Conformal carbon nitride thin film inter-active interphase heterojunction with sustainable carbon enhancing sodium storage performance

Sustainable, high-performance carbonaceous anode materials are highly required to bring sodium-ion batteries to a more competitive level. Here, we exploit our expertise to control the deposition of a nm-sized conformal coating of carbon nitride with tunable thickness to improve the electrochemical performance of anode material derived from sodium lignosulfonate. In this way, we significantly enhanced the electrochemical performances of the electrode, such as the first cycle efficiency, rate-capability, and specific capacity. In particular, with a 10 nm homogeneous carbon nitride coating, the specific capacity is extended by more than 30% with respect to the bare carbon material with an extended plateau capacity, which we attribute to a heterojunction effect at the materials' interface. Eventually, the design of (inter)active electrochemical interfaces will be a key step to improve the performance of carbonaceous anodes with a negligible increase in the material weight.

Fabriker) as a carbon source; urea (4 wt. %, Sigma-Aldrich) and D-glucose anhydrous (4 wt. %, Sigma-Aldrich) as a binder mixture; DI water (10 wt. %); ZnO nanoparticles (24 wt. %, NanoAmor, 20 nm) as porogen were mixed using a commercial kitchen kneader (Bosch, Germany). 1, 2 DI water in drops was slowly added to the mixture as a plasticizer to provide consistency for extrusion. Finally, the low-moisture mixture was extruded and cut in the form of pellets (1 mm in diameter and 1.5 mm in length) using a commercial noodle extruding machine (La Monferrina, Italy) (Figure S1a) following the procedure developed by Brandi et al. 1 The role of glucose/urea binder is to provide crosslinking while carbonizing the formed pellets and generating an intact rigid sulfur-rich carbon. In the absence of glucose/urea binder, the physical mixture will lead to random evaporation of Zn and will not result in the desired interconnected hierarchical material. The extruded pellets were dried at room temperature for 12 hours. The dried pellets were put in an ashing furnace (Nabertherm, Germany), heated to 120 ºC with a heating rate of 3 K min -1 , and kept for 2 hours to crosslink under N 2 atmosphere. Subsequently, the oven was heated to 950 ºC with a heating rate of 3 K min -1 and kept at this temperature for 2 hours to finalize step-wise condensation. At this stage, carbothermal reduction of ZnO nanoparticles to Zn metal occurs, and the Zn metal leaves the framework. In the end, the carbonization yield was measured as 30%. Carbonized pellets (Figure S1b Applying carbon nitride (CN) thin films by chemical vapor deposition (CVD) method. The procedure developed by Giusto et al. 3 was followed as a coating model. PlanarGROW-3S-OS (UK) CVD system with a 3-in. quartz tube was used for the deposition of CN films on LSC. In general, LSC powder (150 mg) was placed horizontally in the center of the second chamber, and a glass boat containing melamine 4 (99%, Sigma-Aldrich) precursor (1 g for LSC/ThinCN, 2 g for LSC/MediumCN, and 5 g for LSC/ThickCN) in the center of the first chamber. The vacuum was pulled down to 10 Torr, and the temperature at the substrate was raised to 550 °C, with 50 sccm nitrogen flow as carrier gas. As soon as the substrate was at 550 °C, the melamine was heated up to 300 °C at a rate of 10 K min -1 and kept for an additional 30 min. The samples were cooled to room temperature naturally. The color of samples changes significantly from black to bluish depending on the precursor amount ( Figure S1c). Physico-chemical characterizations. X-ray photoelectron spectroscopy (XPS) was performed using the CISSY equipment (Helmholtz-Zentrum Berlin, Germany) with a SPECS XR 50 X-ray source using the Mg K  radiation. Photoelectrons were analyzed using a hemispherical analyzer (CLAM4 by VG). Binding energy calibration was performed using clean samples of gold, copper, and silver foil. Indium foil (99.99%, Sigma-Aldrich) was used as a substrate. The Shirley-type background and Lorentzian-Gaussian (mixed) models were used for the fittings. The crystallinity of the material was determined by X-ray diffraction (XRD) using Rigaku SmartLab (Japan, Cu K  , 0.154 nm). Raman spectroscopy was obtained using WITec Alpha 300 (Germany) confocal Raman microscope with an excitation wavelength of 532 nm. Thermo Scientific Nicolet iD7 (USA) spectrometer was used as a Fourier-transform infrared (FTIR) spectrometer. Thermogravimetric analysis (TGA) was conducted using NETZSCH TG-209 Libra (Germany) under a synthetic air and N 2 atmosphere at a heating rate of 10 K min -1 . Physisorption measurements were performed on a Quantachrome Quadrasorb SI (Austria) at 77 K for N 2 and 273 K for CO 2 . Samples were degassed overnight before the measurements. The density functional theory (DFT) method was used to evaluate the pore size distribution (PSD) of the materials employing adsorption isotherms. Inductively coupled plasma optical emission spectrometry (ICP-OES) was conducted with PerkinElmer Optima 8000 (USA). Scanning Electron Microscopy (SEM) imaging was performed using the Zeiss LEO 1550-Gemini (Germany) system with acceleration voltages of 3, 5, and 10 kV. An Oxford Instruments X-MAX (UK) 80 mm 2 detector was used to collect the energy-dispersive X-ray (EDX) data.         (002) of the CN, confirming its layered structure. 6 After deposition on the LSC, the (002) peak of the CN is preserved in the diffraction patterns. Furthermore, it is also observed that peaks from ZnS (See supplementary note 1) become more pronounced due to its recrystallization after CN deposition.

