Preparation of a Flexible Reduced Graphene Oxide-Si Composite Film and Its Application in High-Performance Lithium Ion Batteries

: This paper describes a strategy for preparing free-standing reduced graphene oxide@Si nanoparticles (rGO@Si NPs) composite membranes. Graphene oxide (GO) was reduced and self-assembled synchronously with nanoparticles of silicon (Si NPs) on a metal surface and the composite ﬁlm was subsequently used in a lithium-ion battery (LIB). This work describes several important novel aspects of the reported technology. Firstly, the composite membrane has a ﬂexible self-supporting structure, allowing it to function as an anode material without requiring binders and current collectors. Secondly, the successful assembly of Si NPs and reduced Graphene oxide (rGO) sheets has enabled the production of the rGO@Si NPs composite ﬁlm with high controllability and orderliness. Thirdly, the conductive nature of graphene has signiﬁcantly decreased the resistivity while enhancing the electron transport capacity of the battery anode. Lastly, the robust and ﬂexible structure of the graphene sheet has greatly mitigated the large volume variation in Si NPs during charging or discharging, resulting in the rGO@Si NPs composite ﬁlm exhibiting excellent energy density and high-power density.


Introduction
Lithium-ion batteries (LIBs) are one of the most widely used rechargeable batteries today because of their high energy density, long cycle life, and low self-discharge rate [1][2][3].The energy density of LIBs is mainly determined by its specific capacity, which is dependent on anode/cathode electrode materials [4].Till now, the most commercially available LIBs are made with graphite as the anode material and lithium metal oxides/lithium iron phosphate as the cathode material.The theoretical capacities of these anode and cathode materials are 200 mAh g −1 and 372 mAh g −1 [5,6].The theoretical capacities of these commercially available LIBs are far from meeting the requirements for the demands for higher energy density batteries used in new-type electronic devices, such as electric vehicles.
Silicon (Si) is one of the most promising alternatives to usual anodes for its high specific capacity (4200 mA h g −1 ) [7].Which is attributed to the fact that Si can accommodate 4.4 lithium atoms corresponding to the formation of Li 4.4 Si alloy [8].In addition, Si has a low delithiation potential (~0.3 V) [9].However, the commercialization of Si-based materials are subject to the large volume expansion (>300%) and poor electrical conductivity [10][11][12][13][14][15].This volume expansion, as Li ions enter and leave the Si lattice, shortens the cycle life significantly [16].Moreover, the poor electrical conductivity further contributes to low capacity [17].
In the past few years, a significant number of achievements have been proposed to overcome the drawbacks of Si as anode materials.Moreover, several reviews have explored the development of Si anode materials for LIBs from different perspectives [18].Some Crystals 2023, 13, 547 2 of 13 studies have proved Si with smaller diameters, such as nanowires [19,20] nanorods [21,22], and nanotubes [23,24] can reduce the distance between charge and ion transport, thereby enhancing the battery's performance [25][26][27][28].Furthermore, to prepare Si-based composite with various carbon buffer phases is a more successful method for its higher specific capacity than graphite, and has better capacity retention than bare silicon.Typically, Gong et al. [29] prepared a yolk-shell Si/C composite with graphite, which exhibited a reversible capacity of 600 mA h g −1 after 70 cycles, and the first coulomb efficiency was ~50%.Kim et al. [30] reported a Si/graphene sheet/carbon composite with a capacity of 900 mA h g −1 after 70 cycles, and the first coulomb efficiency of 67.3%.Tang et al. [31] constructed a polymer-derived carbon-coated Si composite, and it showed a reversible capacity of 1460 mA h g −1 after 150 cycles.Lu et al. prepared a Si/C composite with a graphene-based core/double-layer carbon structure, and the capacity retention was 89.5% (1200 mA h g −1 ) after 240 cycles [32].Li et al. prepared a carbon/graphene double-layer coated-silicon composite, it maintained a discharge capacity of 1500 mA h g −1 after 300 cycles [33].As a highly conductive flexible material, rGO has been effectively incorporated in a variety of formats to enhance the cycling stability of silicon.Where rGO provides a conductive channel for a high speed electron-transfer, optimizes electrical contact between the electrode components and also act as a buffer.In addition, the structure of graphene sheets can be modified via various reduction methods [34,35].
In this paper, we have developed a self-supporting Si/C composite film network for LIBs.Sol-gel method, magnesiothermic reduction method, and metal reduction method were employed to obtain a flexible self-supporting structure.At a high current density of 2000 mA g −1 , the specific capacity of the material is as high as 876.4 mA h g −1 .After 100 cycles at a current density of 200 mA g −1 , it can still provide a discharge capacity of 1003.4 mA h g −1 .Although its capacity is slightly lower than that of some Si/C composites, our design has the following advantages: First of all, compared with the traditional Si/C composite electrode material, our material does not need to add any binder, because its self-supporting structure means that the material does not need to adhere to the current collector during use, which simplifies the preparation process, reduces the manufacturing cost, and provides favorable conditions for future commercialization [36].Secondly, due to the conductivity of the self-supporting structural material, our material does not need to use a current collector, thereby reducing the volume and weight of the battery and further improving the energy density [37].Third, compared with the current commercial lithiumion batteries, the composite structure in this paper has high flexibility, can withstand certain bending stress without cracking, and can adapt to different shapes [38].Finally, the flexible material has a unique cross-linked layered structure, which can buffer the volume effect of Si NPs during the cycle and prolong the life and performance of the battery.Therefore, reasonable structural design of silicon nanoparticles and graphene is carried out to form a composite material with unique flexible self-supporting structure and excellent electrochemical performance.Moreover, it is no longer necessary to use a current collector and add bonding, which reduces the process flow and saves the production cost.Therefore, we believe that this unique Si/C composite film material will have a wide application prospect and is expected to become an candidate for future battery materials.

