolymerization and crosslinking reactions, forming a functional composite layer.

[0042] Step S5 involves vacuum drying and roll forming to obtain the high-performance diaphragm.

[0043] Furthermore, in step S2, the dispersion treatment sequentially includes planetary stirring, ultrasonic dispersion, and vacuum degassing, and the solid content of the slurry is 25-45 wt%.

[0044] In step S3, the wet film thickness is 15-50 μm, the dew point of the coating environment is ≤-40℃, and the pre-drying temperature is 40-60℃.

[0045] In step S4, the ultraviolet light wavelength is 350-380 nm, and the light intensity is 50-150 mW / cm². 2 The irradiation time is 15-60 seconds;

[0046] In step S5, the vacuum drying temperature is 60-80℃ and the roller pressure is 0.5-2MPa.

[0047] Furthermore, when the high-performance diaphragm has a double-sided asymmetric structure, a first functional layer containing modified ceramic filler is first coated and cured on one side of the porous support layer. Then, a release film is covered on the surface of the first functional layer for protection, and the diaphragm is flipped over to coat and cure on the other side of the porous support layer to form a second functional layer containing partially lithium-ionized N,N-disubstituted acrylamide-type phosphonic acid ferrocene derivative.

[0048] Furthermore, the preparation method of the modified ceramic filler is as follows: N,N-disubstituted acrylamide-type phosphonic acid ferrocene derivative is dissolved in a mixed solvent consisting of 98-99 parts by volume of anhydrous ethanol and 1-2 parts by volume of deionized water, the pH is adjusted to 6.5-7.5, and then the surface is modified with pre-dried lithium-ion conductor ceramic particles.

[0049] Furthermore, the method for synthesizing the N,N-disubstituted acrylamide-type phosphonic acid ferrocene derivative is as follows:

[0050] Ferrocene formaldehyde and diethyl aminomethylphosphonate were first dehydrated and condensed in the presence of an acidic catalyst and a 4Å molecular sieve to generate an imine intermediate. Then, the imine was reduced with NaBH4 as a reducing agent to obtain a secondary amine intermediate containing diethyl phosphonate groups.

[0051] The secondary amine intermediate was reacted with acryloyl chloride in the presence of an acid-binding agent and a polymerization inhibitor to undergo an N-acrylation reaction, yielding an intermediate containing an acrylamide double bond and a diethyl phosphonate group;

[0052] The diethyl phosphonate group was deprotected with trimethylbromosilane using McKenna, and the target product containing free phosphonate groups was obtained by hydrolysis.

[0053] Application of a high-performance separator in a solid-state lithium metal battery, the battery comprising a positive electrode, a negative electrode, and the high-performance separator disposed between the positive electrode and the negative electrode;

[0054] The active material of the positive electrode is selected from at least one of lithium nickel cobalt manganese oxide, lithium nickel cobalt aluminum oxide, lithium iron phosphate, and lithium-rich manganese-based positive electrode materials;

[0055] The negative electrode is a lithium metal negative electrode or a lithium-containing alloy negative electrode;

[0056] When the high-performance diaphragm has a double-sided asymmetric structure, the first functional layer is oriented towards the negative electrode, and the second functional layer is oriented towards the positive electrode.

[0057] Compared with the prior art, the present invention has the following significant advantages:

[0058] (i) Significantly improved ion conductivity: Thanks to the improved compatibility of the Fc-AAm-PA phosphonic acid group with the ceramic-polymer interface and the continuous Li⁺ transport channels synergistically constructed with the polyether segments, the composite solid membrane described in the example achieves an ion conductivity of 1.6 × 10⁻⁶ at 25 °C. -4 ~1.9×10 -4 S·cm -1 Compared with the unmodified LLZTO control system (8.5×10⁻⁶), -5 S·cm -1 The efficiency was increased by approximately 88% to 124%, and the lithium-ion mobility number increased from 0.40 to 0.52 to 0.58.

[0059] (II) Ferrocene shuttle migration is effectively suppressed: Ferrocene units are doubly locked through chemical anchoring of the ceramic surface by phosphonic acid groups and covalent fixation of the polymer network by acrylamide double bonds. In Example 1, the open-circuit voltage decay after 72 hours was only 32-42 mV, which is within the normal self-discharge level of lithium metal batteries; while the comparative example containing free ferrocene molecules showed a decay as high as 118 mV, indicating that the covalent anchoring strategy effectively suppresses Fc / Fc... + Inhibition of redox shuttle.

[0060] (III) Widening of the electrochemical stability window: The irreversible oxidative decomposition initiation potential in the examples reached 4.7~4.8V (relative to the lithium reference electrode Li / Li). + It can be stably matched with high-voltage cathode materials such as NCM811 (4.3V charging cutoff).

[0061] (iv) Significantly improved long-term cycling stability: After 300 cycles at 0.5C / 0.5C for the NCM811 / Li full cell, the capacity retention rate of the example cell reached 85.8%~88.3%, and the average coulombic efficiency was not less than 99.95%; the Li / Li symmetric cell maintained a capacity of 0.5mA / cm². 2 0.5mAh / cm 2 Under certain conditions, it can be stably cycled for ≥1000 hours; after 300 cycles, the interface impedance growth rate is only 31.8%~52.0%, which is much lower than the 100.5%~263.8% of the comparative example.

[0062] (v) Balancing mechanical and thermal stability: The composite membrane has a tensile strength of 58 MPa, a heat shrinkage rate of as low as 1.8% at 200℃ / 0.5h, and can still release 152 mAh / g at a high rate of 2C (77.9% of the 0.1C capacity).

[0063] In summary, this invention, through the molecular design of a reactive multifunctional monomer (N,N-disubstituted acrylamide-type phosphonic ferrocene derivative, Fc-AAm-PA), the acid-base balance regulation of partial lithiation, dual covalent immobilization, and the synergistic effect of a bifacial asymmetric composite structure, produces a high-performance composite solid-state separator that simultaneously achieves synergistic enhancement of multiple properties in a single separator system, including high ionic conductivity, high mobility number, wide electrochemical window, suppression of lithium dendrites, high-voltage cathode interface resistance, and suppression of redox shuttle. This provides a key separator material solution for high energy density, long lifespan, and high safety lithium metal solid-state batteries. Detailed Implementation

[0064] Experimental Example 1:

[0065] Synthetic method of N,N-disubstituted acrylamide-type phosphonate ferrocene derivatives (Fc-AAm-PA)

[0066] (1) Preparation of intermediate I

[0067] Under nitrogen protection and in the dark, ferrocene formaldehyde (4.28 g, 20 mmol), diethyl aminomethylphosphonate (3.34 g, 20 mmol), anhydrous methanol (80 mL), and glacial acetic acid (0.12 g, 2 mmol, 10 mol%) were added to a 250 mL three-necked flask. Activated 4 Å molecular sieve (5 g) was added to promote the dehydration condensation reaction. The reaction was stirred at room temperature for 4 h to allow the aldehyde and amine to fully condense and form an imine intermediate.

[0068] After the reaction was complete, the system was cooled to 0°C, filtered to remove the molecular sieve, and the filter cake was washed with a small amount of anhydrous methanol (10 mL). Then, NaBH4 (1.13 g, 30 mmol, 1.5 eq, with an integral molecular weight of NaBH4 of 1.5 eq and a relative imine hydrogen negative equivalent of approximately 6 eq) was added in portions at 0°C. The addition rate was controlled to avoid violent exothermic reactions and a large instantaneous release of hydrogen. The reaction was maintained at this temperature for 1 h, then raised to room temperature and stirred for another 2 h to completely reduce the imine to the corresponding secondary amine.

[0069] After the reaction was completed, the solution was quenched with saturated NH4Cl aqueous solution, extracted with dichloromethane (3 × 50 mL), and the organic phases were combined, dried over Na2SO4, filtered, and concentrated under reduced pressure. The crude product was purified by silica gel column chromatography with an elution system of CH2Cl2 / MeOH / Et3N = 95:5:0.5 → 90:10:1 (volume gradient elution). Triethylamine (Et3N) was added to inhibit the adsorption and tailing of amine compounds on the silica gel. The target fractions were combined, concentrated under reduced pressure at 35 °C, and then dragged under high vacuum (≤10 Pa) at 30 °C for 2 h to thoroughly remove residual Et3N and a small amount of MeOH, avoiding interference from residual Et3N on the equivalence of the subsequent acryloyl chloride reaction.

[0070] An orange-yellow oily substance, namely intermediate I, was obtained with a yield of 81.5%.

[0071] The chemical structural formula of intermediate I is: .

[0072] (2) Preparation of intermediate II

[0073] Under nitrogen protection and light-protected conditions, intermediate I (3.65 g, 10 mmol) and triethylamine (2.02 g, 20 mmol, 2.0 eq) were dissolved in anhydrous dichloromethane (60 mL).

[0074] p-hydroxyanisole (MEHQ, approximately 0.05 wt%, based on the target product) was added to the system as a free radical polymerization inhibitor to suppress the self-polymerization of subsequently generated acryloyl groups during the reaction and post-treatment. The system was then cooled to 0°C.

[0075] Acryloyl chloride (1.36 g, 15 mmol, 1.5 eq) was pre-dissolved in anhydrous dichloromethane (10 mL) and added dropwise to the reaction system over approximately 30 min using a constant pressure dropping funnel, while controlling the reaction temperature to not exceed 0 °C.

