Composite nanostructure, multilayer composite, and method for producing composite nanostructure

Abstract

Composite nanostructures are disclosed that include a columnar membrane grown on top of another layer, where the pillars meet at the top to form an arch with optimized properties. These nanostructures, called nanovaults, provide high mechanical stability for the membrane under strong and fluctuating stresses.

Description

[Technical Field]

[0001] This invention relates to materials and applications where surface mechanics play an important role, such as Si anodes for LIBs. [Background technology]

[0002] Alloy-based anode materials are promising alternatives to graphite for high-energy LIBs because they offer up to 10 times the theoretical capacity. However, the large volume change during lithiation prevents the formation of a stable solid electrolyte interface (SEI), leading to electrode breakdown. The low mechanical stability of LIBs using Si anodes poses an obstacle to their practical application, although composite anodes containing Si additives are already commercially available. However, the mechanical stability of Si is a key parameter that limits both the particle size and content of Si in composite anodes. Therefore, increasing the Si content in anodes while maintaining sufficient mechanical stability remains a challenge for high-energy LIB technology. Summary of the Invention [Problem to be solved by the invention]

[0003] The main technical problem is that compressive stress accumulates during lithiation and is subsequently released during delithiation. If this compressive stress exceeds the yield strength, the electrode deforms to accommodate the volume change. Therefore, adjusting the elastic modulus E of the Si anode is of utmost importance. E quantifies the strain (the amount of deformation obtained by dividing the displacement by the reference length ε) of the material when stress (force per area, σ) is applied (E=σ / ε).

[0004] Therefore, a Si electrode with a low E value maximizes anode capacity at the expense of reduced mechanical stability, as the material deforms easily. This issue can be addressed by encapsulating the nanostructured Si with a shell, allowing the formation of a stable SEI while maintaining the electrode integrity, but this results in a decrease in LIB energy density. Another way to improve the mechanical stability of Si-based anodes is to increase the anode's elastic modulus by using new binders for Si composites or by physically suppressing volume expansion. This also limits the energy density of LIBs. [Means for solving the problem]

[0005] The present invention has been made in view of the above circumstances, and to solve the above problems, a first aspect of the present invention is a composite nanostructure comprising a columnar film grown on another layer, the columns contacting each other at the top to form an arched shape with optimized properties.

[0006] A second aspect of the present invention is the composite nanostructure of the first aspect, wherein the columnar film is an amorphous Si film in an annealed state.

[0007] A third aspect of the present invention is the composite nanostructure of the first aspect, wherein the columnar film is formed on a layer of metal nanoparticles.

[0008] A fourth aspect of the present invention is the composite nanostructure of the first aspect, wherein the optimized properties include at least one of low lithium consumption during the formation of the solid electrolyte interface in lithium ion batteries, high coulombic efficiency, and high mechanical stability suitable for any application where the surface of the film is subjected to strong fluctuating stresses.

[0009] A fifth aspect of the present invention is a multilayer composite comprising at least two perpendicular repeats of the composite nanostructure of the first aspect.

[0010] A sixth aspect of the present invention is the multilayer composite of the fifth aspect, wherein said optimized properties compared to those of a monolayer structure include both a higher coulombic efficiency in lithium ion batteries and a mechanical stability of the membrane exhibiting arching reinforcement.

[0011] A seventh aspect of the present invention is a method for producing a composite nanostructure, comprising the steps of: a) depositing nanoparticles from the gas phase onto a substrate; and b) growing a columnar film on the nanoparticle layer, wherein in step b) the column diameter of the columnar film increases with thickness until the columnar heads touch each other and the surface is closed via the formation of an arched structure.

[0012] An eighth aspect of the present invention is the manufacturing method of the seventh aspect, wherein in the step b), the growth of the columnar film is stopped immediately after the arch is formed.

[0013] A ninth aspect of the present invention is the manufacturing method of the seventh aspect, further comprising the step of c) subjecting the columnar film to thermal annealing treatment.

[0014] A tenth aspect of the present invention is the manufacturing method according to the seventh aspect, wherein the columnar film is an amorphous Si film.

[0015] An eleventh aspect of the present invention is the production method of the seventh aspect, wherein the nanoparticles are metal nanoparticles. [Effects of the Invention]

[0016] To address the above challenges, we use a novel nanostructure, called a vault structure, synthesized using nanoparticles. The name derives from its definition in civil engineering: a multi-arch structure supported by pillars, characterized by a high elastic modulus. According to the first through eleventh aspects of the present invention, nanovault structure units are grown using nanoparticles as scaffolds, enabling vertical repetition and enabling multilayer composites that reinforce the optimized properties observed in single layers. As a result, Si anodes with vault structures simultaneously exhibit high mechanical stability and low lithium consumption during SEI formation, successfully resolving two major challenges in the practical application of Si anodes. This optimal electrochemical performance is associated with a distinct change in mechanical behavior when individual Si pillars coalesce to form a closed arch (but beyond that point, no further growth of an amorphous Si film on top occurs). The introduction of vault structures and arching behavior not only opens up many new possibilities for the design of new materials for batteries, but also has applications in other applications where surfaces are subject to strong and fluctuating stresses.

[0017] New Strategy: No binder Solvent-free Design Flexibility High controllability

[0018] Nanostructure Unit: Nanostructures, a well-known technology in civil and architectural engineering Z-axis repeatability Improved characteristics

[0019] Optimized characteristics: Inner Void Sealed Surface High mechanical stability

[0020] Lithium-ion battery application: large capacity Fast charge / discharge rate High Coulombic efficiency [Brief explanation of the drawings]