Preparation of Nano-SiO 2
To prepare the nano-sized SiO 2 material, first, 60 mL of anhydrous ethanol and 6 mL of ethyl silicate (TEOS, Wuhan Canos Technology Co., Ltd., Wuhan, China) solution were combined, stirred for 20 min, and labeled solution A. Next, 60 mL of anhydrous ethanol, 4 mL of deionized water (DI) and 10 mL of ammonia were combined and stirred for 20 min, and labeled solution B. Solution A and solution B were combined and stirred at a constant speed for 5 h in a 40 • C water bath.Then, 4 mL TEOS was added to the mixed solution which was stirred for 4 h.The solution was first washed several times with distilled water until the pH = 7, then several times with anhydrous ethanol (no separating funnel was used in the process), and dried in a vacuum at 60 • C for 12 h to obtain the precursor nano-SiO 2 powder.

Preparation of Si NPs
A magnesium thermal reduction method was used to produce the Si NPs.First, 0.96 g of metal magnesium powder (Tianjin Beichen District Fangzheng Reagent Factory, Tianjin, China) and 0.6 g of the nano-SiO 2 powder were mixed and this mixture was placed into the tube furnace in an Ar environment (Shanghai Lichen Instrument Technology Co., Ltd., Shanghai, China).The material was heated to 650 • C for 5 h at a heating rate of 5 • C min −1 .Then continued to heat up to 850 • C at a rate of 5 • C min −1 and held for 4 h.The material was then cooled naturally under Ar.The resulting product was pickled with hydrochloric acid at a concentration of 2 mol L −1 for 5 h then washed with distilled water and anhydrous ethanol repeatedly.Si NPs was obtained via a vacuum drying of the as-prepared materials at 60 • C for 12 h.

Preparation of rGO@Si NPs Composite Films
The rGO@Si NPs composite films were prepared from a GO aqueous dispersion.A modified version of the Hummers process was used to produce the GO aqueous dispersion (10 mg mL −1 ) [37].The high specific surface area of the GO sheets in the dispersion made it a suitable surfactant to disperse the nanostructured Si NPs.The procedure consisted of first, the precursor solution was prepared by adding 21 mL of the GO dispersion to 90 mg of the Si NPs which was magnetically stirred for 1-2 h at room temperature to ensure uniform dispersion of silicon nanoparticles.Next, a Zn plate was submerged in the aforementioned precursor solution and incubated in the solution at room temperature for one hour.The GO solutions containing Si NPs were reduced on the surface of the Zn plates which formed a film on the metal surface.This rGO@Si composite films film was then rinsed with DI water to remove substances that were physically adsorbed on the surface, and the composite membrane was submerged in DI water for 30 min to further eliminate the physically adsorbed complex.The composite film was then submerged in a solution of 0.5 mol L −1 H 2 SO 4 and placed onto the surface of metallic Zn.The resulting composite film electrode material was peeled off the Zn.The rGO@Si NPs composite films were obtained by immersing the stripped composite membrane in a 0.5 mol L −1 H 2 SO 4 solution to remove the Zn ions contained in the composite film.After further washing and freeze-drying for 24 h, a flexible rGO@Si NPs composite film was obtained and labeled rGO@Si NPs-1.These steps were repeated, and 0, 140, 210, and 315 mg Si NPs were added to 21 mL of GO dispersion to obtain composite film electrode materials, which were labeled rGO (0 mg), rGO@Si NPs-2(140 mg), rGO@Si NPs-3(210 mg), and rGO@Si NPs-4 (315 mg), respectively.