[0076] After the addition was completed, the reaction was continued at 0°C for 1 hour to reduce the probability of side reactions. Then the temperature was raised to room temperature and the reaction was stirred for 8 hours under dark conditions.

[0077] After the reaction was completed, the organic phase was washed successively with 5wt% NaHCO3 aqueous solution (2×30mL) and saturated brine (30mL) to remove acidic byproducts and salts. The mixture was then dried over anhydrous Na2SO4, filtered, and concentrated under reduced pressure at low temperature (≤30℃).

[0078] The crude product was purified by rapid silica gel column chromatography under light-protected conditions, with an elution system of ethyl acetate / petroleum ether = 1:2 → 1:1 (volume ratio).

[0079] An orange viscous liquid, namely intermediate II, was obtained with a yield of 77%.

[0080] The chemical structural formula of intermediate II is: .

[0081] (3) Preparation of Fc-AAm-PA

[0082] Under nitrogen protection and light-protected conditions, intermediate II (4.19 g, 10 mmol) was dissolved in anhydrous dichloromethane (40 mL, moisture ≤ 20 ppm) dried through a 4 Å molecular sieve, and MEHQ (approximately 0.05 wt%) was added to inhibit the free radical polymerization of double bonds in the system.

[0083] All reactor vessels were pre-dried. The system was cooled to 0°C, and freshly distilled trimethylbromosilane (TMSBr, 9.18 g, 60 mmol, 6.0 eq relative to the substrate, i.e., 6.0 eq per phosphonate diethyl ester group PO(OEt)2) was slowly added using a dry syringe. According to the McKenna reaction stoichiometry, 1 eq of TMSBr is required per P–OEt bond, and theoretically 2 eq of TMSBr is required to completely remove one PO(OEt)2 group from two esters; this study used approximately 3 times the excess (6.0 eq) to offset the loss of TMSBr due to volatilization, the consumption of trace amounts of water in the system, and to promote the complete conversion of the monoester intermediate, ensuring complete deprotection.

[0084] After stirring at 0°C for 30 minutes, the mixture was brought to room temperature and stirred for another 12 hours.

[0085] The reaction progress was monitored using TLC (evolving solvent: iPrOH / H2O / NH3·H2O = 7:2:1). When the presence of a monoester intermediate was detected, TMSBr (2-4 eq) was added to the original system, and the reaction was continued for 4-6 hours until the starting material was basically completely converted.

[0086] After the reaction was completed, the solvent and excess TMSBr were removed under low temperature and reduced pressure. Under ice bath conditions, a methanol / water mixed solvent (50 mL, volume ratio 9:1) was slowly added to the resulting residue, and hydrolysis was carried out for 2 h to completely convert the silicon ester intermediate into free phosphonic acid.

[0087] After removing methanol under reduced pressure, the resulting aqueous phase was slowly added dropwise to an excess of cold acetone / ethyl ether mixed solvent (volume ratio 5:5) to induce precipitation. The solid was collected by filtration and washed sequentially with cold acetone and ethyl ether to remove low-molecular-weight byproducts, residual silanols, and polymerization inhibitors.

[0088] The product was dried under vacuum to constant weight to obtain an orange-yellow solid Fc-AAm-PA, with a yield of 82.5%.

[0089] The chemical structural formula of Fc-AAm-PA is: .

[0090] 1 HNMR (DMSO-d6, 400MHz, δ / ppm): 10.12 (br s, 2H, P-OH), 6.45-6.31 (m, 1H, =CH-), 6.16-6.02 (m, 1H, =CH2), 5.77-5.63 (m, 1H, =CH2 ), 4.37-4.26 (m, 2H, Fc-CH2-N), 4.19-4.01 (m, 9H, Cp-H), 3.67-3.56 (m, 2H, N-CH2-P).

[0091] Elemental analysis (C 15 H 18 FeNO4P, M=363.13 g / mol): Theoretical calculated values: C 49.61%, H 4.99%, N 3.86%; Actual values: C 49.38%, H 5.12%, N 3.79%, within the allowable error range (±0.4%).

[0092] Experimental Example 2:

[0093] Preparation of N,N-disubstituted acrylamide-type phosphonate-based ferrocene derivative modified lithium lanthanum zirconium tantalum oxide ceramic filler (denoted as Fc-AAm-PA@LLZTO)

[0094] Weigh out lithium lanthanum zirconium tantalum oxide (LLZTO, with the chemical composition Li). 6.5 La3Zr 1.5 Ta 0.5 O 12 10.0 g of ceramic powder with an average particle size D50 of 300±50 nm was placed in a vacuum dryer at 120 °C for 12 h and then transferred to a nitrogen-protected environment for later use.

[0095] Dissolve 0.30 g of Fc-AAm-PA in 100 mL of a mixed solvent (anhydrous ethanol / deionized water = 99:1, V / V), and add MEHQ as a polymerization inhibitor (0.05 wt% of the mass of Fc-AAm-PA). Adjust the apparent pH of the system to 6.5-7.5 using a trace amount of ammonia or dilute LiOH aqueous solution, avoiding treatment under strong acid or strong alkaline conditions.

[0096] Pre-dried LLZTO powder was added to the above solution and ultrasonically dispersed for 15 min. Then, it was stirred for 2 h at room temperature to 35 °C under nitrogen protection. After treatment, the solid was collected by centrifugation and washed three times with anhydrous ethanol to remove unbound Fc-AAm-PA. The resulting solid was vacuum dried at 60 °C for 12 h to obtain Fc-AAm-PA@LLZTO.

[0097] The apparent loading of Fc-AAm-PA in Fc-AAm-PA@LLZTO was estimated using the TGA difference method. The thermogravimetric behavior of unmodified LLZTO and Fc-AAm-PA@LLZTO was tested under the same conditions, and the difference in weight loss within the range of 150-800℃ was used as an approximation of the organic layer content. The calculation formula is as follows:

[0098] G=W1-W0

[0099] Wherein, G is the apparent loading of Fc-AAm-PA based on the total mass of Fc-AAm-PA@LLZTO, W1 is the weight loss rate of Fc-AAm-PA@LLZTO in the range of 150-800℃, and W0 is the weight loss rate of unmodified LLZTO in the same temperature range.

[0100] It should be noted that this method yields the apparent loading, not the absolute loading of Fc-AAm-PA, and may be affected by the following factors: (a) residual solvent from physical adsorption, moisture, and other small molecules; (b) decomposition of carbonate impurities such as Li2CO3 formed by the dehydration of hydroxyl groups on the LLZTO surface and adsorption of CO2 in this temperature range; (c) during the thermal decomposition of Fc-AAm-PA, elements such as Fe and P are not completely volatilized and remain in the substrate in the form of Fe2O3, phosphate, etc. The subtraction method does not deduct the decomposition residue of the organic part. Therefore, the TGA subtraction value is the apparent loading of this system, used for cross-sectional comparison of the change in the surface organic layer content of LLZTO before and after modification.

[0101] In this batch of experiments, the weight loss rate of unmodified LLZTO was 0.62wt% in the range of 150-800℃, and the weight loss rate of Fc-AAm-PA@LLZTO in the same temperature range was 2.82wt%. Therefore, the apparent loading of Fc-AAm-PA in Fc-AAm-PA@LLZTO was calculated to be 2.20wt%.

[0102] Example 1:

[0103] Preparation of Fc-AAm-PA@LLZTO / polyether acrylate / PI composite solid membrane

[0104] 1.1 Raw Materials and Specifications

[0105] (1) Fc-AAm-PA@LLZTO ceramic packing

[0106] The N,N-disubstituted acrylamide-type phosphonic acid ferrocene derivative modified lithium lanthanum zirconium tantalum oxide ceramic filler (Fc-AAm-PA@LLZTO) was prepared using Experimental Example 2.

[0107] (2) Polyether acrylate monomers and crosslinking agents

[0108] mPEGA (monofunctional reactive monomer): polyethylene glycol monomethyl ether acrylate (also known as methoxy polyethylene glycol acrylate, CAS No. 32171-39-4), with the structural formula CH2=CH-C(O)O-(CH2CH2O). n -CH3, one end is methoxy-terminated and the other end is a polymerizable acrylate group, Mn≈2000;

[0109] PEGDA (difunctional crosslinking agent): polyethylene glycol diacrylate (CAS No. 26570-48-9), structural formula CH2=CH-C(O)O-(CH2CH2O). n -C(O)-CH=CH2, with polymerizable acrylate groups at both ends, Mn≈600;

[0110] The monomers mentioned above are dried with 4Å molecular sieves before use, with a moisture content of ≤50ppm. MEHQ (approximately 100-200ppm, based on monomer mass) is added before use to prevent self-polymerization during storage and batching.

[0111] (3) Film-forming polymers

[0112] The film-forming polymer is PVDF-HFP (vinylidene fluoride-hexafluoropropylene copolymer), wherein the HFP content is 20 mol% and the number average molecular weight Mn ≈ 5 × 10⁻⁶. 5 .

[0113] (4) Lithium salts

[0114] The lithium salt is LiTFSI (lithium bis(trifluoromethanesulfonylimide)), with a purity ≥99.9% and a moisture content ≤50ppm. It is vacuum dried at 80℃ for 12h before use.

