Exfoliation and Reassembly Routes to a Ge/RuO2 Nanocomposite as an Anode for Advanced Lithium-Ion Batteries

Ge/RuO2 nanocomposites were successfully fabricated as anode materials for lithium-ion batteries using RuO2 nanosheets and Ge/GeO2 nanoparticles (NPs). X-ray diffraction (XRD) and X-ray absorption spectroscopy (XAS) analyses showed that elemental Ge nanoparticles were distributed onto the rutile-type RuO2. Transmission electron microscopy images showed well-dispersed Ge nanoparticles embedded in rutile-type RuO2. The Ge/RuO2 nanocomposite maintained higher discharge capacities (471 mA h g−1) after the 90th cycle at 0.1 A g−1 than that (211 mA h g−1) of Ge/GeO2 nanoparticles. The Ge/RuO2 nanocomposite exhibited a higher capacity retention than Ge/GeO2 NPs. These results suggest that the well-dispersed Ge nanoparticles within RuO2 matrices enhance the cycle stability and capacity retention of the anode material.


Introduction
The growing demand for energy-storage devices has triggered extensive research on the development of high-performance metal ion batteries, especially lithium-ion batteries [1][2][3][4][5]. In this regard, the development of next-generation lithium-ion battery (LIB) anode materials is driven by the limited theoretical capacity (372 mA h g −1 ) of graphite, which is a commonly used anode material [6].
Most recent research for obtaining LIBs with enhanced energy density has focused on fabricating alternative anode materials based on group XIV elements and transition metals [7,8]. Germanium is a potential anode material with high theoretical capacity, thermal stability, and good Li-ion conductivity [7][8][9][10][11]. Unfortunately, the large volume changes of Ge-based materials during discharge/charge lead to cracking and pulverization, which are major factors inducing capacity fading in LIBs [12]. Additionally, the oxides of Ru and Ge undergo irreversible conversion reaction with Li to form LiO 2 , which restricts the reversible discharge/charge of the electrodes [13,14]. These defects degrade Li-storage capacities and cycling stabilities. To overcome these drawbacks, various strategies have been developed to fabricate Ge/GeO 2 and RuO 2 composites [10,[15][16][17][18][19]. The fabrication of oxide-free metallic nanoparticles for anode materials is particularly desirable. Typically, the GeO 2 component of the Ge/GeO 2 composite plays an important role in buffering the volume expansion caused by the alloying reaction between Ge and Li. However, the low electrical conductivity of GeO 2 leads to a decrease in the discharge capacity of the Ge/GeO 2 composite. For this reason, dispersing Ge nanoparticles in conductive materials, such as carbon-based materials, is recognized as an effective approach for enhancing the electrochemical performance of Ge nanoparticles [20]. Host materials can withstand the volume expansion of Ge and improve the cycling performance of GeO 2 -free Ge nanoparticles. In this regard, the structural engineering of anode materials has been studied for achieving improved capacity retention. For example, TiO 2 , SnO 2 , and RuO 2 nanosheets 2 of 10 (NSs) comprising unique structures, such as flower-like or hollow spheres, exhibit enhanced electrochemical properties [21,22]. In particular, RuO 2 possesses good chemical stability, high electrical conductivity, and high discharge capacity of 1130 mAh/g [22,23]. These structural and electrochemical advantages render RuO 2 a good candidate as a matrix material for incorporated anode materials. Hence, well-dispersed Ge nanoparticles in structurally modified RuO 2 nanosheets are expected to achieve high cycling performance and capacity retention in LIB anodes.