Materials Characterization
The cross-section morphology and microstructure of the composite film were characterized by field emission scanning electron microscopy (FE-SEM, JSM7500F, Shanghai Baihe Technologies Corporation, Shanghai, China); transmission electron microscope (TEM, JEM-2100F, JEOL Ltd., Tokyo, Japan).The crystal structure of the composite films was characterized using X-ray diffraction (XRD, D8 Advance, XI'AN Wilon Industrial Development INC., Xi'an, China).The elemental composition of the composite film surface was characterized using X-ray photoelectron spectroscopy (XPS, Thermo K-Alpha, Al Kα, Thermo Fisher Scientific (CHINA) Co., Ltd., Waltham, MA, USA).Thermogravimetric analysis (TG) was conducted on the films using a Netzsh TG 209 F1 (Netzsh Scientific Instruments TRADING Ltd., Shanghai, China), from 25 • C to 800 • C at a heating rate of 10 • C min −1 in an ambient air environment.

Electrochemical Measurements
Evaluation of Li + storage performance of the composite films was conducted using coin-type half cells (CR2032, Guangdong Canrd New Energy Technology Co., Ltd., Guangdong, China).The rGO@Si NPs composite films and pure-Si were used as the working electrodes, respectively.A lithium foil was used as the counter electrode.1 M LiPF 6 in ethylene carbonate (EC) and diethyl carbonate (DEC) (1:1 in volume ratio) was used as the electrolyte.A Celgard 2400 film was used as the separator.A CT 2001A (LAND, Wuhan Bateri Technology Co., Ltd., Wuhan, China) automatic battery testing system was employed to assess the cycle performance and rate capabilities of the coin cells over a voltage range of 0.01-2.0V (vs.Li + /Li) at ambient temperature.A CHI760E electrochemical workstation ((Shanghai Chenhua Instrument Co., Ltd., Shanghai, China) was used to conduct cyclic voltammetry (CV) tests at a scanning rate of 0.2 mV s −1 in the voltage window range of 0.01-1.5 V (against Li/Li + ).A CHI660E electrochemical workstation (Shanghai Chenhua Instrument Co., Ltd., Shanghai, China), was used to conduct electrochemical impedance spectra (EIS) testing on the coin cells from 10 −2 -10 5 Hz.The temperature was maintained at 26 • C for the duration of the measurement.

Results and Discussion
The synthesis process used to produce the nano-SiO 2 , Si NPs, and self-supporting rGO@Si NPs composite films is shown in Figure 1a-c.
velopment INC., Xi'an, China).The elemental composition of the composite film surface was characterized using X-ray photoelectron spectroscopy (XPS, Thermo K-Alpha, Al Kα, Thermo Fisher Scientific (CHINA) Co., Ltd., Waltham, MA, USA).Thermogravimetric analysis (TG) was conducted on the films using a Netzsh TG 209 F1 (Netzsh Scientific Instruments TRADING Ltd., Shanghai, China), from 25 °C to 800 °C at a heating rate of 10 °C min −1 in an ambient air environment.

Electrochemical Measurements
Evaluation of Li + storage performance of the composite films was conducted using coin-type half cells (CR2032, Guangdong Canrd New Energy Technology Co., Ltd., Guangdong, China).The rGO@Si NPs composite films and pure-Si were used as the working electrodes, respectively.A lithium foil was used as the counter electrode.1 M LiPF6 in ethylene carbonate (EC) and diethyl carbonate (DEC) (1:1 in volume ratio) was used as the electrolyte.A Celgard 2400 film was used as the separator.A CT 2001A (LAND, Wuhan Bateri Technology Co., Ltd., Wuhan, China) automatic battery testing system was employed to assess the cycle performance and rate capabilities of the coin cells over a voltage range of 0.01-2.0V (vs.Li + /Li) at ambient temperature.A CHI760E electrochemical workstation ((Shanghai Chenhua Instrument Co., Ltd., Shanghai, China) was used to conduct cyclic voltammetry (CV) tests at a scanning rate of 0.2 mV s −1 in the voltage window range of 0.01-1.5 V (against Li/Li + ).A CHI660E electrochemical workstation (Shanghai Chenhua Instrument Co., Ltd., Shanghai, China), was used to conduct electrochemical impedance spectra (EIS) testing on the coin cells from 10 −2 -10 5 Hz.The temperature was maintained at 26 °C for the duration of the measurement.