[0115] (5) Photoinitiator

[0116] Photoinitiator TPO (CAS No. 75980-60-8, 2,4,6-trimethylbenzoyl diphenylphosphine oxide), purity ≥98%.

[0117] (6) Solvent

[0118] The solvent is N-methylpyrrolidone (NMP), the moisture content is ≤100ppm, and it is dried by a 4Å molecular sieve before use.

[0119] (7) Porous support layer

[0120] The porous support layer is a polyimide (PI) porous membrane, and its parameters are as follows:

[0121] Thickness: 12μm;

[0122] Porosity: 55%;

[0123] Average pore size: 1.0 μm;

[0124] Heat resistance temperature: ≥300℃.

[0125] 1.2 Formulation

[0126] Based on the total mass of solids in the functional composite layer (excluding solvents and support layers), the specific components include:

[0127] Table 1 Formulation of Fc-AAm-PA@LLZTO / polyether acrylate / PI composite solid membrane

[0128] Fc-AAm-PA@LLZTO 35 Interface regulation, ion conduction assistance mPEGA 30 <![CDATA[polyether segments, which promote Li + conduction]]> PEGDA 6 Crosslinking agent PVDF-HFP 8 Film formation and flexibility LiTFSI 20 Lithium-ion source TPO 1 Photoinitiator total 100 /

[0129] The Fc-AAm-PA content in the functional composite layer is 35wt%×2.20wt%≈0.77wt%, which is sufficient to provide interface anchoring and ferrocene function, while avoiding the risk of migration of free small molecules.

[0130] 1.3 Preparation steps

[0131] Step 1: Preparation of PVDF-HFP pre-solution

[0132] Add PVDF-HFP to anhydrous NMP and stir at 60°C for 4 hours until completely dissolved to prepare an 8wt% PVDF-HFP / NMP solution. When using, measure according to the mass of PVDF-HFP solid.

[0133] Step 2: Preparation of polymerization precursor slurry

[0134] Under nitrogen protection and yellow light conditions, PVDF-HFP / NMP solution, mPEGA and PEGDA, LiTFSI (stirred until completely dissolved), photoinitiator TPO, and Fc-AAm-PA@LLZTO were added sequentially, followed by dispersion treatments: planetary stirring (800 rpm revolution, 1200 rpm rotation, 30 min), ultrasonic dispersion (300 W, 40 kHz, ice bath, 20 min), and vacuum degassing (≤−0.09 MPa, 15 min). By adjusting the NMP dosage, the slurry solid content was adjusted to approximately 35 wt%, and the viscosity was controlled within the coating range of 300-1000 mPa·s (25 °C), preferably 500 mPa·s (25 °C), to obtain a uniform and stable suspension slurry.

[0135] Step 3: Apply with a scraper

[0136] The polymerization precursor slurry obtained in step 2 was coated onto one side of the PI porous membrane using an automatic doctor blade coater. The coating process was carried out under yellow light and low humidity conditions, with an ambient dew point ≤−40℃. The coating parameters were: wet film thickness 40μm (excluding the thickness of the PI porous support membrane), and coating speed 2m / min.

[0137] After coating, the wet film is pre-dried (the pre-drying temperature and time can be adjusted appropriately according to the solvent evaporation rate, preferably 50℃ for 2-3 minutes) to remove some NMP solvent, improve coating uniformity and dimensional stability, and avoid film flow or local accumulation during subsequent UV curing.

[0138] Step 4: UV in-situ curing

[0139] The pre-dried composite film was placed in a nitrogen atmosphere and in-situ photocured using 365nm ultraviolet light. The curing conditions were: light intensity 100mW / cm². 2 Irradiation time: 30 seconds.

[0140] Under the action of photoinitiator TPO, the acrylate double bonds in mPEGA and PEGDA and the acrylamide double bonds of Fc-AAm-PA on the Fc-AAm-PA@LLZTO surface undergo free radical copolymerization and crosslinking reaction to form a polyether acrylate crosslinking network that covalently anchors the Fc-AAm-PA unit.

[0141] Characterization of UV-cured double bond conversion rate:

[0142] Considering that the functional composite layer contains 35 wt% opaque ceramic filler (Fc-AAm-PA@LLZTO), and the coating still has a certain thickness (approximately 15-18 μm) after pre-drying, the penetration depth of UV light within the coating may be affected by scattering and absorption. To verify the above UV curing conditions (365 nm, 100 mW / cm²), 2 30s, total energy dose 3J / cm 2 To determine whether sufficient cross-linking and curing were achieved, attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) was used to characterize the double bond conversion rate of the functional composite layer before and after curing.

[0143] Test method: The functional composite layer film before UV curing (after pre-drying) and after UV curing was tested using an ATR-FTIR spectrometer (Ge crystal, 4cm resolution). -1 The scan was performed 64 times, from both the upper surface (directly exposed to UV radiation) and the lower surface (close to the PI film) of the functional layer. The scan was conducted at a depth of 1410 cm⁻¹. -1 The in-plane bending vibration peak of the C=C double bond in the acrylate (δ=CH2) was taken as the reaction peak, with a wavelength of 1720 cm⁻¹. -1 The carbonyl C=O stretching vibration peak (which does not participate in the polymerization reaction and whose intensity does not change with the degree of curing) is used as an internal standard peak. The double bond conversion rate is calculated using the following formula:

[0144] α=[1−(A 1410 / A 1720 After curing / (A) 1410 / A 1720 )_before curing]×100%

[0145] Among them, A 1410 and A 1720 1410cm respectively -1 and 1720cm -1The area of ​​the absorption peak at that point.

[0146] In addition, at 810cm -1 An out-of-plane bending vibration peak (ω=CH2) of acrylate can also be observed at 1410 cm⁻¹. This peak weakens significantly and eventually disappears after curing. -1 The calculation results at each location show a consistent trend and can be used as an auxiliary criterion.

[0147] Table 2 Double bond transformation of functional composite layers under different UV curing conditions

[0148] <![CDATA[Example 1 (100mW / cm 2 , 30s, 3J / cm 2 )]]> 94.2 88.7 Meets curing requirements <![CDATA[Example 1 Control Group A (100mW / cm 2 , 15s, 1.5J / cm 2 )]]> 91.8 78.3 Insufficient curing of the lower surface <![CDATA[Example 1 Control Group B (100mW / cm 2 , 60s, 6J / cm 2 )]]> 95.1 92.4 The conversion rate is slightly higher, but it may cause yellowing due to excessive irradiation. <![CDATA[Example 1 Control Group C (50mW / cm 2 , 30 s, 1.5 J / cm 2 )]]> 90.5 76.9 Insufficient energy density <![CDATA[Example 3 - Negative electrode side first functional layer (100mW / cm 2 , 30s)]]> 94.8 90.1 The wet film has a thickness of 25μm, resulting in better penetration. <![CDATA[Example 3 - Positive electrode side second functional layer (100mW / cm 2 , 30s)]]>< 96.5 94.8 Free of ceramic fillers, allowing unobstructed UV penetration.

[0149] Results analysis:

[0150] (1) Curing conditions of Example 1 (100mW / cm) 2 The double bond conversion rate reached 94.2% on the upper surface and 88.7% on the lower surface (PI side) after 30s, with a conversion rate difference of 5.5 percentage points. This difference stems from the light scattering effect of the ceramic filler, which causes UV light to attenuate in the depth direction. However, the 88.7% conversion rate on the lower surface is higher than the generally accepted sufficient curing threshold (≥85%) in the literature, which is sufficient to form a structurally complete cross-linked network. 810cm -1 The out-of-plane bending vibration peak at the =CH2 point basically disappeared on both sides after curing.

[0151] (2) When the UV energy dose drops to 1.5 J / cm 2 In the control groups A and C, the conversion rate of the lower surface dropped to 76.9-78.3%, which poses a risk of insufficient curing and may cause creep or swelling of the functional layer during long-term use.

[0152] (3) When the UV energy dose increases to 6 J / cm 2 In control group B, the conversion rate only increased slightly (from 88.7% to 92.4% on the lower surface), but the coating showed slight yellowing, and excessive irradiation may lead to free radical degradation of the formed polymer network. Taking all factors into account, 3J / cm²... 2 (100mW / cm) 2 The suitable curing conditions for this system are (×30s).

[0153] (4) The second functional layer on the positive electrode side of Example 3 does not contain inorganic ceramic fillers. Its UV penetration is significantly better than that of the negative electrode side containing ceramics. The conversion rate of both the upper and lower surfaces reaches more than 94.8%, and the difference between the upper and lower surfaces is only 1.7 percentage points. The curing uniformity is the best.

[0154] (5) The thickness of the wet film of the first functional layer on the negative electrode side of Example 3 is 25 μm (about 10-12 μm after pre-drying), which is thinner than the 40 μm wet film of Example 1 (about 15-18 μm after pre-drying). The UV light penetration depth is more sufficient, and the lower surface conversion rate is 90.1%, which is better than Example 1 (88.7%).

[0155] In summary, under the preferred UV curing conditions of this invention (365nm, 100mW / cm²), 2 The double bond conversion rate of the 40μm wet film system containing 35wt% ceramic filler was in the range of 88.7-94.2%, which met the requirements for full cross-linking and curing.