[0021] [Figure 1A] FIG. 1A is a schematic diagram showing the columnar growth process of an amorphous Si thin film. [Figure 1B] FIG. 1B is a schematic diagram showing the columnar growth process of an amorphous Si thin film. [Figure 1C] FIG. 1C is a schematic diagram showing the columnar growth process of an amorphous Si thin film. [Figure 1D] FIG. 1D is a schematic diagram showing the mechanical response of the columnar structure of an amorphous Si film to nanoindentation force. [Figure 1E] FIG. 1E is a schematic diagram showing the mechanical response of the vault structure of a Si amorphous film to nanoindentation force. [Figure 1F] FIG. 1F is a schematic diagram showing the mechanical response of the deposited structure of a Si amorphous film to nanoindentation force. [Figure 2A] Figure 2A is a TEM lamella image of sample 54V. [Figure 2B] Figure 2B is a TEM lamella image of sample 216S. [Figure 2C] FIG. 2C is an SEM top view image of sample 54V. [Figure 2D] FIG. 2D is an SEM top view image of sample 216S. [Figure 2E] Figure 2E is an SEM cross-sectional image of sample 54V. [Figure 2F] Figure 2F is an SEM cross-sectional image of sample 216S. [Figure 3A] FIG. 3A shows the surface morphology and elastic modulus of sample 25C. [Figure 3B] FIG. 3B shows the surface morphology and elastic modulus of sample 54V. [Figure 3C] FIG. 3C shows the surface profile and elastic modulus of sample 155S. [Figure 3D] Figure 3D shows a histogram of E values ​​for several samples. [Figure 3E] Figure 3E plots the most weighted E values ​​in the histogram for all samples across thickness h. [Figure 4A] FIG. 4A shows a cross-sectional view along the (110) direction of the growth simulation of Si deposited on nanoparticles. [Figure 4B] FIG. 4B shows a cross-sectional view along the (110) direction of the growth simulation of Si deposited without nanoparticles. [Figure 4C] FIG. 4C shows cross-sectional views of eight examples (eight stages) of growth simulation using nanoparticles, viewed along the (100) direction. [Figure 4D] Figure 4D shows the force-depth curves for eight cases of compression-hold-release loops using a flat diamond-carbon tip at 500K. [Figure 4E] Figure 4E shows the initial linear elastic deformation at a displacement of 2 nm for eight cases, as a linear regression fit (solid line). [Figure 4F] Figure 4F shows the stiffness of the corresponding structure as a function of thickness. [Figure 5A] Figure 5A shows the charge capacity (delithiation of the Si electrode) of Si films grown on TaNS and Cu foil substrates. [Figure 5B] Figure 5B shows the coulombic efficiency of the semiconductor battery for Si films grown on TaNS and Cu foil substrates cycled at 0.5 C and in the voltage range of 0.01 to 1 V. [Figure 5C] Figure 5C shows the charge capacity (delithiation of the Si electrode) of Si films grown on TaNS and Li foil substrates. [Figure 5D] Figure 5D shows the coulombic efficiency of the semiconductor battery at 0.5 C and in the voltage range of 0.01 to 1 V for Si films grown on TaNS and Li foil substrates. [Figure 6] Figure 6 is a schematic diagram of the experimental setup. [Figure 7A] FIG. 7A shows the thickness of Si films deposited on TaNS as a function of sputtering time. [Figure 7B] Figure 7B shows the roughness of Si films deposited on TaNS at different film thicknesses. [Figure 8A]FIG. 8A shows the results of X-ray reflectivity (XRR) measurements of the surface of a plain Cu wafer with a copper oxide layer. [Figure 8B] FIG. 8B shows the results of measuring X-ray reflectivity (XRR) during the growth process of a Si film when Ta nanoparticles were not previously deposited. [Figure 8C] FIG. 8C shows the results of X-ray reflectivity (XRR) measurements during the growth process of a Si film on a Ta nanoparticle (25C) substrate. [Figure 8D] FIG. 8D shows the results of X-ray reflectivity (XRR) measurements during the growth process of a Si film on a Ta nanoparticle (54V) substrate. [Figure 8E] Figure 8E shows the results of X-ray reflectivity (XRR) measurements during the growth process of a Si film on a Ta nanoparticle (110S) substrate. [Figure 8F] FIG. 8F shows the results of X-ray reflectivity (XRR) measurements during the growth process of a Si film on a Ta nanoparticle (216S) substrate. [Figure 9A] FIG. 9A shows the relationship between the atomic percentage concentrations of Si and Ta and the etching time when a sample in which Si and Ta were co-deposited in a matrix was etched by Ar ion etching, as measured by X-ray photoelectron spectroscopy (XPS). [Figure 9B] FIG. 9B shows the Si 2p peak obtained by Ar ion etching of a sample in which Si and Ta were co-deposited in a matrix, as measured by X-ray photoelectron spectroscopy (XPS). [Figure 9C] FIG. 9C shows the 4f peak of Ta obtained by Ar ion etching of a sample in which Si and Ta were co-deposited in a matrix, as measured by X-ray photoelectron spectroscopy (XPS). [Figure 10] FIG. 10 is a scanning electron microscope (SEM) cross-sectional image of sample 25C grown on a Si substrate. [Figure 11A] FIG. 11A shows the elastic modulus histograms of Si sputtered on TaNS for 60 min (sample 110S) for two substrates (Cu, Si). [Figure 11B]FIG. 11B shows the elastic modulus histograms for Si sputtered directly onto a substrate containing no nanoparticles for 60 minutes for two types of substrates (Cu, Si). [Figure 12A] FIG. 12A shows the relationship between elastic modulus and height for Sample 15C and Sample 25C. [Figure 12B] FIG. 12B shows the relationship between modulus and height for Sample 35C and Sample 54V. [Figure 12C] FIG. 12C shows the modulus versus height function for Sample 110S and Sample 155S. [Figure 12D] FIG. 12D shows the relationship between elastic modulus and height for sample 201S. [Figure 13A] Figure 13A shows the shape of the vault structure sample (54V) measured using PF-QNM. [Figure 13B] FIG. 13B shows the elastic modulus of the vault structure sample (54V) measured using PF-QNM. [Figure 13C] FIG. 13C is a histogram of E values ​​of the vault structure sample (54V) measured using PF-QNM. [Figure 13D] Figure 13D shows the shape of the deposited structure sample (155S) measured using PF-QNM. [Figure 13E] Figure 13E shows the elastic modulus of the deposited structure sample (155S) measured using PF-QNM. [Figure 13F] Figure 13F is a histogram of E for the sedimentary structure sample (155S) measured using PF-QNM. [Figure 14A] FIG. 14A shows a cross-sectional view of Si deposition onto a single nanoparticle along the (100) direction by MD simulation. [Figure 14B] FIG. 14B shows a cross-sectional view of Si deposition onto a single nanoparticle along the (110) direction by MD simulation. [Figure 14C] FIG. 14C shows the surface mesh inside the deposited layer (without atoms) in 3D. [Figure 15]Figure 15 shows the state of Si deposited on two adjacent nanoparticles. [Figure 16A] FIG. 16A shows the elastic deformation of the structure shown in FIG. 4C upon unloading. [Figure 16B] FIG. 16B shows the corresponding decompression stiffness of the structure as a function of structure thickness. [Figure 17A] FIG. 17A shows the charge capacity of the Si sample (Si electrode delithiation). [Figure 17B] FIG. 17B shows the coulombic efficiency of the Si sample. [Figure 18A] FIG. 18A is a low-magnification SEM image of Sample 15C after three charge-discharge cycles and scratching with a diamond pen. [Figure 18B] FIG. 18B is a low-magnification SEM image of sample 54V after three charge-discharge cycles and scratching with a diamond pen. [Figure 18C] FIG. 18C is a low-magnification SEM image of sample 155S after three charge-discharge cycles. [Figure 18D] FIG. 18D is a cross-sectional SEM image of Sample 15C, which was subjected to three charge-discharge cycles and diamond scratching to expose the film edge. [Figure 18E] FIG. 18E is a cross-sectional SEM image of sample 54V, which was subjected to three charge-discharge cycles and diamond scratching to expose the film edge. [Figure 18F] FIG. 18F is a cross-sectional SEM image of sample 155S after three charge-discharge cycles. [Figure 18G] FIG. 18G is a high-magnification SEM image of the cycled anode of Sample 15C after three charge-discharge cycles. [Figure 18H] FIG. 18H is a high-magnification SEM image of the cycled anode of sample 54V after three charge-discharge cycles. [Figure 18I] FIG. 18I is a high-magnification SEM image of the cycled anode of sample 155S after three charge-discharge cycles. [Figure 19] FIG. 19 shows the volume expansion rate at 0.5C of the electrode 54V in a lithium half cell. [Figure 20A] Figure 20A is a Nyquist plot of 54V and 15C measured at open circuit after 3 cycles. [Figure 20B] FIG. 20B is a Nyquist plot of 54V and 15C measured at open circuit after 10 cycles. [Figure 20C] FIG. 20C is a Nyquist plot of 54V and 15C measured at open circuit after 50 cycles. [Figure 20D] FIG. 20D is a Nyquist plot of 54V and 15C measured at open circuit after 75 cycles. [Figure 21A] FIG. 21A shows the charge capacity of the Si sample (Si electrode delithiation). [Figure 21B] FIG. 21B shows the coulombic efficiency of the Si sample. [Figure 22A] FIG. 22A is a TEM lamella image of a D54V sample grown on a Si substrate before lithiation-delithiation cycling. [Figure 22B] FIG. 22B is an SEM image of a D54V sample grown on a Si substrate before lithiation-delithiation cycling. [Figure 23A] FIG. 23A shows the elastic modulus of the D54V sample measured using PF-QNM. [Figure 23B] FIG. 23B is a histogram of the elastic modulus of the D54V and 54V samples. [Figure 24A] FIG. 24A shows the charge (delithiation) capacity of the LIB half-cells cycled at 0.5C for the D54V and 54V samples. [Figure 24B] Figure 24B shows the coulombic efficiency of the LIB half-cells cycled at 0.5C for the D54V and 54V samples. [Figure 25] FIG. 25 is an SEM image of sample D54V after three lithiation-delithiation cycles. DETAILED DESCRIPTION OF THE INVENTION