Results and Discussion
The synthesis process used to produce the nano-SiO2, Si NPs, and self-supporting rGO@Si NPs composite films is shown in Figure 1a-c.It was found that the mechanical behavior of the rGO@Si NPs composite membrane made them very flexible so they could be bent and even doubled over with no crevices It was found that the mechanical behavior of the rGO@Si NPs composite membrane made them very flexible so they could be bent and even doubled over with no crevices produced (the illustration in Figure 1c).The Si NPs were grown in situ on the GO sheets, ensuring good physical contact between the two.The additional in-situ growth of Si NPs was limited because there were fewer anchoring Si atoms in the Si-O groups on the surface of GO sheets.The Zn foil reduced the GO sheet because the rGO/GO had a higher reduction potential than Zn/Zn 2+ (0.76 V versus SHE) at pH = 4 (0.4 V against SHE).However, too many Si NPs prevented the GO sheet from self-assembling into a complete composite film, which resulted in the production of cracks in the film [39,40].Figure 2 illustrates the micromorphology of nano-SiO 2 (a) and Si NPs (b), as well as the cross-section of the rGO film (c), rGO@Si NPs-1 (d), -2 (e), and -3 (f).The preparation of rGO@Si NPs-4 Crystals 2023, 13, 547 5 of 13 failed, as evidenced by Figure 2g where it can be seen that the film was completely broken up.The GO sheet layer produced on the Zn foil's surface easily received electrons from the Zn foil during the reduction process.Because the GO sheet was negative, the Zn 2+ ions from the surface of the Zn foil were attracted by the GO sheet [38].This chemically reduced the GO sheets and an incredibly thin layer of rGO developed and adhered to the Zn plate's surface.
was limited because there were fewer anchoring Si atoms in the Si-O groups on the surface of GO sheets.The Zn foil reduced the GO sheet because the rGO/GO had a higher reduction potential than Zn/Zn 2+ (0.76 V versus SHE) at pH = 4 (0.4 V against SHE).However, too many Si NPs prevented the GO sheet from self-assembling into a complete composite film, which resulted in the production of cracks in the film [39,40].Figure 2 illustrates the micromorphology of nano-SiO2 (a) and Si NPs (b), as well as the cross-section of the rGO film (c), rGO@Si NPs-1 (d), -2 (e), and -3 (f).The preparation of rGO@Si NPs-4 failed, as evidenced by Figure 2g where it can be seen that the film was completely broken up.The GO sheet layer produced on the Zn foil's surface easily received electrons from the Zn foil during the reduction process.Because the GO sheet was negative, the Zn 2+ ions from the surface of the Zn foil were attracted by the GO sheet [38].This chemically reduced the GO sheets and an incredibly thin layer of rGO developed and adhered to the Zn plate's surface.Thereafter, electrons were transferred from inside the Zn foil through the layered rGO to the outer GO sheet causing the rGO sheet and the Si NPs to weakly electrostatically repel each other.When the bonding interaction between them was greater than the static repulsive force between the rGO sheet and Si NPs (for example the π-π interaction and hydrophobic), both the rGO sheets and Si NPs cross-linked layer-by-layer to produce the rGO@Si NPs nanostructured composite film with cross-linked porous layers [41].
Figure 2a shows that the grain sizes of nano-SiO2 prepared by a solution gel method were evenly distributed, and the particles had a diameter that varied between 600 and 700 nm.The morphology of Si NPs further obtained by the magnesium thermal reduction method is shown in Figure 2b, where it appeared that there was a small amount of Si NPs particle aggregation.However, the spherical particles were between 40 and 70 nm in size.Figure 2c shows the multi-layer cross-linked pore structure of the cross-sectional view of rGO without nano-silicon particles.The continuous cross-linked structure formed in the Thereafter, electrons were transferred from inside the Zn foil through the layered rGO to the outer GO sheet causing the rGO sheet and the Si NPs to weakly electrostatically repel each other.When the bonding interaction between them was greater than the static repulsive force between the rGO sheet and Si NPs (for example the π-π interaction and hydrophobic), both the rGO sheets and Si NPs cross-linked layer-by-layer to produce the rGO@Si NPs nanostructured composite film with cross-linked porous layers [41].
Figure 2a shows that the grain sizes of nano-SiO 2 prepared by a solution gel method were evenly distributed, and the particles had a diameter that varied between 600 and 700 nm.The morphology of Si NPs further obtained by the magnesium thermal reduction method is shown in Figure 2b, where it appeared that there was a small amount of Si NPs particle aggregation.However, the spherical particles were between 40 and 70 nm in size.Figure 2c shows the multi-layer cross-linked pore structure of the cross-sectional view of rGO without nano-silicon particles.The continuous cross-linked structure formed in the rGO@Si NPs composite film ensured fast electron transport by successfully avoiding stacking rGO sheets, and this occurred throughout the sheet network structure.Figure 2d shows that fewer, uneven Si NPs adhered between rGO layers when the Si NP concentration was 90 mg.As seen in Figure 2e, more Si NPs were embedded between the rGO layers when the Si NP concentration was increased to 140 mg.The Si NPs were also equally distributed over the surface of the rGO sheets.When the amount of Si NPs was increased to 210 mg, particle agglomeration prevented viewing of the lamellar structure of rGO, as shown in Figure 2f.