[0156] Step 5: Vacuum drying and roller pressing

[0157] The UV-cured composite film was placed in a vacuum oven and dried at 70°C until its mass was constant. The residual NMP content was ≤1000ppm based on the total mass of the composite film.

[0158] After drying, the composite membrane is rolled and shaped at a pressure of 1 MPa and a temperature of 35°C to improve the density, thickness uniformity, and bonding stability of the functional composite layer with the PI porous membrane.

[0159] After the above treatment, a dense and uniform Fc-AAm-PA@LLZTO / polyether acrylate / PI composite solid membrane is obtained.

[0160] Example 2:

[0161] Preparation of Fc-AAm-PA@LLZTO / polyether acrylate / PI composite solid membrane containing partially lithium-ionized Fc-AAm-PA units

[0162] This embodiment illustrates that Fc-AAm-PA can not only be used as a ceramic surface modifier, but also as a reactive comonomer to further participate in the construction of polymer networks.

[0163] 2.1 Formula

[0164] Based on the total mass of solids in the functional composite layer (excluding solvents and support layers), the specific components include:

[0165] Table 3 Formulations of Fc-AAm-PA@LLZTO / polyether acrylate / PI composite solid separators containing partially lithium-ionized Fc-AAm-PA units

[0166] Fc-AAm-PA@LLZTO 32 Interface regulation, ion conduction assistance Partially lithium-ionized Fc-AAm-PA 3 Reactive functional monomers and provide ferrocene / phosphonic acid units mPEGA 30 reactive polyether monomers PEGDA 6 Crosslinking agent PVDF-HFP 8 Film-forming polymers LiTFSI 20 lithium salts TPO 1 Photoinitiator total 100 /

[0167] Partially lithium-ionized Fc-AAm-PA was added based on the equivalent mass of unlithiated Fc-AAm-PA. The total Fc-AAm-PA structural unit content in the functional composite layer is approximately 3.70 wt%, including approximately 0.70 wt% (32 wt% × 2.20 wt%) of apparent loading from Fc-AAm-PA@LLZTO and an additional 3 wt% of partially lithium-ionized Fc-AAm-PA.

[0168] The selection criteria for partial lithiation degree:

[0169] The phosphonic acid groups of Fc-AAm-PA are neutralized by LiOH at a rate of 22.0 mol% to form a lithium phosphonate structure (22 mol% refers to 22% of the total molar amount of acidic protons, with 22% of the total acidic protons being neutralized). This ratio was determined through system optimization, mainly based on the following considerations:

[0170] Acid-base balance and material compatibility:

[0171] Each phosphonate group in the Fc-AAm-PA molecule contains two ionizable acidic protons, for a total of two acidic protons per molecule. Complete lithiation (100 mol%) replaces both acidic protons with Li⁺, resulting in a strongly alkaline molecule that may weaken the interfacial anchoring effect between the phosphonate group and the ceramic filler surface. Complete non-lithiation, on the other hand, leaves the phosphonate group with strong acidity, potentially leading to slight proton exchange or corrosion with the LLZTO surface, affecting long-term interfacial stability. Partial lithiation (22.0 mol%) means that, based on the total molar amount of acidic protons in the Fc-AAm-PA molecule, 22.0 mol% is neutralized by LiOH (i.e., approximately 2 × 0.22 = 0.44 acidic protons per molecule are replaced by Li⁺). This allows the pH of the corresponding solution system to be adjusted to near neutral (matching the pH range of 6.5-7.5 in the ceramic modification step), thus achieving a good trade-off between acid-base balance and material compatibility.

[0172] Solubility and dispersibility optimization:

[0173] This lithiation ratio significantly improves the solubility of Fc-AAm-PA in polar solvents such as NMP and ethanol, and avoids the aggregation of unlithiated products due to hydrogen bonding.

[0174] Electrochemical performance trade-offs:

[0175] Comparative half-cell tests with different lithiation levels revealed that the 22 mol% lithiation sample exhibited the highest room-temperature ionic conductivity and the lowest interfacial impedance. Excessive lithiation (<15 mol%) led to acidic residues and increased interfacial side reactions; excessive lithiation (>35 mol%) slightly reduced mechanical strength and tolerance to high cathode voltages due to excessive free Li⁺. This ratio achieved the optimal trade-off between ionic conductivity, interfacial stability, and mechanical properties. These optimizations were derived through orthogonal experiments combined with electrochemical impedance spectroscopy (EIS) and linear sweep voltammetry (LSV) measurements.

[0176] Among them, the electrochemical performance comparison data for different lithiation ratios are as follows:

[0177] To optimize the lithiation ratio, Fc-AAm-PA samples with lithiation degrees of 0 mol% (unlithiated), 10 mol%, 15 mol%, 20 mol%, 22 mol%, 25 mol%, 35 mol%, 40 mol%, and 100 mol% (fully lithiated) were prepared. Composite membranes were then prepared using the same formulation and process as in Example 2 (3 wt% partially lithiated Fc-AAm-PA and 32 wt% Fc-AAm-PA@LLZTO, with the remaining components unchanged) for comparative testing. The testing methods were the same as those used in the performance testing section for Test 1 (EIS ionic conductivity), Test 2 (lithium-ion transference number), and Test 3 (LSV electrochemical stability window).

[0178] Table 4. Effect of different lithiation ratios on the electrochemical performance of the composite membrane

[0179] 0 (Unlithiated) <![CDATA[1.2×10 -4 ]]> 0.46 4.5 Moderate, with slight aggregation 10 <![CDATA[1.4×10 -4 ]]> 0.49 4.6 improve 15 <![CDATA[1.6×10 -4 ]]> 0.53 4.7 good 20 <![CDATA[1.8×10 -4 ]]> 0.56 4.8 good 22 <![CDATA[1.9×10 -4 ]]> 0.58 4.8 good 25 <![CDATA[1.8×10 -4 ]]> 0.57 4.8 good 35 <![CDATA[1.5×10 -4 ]]> 0.52 4.7 good 40 <![CDATA[1.3×10 -4 ]]> 0.48 4.6 high 100 (fully lithium-ionized) <![CDATA[9.5×10 -5 ]]> 0.42 4.4 high

[0180] The trend analysis of the above data is as follows:

[0181] (1) Ionic conductivity and transport number: As the lithiation ratio increased from 0 (0 mol% sample σ was lower than in Example 1, attributed to the combined effects of the aggregation of unlithiated Fc-AAm-PA and the slight decrease in LLZTO content (32 wt% vs 35 wt%)) to 22 mol%, the ionic conductivity increased monotonically, reaching a maximum of 1.9 × 10⁻⁶ at 22 mol%. -4 S·cm -1 The migration number reached 0.58. This is attributed to the lithium phosphonate group (-P-OLi) acting as an anion anchoring site, which effectively binds TFSI. - Anion migration, while providing additional Li +Dissociation and hopping transport sites. When the lithiation ratio exceeds 22 mol%, the conductivity and transport number begin to decrease. This is because: excessive lithium phosphonate salt introduces too many ion pairs, increasing the probability of ion association; the anchoring ability of phosphonate groups decreases, leading to an increase in the ceramic-polymer interface impedance; and the flexibility of highly ionized Fc-AAm-PA molecular chain segments decreases, which is not conducive to Li⁺ transport assisted by chain segment movement.

[0182] (2) Electrochemical stability window: In the range of 15-25 mol%, the irreversible oxidation decomposition initiation potential is stable at 4.7-4.8 V. In low-lithiation (<15 mol%) samples, the residual free phosphonate protons are prone to deprotonation oxidation side reactions at high potentials, resulting in a lower decomposition potential (4.5-4.6 V). In high-lithiation (>35 mol%) samples, the strong ionic characteristics of the lithium phosphonate group make it easy for it to undergo irreversible interfacial reactions with the cathode material at high potentials, and the decomposition potential also decreases.

[0183] (3) Solubility and processability: Unlithiated Fc-AAm-PA exhibits slight aggregation in NMP (attributed to strong hydrogen bonding association of intermolecular phosphonic acid groups), and its solubility is significantly improved after lithiation of 10 mol% or more.

[0184] (4) Although the difference between 22 mol% and 20 mol% and 25 mol% is small (on the edge of the experimental error range), in multiple independent repeated experiments (n=3 per group), the 22 mol% sample showed the highest median conductivity and migration number. Considering that the pH of the solution system corresponding to this ratio is about 6.8 (matching the pH range of 6.5-7.5 in the ceramic modification step), 22 mol% was finally selected as the preferred lithiation ratio.

[0185] It should be noted that the above optimization results are system-specific, meaning the optimal lithiation ratio is related to factors such as the specific type of ceramic filler, polymer matrix composition, and lithium salt concentration. In different formulation systems, the optimal lithiation ratio may fluctuate within the range of 15-30 mol%.

[0186] Coordination and matching with ceramic loading:

[0187] The 22 mol% lithiation-added free functional monomers can precisely control the overall phosphonic acid / amide coordination site density, forming a synergistic Li group with Fc-AAm-PA@LLZTO anchored on the ceramic surface. + Transmission channel.

[0188] Partial lithiation method: Fc-AAm-PA was dissolved in a methanol / water mixed solvent, and LiOH aqueous solution was added at 22.0 mol% of the total molar amount of phosphonic acid protons in the Fc-AAm-PA molecule. The mixture was stirred at room temperature for 2 h (under light-protected conditions to prevent acrylamide double bond polymerization). Subsequently, the solvent was removed under reduced pressure and the mixture was dried under vacuum to obtain partially lithiated Fc-AAm-PA.