[0022] <1. Summary of the Invention> Nanomaterials that repeatedly swell and deswell have the advantage of being able to accommodate large volume changes due to their internal voids. However, this flexibility comes at the cost of reduced mechanical stability, leading to component degradation and even failure. Here, we identify optimal components for the anode of Si-based lithium-ion batteries (LIBs), fabricate them using ligand- and spill-free cluster beam deposition, and demonstrate their robustness through atomic-level computer simulations. To achieve this, we grew pillar-shaped amorphous Si films on Ta nanoparticle scaffolds via their shadowing effect. Quantitative nanomechanical mapping using PeakForce revealed the critical point where the mechanical behavior changes as the pillars contact to form a vault structure. We found that maximizing mechanical strength is due to arching, a well-known civil engineering concept. This vault nanostructure seals the electrode surface, reduces the electrode-electrolyte interface, and distributes lithiation stress. Furthermore, vertically repeating the nanostructures to form a double-layered structure, like an aqueduct, improved the capacity stability and Coulombic efficiency of LIBs. These results demonstrate the successful use of a macroscale strategy to achieve mechanically stable nanostructures by exploiting arching action at the nanoscale.

[0023] <2. Detailed Description of the Drawings> Figures 1A–1F show the design principles of the TaNS-Si amorphous film composite anode and its structural and mechanical relationships. Figures 1A–1C show a schematic representation of the growth process, which consists of three steps: TaNS deposition, columnar growth of the Si amorphous film utilizing the shadowing effect of TaNS, and thermal annealing at 150°C to improve mobility and eliminate voids on the open surface. The mechanical response to nanoindenter force is also shown for three structures: columnar (Figure 1D), vault (Figure 1E), and stack (Figure 1F). The nanoindenter for PF-QNM measurements is shown on top of the structure, applying a compressive force to the Si sample. In Figures 1D-1F The white arrow indicates the film stress not due to indentation. Open arrow in Figure 1E indicates the film stress measured by the nanoindenter. In Figure 1D The dashed line represents the deformation of the sample. The gray in Figure 1E The small arrows indicate the force distribution within the cylinder.

[0024] Figures 2A-2F show the structural characteristics of the Si thin film grown on TaNS. TEM lamella image of the sample: 54 in Figure 2A. V and 216 in Figure 2B. S The nanoparticle nature of TaNS and the amorphous nature of Si are shown. The observed Pt layer was deposited during the lamella fabrication. SEM top view of the sample: 54 in Figure 2C. V and 216 in Figure 2D. S , and the corresponding cross-sectional view: 54 in FIG. 2E V (Enlarged inside) and 216 in Fig. 2F S .54 V corresponds to a vault structure, with the capitals touching each other. S In the structure, the pillars have already coalesced into a continuous film with Volmer-Weber growth, as indicated by the dome-shaped morphology. The substrate used for the sample preparation for this structure analysis was Si(111).

[0025] Figures 3A to 3E show the results of mechanical properties measured by PF-QNM for Si films of different thicknesses grown on TaNS. The substrate used for sample preparation was Si(111), but similar results were obtained with a Cu foil substrate (Figure 11). Figure 3A shows the mechanical properties of sample 25. C Figure 3B shows the surface shape and elastic modulus of sample 54. V Figure 3C shows the surface shape and elastic modulus of sample 155. S The surface topography and modulus of elasticity of the three specimens are shown. For ease of comparison, the same scale is used for the E-value mapping of the three specimens. VThe E values ​​for the SiO2 films saturate for a high percentage of the mapping. Figure 3D shows a histogram of E values ​​for several samples. The E values ​​increase from columnar to vault structures (top histogram) and decrease for the stacked structures (bottom histogram). Figure 3E plots the most weighted E values ​​in the histograms against film thickness h for all samples. The change in E values ​​with film thickness h corresponds to the transition from columnar to Volmer-Weber growth of Si, the region where the columnar heads meet to form the vault structure.

[0026] Figures 4A–4F show the correlation between morphology and mechanical properties obtained by MD simulations. Figure 4A shows the columnar growth of Si deposited on Ta nanoparticles (replicated in the (100) direction due to periodic boundary conditions) at 500 K, forming a vault structure. The simulation box is sliced ​​along the (110) direction for easier observation. The right image shows only the surface mesh slice (not including atoms) within the deposited layer, revealing the presence of voids. Figure 4B shows a similar simulation performed without TaNS. In Figure 4B, we can see that a deposition structure forms from the start of deposition, with voids spreading throughout. Figure 4C shows eight cases (eight stages) selected from the growth simulation with nanoparticles sliced ​​along the (100) direction. Figure 4D shows the evolution of the compression depth curve during a compression-hold-release loop using a flat diamond-carbon tip at 500 K. Figure 4E shows the initial linear elastic deformation within a displacement of 2 nm, as fitted by linear regression (solid line). Figure 4F shows the stiffness of the corresponding structure as a function of thickness, which is in good agreement with the PF-QNM experimental values ​​in Figure 3C, clearly demonstrating the stiffness of the vault structure.

[0027] Figures 5A–5D show a comparison of the electrochemical properties of Si films grown on TaNS and Cu foil substrates. Figures 5A–5B show a comparison of representative columnar, vault, and stacked structures, while Figures 5C–5D show a comparison of single-vault and double-vault structures containing the same amount of Si. The anodes were assembled into half-cells using Li foils as the reference and counter electrodes and 1.0 M LiPF in a 50:50 (w / w) mixture of EC:DEC (ethylene carbonate:diethyl carbonate) as the electrolyte. Figures 5A and 5C show the charge capacity (delithiation of the Si electrode) of the half-cell cycled between 0.01 and 1 V at 0.5 C, and Figures 5B and 5D show the coulombic efficiency under the same conditions.