In this study, we incorporated Ge nanoparticles (NPs) into matrices of layered RuO 2 via successive exfoliation and reassembly processes as shown in Figure 1. Notably, nanosized Ge nanoparticles can be obtained by the dissolution of GeO 2 from the Ge/GeO 2 composite because a basic aqueous solution of tetrabutylammonium hydroxide (TBA + OH − ) for the exfoliation of layered RuO 2 could dissolve GeO 2 by an acid-base reaction [24,25]. Subsequent thermal treatments transformed the layered RuO 2 into a rutile form, which is thermodynamically stable and has a metal ion-permeable channel that enhances the Li-ion's conductivity [26,27]. Therefore, the Ge-dispersed rutile-structured RuO 2 can contribute to enhancing electrochemical performances through a combination of inherent Li-storage capacities and structural advantages. Therefore, the developed strategy provides a new approach for synthesizing advanced anode materials comprising rutile-structured RuO 2 -based Ge composites. volume expansion of Ge and improve the cycling performance of GeO2-free Ge nanoparticles. In this regard, the structural engineering of anode materials has been studied for achieving improved capacity retention. For example, TiO2, SnO2, and RuO2 nanosheets (NSs) comprising unique structures, such as flower-like or hollow spheres, exhibit enhanced electrochemical properties [21,22]. In particular, RuO2 possesses good chemical stability, high electrical conductivity, and high discharge capacity of 1130 mAh/g [22,23]. These structural and electrochemical advantages render RuO2 a good candidate as a matrix material for incorporated anode materials. Hence, well-dispersed Ge nanoparticles in structurally modified RuO2 nanosheets are expected to achieve high cycling performance and capacity retention in LIB anodes.
In this study, we incorporated Ge nanoparticles (NPs) into matrices of layered RuO2 via successive exfoliation and reassembly processes as shown in Figure 1. Notably, nanosized Ge nanoparticles can be obtained by the dissolution of GeO2 from the Ge/GeO2 composite because a basic aqueous solution of tetrabutylammonium hydroxide (TBA + OH − ) for the exfoliation of layered RuO2 could dissolve GeO2 by an acid-base reaction [24,25]. Subsequent thermal treatments transformed the layered RuO2 into a rutile form, which is thermodynamically stable and has a metal ion-permeable channel that enhances the Li-ion's conductivity [26,27]. Therefore, the Ge-dispersed rutile-structured RuO2 can contribute to enhancing electrochemical performances through a combination of inherent Li-storage capacities and structural advantages. Therefore, the developed strategy provides a new approach for synthesizing advanced anode materials comprising rutile-structured RuO2-based Ge composites.

Results and Discussion
XRD analyses were used to characterize the crystal structures of the samples during the reactions (Figure 2). The XRD pattern of the Ge/GeO2 NPs displays characteristic peaks of diamond-type cubic Ge and hexagonal GeO2 structures (Figure 2a). The XRD peaks assigned to GeO2 were not observed in the XRD pattern of the as-prepared Ge/RuO2 (Figure 2b), suggesting that the basic TBA + solution containing exfoliated RuO2 dissolved GeO2 from the Ge/GeO2 NPs. Subsequent thermal treatments at 450 °C for 2 h resulted in the appearance of sharp peaks at 28.  Figure 2c). Therefore, reassembled RuO2 was transformed into rutile-type RuO2. Despite the thermal treatment, the crystal structure of the metallic Ge was still maintained without a decrease in the relative intensity of the peaks. These results indicate that Ge nanoparticles were successfully incorporated into the matrices of rutile-type RuO2.

Results and Discussion
XRD analyses were used to characterize the crystal structures of the samples during the reactions (Figure 2). The XRD pattern of the Ge/GeO 2 NPs displays characteristic peaks of diamond-type cubic Ge and hexagonal GeO 2 structures (Figure 2a). The XRD peaks assigned to GeO 2 were not observed in the XRD pattern of the as-prepared Ge/RuO 2 (Figure 2b), suggesting that the basic TBA + solution containing exfoliated RuO 2 dissolved GeO 2 from the Ge/GeO 2 NPs. Subsequent thermal treatments at 450 • C for 2 h resulted in the appearance of sharp peaks at 28.  Figure 2c). Therefore, reassembled RuO 2 was transformed into rutile-type RuO 2 . Despite the thermal treatment, the crystal structure of the metallic Ge was still maintained without a decrease in the relative intensity of the peaks. These results indicate that Ge nanoparticles were successfully incorporated into the matrices of rutile-type RuO 2 .