[0189] 2.2 Preparation steps

[0190] Except for the addition of 3wt% of partially lithium-ionized Fc-AAm-PA during the slurry preparation process, the other steps (PVDF-HFP presol solution preparation, polymerization precursor slurry preparation, doctor blade coating, UV in-situ curing, vacuum drying and roll forming) are the same as in Example 1.

[0191] After the above treatment, a dense and uniform composite solid membrane containing partially lithium-ionized Fc-AAm-PA units was obtained. This design further improves the ion transport performance and interfacial stability of the membrane by covalently fixing additional functional units.

[0192] Example 3:

[0193] Preparation of double-sided asymmetric Fc-AAm-PA modified polyether acrylate / PI composite solid membrane

[0194] This embodiment illustrates a double-sided asymmetric composite separator designed to meet the interface requirements of the lithium anode side and the high-voltage cathode side, respectively.

[0195] 3.1 Diaphragm Structure

[0196] A first functional layer facing the negative electrode is formed on one side of the PI porous membrane. The first functional layer contains Fc-AAm-PA@LLZTO, a polyether acrylate crosslinking network, and a lithium salt to promote uniform Li⁺ transport and suppress lithium dendrites.

[0197] A second functional layer facing the positive electrode is formed on the other side of the PI porous membrane. The second functional layer contains partially lithium-ionized Fc-AAm-PA, a polyether acrylate crosslinking network, and lithium salt, which is used to stabilize the high-voltage positive electrode side CEI (positive electrode interface membrane) and improve the high-voltage positive electrode interface stability and membrane thermal stability.

[0198] 3.2 Formulation of the first functional layer on the negative electrode side

[0199] Based on the total mass of the solid content of the first functional layer (excluding solvent and support layer), the formulation of the first functional layer on the negative electrode side of the double-sided asymmetric Fc-AAm-PA modified polyether acrylate / PI composite solid separator is the same as the formulation of the functional composite layer of the Fc-AAm-PA@LLZTO / polyether acrylate / PI composite solid separator in Example 1.

[0200] 3.3 Formulation of the second functional layer on the positive electrode side

[0201] Based on the total mass of solids in the second functional layer (excluding solvent and support layer), the specific components include:

[0202] Table 5 Formulation of the second functional layer on the positive electrode side of the double-sided asymmetric Fc-AAm-PA modified polyether acrylate / PI composite solid separator

[0203] Partially lithium-ionized Fc-AAm-PA 35 mPEGA 30 PEGDA 6 PVDF-HFP 8 LiTFSI 20 TPO 1 total 100

[0204] Partially lithium-ionized Fc-AAm-PA was added based on the equivalent mass of unlithiated Fc-AAm-PA, and 22.0 mol% of the total molar amount of phosphonic acid protons in the Fc-AAm-PA molecule was added to an aqueous LiOH solution. The mixture was stirred at room temperature for 2 hours. Subsequently, the solvent was removed and the mixture was dried under vacuum to obtain partially lithium-ionized Fc-AAm-PA.

[0205] Design specifications regarding the absence of inorganic ceramic fillers in the second functional layer on the positive electrode side:

[0206] The second functional layer on the positive electrode side uses 35wt% partially lithium-modified Fc-AAm-PA to replace the Fc-AAm-PA@LLZTO ceramic filler. It is essentially a pure polymer / organic functional layer system, and its design is based on the following:

[0207] (1) The interface requirements on the positive electrode side are fundamentally different from those on the negative electrode side. The main challenge on the negative electrode side is the mechanical suppression of lithium dendrites, which requires high-modulus ceramic fillers to provide a physical barrier; while the main challenge on the positive electrode side is the stability of the CEI film under high voltage and the interfacial electrochemical compatibility, which requires functional chemical species (such as lithium phosphonate groups and Fc / Fc⁺ redox pairs) for interfacial regulation, rather than mechanical reinforcement.

[0208] (2) After 35wt% of partially lithium-ionized Fc-AAm-PA in the positive electrode side functional layer is copolymerized into the polyether acrylate crosslinking network through UV curing, its high-density lithium phosphonate groups and amide groups form a large number of hydrogen bonds and ion-dipole interactions with the polymer segments, which significantly improves the cohesive energy density and glass transition temperature of the crosslinking network, partially compensating for the lack of rigidity caused by the absence of ceramic filler. The positive electrode side functional layer mainly undertakes the interface regulation function. Although its self-supporting strength is relatively low, it does not affect the overall mechanical properties of the composite membrane.

[0209] (3) According to the measured data, the overall tensile strength of Example 3 is 58 MPa, which is higher than that of Example 1 (52 MPa) and Example 2 (55 MPa). It should be noted that the tensile strength test result is the comprehensive performance of the composite membrane as a whole (including the PI porous support membrane and the double-sided functional layer). Among them, the PI porous membrane (12 μm, tensile strength of 48 MPa alone) provides the main mechanical load as a rigid skeleton. The tensile strength of Example 3 is the highest (58 MPa), which is mainly attributed to the constraint effect of the cross-linked polymer layer applied to both sides of the PI membrane by the double-sided coating structure, forming a sandwich structure, which effectively inhibits the necking and pore deformation of the PI membrane during the stretching process. The tensile strength of the self-supporting film of the single positive electrode side functional layer (excluding the PI membrane and the negative electrode side functional layer) is about 18-22 MPa. Although it is lower than the tensile strength of the self-supporting film of the negative electrode side functional layer containing ceramic filler (about 28-32 MPa), the positive electrode side functional layer does not bear the main mechanical load function in the composite membrane. Its core role is to provide electrochemical interface regulation.

[0210] (4) Regarding the elongation at break, Example 3 (8%) was lower than Example 1 (12%) and Example 2 (10%), indicating that the double-sided coating cross-linked structure increases the overall rigidity and reduces the toughness. However, the elongation at break of 8% still meets the flexibility requirements of the winding process and battery assembly.

[0211] 3.4 Preparation steps

[0212] Step 1: Fabrication of the first functional layer on the negative electrode side

[0213] The first functional layer slurry on the negative electrode side was prepared according to the method of Example 1 and coated on one side (negative electrode side) of the PI porous membrane. The wet film thickness was 25 μm. After pre-drying and UV curing, the first functional layer was formed.

[0214] Step 2: Protection and flipping of the cured side

[0215] The single-sided cured composite film obtained in step 1 is transferred to a clean photolithography worktable. A clean polyester (PET) release film (25 μm thick, surface roughness Ra ≤ 0.1 μm) is then applied to the surface of the first functional layer to protect the cured functional layer from mechanical scratches, contamination, and deformation or damage caused by contact pressure from the coating platform during subsequent coating operations. The PET release film is adhered to the first functional layer by electrostatic adsorption without the use of any adhesive, ensuring that it can be peeled off without damage later.

[0216] Flip the coated composite film so that the uncoated side (positive electrode side) of the PI porous membrane faces upward and the PET release film side faces downward, and place it flat on the coating platform. A clean silicone rubber pad (Shore hardness A40-A60) is pre-laid on the surface of the coating platform to provide uniform elastic support and avoid local stress concentration on the first functional layer caused by the rigid platform.

[0217] Step 3: Fabrication of the second functional layer on the positive electrode side

[0218] The second functional layer slurry on the positive electrode side was prepared according to the method of Example 1 (except that 35wt% of Fc-AAm-PA@LLZTO was replaced with 35wt% of partially lithium-ionized Fc-AAm-PA, the rest of the slurry preparation method was the same as in Example 1), and it was coated on the other side (positive electrode side) of the PI porous membrane, with a wet film thickness of 25μm. During the coating process, the pressure of the coating doctor blade was controlled to not exceed 0.5N / cm to reduce the pressure on the underlying structure.

[0219] After pre-drying and UV curing, a second functional layer is formed. During the UV curing process, ultraviolet light only irradiates from the positive electrode side. The shielding effect of the PET release film and PI porous film on ultraviolet light can effectively avoid unintended secondary initiation by residual photoinitiators in the first functional layer.

[0220] Step 4: Release film peeling, vacuum drying and roller pressing for shaping

[0221] After UV curing, peel off the PET release film from the surface of the first functional layer on the negative electrode side. Place the double-sided cured composite film in a 70℃ vacuum oven and dry until the mass is constant. The residual NMP content is ≤1000ppm based on the total mass of the composite film.

[0222] After drying, the composite film is rolled and shaped at a pressure of 1 MPa and a temperature of 35°C. During the rolling process, both the upper and lower roller surfaces are covered with a clean PET release film to prevent the functional layer from directly contacting the roller surface and causing adhesion or surface defects.

[0223] After the above treatment, a double-sided asymmetric Fc-AAm-PA modified polyether acrylate / PI composite solid membrane was obtained.

[0224] Comparative Example 1:

[0225] Unmodified LLZTO composite solid membrane: Except for using LLZTO that has not been modified by Fc-AAm-PA to replace Fc-AAm-PA@LLZTO in Example 1, the other formulations, support membrane, slurry preparation, coating, UV curing, vacuum drying and rolling processes are the same as in Example 1.