[0028] A schematic diagram of the experimental setup is shown in Figure 6. The layered structure was achieved by sequentially depositing a Ta nanoparticle scaffold using magnetron sputtering inert gas condensation and an amorphous Si film using RF sputtering. Figure 6 was created by Pavel Puchenkov using Blender 2.8 (see www.blender.org).

[0029] Figures 7A-7B show the thickness and roughness of Si films deposited on TaNS measured by XRR. Data and experimental details related to the measurements in Figures 7A-7B are provided in Figures 8A-8F. Figure 7A shows the thickness (h) of Si films as a function of sputtering time (t). Square dots represent the measured data, and round dots represent the interpolated thickness at different times. The thickness-time plots are fitted to the linear function provided. Figure 7B shows the roughness of Si films deposited at different thicknesses.

[0030] Figures 8A–8F show the X-ray reflectivity (XRR) measurements of several samples. The substrate used for these measurements was Cu(100). XRR measurements were performed using a Bruker D8 Discover instrument (Bruker AXS GmbH, Karlsruhe, Germany) equipped with a Cu X-ray source with a wavelength of λ = 1.54 Å operating at 1600 W and a Goebel mirror. To minimize the irradiated area at the sample position, a 0.05 mm slit was used to reduce the beam size in the reflecting plane. After careful sample alignment, data were collected from 0.2° to 3° (2θ) at 0.01° intervals. Experimental XRR data were fitted using GenX 2.4.10 software (v.2.4.10, http: / / genx.sf.net). In Figure 8A, a bare Cu wafer surface with a native copper oxide layer is modeled. In Figure 8B, reflection fringes at 2θ > 0.8° are observed for the Si film grown without prior deposition of Ta nanoparticles. These reflection fringes are not observed for the Si film grown on the Ta nanoparticle-coated substrate in Figure 8 (Figures 8C to 8F). Therefore, these reflection fringes are absent on the Ta nanoparticle-coated substrate and are attributed to the very smooth Cu / Si interface. The silicon in Figures 8B and 8D was grown using the same Si deposition time. Figures 8C, 8D, 8E, and 8F show the measurement results for Si thin film samples grown using different Si deposition times.

[0031] 9A to 9C are diagrams showing the estimated proportion of Ta in a Si sample using X-ray photoelectron spectroscopy (XPS). For this measurement, a matrix of samples was prepared by co-evaporating Si and Ta for 60 minutes (thickness of the sample was 110 mm). SXPS spectra were acquired using a Kratos AXIS Ultra DLD photoelectron spectrometer with an Al Ka ​​(1486.6 EV) source and a base pressure of 10-10 mbar. To obtain representative data, XPS measurements were performed with Ar ion etching for various times (ion energy 3 keV). Figure 9A shows the relationship between the atomic percentages of Si and Ta and etching time, estimated from the Si 2p peak in Figure 9B and the Ta 4f peak in Figure 9C. The atomic percentage of Si is approximately 95%, while that of Ta is approximately 0.1%. The difference between the total and 100% is due to O and C (present only in the first measurement). Oxidation of Si may occur during sample transport from the glove box to the XPS instrument.

[0032] Figure 10 shows the structure of sample 25 grown on a Si substrate. C This is a cross-sectional image taken by a scanning electron microscope (SEM).

[0033] 11A-11B show the influence of the substrate on the elastic modulus measurement. FIG. 11A shows the Si (sample 110) sputtered on TaNS for 60 minutes. S ), and Figure 11B shows the elastic modulus histogram of Si sputtered for 60 min directly onto a nanoparticle-free substrate. The E value of the thin film is independent of the substrate when TaNS is present. However, when deposited directly onto Cu or Si(111), the E value is strongly dependent on the substrate. This demonstrates that TaNS (and not the substrate) controls the Si growth when TaNS is present.

[0034] Figures 12A-12D show the modulus vs. height function for all samples. The correlation coefficient (CC) is displayed within each graph. The correlation coefficient for randomly distributed X and Y is a dimensionless indicator of the linear relationship between X and Y and is defined as follows:

number

[0035] 13A to 13F show the mechanical properties of the unannealed sample measured using PF-QNM. The sample fabrication process was carried out in step 3, i.e., at an Ar pressure of 8 × 10 -3 13A, 13B, and 13C are the same as those in FIG. 3, except that the thermal annealing step at 150° C. for 60 minutes at 500 mBar was omitted. V 13D, 13E, and 13F are histograms of surface topography, modulus, and E values ​​for the deposited structure (155 S ) is a histogram of the surface profile, elastic modulus, and E value for the former sample. (54 V ) (average value of about 60-70 GPa), the average value of the thermally annealed samples value Approximately 120 GPa (See Figure 3D) On the other hand, the latter sample shows a large decrease in E value. (155 S ) The control experiment showed that the annealing treatment had no significant effect on the strength of the vault structure, with the average strength being approximately 40-50 GPa, similar to that of the annealed samples. This control experiment confirmed that the thermal annealing treatment, which leads to an increase in the stiffness of individual columns through the migration and subsequent disappearance of voids, is important for strengthening the vault structure.

[0036] Figures 14A–14C show the deposition of Si on a single Ta nanoparticle. To clearly examine the size effect, the simulation box and nanoparticle are larger than those shown in Figures 4A–4F (side length: 16.4 nm, diameter: 8 nm, respectively). Other conditions are the same as in Figures 4A–4F (i.e., the simulation box is replicated along the (100) direction for periodic boundary conditions, and the temperature is 500 K). Again, the Si film follows a columnar growth pattern, forming a vault structure. For clearer observation, Figure 14A slices the simulation box along the (100) direction, and Figure 14B slices the simulation box along the (110) direction. Figure 14C shows a 3D surface mesh of the interior (i.e., atom-free) of the deposited layer, revealing the presence of voids.

[0037] Figure 15 shows Si deposited on two adjacent Ta nanoparticles. To avoid the symmetry seen in Figures 4A–4F, in this set of simulations, the two Ta nanoparticles are intentionally placed in a simulation box twice as long. The rest of the setup is the same as in Figures 4A–4F (i.e., the simulation box is replicated along the (100) direction for periodic boundary conditions, and the temperature is 500 K). Here, too, the Si film follows a columnar growth pattern, eventually forming vault and stack structures. The simulation box is sliced ​​along the (100) direction for clearer observation. Each image is taken from a section of the video where the growth process is clearly visible.

[0038] Figures 16A-16B show the elastic deformation during unloading. In Figure 16A, the stiffness of the corresponding structure is extracted from the slope of the solid line, which is a linear approximation of the curve, as shown in Figure 4C. In Figure 16B, the decompression stiffness as a function of the structure's thickness decreases monotonically, unlike under load. The red dashed line shows the nonlinear approximation of [Equation 2] below.

[0039] Verification of porosity obtained by MD simulation: The porosity of the deposited amorphous silicon structure is strongly influenced by sputtering conditions and temperature, as these parameters determine the mobility of adatoms. The porosity obtained by MD simulation can be estimated by comparing the number density of Si atoms in the porous structure, whose volume can be measured from surface structure analysis, with the reference number density of the bulk amorphous silicon sample. The resulting porosity was approximately 0.3.