X-ray absorption spectral (XAS) analysis at the Ge K-edge was used to confirm the absence of GeO 2 in the as-prepared Ge/RuO 2 ( Figure 3). The normalized XANES spectrum of the Ge/GeO 2 NPs developed a shoulder peak around 11,105 eV and an edge peak around 11,110 eV, corresponding to the absorption peaks of Ge 0 and Ge 4+ , respectively ( Figure 3a) [28]. However, the absorption peak of Ge 4+ was not observed in the XANES spectrum of as-prepared Ge/RuO 2 . Furthermore, Fourier transforms (FTs) of the k 3 -weighted EXAFS spectra of Ge/GeO 2 NPs and as-prepared Ge/RuO 2 were performed to determine the local structural variation around Ge ions ( Figure 3b). The first and third FT peaks around 1.3 Å and 2.8 Å (non-phase-shift-corrected) are assigned to the (Ge-O) and (Ge-Ge) bonds of GeO 2 , whereas the second FT peak around 2.2 Å is assigned to the (Ge-Ge) bond of metallic Ge [28]. The peaks of GeO 2 around 1.3 Å and 2.8 Å are clearly absent in the spectrum of as-prepared Ge/RuO 2 . Therefore, these results reveal that metallic Ge nanoparticles were included within RuO 2 , which is consistent with XRD results. X-ray absorption spectral (XAS) analysis at the Ge K-edge was used to confirm the absence of GeO2 in the as-prepared Ge/RuO2 ( Figure 3). The normalized XANES spectrum of the Ge/GeO2 NPs developed a shoulder peak around 11,105 eV and an edge peak around 11,110 eV, corresponding to the absorption peaks of Ge 0 and Ge 4+ , respectively (Figure 3a) [28]. However, the absorption peak of Ge 4+ was not observed in the XANES spectrum of as-prepared Ge/RuO2. Furthermore, Fourier transforms (FTs) of the k 3weighted EXAFS spectra of Ge/GeO2 NPs and as-prepared Ge/RuO2 were performed to determine the local structural variation around Ge ions ( Figure 3b). The first and third FT peaks around 1.3 Å and 2.8 Å (non-phase-shift-corrected) are assigned to the (Ge-O) and (Ge-Ge) bonds of GeO2, whereas the second FT peak around 2.2 Å is assigned to the (Ge-Ge) bond of metallic Ge [28]. The peaks of GeO2 around 1.3 Å and 2.8 Å are clearly absent in the spectrum of as-prepared Ge/RuO2. Therefore, these results reveal that metallic Ge nanoparticles were included within RuO2, which is consistent with XRD results. The SEM images in Figure 4a,b, were used to analyze the morphology of the as-prepared Ge/RuO2 and Ge/RuO2 nanocomposite. These SEM images show that the planar morphology of the as-prepared Ge/RuO2 collapsed after thermal treatment, which means that layered RuO2 was transformed into rutile-type RuO2. The TEM images of the Ge/RuO2 The SEM images in Figure 4a,b, were used to analyze the morphology of the asprepared Ge/RuO 2 and Ge/RuO 2 nanocomposite. These SEM images show that the planar morphology of the as-prepared Ge/RuO 2 collapsed after thermal treatment, which means that layered RuO 2 was transformed into rutile-type RuO 2 . The TEM images of the Ge/RuO 2 nanocomposite show well-dispersed Ge nanoparticles on the rutile-type RuO 2 (Figure 4c). Furthermore, the lattice fringes at 0.22, 0.25, and 0.20 nm are attributed to the (220) and (101) planes of rutile-type RuO 2 and the (220) plane of metallic Ge, respectively (Figure 4d). These results are in good agreement with the XRD pattern of the Ge/RuO 2 nanocomposite. The SEM images in Figure 4a,b, were used to analyze the morphology of the as-prepared Ge/RuO2 and Ge/RuO2 nanocomposite. These SEM images show that the planar morphology of the as-prepared Ge/RuO2 collapsed after thermal treatment, which means that layered RuO2 was transformed into rutile-type RuO2. The TEM images of the Ge/RuO2 nanocomposite show well-dispersed Ge nanoparticles on the rutile-type RuO2 (Figure 4c). Furthermore, the lattice fringes at 0.22, 0.25, and 0.20 nm are attributed to the (220) and (101) planes of rutile-type RuO2 and the (220) plane of metallic Ge, respectively ( Figure  4d). These results are in good agreement with the XRD pattern of the Ge/RuO2 nanocomposite.  