The TEM images of the materials in Figure 3a,b clearly reveal that the Si NPs were embedded in the 2D lamellar structure of rGO.In addition, the TEM images of the selected areas and the corresponding element mapping showed that the Si NPs were evenly embedded in the rGO lamellar structure.Combining this information with the SEM images it appeared that the structure of lamellar graphene was evenly coated.Therefore, the even coating of rGO improved the conductivity of the Si NPs.Furthermore, from the HRTEM images shown in Figure 3c, the (111) crystal surface of silicon can be clearly seen, and the lattice spacing was calculated to be 0.31 nm.This lamellar structure possessed a degree of toughness, which could withstand large volume changes of Si during cycling to enhance the cycle stability.
rGO@Si NPs composite film ensured fast electron transport by successfully avoiding stacking rGO sheets, and this occurred throughout the sheet network structure.Figure 2d shows that fewer, uneven Si NPs adhered between rGO layers when the Si NP concentration was 90 mg.As seen in Figure 2e, more Si NPs were embedded between the rGO layers when the Si NP concentration was increased to 140 mg.The Si NPs were also equally distributed over the surface of the rGO sheets.When the amount of Si NPs was increased to 210 mg, particle agglomeration prevented viewing of the lamellar structure of rGO, as shown in Figure 2f.
The TEM images of the materials in Figure 3a,b clearly reveal that the Si NPs were embedded in the 2D lamellar structure of rGO.In addition, the TEM images of the selected areas and the corresponding element mapping showed that the Si NPs were evenly embedded in the rGO lamellar structure.Combining this information with the SEM images it appeared that the structure of lamellar graphene was evenly coated.Therefore, the even coating of rGO improved the conductivity of the Si NPs.Furthermore, from the HRTEM images shown in Figure 3c, the (111) crystal surface of silicon can be clearly seen, and the lattice spacing was calculated to be 0.31 nm.This lamellar structure possessed a degree of toughness, which could withstand large volume changes of Si during cycling to enhance the cycle stability.In addition to the characterization of the appearance of the sample, the composition, content, and valence state of the atoms contained in the sample were analyzed.The XRD patterns of the prepared Si NPs, rGO, and rGO@Si NPs composites are shown in Figure 4a.The characteristic diffraction peaks of the rGO@Si NPs sample were located at 28.3°, 47.0°, 56.2°, 69.3°, and 76.5° (PDF#27-1402), which demonstrated that the Si crystal structure was retained during the production process.In addition, the XRD results of rGO show that there is an obvious envelope peak at about 24°, which corresponded to the (002) crystal face of graphite, which also suggested that the prepared rGO was composed of an indeterminate type of carbon [42].The presence of this packet peak can also be observed at the same position (2θ ≈ 24°) in the images of the rGO@Si NPs composites, which suggests the effective introduction of carbon (rGO) and the successful preparation of the composites.In addition to the characterization of the appearance of the sample, the composition, content, and valence state of the atoms contained in the sample were analyzed.The XRD patterns of the prepared Si NPs, rGO, and rGO@Si NPs composites are shown in Figure 4a.The characteristic diffraction peaks of the rGO@Si NPs sample were located at 28.3 • , 47.0 • , 56.2 • , 69.3 • , and 76.5 • (PDF#27-1402), which demonstrated that the Si crystal structure was retained during the production process.In addition, the XRD results of rGO show that there is an obvious envelope peak at about 24 • , which corresponded to the (002) crystal face of graphite, which also suggested that the prepared rGO was composed of an indeterminate type of carbon [42].The presence of this packet peak can also be observed at the same position (2θ ≈ 24 • ) in the images of the rGO@Si NPs composites, which suggests the effective introduction of carbon (rGO) and the successful preparation of the composites.