[0040] In the elastic region, the deposited structure can be approximated by a spring, whose stiffness is inversely proportional to its length:

number

number

[0041] 17A-17B show the LIB performance of Si samples, showing different behavior depending on the structure and E characteristics. Fig. 17A shows the charge capacity (Si electrode delithiation), and Fig. 17B shows the coulombic efficiency.

[0042] 18A to 18I show the results of the test on sample 15, which was charged and discharged three times between 0.01 and 1 V at 0.5 C. C (Figures 18A, 18D, 18G), Sample 54 V (Figures 18B, 18E, 18H), and sample 155 S The SEM images of the electrodes (Figures 18C, 18F, and 18I) are shown. Imaging the physical state of the electrodes after cycling allows us to explain their electrochemical behavior. After the electrodes were peeled, the half-cell was opened in an Ar glove box, and the anode was washed with dimethyl carbonate and dried under high vacuum. The low-magnification images (Figures 18A-18C) show that the 15C and 54 V Ripples (wave shapes) formed by diamond pen scratches can be seen on the surface. S Similar scratching was not required, and many cracks in the film forming islands were observed. Cross-sections (Figures 18D-18F) show columnar and vault structures at the edges of the ripples, 155 S The peeled membrane is seen at the edge of the cracked island. S The high coulombic efficiency observed at 15°C, accompanied by a significant capacity loss, is due to loss of active material rather than exposure of the fresh electrode to electrolyte. High-magnification top-view images of the anode during cycling (Figures 18G-18I) confirm the formation of pores. These pores are particularly evident at 15°C. C and 54 V These structures are predominant and are thought to be due to channels created by the electrolyte during cycling, possibly related to the activation phenomenon (during activation, more active material is involved in the lithiation process during cycling).

[0043] Figure 19 shows the electrode structure of a lithium half-cell estimated by VL Chevrier et al. (VL Chevrier, L. Liu, DB Le, J. Lund, B. Molla, K. Reimer, LJ Krause, LD Jensen, E. Figgemeier, KW Eberman J. Electrochem. Soc. 2014, 161, A783). V This shows the volume expansion at 0.5°C.

[0044] Figures 20A-20D show the 54 V and 15 C The resistance at high frequencies is related to the electrode / electrolyte interface resistance (SEI and charge transfer), and is C So 54 V This is three times higher than the sample with a columnar structure, which confirms that the high electrode / electrolyte interface increases the SEI.

[0045] Figures 21A-21B show the LIB performance of Si samples, showing different behaviors depending on the structure and E characteristics. Figure 21A shows the charge capacity (Si electrode delithiation), and Figure 21B shows the Coulombic efficiency. Although the Coulombic efficiency value is close to 100%, it is not significant because the decrease in capacity suggests that the sample has fractured.

[0046] 22A-22B show the D54 grown on a Si substrate. V The characteristics of the sample before lithiation-delithiation cycles are shown in Figure 22A, which is a TEM lamella image, and Figure 22B, which is an SEM image. The stacked structure (110 S ) and a double vault structure is clearly visible.

[0047] Figures 23A-23B show the D54 V Figure 23 shows the mechanical properties measured using PF-QNM for the sample. Figure 23A shows the modulus (same scale as used in Figure 3, so most of the mapping is saturated), and Figure 23B shows the corresponding histogram. The histogram shows the D54 V and 54 V Comparison between samples is shown.

[0048] Figure 24A shows the D54 V and 54 V Figure 24B shows the charge (delithiation) capacity of a LIB half-cell cycled at 0.5C for the sample, and Figure 24B shows the coulombic efficiency of the same sample.

[0049] FIG. 25 shows the results of sample D54 after three lithiation-delithiation cycles. V This is an SEM image of a sample grown on Cu foil so that the anode operates as a half-cell. The bilayer structure is maintained after charge-discharge cycling.

[0050] <3. Overview of design strategy and vault structural concept>. The sample was grown directly on the substrate by sequential and independently controlled cluster beam deposition (CBD) of Ta nanoparticles and RF sputtering of Si thin film (Figure 6). This configuration enabled the fabrication of binder-free, high-purity films (grown under high-vacuum conditions) with precise control over the film thickness, nanoparticle size, and nanoparticle shape. First, crystalline Ta nanoparticles were evaporated (Figure 1A, Step 1) to form a porous nanoparticle film. This film served as a nanoscaffold (referred to as TaNS) for fabricating Si anodes. Next, to take advantage of the shadowing effect of TaNS, Si films of various thicknesses were sputtered obliquely at an acute angle onto the TaNS. This resulted in columnar growth of Si in the initial stage (Figure 1A, Step 2). The diameter of the cylinders increased with thickness, and the vertices of the cylinders contacted each other, closing the surface and forming a vault-like structure (Figure 1B). Further deposition of Si resulted in the formation of continuous amorphous films, called stacked structures, through Volmer-Weber (island) growth (Fig. 1C). Subsequent thermal annealing (Fig. 1A, step 3) enhanced the mobility of defects at the surface of each structure, ultimately leading to their annihilation. In the case of the columnar structures, this process improved their rigidity. In contrast, in the stacked structures, the barrier to void movement to the free surface was significantly higher, so the voids remained confined within the Si layer, resulting in a sponge-like, porous film.

[0051] Variations in the structure of Si films significantly affect their mechanical properties. In our PeakForce quantitative nanomechanical mapping (PF-QNM) measurements, a nanoindenter applied a vertical force (Figures 1D–1F, gray arrows) to the top of each structure. In the columnar structures, compressive stress maintains each column's shape and individuality (Figure 1D, open arrows). The columns deform easily until they contact each other, where a clamping effect prevents further deformation. Therefore, as the film thickness increases, the columns become closer together, resulting in an increased E value. As an extreme example, in the vault structure (Figure 1E), the nanoindentation force immediately clamps the column without requiring any initial deformation, resulting in a high E value. This effect is similar to the arching action described in civil engineering, where an arch pushes outward, transmitting stress to the ground, which then generates a counterforce. In the deposited structure (Figure 1F), the E value is related to the surface domes, which facilitate the diffusion of Si atoms at the tops of the hills into the valleys under the nanoindentation force. They are also affected by pores that soften the Si layer. As a result, a low E value is recorded.

[0052] The vault structure can be used as a nanostructural unit that can dissipate lithiation (or other) stresses while avoiding the cracking observed in electrodes based on stacked structures. Vertically repeating the vault structure (i.e., repeatedly stacking vault nanostructures on top of another layer, with each layer forming a single vault nanostructural unit) can create thin-film electrodes with optimized mechanical and surface stability during battery cycling while increasing the amount of Si active material. Most importantly, the concept of nanovault architecture as a repeating nanostructural unit is broadly applicable to the design of novel materials requiring high stress tolerance.

[0053] 4. Correlation between morphology and mechanical properties In the following, the name of the sample is h XThe format is as follows: h is the film thickness (unit: nm), and X indicates the type of structure as follows: C (columnar structure), V (vaulted structure), S (sedimentary structure).