Figure 5a-c show the galvanostatic charge/discharge curves of the Ge/GeO 2 NPs and Ge/RuO 2 nanocomposite versus Li at a current density of 100 mA g −1 in the range of 0.01 and 2 V. The initial discharge/charge capacities of the Ge/GeO 2 NPs and Ge/RuO 2 nanocomposite were 1309/913 mA h g −1 and 1447/814 mA h g −1 , corresponding to initial coulombic efficiencies of 69.7% and 56.2%, respectively. The initial irreversible capacity is derived from the formation of the solid electrolyte interphase (SEI). Nevertheless, the average CE value of the Ge/RuO 2 nanocomposite after the first cycle was 97.7%, which is slightly higher than that of Ge/GeO 2 NPs (97.6%). As a result, the discharge capacity of the Ge/RuO 2 nanocomposite after the 90th cycle was 471 mA h g −1 , which is significantly higher than that of the Ge/GeO 2 NPs (211 mA h g −1 ). Figure 5d shows the rate capability of the Ge/GeO 2 NPs and Ge/RuO 2 nanocomposite at different current densities of 100, 300, 500, 700, 900, and 1000 mA/g for five successive cycles. After applying the current density of 1000 mA/g, the current density was reverted to the 100 mA/g. After the current density returned to 100 mA/g, the capacity retention of the Ge/RuO 2 nanocomposite was 88%, whereas that of the Ge/GeO 2 NPs was 58%. Therefore, the results indicate that rutile-type RuO 2 prevents large volume changes, which contributes to not only good cycle stability, but also the capacity retention of anode materials. 500, 700, 900, and 1000 mA/g for five successive cycles. After applying the current density of 1000 mA/g, the current density was reverted to the 100 mA/g. After the current density returned to 100 mA/g, the capacity retention of the Ge/RuO2 nanocomposite was 88%, whereas that of the Ge/GeO2 NPs was 58%. Therefore, the results indicate that rutile-type RuO2 prevents large volume changes, which contributes to not only good cycle stability, but also the capacity retention of anode materials. As shown in Figure 6, EIS analysis was used to determine the electrochemical properties of the Ge/GeO2 NPs and Ge/RuO2 nanocomposite. The semicircle in the Nyquist plot is related to charge-transfer resistances, whereas the straight line is related to Li-ion diffusion (Figure 6a). At high frequencies, the radius of the semicircle of the Ge/RuO2 nanocomposite is much smaller than that of the Ge/GeO2 NPs. In other words, the chargetransfer impedance (Rct) of the Ge/RuO2 nanocomposite is significantly lower than that of Ge/GeO2 NPs, which means that the Ge/RuO2 nanocomposite has good electrical conductivity when compared with Ge/GeO2 NPs. These results indicate that rutile-type RuO2 facilitates charge transfer and enhances electrical conductivities. At low frequency, straight As shown in Figure 6, EIS analysis was used to determine the electrochemical properties of the Ge/GeO 2 NPs and Ge/RuO 2 nanocomposite. The semicircle in the Nyquist plot is related to charge-transfer resistances, whereas the straight line is related to Li-ion diffusion (Figure 6a). At high frequencies, the radius of the semicircle of the Ge/RuO 2 nanocomposite is much smaller than that of the Ge/GeO 2 NPs. In other words, the chargetransfer impedance (R ct ) of the Ge/RuO 2 nanocomposite is significantly lower than that of Ge/GeO 2 NPs, which means that the Ge/RuO 2 nanocomposite has good electrical conductivity when compared with Ge/GeO 2 NPs. These results indicate that rutile-type RuO 2 facilitates charge transfer and enhances electrical conductivities. At low frequency, straight lines are observed in the Warburg plots, where the slope indicates the ionic conductivity (Figure 6b). The slope of the plot for the Ge/RuO 2 nanocomposite (82.5) corresponded to a lower Warburg coefficient than that of the Ge/GeO 2 NPs (589.3). This result demonstrates that the Ge/RuO 2 nanocomposite has good Li-ion conductivity. Furthermore, the Warburg coefficient of the Ge/RuO 2 nanocomposite was lower than that of the as-prepared Ge/RuO 2 ( Figure S1). This phenomenon indicates that the combination of well-dispersed Ge and rutile-type RuO 2 with an ion-permeable channel contributes to higher Li-ion conductivities compared with Ge/GeO 2 NPs and as-prepared Ge/RuO 2 .