Supplementary note 4.
It is worth underlining that the composite material has a considerably higher thermal stability in the air, with ca. 24% leftover at 900°C for the LSC/ThinCN material, with respect to only 10% leftover for the pristine LSC carbon ( Figure S6). This is especially remarkable if we consider that the CN thin films are quantitatively decomposed at 650°C in air. 7 This is typical for composite materials with intimate contact, resulting in a higher thermal resistance than its single counterparts. 7      20 Figure S15. The first cycles of the half-cells with initial Coulombic efficiencies (ICEs).

Supporting note 5.
To prove that the improvement in sodium storage performance is a matter of conformal CN coating, we exfoliated the pristine LSC with the post-treatment in CVD without melamine precursor to see the heat treatment effect on electrochemical performance. By doing so, we observe that heat treatment even decreases the capacity of the pristine LSC, keeping the ICE same, confirming once more the significant improvement in electrochemical performance is due to the conformal CN layer.
Moreover, pure bulk CN shows a negligible contribution to sodium energy storage ( Figure S16) as well as high electrical resistivity, thus making it an incompatible active material for sodium storage under the same electrochemical conditions. This points further to a synergetic contribution between the LSC/CN rather than the properties of the single material.  Supporting note 6. We confirmed that the potential response behaves linearly with respect to the square root of the step time ( Figure S19). Hence, we used the first-order approximation derived by Weppner and Huggins, 14 which is valid for both spherical and planar geometries. 15 Still, this approach requires a series of assumptions; 16 i.
Step time is significantly less than the effective diffusion time.
ii. Transient data must be large enough not to include ohmic and kinetic overpotential.
Based on the assumptions, diffusion coefficients from GITT measurements can be calculated using the following simplified equation (eq. 1). (1) Where τ is the pulse duration, and are the actual and molar mass of the active material, is the molar volume, and S is the surface area of the electrodes. Hence, represents the electrode ( ) volume, and dividing it by the S gives electrode thickness.
(change of the steady-state voltage during ∆ a single-step GITT experiment) and (change of cell voltage during a constant current pulse) can be ∆ extracted from the typical GITT curve of the material ( Figure S18). 17,18 The diffusion coefficients were calculated from 1.0 V to 0.2 V (vs. Na + /Na) with a step size of 0.05 V (Figure 3e).