[0054] Nanoporous TANS is a crystalline Ta nanoparticle (diameter 3 nm, narrow particle size distribution) from It is composed of CBD, which provides a soft landing and maintains its properties. 。 Transmission electron microscopy (TEM) lamella images (Figures 2A and 2B) show that the TaNS is approximately 10 nm thick, confirming the amorphous nature of the overlaid Si layer. The thickness of the Si thin film increases linearly with time at a rate of 1.69 nm / min (via XRR). Figure 7A ) Estimated by X-ray photoelectron spectroscopy (XPS, Figures 9A-9C), the Ta content is less than 0.5 atomic %, but the Ta is in a granular form (Figure 2 C~ 2D), resulting in high roughness in the Si film (in the range of 3.3–3.8 nm compared to 0.8 nm for Si grown directly on the substrate, Figure 7B ) occurred.

[0055] Sample 54 V As shown by cross-sectional scanning electron microscopy (SEM, Fig. 2E), the TaNS exhibits columnar structures of increasing diameter, resembling inverted truncated cones that meet at the top, forming a vault structure. This columnar structure (which begins to form early in the Si growth stage, Fig. 10, 25C) is due to the shadowing effect of TaNS, as it perturbs the incident beam of Si atoms. The columnar structure is formed by 216 S This is also observed at the bottom of the vault layer (Figure 2F). However, when additional Si is deposited on top of the vault layer, the amorphous nature of the sputtered Si causes the Si layer to become continuous without long-range structural ordering. A dome then forms on top of the continuous film, reducing the local surface energy.

[0056] The domes initially endure compressive stress, but at a certain limit, this stress changes to tensile stress. This causes Si adatoms to diffuse from the hilltops to the valleys. As the domes coalesce, the thickness of the continuous film increases. This is a typical phenomenon in Volmer-Weber (island) thin film growth.

[0057] Sample 25 C , Sample 54 V , sample 155 S Surface morphology and elastic modulus mapping of TaNS samples were performed by PF-QNM using an atomic force microscope (AFM) operated in PeakForce tapping mode, measuring nanoscale roughness (Figure 3A). For these measurements, Si samples were fabricated on Si(111), but E measurements are valid for any substrate when TaNS is deposited between the substrate and sputtered Si (Figures 11A-11B). Samples with all deposition structures showed very strong correlations (approximately 1) between surface morphology and E value characteristics (Figures 12A-12D). Thus, E value roughness is related to surface morphology. However, the correlation is reduced for vault structures (0.8) and significantly reduced for columnar structures (dispersed in the range 0-0.6). This suggests that factors other than surface morphology are involved in E value roughness, possibly due to the presence of honeycomb-like density-depleted regions (void networks) between the columnar structures and the void network. V This is likely related to the presence of an incomplete arch.

[0058] The distribution of the mapped E values ​​is shown in a histogram (Fig. 3 D ), and it can be seen that the film thickness of each sample correlates with the peak of the E value (Fig. 3 E The sample with a columnar structure exhibits a polynomial increase with film thickness, and C and 25 C At pressures of approximately 40 GPa, the E values ​​were almost the same, which is probably related to TaNS. The E value increased with film thickness, and the pillars came into contact with each other to form multiple arches, showing very high E values ​​up to 250 GPa. In the deposition structure, E values ​​similar to those of the columnar structure (approximately 40 GPa) were observed regardless of film thickness.

[0059] In a control experiment in which step 3 of the fabrication process was omitted, the migration and subsequent disappearance of voids confirmed the importance of thermal annealing in strengthening the columnar (not vault) structure (Figures 13A-13F). The strong correlation with topography mapping suggests that the E value measurements are limited to the dome-like structure of the Si surface. This differs from the columnar and vault structure samples, in which the granular surface represents the stigma.

[0060] 5. Explanation of structural and mechanical properties through atomic simulations We performed molecular dynamics (MD) simulations to mimic the film deposition process of TaNS-Si film composites on a rotating substrate holder. As shown in the left panel of Figure 4A (Figures 14A-14C, Figure 15), in the presence of Ta nanoparticles, the nanoparticles were shadowed by the deposited Si atoms, leading to the formation of columnar structures on the Ta nanoparticles. As the deposition time increased, the columns merged, first forming vault structures and finally stacked structures, in good agreement with the experimental results. In contrast, in a control simulation without nanoparticles (Figure 4B), stacked structures were formed from the early stages of deposition. This clearly demonstrated that the shadowing effect of TaNSs is essential for the formation of the vault structures.

[0061] As shown by the gray surface meshes in the right panels of Figures 4A and 4B, surface structure analysis highlighted the location and size of voids within the deposited structures. Within the same simulation, large voids formed only within the deposited region, increasing the estimated porosity from 0.09 to 0.3. This can be explained by the finite-size effect of the columnar structure. Small vacancies near the open surface can be quickly filled by a few displacements of surface atoms or newly deposited atoms. On the other hand, large vacancies formed below the surface of the deposited structure require the collective migration of Si atoms, which is significantly slower. Furthermore, because the activation enthalpy (migration barrier) for self-diffusion in amorphous Si is approximately 2.7 eV, we found that once large voids formed in the deposited region, they remained stable even under annealing conditions.

[0062] The mechanical properties obtained from the PF-QNM measurements were explained by simulation of a compression-hold-release loop. As shown in Figure 4C, we selected eight examples (eight stages) using nanoparticles from the first simulation group. The evolution of the feedback force can be tracked with time and the displacement of the plate on the Si layer. As shown by the dashed lines in Figure 4D, the feedback force is clearly differentiated for each stage. From the force-displacement curves, we extracted the stiffness of the structure during the loading stage (shown by the solid line in Figure 4E).

[0063] As shown in Figure 4F, Clearly, the stiffness reached its maximum when the pillar-like structure evolved into a vault structure. This is very similar to the PF-QNM measurement results in Figure 3C. This can be explained by the fact that in the initial elastic deformation region (1 nm to 2 nm displacement), the vault structure (i.e., the 20.8–26.3 nm cases) reaches the largest contact area, combines the smallest porosity with a relatively small thickness. The stiffness extracted from the decompression stage (where the contact area is almost the same for all structures) is 0.3, indicating a spring-like behavior (Figures 16A–16B).

[0064] 6. Vault structure as a component of LIB electrodes The commonly reported failure of Si anodes is associated with volume expansion (300–400%) due to mechanical stress accumulated in the electrode. This effect limits the thickness and size of Si thin films and nanoparticles used as anodes in LIBs. Although Si thin films sputtered on rough substrates can increase the electrode stability during cycling, the film thickness remains limited.

[0065] The improved mechanical strength provided by the arching action helps overcome this limitation. C , Sample 54 V , sample 155 SCharging cycles at 0.5 C (Figure 5A) are representative of each sample type (Figures 17A-17B), demonstrating the crucial role of nanostructure. V showed the highest capacity in the first 60 cycles, and 15 C Both samples show an initial capacity increase associated with the activation phenomenon ( Figure 17A ). 54 V Although the 155% expansion rate (Fig. 19) was achieved, the capacity retention was similar to that of the columnar structure, but the coulombic efficiency was higher. S A fast capacity fade was observed, especially in the first 15 cycles, which was attributed to the loss of active material due to delamination from the substrate, as evidenced by SEM (Figures 18A-18I). S The Coulomb efficiency of sample 54 was as high as 96–100% (Fig. 5B). V is 85-96%, 15 C This can also be explained by the fact that the ratio of the solubility of the sample 54 is 60-85%. V and sample 15 C The Coulombic efficiency of 54 reflects the irreversible consumption of lithium due to side reactions during charging (usually SEI formation), and the low values ​​during cycling represent exposure of fresh electrodes to the electrolyte due to crushing. V The vault structure of the present invention creates a seal between the electrode and the electrolyte, reducing lithium consumption due to side reactions.

