Boron Oxide Enhancing Stability of MoS2 Anode Materials for Lithium-Ion Batteries

Molybdenum disulfide (MoS2) is the most well-known transition metal chalcogenide for lithium storage applications because of its simple preparation process, superior optical, physical, and electrical properties, and high stability. However, recent research has shown that bare MoS2 nanosheet (NS) can be reformed to the bulk structure, and sulfur atoms can be dissolved in electrolytes or form polymeric structures, thereby preventing lithium insertion/desertion and reducing cycling performance. To enhance the electrochemical performance of the MoS2 NSs, B2O3 nanoparticles were decorated on the surface of MoS2 NSs via a sintering technique. The structure of B2O3 decorated MoS2 changed slightly with the formation of a lattice spacing of ~7.37 Å. The characterization of materials confirmed the formation of B2O3 crystals at 30% weight percentage of H3BO3 starting materials. In particular, the MoS2_B3 sample showed a stable capacity of ~500 mAh·g−1 after the first cycle. The cycling test delivered a high reversible specific capacity of ~82% of the second cycle after 100 cycles. Furthermore, the rate performance also showed a remarkable recovery capacity of ~98%. These results suggest that the use of B2O3 decorations could be a viable method for improving the stability of anode materials in lithium storage applications.


Introduction
Low-dimensional layered structures of transition metal chalcogenides (TMCs) have attracted increased attention because of their superior properties, such as high conductivity, high stability, easy processing, and easy computing, in two-dimensional (2D) structures [1][2][3][4][5][6]. Therefore, various research has been undertaken to utilize TMCs in applications that traditionally used graphene materials [7][8][9][10]. Among them, MoS 2 is the most well-known TMC material. MoS 2 nanosheets (NSs) can be easily obtained through either top-down approaches, such as scotch tape, sonication, and chemical exfoliation, or bottom-up approaches, such as hydrothermal, chemical vapor deposition, and microwave-assisted methods. 2D MoS 2 NSs possess high conductivity, flexibility, and a large surface area, thereby making them potential candidates for anode materials in lithium storage applications. The MoS 2 NSs have a theoretical capacity of~670 mAh·g −1 , which is twice that of graphite (~372 mAh·g −1 ) [11]. However, previous reports have shown that the MoS 2 NS anodes undergo fast degradation due to the dissolution of sulfur atoms and dislocation of MoS 2 nanosheets during the cycling process [12,13]. Moreover, the conversion reaction of MoS 2 to form Li 2 S, the solid electrolyte interface (SEI) layer, and the degradation of the electrolyte resulted in the formation of a gel-like polymeric layer, which led to fast capacity fading [14]. Many attempts have been made to enhance the stability of MoS 2 NSs based on the use of graphene/carbon nanotube (CNT)/carbon cloth as skeletons, carbon coating layers, or the addition of foreign materials (such as TiO 2 , MnO, Ag, and Sn) to prevent the restacking of MoS 2 and co-contribute to the electrochemical conversion reaction with lithium [11,[14][15][16][17][18][19][20]. For example, Kong et al. demonstrated that MoS 2 nanoplates, with coverage of rolled-up graphene layers, form a core-shell MoS 2 @graphitic nanotube, which showed a high rate performance and high capacity without using a binder [21]. Yoo et al. used CNTs as skeletons to grow MoS 2 via microwave irradiation [22]. The cylindrical-structured MoS 2 on CNTs exhibited advantageous electrochemical properties, such as high rate and high stability, as anode materials in lithium-ion batteries (LIBs). Ren et al. combined both graphene and CNTs as a frame structure for the decoration of MoS 2 nanoparticles (NPs), which delivered a high reversible capacity of~600 mAh·g −1 for 200 cycles [23]. Qu et al. decorated Fe 2 O 3 NPs on MoS 2 NSs via a hydrothermal method and sintering process, in which the anodes exhibited high-rate performances and a high reversible capacity of 900 mAh·g −1 [24]. Zhao et al. prepared the composition MoO 3 /MoS 2 , which has coresheath structure, via a sulfurization technique [25]. The MoO 3 /MoS 2 core-sheath anodes exhibited a negative fading phenomenon and achieved a capacity of~1500 mAh·g −1 after 150 cycles. Even though many attempts on improving the electrochemical performance of MoS 2 NS have been made, the mechanisms are still not clearly revealed and further improvement in stability is still needed to meet the requirements of practical applications.
Recently, lithium nickel cobalt manganese oxide (NMC) cathode materials have been effectively enhanced their stability performance by using boron compounds, such as cobalt boride (Co x B) and B 2 O 3 , for surface modifications [26][27][28]. Yoon et al. revealed that Co x B metallic glass in Ni-rich NMC can effectively enhance the stability of cathode materials via reactive wetting [26]. Li et al. utilized B 2 O 3 as a surface-modification material to enhance the performance of the NMC111 cathode [27]. The use of B 2 O 3 also resulted in graphene combined with a MoS 2 hierarchical structure, which improved the photo/electro properties of the graphene/MoS 2 composition for bio applications [29]. Riyanto et al. reported that a boron-doped graphene quantum structure with MoS 2 could deliver a high capacity of 1000 mAh·g −1 [30]. However, the effect of B 2 O 3 in lithium-ion batteries has not been investigated. B 2 O 3 is a low-cost material with low environmental pollution and easy processing, and it plays an important role in many applications such as thermochemical energy storage, the addition of glass fibers, and the synthesis of boron compound materials such as BN [31,32]. B 2 O 3 is believed to enhance the electrochemical properties of MoS 2 as it is conducted on 2D graphene materials.
In this study, we report the use of boron-oxide-nanoparticle-decorated MoS 2 NSs as anode materials in LIBs. The MoS 2 NSs were prepared using a chemical exfoliation method, and the decoration of B 2 O 3 was carried out using a facile sintering technique. The results showed enhanced cycling stability in the MoS 2 anode when B 2 O 3 formed a crystal structure, delivering a reversible capacity of~500 mAh·g −1 . These results suggest that the use of B 2 O 3 can be a viable strategy for stabilizing anode materials for lithium storage applications.

