Negating Na‖Na3Zr2Si2PO12 interfacial resistance for dendrite-free and “Na-less” solid-state batteries

Solid electrolytes hold promise in safely enabling high-energy metallic sodium (Na) anodes. However, the poor Na‖solid electrolyte interfacial contact can induce Na dendrite growth and limit Na utilization, plaguing the rate performance and energy density of current solid-state Na-metal batteries (SSSMBs). Herein, a simple and scalable Pb/C interlayer strategy is introduced to regulate the surface chemistry and improve Na wettability of Na3Zr2Si2PO12 (NZSP) solid electrolyte. The resulting NZSP exhibits a perfect Na wettability (0° contact angle) at a record-low temperature of 120 °C, a negligible room-temperature Na‖NZSP interfacial resistance of 1.5 Ω cm2, along with an ultralong cycle life of over 1800 h under 0.5 mA cm−2/0.5 mA h cm−2 symmetric cell cycling at 55 °C. Furthermore, we unprecedentedly demonstrate in situ fabrication of weight-controlled Na anodes and explore the effect of the negative/positive capacity (N/P) ratio on the cyclability of SSSMBs. Both solid-state Na3V2(PO4)3 and S full cells show superior electrochemical performance at an optimal N/P ratio of 40.0. The Pb/C interlayer modification demonstrates dual functions of stabilizing the anode interface and improving Na utilization, making it a general strategy for implementing Na metal anodes in practical SSSMBs.

Pb/C surface modification. 10 μL aqueous solution of saturated Pb(Ac) 2 ·3H 2 O was drop on the surface of a NZSP pellet and distrusted evenly using a brush. For samples used in symmetrical cell testing, both sides of NZSP pellets were coated with Pb(Ac) 2 ·3H 2 O using the same method. After drying, the coated NZSP pellets were transferred into a tube furnace and heated at 550 °C for 5 h under Ar flow. The carbon-coated NZSP was prepared by following the same protocol but with a NaAc·3H 2 O precursor. To reduce exposure to air, the quartz tube containing the treated pellets was transferred into a glove box immediately after the samples were cooled to 90 °C.