[0066] Electrochemical impedance spectroscopy (EIS, Figures 20A-20D) revealed that the resistance associated with electrode / electrolyte interfacial phenomena (SEI and charge transfer) was 15 C So 54 V It was confirmed that the number of cases was three times higher than that of the 54 V The sealed surface of the ZnO electrode allows for high-rate charge / discharge rates similar to those of columnar structures while protecting the surface (Charge / discharge and Coulombic efficiency plots at 5 C are shown in Figures 21A-21B). This is consistent with recent chemical-mechanical models that showed that anode materials with high E values ​​can sustain high lithiation stresses before collapsing.

[0067] Because the vault structure is generated by TaNS, it can be constructed substrate-independently. This was demonstrated by depositing a second layer of TaNS on top of an existing Si vault structure, followed by further deposition of Si. The second TaNS layer prevents the pillars from coalescing towards the deposited structure, instead generating a second vault structure (Figures 21A-21B). This is referred to as a double 54 V D54 named after the layer V It is written as D54. V The sample also exhibits a heterogeneous E value profile, with values ​​in the same range as observed for the single layer (Figures 23A-23B).

[0068] To exclude the film thickness effect, sample D54 V The electrochemical performance of the deposited sample 110 containing the same amount of Si was compared. S (i.e., TaNS monolayer) (Figures 5C-5D). V The initial capacity of 110 was 3230 mAh / g, which increased by about 5% in the first cycle due to activation, and then showed 88% capacity retention after 100 cycles (2832 mAh / g). S The sample showed a strong activation phenomenon, reaching a maximum capacity of 2700 mAh / g after 20 cycles and decreasing by 45% (1477 mAh / g) after 100 cycles. V is a single layer 54 V The electrochemical performance is also improved compared to the previous study, with comparable capacity for the first 50 cycles, but higher capacity retention and coulombic efficiency, reaching values ​​of 90–98% (Figures 24A–24B). The bilayer structure strongly improves the mechanical stability of the electrode, and this structure is maintained even after charge–discharge cycling (Figure 25).

[0069] The electrochemical response of the studied electrodes is strongly linked to their mechanical and structural properties. The single-layer vault structure exhibits the highest E value due to its arching effect, which can transfer compressive stress to the pillars. This disperses stress from the lithiation process and prevents film fracture, which would otherwise compromise the stacked structure. Furthermore, sealing the electrode significantly reduces side reactions and SEI formation, which are problematic in pillar-type structures. After 20 cycles, the capacity decreases, with a maximum Coulombic efficiency of 95%. This is likely due to the fact that, despite the clamping effect in the lateral direction of the film, there is free space for expansion in the vertical direction, which may lead to mechanical failure in the long term. However, repeating the vault structure imparts a new clamping effect in the vertical direction, improving both capacity retention and Coulombic efficiency and overall mechanical stability.

[0070] <7. Conclusion> A vault structure with arching action has been introduced for the first time at the nanoscale. While this structure demonstrates the potential for a new design of Si anodes for LIBs, it can also be applied to other materials and applications where surface mechanics play an important role. The arching action is observed when the pillars come into contact with each other during columnar growth, forming a vault-like seal around the anode. This favors stress distribution, preventing cracking of the Si electrode during cycling and effectively suppressing capacity fading. The formation of a SEI is significantly reduced compared to a columnar structure, minimizing lithium consumption and improving Coulombic efficiency while maintaining the advantages of the columnar structure, such as high charge / discharge rates. Repeating this nanostructure unit along the vertical axis results in a material with a stable surface that can effectively release imposed stress. This is confirmed by the construction of a double aqueduct-like vault structure, which improved sealing, higher Coulombic efficiency, and mechanical stability, indicating the reinforcement of the arching action. Most importantly, this novel nanostructure design has room for further optimization. For example, different deposition techniques can be used to scale up production, and the nanoparticle scaffolding material can be modified.

[0071] The nanostructures, called nanovaults, are grown using scalable physical methods. The nanovault structure is made up of columnar membranes (the columns meet at the top to form an arch) with optimized properties. The presence of extensive nanovoids and sealed surfaces provides the material with high mechanical stability for membranes subjected to strong and fluctuating stresses. Growing nanovault structures using nanoparticle scaffolding allowed for vertical repetition of the composite in multilayers, reinforcing the optimized properties observed in the monolayer. When the nanovault structure was applied to a Si film, which is the anode material of lithium-ion batteries, high Coulomb efficiency, fast charge / discharge response, and improved cycle performance were confirmed.

[0072] <8. Experiment> <8-1. Sample preparation> All samples were prepared using a vapor deposition system (Mantis Deposition Ltd) under high vacuum (2.0 × 10 -8 The nanoparticles were deposited on a rotating holder (2 rpm for all depositions) at a pressure of 1.5 × 10 mbar, and were arranged to ensure uniform film deposition. For the deposition of Ta nanoparticles, an Ar gas flow rate of 60 standard cubic centimeters per minute, a DC magnetron power of 45 W, and a coagulation zone length of 100 mm were selected. The Si thin films were deposited using a 110 W RF sputtering source at an Ar pressure of 2.1 × 10 -3 Depositions were performed using a 1000 keV (1000 keV) argon (1000 keV) solution at 1000 keV / cm². Magnetron sputtering targets, silicon (n-type, >99.999% purity, resistivity <0.001 Wm) and tantalum (>99.95% purity), were purchased from Kurt J. Lesker. All depositions were performed at ambient temperature (approximately 298 K, measured with a substrate holder thermocouple) and no external bias was applied to the substrate. Finally, all anodes were deposited at 8 × 10 -3 Annealing was performed at 150°C for 60 min under an Ar pressure of 1 mbar.

[0073] <8-2. Characterization using FIB-SEM and TEM lamellae> Cross-section and sample surface imaging were performed using an FEI Helios G3 UC FIB-SEM, combining focused ion beam (FIB) milling and SEM. TEM lamellae were fabricated using the same FIB-SEM system equipped with a Pt deposition needle and an OMNIPROBE® extraction needle. The lamellae were fabricated using the conventional "H-bar" technique by cutting two grooves into the sample and in situ depositing a thin Pt strip into them to protect the desired surface area during milling. The resulting thin slab was thinned to approximately 40 nm to allow passage of the TEM beam. The extraction needle was used to transfer the lamella onto the top of a TEM half-grid, where it was deposited with the Pt deposition needle. TEM lamellae were imaged using an FEI Titan Environmental TEM equipped with a spherical aberration image corrector at an operating voltage of 300 kV.