Exfoliation of MoS 2 NSs
The exfoliation of the MoS 2 NSs was performed according to the method outlined in previous reports [16,33,34]. In brief, 1.0 g of MoS 2 powder and 3 mL of butyllithium/hexane were mixed in a 10-mL vessel (placed in a glove box) to prevent the self-heating of butyllithium. The 1.6 M butyllithium/hexane was prepared by diluting the delivered 2.5 M butyllithium/hexane solution into hexane solvent. The mixture was maintained for 2 days to form Li x MoS 2 . Li x MoS 2 was then collected via centrifugation to remove the hexane and residual butyllithium. The obtained Li x MoS 2 was added to 200 mL of deionized (DI) water and placed in a sonication bath for 2 h to exfoliate MoS 2 . Finally, 1T-MoS 2 was washed with DI water four times to remove lithium ions and then freeze-dried using a Labconco freeze dryer (Labconco Corp., Kansas, MO, USA).

Preparation of Boron Oxide Decorated MoS 2 NS
For boron oxide decoration, the MoS 2 NSs were collected after washing four times with DI water. The amount of MoS 2 was determined by weighing the same amount of MoS 2 NS in the solution after freeze-drying. The boric acid to MoS 2 NS weight ratios were approximately 10, 20, and 30%. The mixtures were dispersed in DI water by sonication for 1 h, then freeze-dried, and sintered at 400 • C for 2 h in a tube furnace under Ar gas. The collected powder was denoted as MoS 2 _B1, -B2, and -B3 with increasing amounts of boric acid (10, 20, and 30 wt%, respectively).

Material Characterization
The structure of the materials was measured by X-ray diffraction (XRD) (D/MAX-2200 Rigaku Tokyo, Japan) over the 2θ range of 10-70 • . The morphologies, sizes, and detailed structures of B 2 O 3 decorated MoS 2 NS were analyzed using scanning electron microscopy (SEM) (Hitachi S4700, Tokyo, Japan) and transmission electron microscopy (TEM, TECNAI G2F30, FEI Corp., Hillsboro, OR, USA).