Preparation of NVP/C cathode.
In a typical synthesis, 15 mmol Na 2 CO 3 (99.9%, Aladdin), 10 mmol V 2 O 5 (99%, Energy Chemical), 30 mmol NH 4 H 2 PO 4 (99.99%, Aladdin), and 1 g C 6 H 12 O 6 (99%, Sinopharm Chemical Reagent Co., Ltd.) were ball-milled for 12 hours at 300 rpm in ethanol using an alumina jar. The obtained mixture was uniaxially pressed into a pellet that was subsequently annealed at 450 °C for 4 h under Ar flow. After cooling, the precursor pellet was crushed and ball-milled again in ethanol at 300 rpm for 12 hours using an alumina jar. Finally, the NVP/C sample was produced by calcinating the dry precursor powders at 800 °C for 24 h in Ar. To fabricate NVP/C electrodes, a slurry of the NVP/C powder, Super P (MTI Corp., Shenzhen), and PVDF (Kynar HSV 900) at a weight ratio of 70: 20: 10 was cast onto carboncoated Al foil (MTI Corp., Shenzhen, 18 mm in thickness). The electrode material loading is about 2-3 mg cm -2 .
Preparation of S cathode. The Fe 3 C-and nitrogen-doped carbon at activated porous carbon cloth (Fe 3 C-NC@ACC) matrix was prepared using a previously reported method. [1] The activated porous carbon cloth was prepared by a simple alkaline activation treatment on commercial carbon cloth. Typically, a piece of ACC was immersed in 40 mL DI water that contains 0.5 mL pyrrole monomer and 0.8 mL hydrochloric acid solution. 1.0 g of potassium ferricyanide was added into 10 mL DI water under stirring. After cooling down to about 4 °C, 3 both solutions were mixed together and kept at 4 °C for 4 h. The resultant carbon cloth was washed by DI water and dried at 60 °C, followed by annealing at 800 °C for 2 h in N 2 .
Subsequently, sulfur was loaded on the carbon composite scaffold according to a simple inside encapsulation method to obtain S/Fe 3 C-NC@ACC. The carbon matrix was immersed in the sulfur/CS 2 solution until the mixture was completely dry. Then, the mixture was transferred into a sealed glass bottle filled with argon and heated at 155 °C for 10 h. The sulfur contents in the composite sample were about 1.0 mg/cm 2 . Finally, the sulfur cathode in solid state was prepared by mixing 80 wt% S/Fe 3 C-NC@ACC and 20 wt.% PEO10-NaFSI in acetonitrile. The PEO10-NaFSI was prepared by mixing 40 wt.% NaFSI and 60 wt.% PEO (M.W.: 5,000,000, Sigma-Aldrich) in acetonitrile and stirred at 60 °C for at least 24 h in Ar-filled glove box.
Physicochemical characterizations. XRD patterns were collected using an X-ray diffractometer (SmartLab9 KW) at a scan rate of 10° min -1 in the 2θ range of 10-80°. The microstructural and composition analysis of the powders and fractured cross-sections of NZSP pellets were conducted on a field emission scanning electron microscope (SEM, Hitachi S4800) equipped with an energy dispersive spectrometer (EDS). Surface chemical states of NZSP pellets were identified by X-ray photoelectron spectroscopy (XPS, PHI QuanteraII) on an Xray photoelectron spectrometer with Al Kα as the X-ray source, and all binding energies of samples were corrected by referencing the C 1s peak to 284.8 eV. An in-situ heating kit (Beijing Scistar Technology) was used for the in-situ Raman study on the decomposition process of Pb(Ac) 2 ·3H 2 O coated on NZSP pellets. Raman spectra were collected on a Via-Reflex Raman spectrometer (HORIBA XploRA PLUS, Japan) equipped with a 50 -LWD objective and laser × at 638 nm with 8 mW laser power. The time acquisition was 10 s and 10 scans were recorded to improve the signal-to-noise ratio. For contact-angle measurements, NZSP pellets were preheated on a hot plate at 120 °C inside the glovebox, and a small piece of Na metal (oxidation layer pre-removed) was placed on top of NZSP pellets to allow the observation of wetting 4 phenomena.
Electrochemical Testing. The ionic conductivity of NZSP pellets was determined by AC impedance measurements conducted on a Gamry REF 600+ potentiostat/galvanostat (Gamry, USA) using an Au||NZSP||Au configuration from 5 MHz to 1 Hz with an amplitude of 10 mV.
For the assembly of Na||NZSP||Na symmetric cells, two Na disks (diameter: 9 mm) were handpressed on both sides of the NZSP pellet. For the assembly of Na||Pb/C@NZSP||Na cells, Na was preloaded on both sides of the Pb/C@NZSP pellet by a molten Na infusion process. All Na symmetric cells were assembled using a pressure-controllable Swagelok-type cell holder (EQ-PSC, MTI Corp.). The stacking pressure control was enabled with the help of a uniaxial press equipped with a pressure sensor (YLJ-5T, MTI Corp.). To be specific, the pressure sensor reads the mass load (m) in kg, which can be converted to the pressure applied on the NZSP pellet

Density Functional Theory (DFT) Calculations. All DFT calculations were performed with
the VASP code, [2] Projector Augmented Wave (PAW) methods were used for the pseudopotentials. [3] The generalized gradient approximation (GGA)with the Perdew-Burke-Emzerhof (PBE) functional was employed to describe the exchange correlation energy. [4] The energy cutoff was set to 450 eV. The convergence criterion for the electronic self-consistency loop and atomic forces was set to 10 -5 eV and 0.05 eV Å -1 , respectively. To simulate the interface of Na/Pb, Na/NZSP, Na/Na 2 CO 3 and Na/C, the lattice constant of four layers of Na (001) slab was adjusted to adapt the dimensions of Pb (001), NZSP (001), Na 2 CO 3 (001) and C (001). The number of atoms in these interfaces and the lattice parameters for the interfaces are listed in Table S3. A vacuum region of 15 Å was set along the z direction and the back half atoms that are farthest away from the interface in each slab were fixed for all the heterostructure geometry optimizations. A 1×1×1 k-point mesh was used for the Brillouin Zone sampling to speed the optimization process.                   Fig. 3b. Therefore, the R int. of 3.72 Ω was estimated by subtracting the total cell resistance (R b + R gb ) of 170.43 Ω by the total resistance of the NZSP pellet (166.71 Ω, Fig. S1).