Raman spectroscopy was conducted on the experimental materials to study the structural aspects of the Si NPs, rGO, and rGO@Si NPs composite films, and the results are shown in Figure 4b.Peaks at 290, 513, and 926 cm −1 are indicative of the pure Si vibrational band and its resonances [43].The rGO@Si NPs composite had two different peaks.The vibration peaks at 1354 cm −1 and 1588 cm −1 corresponded to D-band and G-band, respectively.The D-band peak corresponds to the defect and disorder degree of the carbon structure, while the G-band corresponds to the sp2 bond carbon atom in the graphite structure [44].As illustrated in Figure 4, the basic amorphous structure of the rGO sheet coated on Si NPs was determined by the peak ratio of D peak intensity and G peak intensity.These values, as shown in Figure 4c, for the rGO@Si NPs-1, -2, and -3, were 1.072, 1.121, and 1.067 respectively, indicating that a for all these samples, a sheet of amorphous rGO was coated on the Si NPs.Raman spectroscopy was conducted on the experimental materials to study the structural aspects of the Si NPs, rGO, and rGO@Si NPs composite films, and the results are shown in Figure 4b.Peaks at 290, 513, and 926 cm −1 are indicative of the pure Si vibrational band and its resonances [43].The rGO@Si NPs composite had two different peaks.The vibration peaks at 1354 cm −1 and 1588 cm −1 corresponded to D-band and G-band, respectively.The D-band peak corresponds to the defect and disorder degree of the carbon structure, while the G-band corresponds to the sp2 bond carbon atom in the graphite structure [44].As illustrated in Figure 4, the basic amorphous structure of the rGO sheet coated on Si NPs was determined by the peak ratio of D peak intensity and G peak intensity.These values, as shown in Figure 4c, for the rGO@Si NPs-1, -2, and -3, were 1.072, 1.121, and 1.067 respectively, indicating that a for all these samples, a sheet of amorphous rGO was coated on the Si NPs.
The elemental composition of rGO@Si NPs composite was further analyzed using XPS tests.As illustrated in Figure 4, the survey clearly showed the Si, O, and C peak locations (d).The silicon oxide species generated during sample preparation and testing were responsible for the O [45]. Figure 4e shows the high-resolution Si 2p XPS spectra of rGO@Si NPs composite, which can be deconvoluted into Si 0+ (99.08 eV), Si 2+ (101.8 eV), and Si 4+ (103.1 eV), respectively [46].The peak area of Si 2+ and Si 4+ is larger than that of Si 0+ because of the surface oxidation of Si NPs [47].According to Figure 4f, which corresponds to the C 1s fitting, the groups C-C, C-O, C=O, and C-C=O were present with binding energies at 284.7, 284.9, 286.1, and 285.5 eV [46].
The mass percentages of the Si NPs, rGO, rGO@Si NPs-1, -2, and -3 were evaluated by TG testing (Figure 5).As shown, there was a significant mass decrease in the range of 500 °C to 650 °C which was attributed to the decomposition of rGO because the Si content did not change substantially with the temperature [48].After 700 °C, the curve had a slight upward trend due to the progressive oxidation of the Si NPs [49].Based on this analysis the mass of C in rGO@Si NPs-1, -2, and -3 was around 70.1, 60.3, and 50.1 wt.%, respectively.Therefore, the mass of Si in rGO@Si NPs-1, -2, and -3 was about 29.9, 39.7, and 49.9 wt.%; this is in agreement with expectations, which can also be clearly seen in the SEM in Figure 4d-f.The elemental composition of rGO@Si NPs composite was further analyzed using XPS tests.As illustrated in Figure 4, the survey clearly showed the Si, O, and C peak locations (d).The silicon oxide species generated during sample preparation and testing were responsible for the O [45]. Figure 4e shows the high-resolution Si 2p XPS spectra of rGO@Si NPs composite, which can be deconvoluted into Si 0+ (99.08 eV), Si 2+ (101.8 eV), and Si 4+ (103.1 eV), respectively [46].The peak area of Si 2+ and Si 4+ is larger than that of Si 0+ because of the surface oxidation of Si NPs [47].According to Figure 4f, which corresponds to the C 1s fitting, the groups C-C, C-O, C=O, and C-C=O were present with binding energies at 284.7, 284.9, 286.1, and 285.5 eV [46].
The mass percentages of the Si NPs, rGO, rGO@Si NPs-1, -2, and -3 were evaluated by TG testing (Figure 5).As shown, there was a significant mass decrease in the range of 500 • C to 650 • C which was attributed to the decomposition of rGO because the Si content did not change substantially with the temperature [48].After 700 • C, the curve had a slight upward trend due to the progressive oxidation of the Si NPs [49].Based on this analysis the mass of C in rGO@Si NPs-1, -2, and -3 was around 70.1, 60.3, and 50.1 wt.%, respectively.Therefore, the mass of Si in rGO@Si NPs-1, -2, and -3 was about 29.9, 39.7, and 49.9 wt.%; this is in agreement with expectations, which can also be clearly seen in the SEM in Figure 4d-f.
To assess their electrochemical performance, all the experimental composites were used to fabricate half-cells in a glove box and tested for their electrochemical characteristics.Figure 6 shows the electrochemical performance of the Si NPs, rGO@Si NPs-1, -2, and -3.The first three CV cycles of the rGO@Si NPs-2 composite film electrode's CV curves between 0.01 and 1.5 V at a scan rate of 0.2 mV s −1 are shown in Figure 6a.The two peaks for the lithiation process are visible in these curves.An SEI film was formed on the surface of the composite, which led to the initial irreversible capacity, associated with the reduction peak at 0.3 V.The peak at 0.1 V corresponded to the lithium reaction (Li → Li x Si).The corresponding oxidation peak at 0.6 V was related to the transition from high lithiated Si to less lithiated Si and the delithium reaction (Li x Si → Li) of Si [50].To assess their electrochemical performance, all the experimental composites were used to fabricate half-cells in a glove box and tested for their electrochemical characteristics.Figure 6 shows the electrochemical performance of the Si NPs, rGO@Si NPs-1, -2, and -3.The