Electrochemical Measurements
To evaluate the electrochemical performance of the materials and their lithium storage capability, the materials were assembled as working electrodes in half-cell LIBs using a coin-type cell (CR 2032, Rotech Inc., Gwangju, Korea) with a lithium reference electrode. The active material was mixed with PVDF and carbon super P at a weight ratio of 70:15:15 in a NMP solution to form a slurry. The working electrode was prepared by casting the slurry on a copper electrode, using the doctor blading method, followed by drying in a vacuum oven at 70 • C for 24 h. The battery structures were assembled under Ar gas in a glovebox with positive pressure. The separator and electrolyte were polyethylene and 1 M LiPF 6 in ethylene carbonate/diethylene carbonate (EC: DEC = 1:1 by volume). The galvanostatic electrochemical charge-discharge performances of the cells were measured using a battery cycle tester (WBCS3000, WonAtech, Seocho-gu, Seoul, Korea) across the voltage range of 0.01-3.0 V versus Li/Li + . Cyclic voltammetry (CV) tests, across a voltage range of 0.01-3.0 V, and electrochemical impedance spectroscopy (EIS), over a frequency range of 100 kHz to 0.1 Hz, were performed using ZIVE MP1 (WonAtech, Seocho-gu, Seoul, Korea). All the specific capacities were calculated based on the weights of the active materials. Figure 1 shows the XRD patterns of the MoS 2 NSs and MoS 2 _B1, -B2, and -B3 samples synthesized with 10, 20, and 30 wt% boric acid. The MoS 2 NS exhibited a main peak at~14.2 • , indicating the main orientation of the (002) plane in the 2D structure, as per JCPDS #37-1492. The other weak peaks of MoS 2 indicated the presence of multiple layers of these materials. These results are consistent with MoS 2 NSs synthesized by various methods, such as hydrothermal or sonication methods [19,20,35]. The B 2 O 3 at lower concentrations of 10% and 20% did not exhibit the peak of boric oxide, which can be due to the amorphous structures on the MoS 2 NS surface. When increasing the boric acid to 30 wt%, the crystallinity of B 2 O 3 was observed. The structure of B 2 O 3 matched the cubic structure of B 2 O 3 in JCPDS card #06-0297 with a high lattice constant (a = 10.05 Å). This lattice constant was sufficiently high compared to the 0.76 Å of lithium ion. Therefore, B 2 O 3 coverage on MoS 2 may not affect lithiation/delithiation. In addition, the XRD patterns of the MoS 2 _B1, -B2 and B3 samples show a broad peak at~12 • . According to Bragg's law, the lattice spacing can be calculated from the equation d = λ/2sinθ, where λ is the X-ray wavelength and θ is the diffraction angle. Therefore, the lattice spacing of this peak is~7.37 Å, and this can be attributed to the expansion of the MoS 2 layers or the stacking of MoS 2 NSs with B 2 O 3 NPs. This stacking layer had a large lithium-ion radius, thereby generating a facile path for the insertion/desertion of these ions. -B2 and B3 samples show a broad peak at ~12°. According to Bragg's law, the lattice spacing can be calculated from the equation = /2 , where is the X-ray wavelength and is the diffraction angle. Therefore, the lattice spacing of this peak is ~7.37 Å, and this can be attributed to the expansion of the MoS2 layers or the stacking of MoS2 NSs with B2O3 NPs. This stacking layer had a large lithium-ion radius, thereby generating a facile path for the insertion/desertion of these ions. To confirm the morphologies of the MoS2 NSs and their B2O3 decorations, the materials were subjected to SEM and TEM measurements, as shown in Figure 2. As seen in Figure 2a, the MoS2 NSs were exfoliated from the bulk material to nanosheets with a wide size ranging from 200 nm to a few micrometers. The size diversity is due to the strong reaction of intercalated lithium between MoS2 layers and DI water, which