[0074] <8-3. Measurement in PF-QNM> The surface morphology and elastic modulus of a Si sample were measured using an AFM (Multimode 8, Bruker) operating in peak force tapping mode. Sample imaging and PF-QNM™ measurements were performed using a Bruker diamond-tipped ultra-high force cantilever (DNISP-HS) (resonant frequency approximately 71.5 kHz, spring constant 432 N / m, nominal tip radius approximately 40 nm). The standard relative method was used to measure the E value, with fused silica (nominal E value: approximately 72 GPa) used as the reference sample. For this method, the cantilever deflection sensitivity was first calibrated by approximating the linear portion of the force-distance curve in ramp mode for a hard sapphire sample. The spring constant value for the measurement was set to 432 N / m from the calibration sheet provided by the manufacturer. Next, a reference sample (quartz) was loaded and PF-QNM measurements were performed, and the peak force setpoint was adjusted to obtain the desired deformation (1–2 nm). The tip radius parameter was then adjusted to accurately measure the elastic modulus of the reference sample (72 GPa). PF-QNM measurements were then performed on the TaSi sample, and the peak force setpoint was adjusted to match the deformation of the reference sample. AFM images (512 × 512 pixels) were acquired at a scan rate of 0.5 Hz and analyzed using Nanoscope Analysis (Ver. 9) software. No reduction processing was performed on the modulus mapping images or data, but quantitative property measurements were performed using standard relative methods. Tip cleaning was performed by forming an indentation on the gold surface and individually calibrating the tip for each sample.

[0075] In PF-QNM, the z-piezo sensor struck the sample surface, and the force-distance curve was measured for each imaging pixel. From the force-distance curve, the DMT model was applied to the section where the tip was in contact with the sample, and the E value was calculated using the following equation.

number

[0076] Reduced elastic modulus (E r ) is related to the Young's modulus (E) of the sample as follows:

number

[0077] <8-4. Electrochemical characterization> For electrochemical characterization, samples were prepared on copper foil (0.25 mm thick, Puratronic 99.9985%, Alfa Aesar). Electrochemical characterization was performed using a two-electrode Swagelok cell with metallic lithium foil as the reference and counter electrodes. The electrolyte consisted of ethylene carbonate (EC, >99%), diethyl carbonate (DEC, >99%), and lithium hexafluorophosphate (LiPF6, >99.99%) purchased from Sigma-Aldrich. The separator consisted of lithium foil (0.38 mm thick, 99.9%) from MTI Corporation and Celgard (25 mm thick). The electrolyte was a 50:50 (w / w) mixture of EC and DEC with 1.0 M LiPF6. All cells were assembled in an Ar glove box (UNICO) with O2 and humidity below 0.25 ppm. Charge-discharge measurements were performed using two 8-channel battery analyzers (0.005–1 mA and 0.02–10 mA, up to 5 V, MTI) in the voltage range of 0.01–1 V. For charge-discharge rate calculations, 1 C was defined as 3579 mAh / g. After three cycles at 0.5 C, the half-cell was opened in an Ar glove box, and the anode was washed three times with dimethyl carbonate (DMC, >99%) and then vacuumed under high vacuum (1.0 × 10 -6 The sample was dried under 1000 mbar for more than 12 hours. Before being introduced into the FIB-SEM, the sample was taken outside and scratched with a diamond pen.

[0078] <8-5.Calculation method> We performed two-group MD simulations with the objectives of (i) elucidating the formation mechanism of the vault structure and (ii) explaining the changes in the mechanical properties of the structure with the growth stage.

[0079] In the first group, we simulated the deposition of silicon layers with and without nanoparticle scaffolds. First, amorphous Si substrates were fabricated by a rapid heating (3000 K, 100 ps)-quenching (500 K, 100 ps) process in an isothermal-isobaric ensemble at 0 Bar. The size of the annealed simulation cell was initially 109 × 109 × 55 Å. Next, the top surface was opened, an atomic layer within 6 Å was pinned to the bottom surface, and an additional 50 ps of relaxation was performed in the canonical ensemble. For the nanoparticle-scaffolded structures, nanoparticles were spontaneously deposited on the amorphous silicon substrate. 5-nm-diameter rhombic lattice silicon nanoparticles were placed horizontally at the (0,0) position 15 Å above the surface.

[0080] Silicon nanoparticles were annealed at 500 K for 50 ps, ​​then accelerated to an additional velocity of 20 m / s to land on the substrate, where they were allowed to relax for another 50 ps. Film growth was simulated by adding new Si atoms from the top of the cell every 200 MD steps. A total of 215,563 Si atoms were added (at an average deposition rate of 1.1 nm / ns). To mimic the rotating substrate of the experimental setup, the initial velocity of the deposited atoms was set to rotate at 1 rev / ns, and 20 ps steps were performed. The angle of incidence was 30° from the surface, and the total velocity was 1,000 m / s. No velocity scaling was applied to the non-deposited atoms. The temperature of the deposited atoms (except those anchored to the bottom surface) was controlled by applying a Langevin thermostat to a group of deposited atoms located 1 nm below the open surface. This group was updated every 2.8 ns, and the deposited atoms were measured after the update. The simulation lasted approximately 40 ns with a 1 fs time step. We also performed benchmark simulations using the same procedure as above, with varying nanoparticle and cell sizes and temperatures, or with two nanoparticles explicitly used.

[0081] The second group of simulations involved mechanical measurements with a simulated AFM tip. Deposited structures of different thicknesses (after deposition times of 5, 10, ..., and 40 ns) were selected and relaxed at 500 K for 200 ps. A flat diamond-carbon plate with a thickness of 5 Å was then placed 5 Å above the surface. Each AFM measurement was simulated (0.01 nm / ps for 400 ps) until a loading depth of approximately 3.5 nm was reached, followed by a step of holding the tip at a fixed position for 600 ps. Finally, the tip was retracted for 400 ps at a rate of 0.002 nm / ps. The change in feedback force was recorded.

[0082] All simulations were performed using the classical MD code LAMMPS, and results were visualized using OVITO. We used an environment-dependent interatomic potential (EDIP) parameterized for silicon. EDIP was tested for bulk phases, defects, and phase transitions, and significantly outperformed other existing potentials for silicon, including the common Stillinger-Weber and Tersoff potentials. EDIP has been used to study liquid-amorphous transitions, self-diffusion, crystal plasticity, brittle fracture, solid-phase epitaxial growth, and amorphous structures. For C-Si interactions, we used a purely repulsive Ziegler-Biersack-Littmark (ZBL) potential, which allowed us to obtain non-sticky feedback force curves.

Claims

1. A composite nanostructure, comprising: A columnar film of amorphous Si grown on a layer of Ta nanoparticles, which are metal nanoparticles, in which the columns contact each other at the top to form an arch shape. A composite nanostructure having:

2. A multilayer composite comprising at least two repeats of the composite nanostructure of claim 1 in a direction perpendicular to the surface of a substrate.

3. A method for producing a composite nanostructure, comprising: a) depositing metal nanoparticles, Ta nanoparticles, from a vapor phase onto a substrate; b) growing a columnar film of amorphous Si on the Ta nanoparticle layer; and In step b), the diameter of the pillars of the pillar-shaped membrane increases with thickness until the capitals come into contact with each other, and the surface closes via the formation of an arched structure.

4. The manufacturing method according to claim 3, In the step b), the growth of the columnar film is stopped immediately after the arch is formed.

5. The manufacturing method according to claim 3, c) a step of thermally annealing the columnar film The method of manufacturing further comprises:

Images