To understand the electrochemical process of Li-storage by the Ge/GeO 2 NPs and Ge/RuO 2 nanocomposite, CV data were acquired at a scan rate of 0.1 mV s −1 in the range of 0.01-2.0 V (Figure 7). The broad peak at 0.74 V in the first cathodic scan for the Ge/RuO 2 nanocomposite is assigned to the formation of an SEI and Li x RuO x [29]. Under 0.5 V, alloying peaks were observed for both samples. In the following anodic scan, the peaks in the range of 0.4-0.6 V and at 1.2 V indicate de-alloying and the reoxidation of Ge, respectively [10,19]. The overall CV profiles of the Ge/RuO 2 nanocomposite are similar to those of Ge/GeO 2 NPs, indicating that most of lithiation/delithiation processes occur on Ge nanoparticles. lines are observed in the Warburg plots, where the slope indicates the ionic conductivity (Figure 6b). The slope of the plot for the Ge/RuO2 nanocomposite (82.5) corresponded to a lower Warburg coefficient than that of the Ge/GeO2 NPs (589.3). This result demonstrates that the Ge/RuO2 nanocomposite has good Li-ion conductivity. Furthermore, the Warburg coefficient of the Ge/RuO2 nanocomposite was lower than that of the as-prepared Ge/RuO2 ( Figure S1). This phenomenon indicates that the combination of well-dispersed Ge and rutile-type RuO2 with an ion-permeable channel contributes to higher Liion conductivities compared with Ge/GeO2 NPs and as-prepared Ge/RuO2. To understand the electrochemical process of Li-storage by the Ge/GeO2 NPs and Ge/RuO2 nanocomposite, CV data were acquired at a scan rate of 0.1 mV s −1 in the range of 0.01-2.0 V (Figure 7). The broad peak at 0.74 V in the first cathodic scan for the Ge/RuO2 nanocomposite is assigned to the formation of an SEI and LixRuOx [29]. Under 0.5 V, alloying peaks were observed for both samples. In the following anodic scan, the peaks in the range of 0.4-0.6 V and at 1.2 V indicate de-alloying and the reoxidation of Ge, respectively [10,19]. The overall CV profiles of the Ge/RuO2 nanocomposite are similar to those of Ge/GeO2 NPs, indicating that most of lithiation/delithiation processes occur on Ge nanoparticles. For detailed electrochemical analysis, CV data were acquired at scan rates ranging from 0.1 to 0.8 mV s −1 (Figure 8a,b). The current is related to the scan speed, as expressed by Equations (1) and (2):  To understand the electrochemical process of Li-storage by the Ge/GeO2 NPs and Ge/RuO2 nanocomposite, CV data were acquired at a scan rate of 0.1 mV s −1 in the range of 0.01-2.0 V (Figure 7). The broad peak at 0.74 V in the first cathodic scan for the Ge/RuO2 nanocomposite is assigned to the formation of an SEI and LixRuOx [29]. Under 0.5 V, alloying peaks were observed for both samples. In the following anodic scan, the peaks in the range of 0.4-0.6 V and at 1.2 V indicate de-alloying and the reoxidation of Ge, respectively [10,19]. The overall CV profiles of the Ge/RuO2 nanocomposite are similar to those of Ge/GeO2 NPs, indicating that most of lithiation/delithiation processes occur on Ge nanoparticles. For detailed electrochemical analysis, CV data were acquired at scan rates ranging from 0.1 to 0.8 mV s −1 (Figure 8a,b). The current is related to the scan speed, as expressed by Equations (1)  For detailed electrochemical analysis, CV data were acquired at scan rates ranging from 0.1 to 0.8 mV s −1 (Figure 8a,b). The current is related to the scan speed, as expressed by Equations (1) and (2): where a and b are adjustable parameters. b-values describe electrochemical behavior, such as diffusion-controlled and