first three CV cycles of the rGO@Si NPs-2 composite film electrode's CV curves between 0.01 and 1.5 V at a scan rate of 0.2 mV s −1 are shown in Figure 6a.The two peaks for the lithiation process are visible in these curves.An SEI film was formed on the surface of the composite, which led to the initial irreversible capacity, associated with the reduction peak at 0.3 V.The peak at 0.1 V corresponded to the lithium reaction (Li → LixSi).The corresponding oxidation peak at 0.6 V was related to the transition from high lithiated Si to less lithiated Si and the delithium reaction (LixSi → Li) of Si [50].To assess their electrochemical performance, all the experimental composites were used to fabricate half-cells in a glove box and tested for their electrochemical characteristics.Figure 6 shows the electrochemical performance of the Si NPs, rGO@Si NPs-1, -2, and -3.The first three CV cycles of the rGO@Si NPs-2 composite film electrode's CV curves between 0.01 and 1.5 V at a scan rate of 0.2 mV s −1 are shown in Figure 6a.The two peaks for the lithiation process are visible in these curves.An SEI film was formed on the surface of the composite, which led to the initial irreversible capacity, associated with the reduction peak at 0.3 V.The peak at 0.1 V corresponded to the lithium reaction (Li → LixSi).The corresponding oxidation peak at 0.6 V was related to the transition from high lithiated Si to less lithiated Si and the delithium reaction (LixSi → Li) of Si [50].To illustrate the electrochemical performance more comprehensively, the first discharge and charge pattern of Si NPs, rGO@Si NPs-1, -2, and -3 electrodes at a current density of 200 mA g −1 , with the voltage range 0.01-2 V are shown in Figure 6b.The first cycle discharge and charge capacities of Si NPs, rGO@Si NPs-1, rGO@Si NPs-2, and rGO@Si NPs-3 electrode cells were 4237/2410, 2543/2073, 3015/2375, and 3687/2908 mA h g −1 with coulombic efficiencies (CE) of 56.9%, 78.8%, 78.7%, and 81.5%.The low CE was attributed to the formation of the SEI film during the process of circulation [43,51].The rGO@Si NPs composite film exhibited a clear voltage plateau around 0.1 V, which corresponded to the Li-Si alloy formation during the first discharge.The CE of rGO@Si NPs composite films Crystals 2023, 13, 547 9 of 13 were considerably higher than those of the Si NPs.This result showed that the composite with lamellar rGO can buffer the volume expansion of Si NPs during charge and discharge.
The electrochemical performance measurements of the Si NPs, rGO@Si NPs-1, rGO@Si NPs-2, and rGO@Si NPs-3 were obtained, as shown in Figure 6.As shown, the stability of the Si NPs was very poor, and an attenuation was evident, which dropped to almost zero capacity within 20 cycles, as has been previously reported [52].However, during our tests, the rGO@Si NPs-2 composite membrane electrode exhibited excellent capacity retention.The discharge capacity was maintained at 1003.4 mA h g −1 even at a current density of 200 mA g −1 , and it remained steady after 100 cycles.The electrochemical performance of the other two materials, which contained varying amounts of silicon, was noticeably worse than rGO@Si NPs-2.Despite the low Si content in the rGO@Si NPs-1 composite film, it also exhibited good cycling stability, except for a low capacity, with a value of 365.2 mA h g −1 after 100 cycles.The small proportion of Si NPs and uneven distribution between rGO sheets produced a low capacity as shown in Figure 2d.Nevertheless, the initial capacity of rGO@Si NPs-3 composite film was high but declined throughout the test, with a value of 534.2 mA h g −1 after 100 cycles, due to the high content and uneven distribution of Si NPs, and some agglomeration, as shown in Figure 2f.In addition, the reversible capacity was reduced to 600 mA h g −1 after 30 cycles compared to the previously designed porous C/Si composite [53].It is due to the use of a graphene layered cross-linked structure to buffer the large bulk effect of the silicon nanoparticles during cycling due to the embedding and detachment of lithium ions, thus improving the charge migration rate and ion transport rate, which in turn improves the reversible capacity of the composite.
The multiplier performance is a key factor in the application of LIBs. Figure 6d shows that the average discharge capacity of all samples such as rGO@Si NPs-1 decreased when the current density rose from 100 mA g −1 to 2000 mA g −1 .The average discharge capacity of rGO@Si NPs-2 was still 876.4mA h g −1 .However, the average discharge capacities of rGO@Si NPs-1 and rGO@Si NPs-3 were only 334.2 and 653.6 mA h g −1 , respectively, even at a high current density of 2000 mA g −1 .Additionally, when the current density was returned to 0.1 mA g −1 , the discharge capacities of rGO@Si NPs-1, rGO@Si NPs-2, and rGO@Si NPs-3 showed recovery and remained steady.However, the rGO@Si NPs-2 recovery was better at 1188.38 mA h g −1 , a clear advantage.The performance of rGO@Si NPs-1 and NPs-3 was worse, recovering at only 772.1 and 970.74 mA h g −1 , respectively.