[0226] Based on the total mass of solid content in the functional composite layer (excluding solvent and support layer), the formulation of Comparative Example 1 is as follows:

[0227] Table 6 Formulation of functional composite layers in unmodified LLZTO composite solid membranes

[0228] Unmodified LLZTO 35 mPEGA 30 PEGDA 6 PVDF-HFP 8 LiTFSI 20 TPO 1 total 100

[0229] Comparative Example 2:

[0230] Ceramic-filler-free polyether acrylate / PI membrane: Except for the absence of Fc-AAm-PA@LLZTO, the other components and preparation steps are basically the same as in Example 1. To maintain the film-forming properties and solid content stability of the functional layer, the ratio of mPEGA and PVDF-HFP is appropriately increased.

[0231] Based on the total mass of solid content in the functional composite layer (excluding solvent and support layer), the formulation of Comparative Example 2 is as follows:

[0232] Table 7 Formulations of functional composite layers for polyether acrylate / PI membranes without ceramic fillers

[0233] mPEGA 55 PEGDA 8 PVDF-HFP 16 LiTFSI 20 TPO 1 total 100

[0234] Comparative Example 3:

[0235] Composite solid membrane with added free ferrocene small molecules: Fc-AAm-PA@LLZTO was not used, but unmodified LLZTO was used, and free ferrocene small molecules with a content basically the same as that in Example 1 were added. The rest of the formulation and preparation steps were the same as in Example 1.

[0236] According to the molecular formula C of Fc-AAm-PA 15 H 18 Based on FeNO4P (363.13 g / mol), the theoretical mass fraction of Fe is approximately 15.38%. In Example 1, the content of Fc-AAm-PA@LLZTO is 35 wt%, and the apparent loading of Fc-AAm-PA in Fc-AAm-PA@LLZTO is 2.20 wt%. Therefore, the actual content of Fc-AAm-PA introduced into the functional composite layer is approximately 0.77 wt%, corresponding to an Fe content of approximately 0.12 wt%. In Comparative Example 3, when approximately 0.40 wt% of free ferrocene is added, the Fe content is approximately 0.12 wt%, which is basically consistent with the Fe content in Example 1.

[0237] Based on the total mass of solid content in the functional composite layer (excluding solvent and support layer), the formulation of Comparative Example 3 is as follows:

[0238] Table 8 Formulations of functional composite layers in composite solid membranes with added free ferrocene small molecules

[0239] Unmodified LLZTO 34.6 Free ferrocene (CAS No. 102-54-5) 0.40 mPEGA 30 PEGDA 6 PVDF-HFP 8 LiTFSI 20 TPO 1 total 100

[0240] In this comparative example, ferrocene exists in small molecule form, is not anchored to the ceramic surface by phosphonic acid groups, and is not covalently fixed in the polymer network by polymerizable double bonds.

[0241] Performance testing

[0242] Test 1: Ionic Conductivity Test

[0243] The membrane sample was cut into circular pieces (16 mm in diameter) and sandwiched between two stainless steel blocking electrodes to assemble an SS / membrane / SS battery. Electrochemical impedance spectroscopy (EIS) was used for testing, with a frequency range of 1 MHz to 0.1 Hz and a perturbation voltage of 10 mV.

[0244] The ionic conductivity σ is calculated using the following formula:

[0245] σ = L / (R×A)

[0246] Where: σ is the ionic conductivity (S·cm) -1 L is the membrane thickness (cm), R is the bulk impedance obtained from EIS fitting (Ω), and A is the effective electrode area (cm²). 2 ).

[0247] The test temperatures included 25℃ and 60℃.

[0248] Test 2: Lithium-ion transference number

[0249] Lithium-ion transference number was determined using a Li / membrane / Li symmetric cell via the Bruce-Vincent method. Initial current I0 and steady-state current I0 were measured at a polarization voltage of 10 mV. SS EIS tests were performed before and after polarization at a temperature of 25℃ to obtain the initial interface impedance R0 and the steady-state interface impedance R. SS .

[0250] The lithium-ion transport number t_Li⁺ is calculated using the following formula:

[0251] t_Li + =[I SS (ΔV-I0R0)] / [I0(ΔV-I SS R SS )]

[0252] Where ΔV is the polarization voltage (10mV).

[0253] Test 3: Electrochemical Stability Window Test

[0254] Linear sweep voltammetry (LSV) tests were performed using an SS / membrane / Li battery. The test conditions were as follows:

[0255] Scan rate: 0.5 mV / s;

[0256] Voltage range: from open circuit potential to 6.0V (with lithium metal potential as the reference of 0V, starting from the natural voltage when the battery is at rest, scan in the positive direction to 6.0V to observe whether the separator will undergo irreversible oxidation decomposition under high voltage).

[0257] Test temperature: 25℃.

[0258] For samples containing ferrocene units, the reversible redox peak of Fc / Fc⁺ at approximately 3.5 V is not considered the initiation potential for irreversible oxidative decomposition of the material. The potential corresponding to a significant and sustained increase in current density after excluding this reversible peak is taken as the initiation potential for irreversible oxidative decomposition.

[0259] Test 4: Mechanical Properties

[0260] Tensile tests were performed using a universal testing machine (referring to GB / T 1040.3-2006 "Determination of tensile properties of plastics - Part 3: Test conditions for films and sheets"). The diaphragm was cut into 10mm × 50mm strips (at least 5 per group), and the tensile rate was 5mm / min. Tensile strength and elongation at break were tested. The test temperature was 25℃.

[0261] Test 5: Heat Shrinkage Rate

[0262] Cut the diaphragm into 3cm × 3cm square samples and place them in:

[0263] Treat in an oven at 150℃ for 1 hour;

[0264] Treat in a 200℃ oven for 0.5 hours.

[0265] After cooling to room temperature, measure the change in membrane area and calculate the heat shrinkage rate using the following formula:

[0266] Thermal shrinkage rate = (S0 - S1) / S0 × 100%

[0267] Where S0 is the area before heat treatment and S1 is the area after heat treatment.

[0268] Test 6: Li / Li Symmetrical Cell Cycle Test

[0269] A Li / separator / Li symmetric cell was assembled by sandwiching a separator between two lithium metal sheets (50 μm thick). The test conditions were as follows:

[0270] Current density: 0.5 mA / cm² 2 ;

[0271] Surface capacity: 0.5mAh / cm² 2 (1 hour per half cycle);

[0272] Test temperature: 25℃;

[0273] Cycle method: constant current deposition / stripping.

[0274] Record the change in polarization voltage over time, as well as the battery short-circuit time.

[0275] Test 7: NCM811 / Li Full Cell Rate Performance Test

[0276] Using NCM811 positive electrode (area loading ~3.0 mg / cm³) 2 The active material, NCM811:conductive carbon black:PVDF=8:1:1 mass ratio, aluminum foil current collector, lithium metal negative electrode sheet (thickness 50μm) and the separator to be tested were used to assemble a CR2032 coin cell.

[0277] At 25°C, the circuit was charged and discharged for 5 cycles each at rates of 0.1C, 0.2C, 0.5C, 1C, and 2C, followed by 5 cycles at 0.1C (1C = 200mAh / g). The voltage range was 2.8-4.3V.

[0278] Test 8: NCM811 / Li Full Cell Long Cycle Test

[0279] Battery assembly is the same as in test seven. Test conditions are as follows:

[0280] Voltage range: 2.8-4.3V;

[0281] Test temperature: 25℃;

[0282] Cycle rate: 0.5C charge / 0.5C discharge;

[0283] Number of cycles: 300.

[0284] Record the discharge specific capacity during the first cycle, the capacity retention rate after 300 cycles, and the average coulombic efficiency.

[0285] Test 9: Self-discharge and Ferrocene migration risk test

[0286] The NCM811 / Li battery (assembled as in Test 7) ​​was charged to 4.3V at 0.1C and then left to stand at 25℃ for 72 hours. The open circuit voltage (OCV) decay value was recorded.

[0287] A greater voltage decay indicates the possible presence of significant redox shuttle, self-discharge, or interfacial side reactions in the system. This test aims to quantitatively evaluate the inhibitory effect of the covalent anchoring strategy on the transmembrane shuttle migration of ferrocene.

[0288] Test 10: EIS Impedance Test After Cycling

[0289] EIS testing (frequency range 1MHz-0.1Hz, disturbance voltage 10mV, 25℃) was performed on batteries that had completed eight long cycles (from the same batch, after resting for 1 hour after the 300th cycle) and compared with the EIS before cycling (after stabilization in the 3rd cycle).

[0290] The performance test results are shown in Table 9-17 below.