pseudocapacitive behavior [30]. Figure 8c,d show the log(i) versus log(v) plot, where the slope indicates the b-value. The b-value for the Ge/RuO 2 nanocomposite was higher than that of the Ge/GeO 2 NPs, indicating that the contribution of the pseudocapacitive process of the Ge/RuO 2 nanocomposite was higher than that of Ge/GeO 2 NPs. Figure 7b shows the capacitive contribution at scan rates of 0.1, 0.3, 0.5, and 0.8 mV s −1 , respectively. The percentage of pseudocapacitive contribution increased with increasing scan speeds. The overall contribution ratios of the pseudocapacitance in the Ge/RuO 2 nanocomposite are higher than those of the Ge/GeO 2 NPs, which indicates that the Ge/RuO 2 nanocomposite's grain boundaries provide additional active sites for surface Li storage [31]. Therefore, these results reveal that the improved pseudocapacitive property of the Ge/RuO 2 nanocomposite contributes to good cycle stability and enhances the rate performance of the electrode [32][33][34]. In conclusion, we determined that the synergic performance of rutile RuO 2 and well-dispersed Ge NPs have key roles in the enhancement of ion conductivity, cyclability, and retention ability of the Ge/RuO 2 nanocomposite.
Ge/RuO2 nanocomposite are higher than those of the Ge/GeO2 NPs, which indicates that the Ge/RuO2 nanocomposite's grain boundaries provide additional active sites for surface Li storage [31]. Therefore, these results reveal that the improved pseudocapacitive property of the Ge/RuO2 nanocomposite contributes to good cycle stability and enhances the rate performance of the electrode [32][33][34]. In conclusion, we determined that the synergic performance of rutile RuO2 and well-dispersed Ge NPs have key roles in the enhancement of ion conductivity, cyclability, and retention ability of the Ge/RuO2 nanocomposite.

Preparation of Ge/GeO 2 NPs
Commercially available Ge/GeO 2 nanoparticles (NPs) were ground in a mortar for 30 min to obtain a fine powder. The obtained powder (0.3 g) was added to a mixture of 15 mL distilled water and 15 mL acetone, followed by ultrasonication for 30 min (3 s on followed by 1 s off) to obtain a colloidal suspension of Ge/GeO 2 NPs.

Synthesis of the Exfoliated RuO 2 NSs
A colloidal suspension of exfoliated RuO 2 nanosheets (NSs) was synthesized according to the procedure shown in Figure 1. Layered potassium ruthenate (K 0.2 RuO 2.1 ·nH 2 O) was synthesized using a solid-state reaction. In a mortar, potassium carbonate (K 2 CO 3 ) was mixed with ruthenium dioxide (RuO 2 ) in a molar ratio of 5:8. This mixture was pelletized and calcined for 12 h at 850 • C under Ar atmosphere. The obtained sample was washed with distilled water to remove water-soluble impurities. Potassium ruthenate was subjected to a proton exchange reaction by exchanging potassium ions with protons in the ruthenate layer in 1 M HCl aqueous solution at 60 • C for 72 h. During this proton exchange reaction, the 1 M HCl aqueous solution was replaced with a fresh batch every 24 h. The layered protonic ruthenate (H 0.2 RuO 2.1 ·nH 2 O) was also subjected to ion-exchange reaction in 50% aqueous ethylamine (EA) solution at room temperature for 24 h to obtain ethylammonium (EA + )-intercalated ruthenate. The EA + -intercalated ruthenate was collected by centrifugation, washed with distilled water, and then reacted with a 10% tetrabutylammonium hydroxide (TBA + OH − ) aqueous solution for 120 h. Thus, exfoliated RuO 2 NSs stably suspended in an aqueous solution of tetrabutylammonium hydroxide were prepared.