Figure 7 is the impedance curve of the sample before the cycle, and the equivalent circuit is shown in Figure 7 (illustrated).R ct represents the charge transfer resistance corresponding to the semicircle in the high frequency region, and W 0 (Warburg impedance) is related to the diffusion resistance of lithium ion in electrolyte and active material, corresponding to the straight line in the low-frequency region.In addition, R s represents the intersection of the corresponding curve and the real axis of the internal resistance, mainly related to the electrolyte and electrode material [54,55].The excellent conductivity of graphene resulted in a significant decrease in the R ct of the composites (177 Ω).The cross-linked layered structure of graphene in the composite also provides ample channels for transporting lithium ions, resulting in a much smaller Li + diffusion resistance (W 0 ) compared to Si NPs, which can be clearly seen in the straight lines in the low-frequency region.In addition, the R s of the rGO@Si NPs-2 (8.75 Ω) was significantly lower than that of Si NPs (18.2 Ω), indicating that the resistance of the composite itself was improved.This improvement can be attributed to the exceptional conductivity of the two-dimensional lamellar graphene and the effective combination between the two materials [56].
In addition, the Rs of the rGO@Si NPs-2 (8.75 Ω) was significantly lower than that of Si NPs (18.2 Ω), indicating that the resistance of the composite itself was improved.This improvement can be attributed to the exceptional conductivity of the two-dimensional lamellar graphene and the effective combination between the two materials [56].In this regard, rGO@Si NPs-2 had the best electrochemical performance for the following reasons: first, compared with rGO@Si NPs-3, the Si NPs were evenly distributed in the rGO sheets, and no agglomeration was observed.Second, compared to rGO@Si NPs-1, the content of Si NPs was higher, so the available capacity was higher than rGO@Si NPs-1.Third, because graphene had good conductivity, its addition improved the ion diffusion of the composite film.Last but not least, the special 2D lamellar structure of graphene had was strong and tough which effectively alleviated the large expansion of Si NPs during the charge and discharge process.A comparison with other Si/C anode materials is shown in Figure 8, the rGO@Si NPs composite film electrode materials exhibit excellent cycling stability [32,[57][58][59][60][61][62][63][64][65].In this regard, rGO@Si NPs-2 had the best electrochemical performance for the following reasons: first, compared with rGO@Si NPs-3, the Si NPs were evenly distributed in the rGO sheets, and no agglomeration was observed.Second, compared to rGO@Si NPs-1, the content of Si NPs was higher, so the available capacity was higher than rGO@Si NPs-1.Third, because graphene had good conductivity, its addition improved the ion diffusion of the composite film.Last but not least, the special 2D lamellar structure of graphene had was strong and tough which effectively alleviated the large expansion of Si NPs during the charge and discharge process.A comparison with other Si/C anode materials is shown in Figure 8, the rGO@Si NPs composite film electrode materials exhibit excellent cycling stability [32,[57][58][59][60][61][62][63][64][65].
In addition, the Rs of the rGO@Si NPs-2 (8.75 Ω) was significantly lower than that of Si NPs (18.2 Ω), indicating that the resistance of the composite itself was improved.This improvement can be attributed to the exceptional conductivity of the two-dimensional lamellar graphene and the effective combination between the two materials [56].In this regard, rGO@Si NPs-2 had the best electrochemical performance for the following reasons: first, compared with rGO@Si NPs-3, the Si NPs were evenly distributed in the rGO sheets, and no agglomeration was observed.Second, compared to rGO@Si NPs-1, the content of Si NPs was higher, so the available capacity was higher than rGO@Si NPs-1.Third, because graphene had good conductivity, its addition improved the ion diffusion of the composite film.Last but not least, the special 2D lamellar structure of graphene had was strong and tough which effectively alleviated the large expansion of Si NPs during the charge and discharge process.A comparison with other Si/C anode materials is shown in Figure 8, the rGO@Si NPs composite film electrode materials exhibit excellent cycling stability [32,[57][58][59][60][61][62][63][64][65].

Conclusions
In summary, this study has successfully prepared rGO@Si NPs composite film by metal reduction film formation method and demonstrated the excellent electrochemical performance of this composite film as an electrode material in Li-ion batteries.Although

Figure 3 .
Figure 3. TEM images of the rGO@Si NPs with different multiples (a-c).Selected areas (d) and the elemental mappings of its Si and C elements (e,f).

Figure 3 .
Figure 3. TEM images of the rGO@Si NPs with different multiples (a-c).Selected areas (d) and the elemental mappings of its Si and C elements (e,f).

Figure 7 .
Figure 7. EIS curves of Si NPs and rGO@Si NPs-2 composite material before cycling.Illustration: the equivalent circuit diagram.

Figure 7 .
Figure 7. EIS curves of Si NPs and rGO@Si NPs-2 composite material before cycling.Illustration: the equivalent circuit diagram.

Figure 7 .
Figure 7. EIS curves of Si NPs and rGO@Si NPs-2 composite material before cycling.Illustration: the equivalent circuit diagram.