[0291] Table 9. Diaphragm Thickness and Mechanical Properties

[0292] Example 1 24 52 12 Example 2 24 55 10 Example 3 28 58 8 Comparative Example 1 24 46 14 Comparative Example 2 22 38 20 Comparative Example 3 24 45 15 PI porous membrane 12 48 28

[0293] Table 10 Ionic conductivity and lithium-ion transference number

[0294] Example 1 <![CDATA[1.6×10 -4 ]]> <![CDATA[6.5×10 -4 ]]> 0.52 Example 2 <![CDATA[1.9×10 -4 ]]> <![CDATA[7.8×10 -4 ]]> 0.58 Example 3 <![CDATA[1.8×10 -4 ]]> <![CDATA[7.2×10 -4 ]]> 0.55 Comparative Example 1 <![CDATA[8.5×10 -5 ]]> <![CDATA[3.7×10 -4 ]]> 0.40 Comparative Example 2 <![CDATA[5.6×10 -5 ]]> <![CDATA[2.4×10 -4 ]]> 0.35 Comparative Example 3 <![CDATA[9.0×10 -5 ]]> <![CDATA[3.9×10 -4 ]]> 0.39

[0295] In Example 2, an additional 3 wt% partially lithium-ionized Fc-AAm-PA (containing lithium phosphonate groups) was introduced, which, after copolymerization into the polymer network, provided more Li⁺ coordination sites and anion anchoring sites. Therefore, the ionic conductivity and lithium-ion transference number were the highest among the examples. In Example 3, the negative electrode side functional layer had the same formulation as Example 1, while the positive electrode side contained 35 wt% partially lithium-ionized Fc-AAm-PA. Its overall conductivity and transference number were between those of Examples 1 and 2. In Comparative Example 3, free ferrocene did not participate in ion conduction, and its conductivity was close to that of Comparative Example 1 (unmodified LLZTO).

[0296] Table 11 Thermal Shrinkage Rate

[0297] Example 1 0.8% 2.7% Example 2 0.7% 2.4% Example 3 0.5% 1.8% Comparative Example 1 1.2% 4.0% Comparative Example 2 1.8% 5.5% Comparative Example 3 1.3% 4.2%

[0298] All samples used a PI porous membrane (heat resistant ≥300℃) as a framework, and the shrinkage rate at 150℃ was <2%. Example 3, with its double-sided coating structure, formed cross-linked functional layers on both sides of the PI membrane, resulting in the lowest thermal shrinkage rate. Comparative Example 2, although lacking ceramic fillers, still possessed good dimensional stability due to the PI framework; however, partial softening of the polymer functional layer at 200℃ for 0.5h led to an overall shrinkage rate higher than the example (5.5%), but the rigid framework of the PI membrane limited further shrinkage. Fc-AAm-PA modification improved the thermal stability of the functional layer through a covalent cross-linked network; the shrinkage rate of the examples at 200℃ was lower than that of the comparative examples.

[0299] Table 12 Electrochemical stability window

[0300] Example 1 ~3.5 4.7 Example 2 ~3.5 4.8 Example 3 ~3.5 4.8 Comparative Example 1 No observations were made. 4.3 Comparative Example 2 No observations were made. 4.2 Comparative Example 3 ~3.5 4.0

[0301] In all examples containing Fc-AAm-PA, a reversible redox peak of Fc / Fc⁺ was observed at ~3.5V (relative to the lithium reference electrode). This peak is an inherent characteristic of the ferrocene structure and does not represent irreversible decomposition of the material. In the examples, Fc-AAm-PA is anchored to the LLZTO surface via phosphonic acid groups and / or covalently copolymerized into the polymer network via acrylamide double bonds. + Oxidation prevents transmembrane migration, thus not leading to a sustained irreversible current. Excluding this reversible peak, the irreversible oxidation decomposition initiation potential in Example 1 is 4.7V. In Examples 2 and 3, the decomposition potential is further increased to 4.8V due to the improved polymer-electrode interface stability caused by the additional partially lithium-lithiated lithium phosphonate groups. In Comparative Example 3, free ferrocene is oxidized to Fc at ~3.5V. + It can then freely diffuse to the negative electrode and be reduced back to Fc, forming a redox shuttle current; at the same time, Fc⁺ can undergo further irreversible oxidation at a higher potential (Fc + →Fc 2+ (Ligand decomposition) results in an irreversible decomposition initiation potential of only 4.0V.

[0302] Table 13 Cycle performance of Li / Li symmetric cells (0.5 mA / cm) 2 0.5mAh / cm 2 (25℃)

[0303] Example 1 58 72 ≥800 Example 2 48 58 ≥900 Example 3 45 55 ≥1000 Comparative Example 1 88 / ~350 Comparative Example 2 120 / ~180 Comparative Example 3 95 / ~280

[0304] In Example 2, the additional partial lithium-ionized Fc-AAm-PA copolymerization into the network provides more uniform Li⁺ deposition sites and a higher t_Li⁺ (0.58), effectively reducing concentration polarization and dendrite nucleation driving force. The initial polarization voltage (48mV) and stable cycling time (≥900h) are both superior to those of Example 1. In Example 3, the first functional layer on the negative electrode side contains Fc-AAm-PA@LLZTO ceramic filler to provide mechanical suppression, combined with high ionic conductivity (1.8×10⁻⁶). -4 S·cm -1 It exhibits a high mobility number (0.55), the lowest polarization voltage (45mV), and the longest cycle time (≥1000h). Comparative Example 2, without the mechanical reinforcement and ion conduction assistance of ceramic fillers, has the weakest dendrite suppression ability; in Comparative Example 3, free ferrocene migrates to the lithium surface to form an uneven deposition layer, which actually accelerates polarization fluctuations.

[0305] Table 14. Rate performance of NCM811 / Li full cells (2.8-4.3V, 25℃, discharge specific capacity / mAh·g) -1 )

[0306] Example 1 192 186 176 160 140 190 Example 2 195 190 182 170 152 193 Example 3 194 189 180 168 150 192 Comparative Example 1 188 178 164 142 115 184 Comparative Example 2 184 172 152 128 96 178 Comparative Example 3 186 175 160 138 108 176

[0307] Among them, the rate performance is consistent with the trends of ion conductivity and transport number. Example 2: Ion conductivity (1.9 × 10⁻⁶) - 4 S·cm -1 The ferrocene exhibited the highest values ​​for both the ferrocene concentration and the migration number (0.58), maintaining optimal capacity at high rates. At 2C, it retained 152 mAh / g (77.9% of 0.1C), compared to only 96 mAh / g (52.2% of 0.1C) in Comparative Example 2. The capacity recovery rates of Examples 1-3 and Comparative Examples 1 and 2 after returning to 0.1C were all >95% (98.9%, 99.0%, 99.0%, 97.9%, and 96.7%, respectively), indicating that high-rate cycling did not cause significant irreversible structural damage to these samples. The capacity of Comparative Example 3 at 0.1C (176 mAh / g) was lower than its initial 0.1C capacity (186 mAh / g), with a recovery rate of only 94.6%, the lowest among all samples. This is presumably related to the accelerated migration of free ferrocene to the cathode surface during high-rate cycling, leading to irreversible oxidation and the accumulation of interfacial side reactions.

[0308] Table 15 Long-term cycle performance of NCM811 / Li full cells (0.5C, 2.8-4.3V, 25℃)

[0309] Example 1 176 151 85.8 99.95 Example 2 182 160 87.9 99.96 Example 3 180 159 88.3 99.96 Comparative Example 1 164 122 74.4 99.90 Comparative Example 2 152 98 64.5 99.85 Comparative Example 3 160 105 65.6 99.86

[0310] Among them, the 300-cycle retention rates of Examples 2 and 3 were close (87.9% and 88.3%, respectively), both significantly better than Example 1 (85.8%), demonstrating the contribution of the additional partially lithium-ionized Fc-AAm-PA to cycle stability. Example 3 was slightly better than Example 2, thanks to its bi-sided asymmetric structure: the ceramic layer on the negative electrode side suppressed lithium dendrites and stabilized the SEI (negative electrode interface film), while the lithium phosphonate functional layer on the positive electrode side stabilized the CEI (positive electrode interface film), with each side adapting to its respective interface requirements. The capacity retention rate of Comparative Example 3 (65.6%) was even lower than that of Comparative Example 1 (74.4%), indicating that the cross-membrane shuttle of free ferrocene caused continuous loss of the positive electrode interface and active lithium.

[0311] Table 16 Self-discharge and ferrocene migration risk test (4.3V full charge, 25℃ standing for 72h)

[0312] Example 1 42 Normal self-discharge level Example 2 38 Normal self-discharge level Example 3 32 Lowest self-discharge Comparative Example 1 55 No Fc shuttle, slightly higher than the example Comparative Example 2 62 There are many side effects in the interface. Comparative Example 3 118 Severe self-discharge, free Fc redox shuttle

[0313] This test is the core experiment for evaluating the effectiveness of the covalent anchoring strategy. The 72hOCV decay (118mV) of Comparative Example 3 is significantly greater than all other samples, possibly attributed to the redox shuttle effect of free ferrocene. Fc is oxidized to Fc⁺ on the positive electrode surface, diffuses through the separator to the negative electrode, and is reduced back to Fc by lithium. This repeated shuttle continuously consumes charge, leading to severe self-discharge. In contrast, in Example 1, Fc-AAm-PA is anchored to the LLZTO ceramic surface via phosphonate groups and copolymerized into the crosslinked network via acrylamide double bonds. The ferrocene units are covalently fixed and cannot freely migrate and shuttle, resulting in an OCV decay of only 42mV, which is within the normal self-discharge level of lithium metal batteries. Examples 2 (38mV) and 3 (32mV) showed better performance because the additional partial lithiation of Fc-AAm-PA was also fixed through covalent copolymerization, and the lithium phosphonate group improved the interfacial stability and reduced self-discharge caused by side reactions of SEI (negative electrode interfacial film) / CEI (positive electrode interfacial film). Comparative Example 1 (55mV) did not have the ferrocene shuttle problem, but the poor compatibility between the unmodified LLZTO and the polymer interface led to more interfacial side reactions.