Synthesis of Ge/RuO 2 Nanocomposites
Ge/RuO 2 nanocomposites were prepared according to the procedure shown in Figure 1. A colloidal suspension of RuO 2 NSs (4 g/L) and colloidal suspension of Ge/GeO 2 nanoparticles (10 g/L) were mixed in a flask at a mass ratio of 3:7, and the mixture was stirred for 3 h. From this synthesis process, GeO 2 NPs were dissolved in a basic aqueous solution of tetrabutylammonium hydroxide (TBA + OH − ). The mixture was collected via centrifugation (15,000 rpm for 5 min) to obtain as-prepared Ge/RuO 2 nanocomposites. The as-prepared Ge/RuO 2 nanocomposites were dried at room temperature and then heat-treated at 450 • C for 2 h. In this synthesis process, RuO 2 NSs were crystallized to the rutile phase, and Ge/RuO 2 nanocomposites were obtained, in which Ge nanoparticles were homogeneously distributed onto the matrices of RuO 2 .

Structural Characterization
The crystalline phases of the samples were characterized by powder X-ray diffraction (PXRD; Bruker D2 Phaser, Billerica, MA, USA) with Cu-Kα radiation (λ = 1.54056 Å). The morphologies and structures of the products were characterized by field-emission scanning electron microscopy (FE-SEM, Hitachi SU8220, HITACHI, Japan) and field-emission transmission electron microscopy (FE-TEM, Titan G2 ChemiSTEM Cs Probe, FEI Company, Eindhoven, the Netherlands). X-ray absorption fine structure (XAFS) spectra of the powder samples were obtained using the 8C NanoProbe XAFS beamline (BL8C) at the Pohang Accelerator Laboratory (PAL, Pohang, Korea). XAFS spectra of the powder samples were collected in transmission mode at room temperature. The collected XAFS spectra were analyzed using the IFEFFIT software package. IFEFFIT's project pages and information are at SourceForge.net, http://sourceforge.net/projects/ifefit/, and http://cars.uchicago.edu/ifefit (accessed on 5 September 2022) [35,36].

Electrochemical Characterization
The anode material was prepared by mixing the synthesized active material, conductive carbon (Super P), and a binder (polyacrylic acid) in NMP solvent in a weight ratio of 7:2:1. The slurry of anode material was deposited on the copper current collector at a mass loading of approximately 1.5 mg cm −2 using the doctor blade method. This electrode was dried in an oven at 60 • C for 5 h (under ambient atmosphere) and dried again in a vacuum oven at 100 • C for 12 h. Subsequently, this active electrode was transferred to a glovebox filled with Ar. The CR2032 coin-type cell for the electrochemical experiment was assembled in a glovebox filled with high-purity argon. The prepared anode and Li metal counter electrodes were electronically separated using a 3501-type separator (Celgard 3501). A solution of LiPF 6 (1 M) in propylene carbonate/fluoroethylene carbonate (98:2, w/w) was used as the electrolyte. Galvanostatic charge/discharge tests were performed using a battery tester (Maccor K4300, Tulsa, OK, USA) in a potential window of 0.01-2.0 V versus Li + /Li. Cyclic voltammetry (CV) measurements were performed using a multi-channel potentiostat (WonATech WMPG1000, Seoul, Korea) in the potential window of 0.01-2.0 V versus Li + /Li. Electrochemical impedance spectroscopy (EIS) measurements were conducted