[0314] Table 17 Changes in EIS impedance of the full cell before and after cycling

[0315] Example 1 18.5 22.3 45.2 68.7 52.0 Example 2 15.8 18.6 38.5 52.1 35.3 Example 3 16.2 19.1 36.8 48.5 31.8 Comparative Example 1 28.6 38.5 82.4 165.2 100.5 Comparative Example 2 36.2 52.8 105.6 248.3 135.1 Comparative Example 3 26.8 48.5 78.5 285.6 263.8

[0316] Where: R_b is the real-axis intercept at the starting point of the high-frequency semicircle, corresponding to the bulk impedance (including the membrane bulk resistance and electrode electronic resistance); R_int is the total diameter of the high-frequency and mid-frequency semicircles, corresponding to the combined contribution of the SEI / CEI interface film and charge transfer impedance. The impedance value is obtained by fitting the equivalent circuit R_b-(R_SEI / CPE_SEI)-(R_CT / CPE_CT)-W using ZView software, where R_int=R_SEI+R_CT.

[0317] The interfacial impedance growth rate of Comparative Example 3 reached a high of 263.8%, far exceeding that of Example 1 (52.0%) and Comparative Example 1 (100.5%). Notably, the initial interfacial impedance of Comparative Example 3 before cycling (78.5Ω) was close to that of Comparative Example 1 (82.4Ω), indicating that the free ferrocene did not significantly deteriorate the interface in the initial state. However, after 300 cycles, the interfacial impedance of Comparative Example 3 (285.6Ω) increased most dramatically, and was higher than that of Comparative Example 2 (248.3Ω) without ceramic filler. This indicates that the redox shuttle of free ferrocene led to a severe accumulation of interfacial side reactions during long-term cycling.

Claims

1. A high-performance diaphragm, characterized in that, Includes a porous support layer and a functional composite layer coated on at least one side of the porous support layer; The functional composite layer comprises modified ceramic filler, polyether acrylate monomers, crosslinking agent, film-forming polymer, lithium salt, and photoinitiator. The content of each component, based on the total solid content of the functional composite layer by mass, is as follows: Modified ceramic filler 25-45 wt%; Polyether acrylate monomers 25-40 wt%; Crosslinking agent 3-10wt%; Film-forming polymer 5-15 wt%; Lithium salt 15-25 wt%; Photoinitiator 0.5-2wt%; The modified ceramic filler is obtained by anchoring N,N-disubstituted acrylamide-type phosphonic acid-based ferrocene derivatives onto the surface of lithium-ion conductor ceramic particles via phosphonic acid groups; the apparent loading of the derivative in the modified ceramic filler is 1.5-3.0 wt%. The chemical structural formula of the derivative is: ; The acrylamide double bonds of the derivatives in the modified ceramic filler are covalently copolymerized with the polyether acrylate crosslinking network, so that the ferrocene units are simultaneously anchored to the ceramic surface by phosphonic acid groups and connected to the polymer network by carbon-carbon covalent bonds.

2. The diaphragm according to claim 1, characterized in that, The porous support layer is a polyimide porous membrane with a thickness of 10-15 μm, a porosity of 50-65%, and an average pore size of 0.8-1.5 μm. The lithium-ion conductor ceramic particles are selected from at least one of garnet-type lithium lanthanum zirconium oxide ceramics, NASICON-type lithium-ion conductor ceramics, and perovskite-type lithium-ion conductor ceramics. The polyether acrylate monomer is polyethylene glycol monomethyl ether acrylate with a number average molecular weight of 480-20000. The crosslinking agent is polyethylene glycol diacrylate with a number average molecular weight of 200-2000; The film-forming polymer is a vinylidene fluoride-hexafluoropropylene copolymer, wherein the HFP content is 15-25 mol% and the number average molecular weight Mn is 4 × 10⁻⁶. 5 -8×10 5 ; The lithium salt is selected from at least one of lithium bis(trifluoromethanesulfonyl)imide, lithium bis(fluorosulfonyl)imide, and lithium difluorooxalateborate.

3. The diaphragm according to claim 1, characterized in that, The functional composite layer also contains partially lithium-ionized N,N-disubstituted acrylamide-type phosphonate ferrocene derivatives as reactive comonomers. Partial lithiation refers to the replacement of 15-35 mol% of the acidic protons in the phosphonate group of N,N-disubstituted acrylamide-type phosphonate-based ferrocene derivatives with lithium ions to form a lithium phosphonate structure. The content of partially lithium-ionized N,N-disubstituted acrylamide-type phosphonic ferrocene derivatives is 1-5 wt% based on the total solid content of the functional composite layer.

4. The diaphragm according to claim 3, characterized in that, The diaphragm has a double-sided asymmetric structure, comprising: A first functional layer coated on the porous support surface facing the negative electrode side, the first functional layer comprising modified ceramic filler, polyether acrylate crosslinking network and lithium salt; And a second functional layer coated on the porous support layer facing the positive electrode side, the second functional layer comprising a partially lithium-ionized N,N-disubstituted acrylamide-type phosphonic ferrocene derivative, a polyether acrylate crosslinking network and a lithium salt, and the second functional layer does not contain inorganic ceramic fillers.

5. A method for preparing a high-performance separator as described in any one of claims 1-4, characterized in that, Includes the following steps: Step S1: Dissolve the film-forming polymer in an organic solvent to prepare a pre-solution; Step S2: Under an inert atmosphere and light-protected conditions, the presol is mixed with polyether acrylate monomers, crosslinking agents, lithium salts, photoinitiators, and modified ceramic fillers, and then dispersed and degassed to obtain a polymerization precursor slurry. Step S3: The polymerization precursor slurry is coated on at least one side of the porous support layer and pre-dried. Step S4: In-situ UV curing is performed under an inert atmosphere to allow the unsaturated double bonds on the N,N-disubstituted acrylamide-type phosphonic acid ferrocene derivatives on the surface of polyether acrylate monomers, crosslinking agents, and modified ceramic fillers to undergo free radical copolymerization and crosslinking reactions, forming a functional composite layer. Step S5 involves vacuum drying and roll forming to obtain the high-performance diaphragm.

6. The preparation method according to claim 5, characterized in that, In step S2, the dispersion process sequentially includes planetary stirring, ultrasonic dispersion, and vacuum degassing, and the solid content of the slurry is 25-45 wt%. In step S3, the wet film thickness is 15-50 μm, the dew point of the coating environment is ≤-40℃, and the pre-drying temperature is 40-60℃. In step S4, the ultraviolet light wavelength is 350-380 nm, and the light intensity is 50-150 mW / cm². 2 The irradiation time is 15-60 seconds; In step S5, the vacuum drying temperature is 60-80℃ and the roller pressure is 0.5-2MPa.

7. The preparation method according to claim 5, characterized in that, When the high-performance diaphragm has a double-sided asymmetric structure, a first functional layer containing modified ceramic filler is first coated and cured on one side of the porous support layer. Then, a release film is covered on the surface of the first functional layer for protection, and the diaphragm is flipped over to coat and cure on the other side of the porous support layer to form a second functional layer containing partially lithium-ionized N,N-disubstituted acrylamide-type phosphonic acid ferrocene derivative.

8. The preparation method according to claim 5, characterized in that, The modified ceramic filler is prepared by dissolving an N,N-disubstituted acrylamide-type phosphonic acid ferrocene derivative in a mixed solvent consisting of 98-99 parts by volume of anhydrous ethanol and 1-2 parts by volume of deionized water, adjusting the pH to 6.5-7.5, and then performing surface modification treatment with pre-dried lithium-ion conductor ceramic particles.

9. The preparation method according to claim 5, characterized in that, The method for synthesizing the N,N-disubstituted acrylamide-type phosphonic ferrocene derivative is as follows: Ferrocene formaldehyde and diethyl aminomethylphosphonate were first dehydrated and condensed in the presence of an acidic catalyst and a 4Å molecular sieve to generate an imine intermediate. Then, the imine was reduced with NaBH4 as a reducing agent to obtain a secondary amine intermediate containing diethyl phosphonate groups. The secondary amine intermediate was subjected to N-acrylation reaction with acryloyl chloride in the presence of an acid-binding agent and a polymerization inhibitor to obtain an intermediate containing an acrylamide double bond and a diethyl phosphonate group. The diethyl phosphonate group was deprotected with trimethylbromosilane using McKenna, and the target product containing free phosphonate groups was obtained by hydrolysis.

10. The application of a high-performance separator in solid-state lithium metal batteries, characterized in that, The battery includes a positive electrode, a negative electrode, and a high-performance separator as described in any one of claims 1-4 disposed between the positive electrode and the negative electrode; The active material of the positive electrode is selected from at least one of lithium nickel cobalt manganese oxide, lithium nickel cobalt aluminum oxide, lithium iron phosphate, and lithium-rich manganese-based positive electrode materials; The negative electrode is a lithium metal negative electrode or a lithium-containing alloy negative electrode; When the high-performance diaphragm is the double-sided asymmetric structure as described in claim 4, the first functional layer is arranged facing the negative electrode, and the second functional layer is arranged facing the positive electrode.

Images