broke the NSs into smaller structures and the random shape of the bulk materials. This result is consistent with previous reports of MoS2 NSs prepared using the liquid exfoliation method [16,34,36]. Moreover, the MoS2 NSs with low amounts of B2O3 (10 and 20 wt% of boric acid) show a surface with tiny spots or blurred surface on the MoS2 NS, which are the amorphous structure B2O3 NP decorations, as illustrated in Figure 2b,c. The MoS2_B2 sample had larger B2O3 particles on its surface. When the B2O3 increased to 30 wt%, the SEM image in Figure 2d reveals B2O3 NPs with sizes in the range of ~10-20 nm. The crystallinity of B2O3 depends on the amount of boric acid, which could be due to the large surface area of MoS2 NS. At low concentration, the sintering of low amount of boric acid on MoS2 created imperfect lattices, leading to the low crystalline structure or amorphous structure of B2O3. On the other hand, when the concentration of boric acid was high enough (>30 wt%), the complete lattices of B2O3 NPs formed, indicating the high crystalline structure of B2O3 NPs. Therefore, it is suggested that a low amount of B2O3 only forms an amorphous structure and a high amount of B2O3 (>30 wt%) is sufficient to form a crystalline structure on the surface of MoS2. To confirm the morphologies of the MoS 2 NSs and their B 2 O 3 decorations, the materials were subjected to SEM and TEM measurements, as shown in Figure 2. As seen in Figure 2a, the MoS 2 NSs were exfoliated from the bulk material to nanosheets with a wide size ranging from 200 nm to a few micrometers. The size diversity is due to the strong reaction of intercalated lithium between MoS 2 layers and DI water, which broke the NSs into smaller structures and the random shape of the bulk materials. This result is consistent with previous reports of MoS 2 NSs prepared using the liquid exfoliation method [16,34,36]. Moreover, the MoS 2 NSs with low amounts of B 2 O 3 (10 and 20 wt% of boric acid) show a surface with tiny spots or blurred surface on the MoS 2 NS, which are the amorphous structure B 2 O 3 NP decorations, as illustrated in Figure 2b,c. The MoS 2 _B2 sample had larger B 2 O 3 particles on its surface. When the B 2 O 3 increased to 30 wt%, the SEM image in Figure 2d  TEM measurements were conducted to further reveal the structure of the MoS2 NSs and B2O3 NPs on the MoS2. Figure 2e shows a high-resolution TEM (HRTEM) image of the MoS2 NSs. The surface image clearly shows a lattice plane spacing of approximately 0.264 nm, which corresponds to the (101) plane of MoS2. Thus, MoS2 NSs with high crystallinity were obtained. However, in the MoS2_B1 samples, the MoS2 NSs were hindered by a blurred surface, which indicated the amorphous structure of B2O3, as illustrated in Figure  2f. The blurred surface area increased in MoS2_B2 owing to the increasing amount of B2O3 amorphous structure, as shown in Figure 2g. In addition, crystalline B2O3 was observed in the MoS2_B3 samples (Figure 2h). The lattice spacing was measured as 0.211 nm, which corresponds to the d-spacing of the B2O3 crystal. These results strongly indicated the presence of well-decorated B2O3 NPs on the MoS2 NS surface.

Results and Discussion
The electrochemical properties of B2O3-decorated MoS2 were recorded by CV tests at a low scanning rate of 0.1 mV·s −1 , in the range of 0.0-3.0 V (vs. Li/Li + ) (Figure 3). The reaction at the anode can be expressed by the following equation: For lithiation: Li MoS + (4 − x)Li + (4 − x)e → Mo + 2Li S For delithiation: Li S → S + 2Li +2e - Finally, the solid electrolyte reaction at first cycles: Li + e + electrolyte → SEI As shown in Figure 3a, the bare MoS2 materials show a cathodic peak in the first cycle To further observe the effect of B2O3 on the MoS2 materials, the initial voltage profiles of B2O3 decorated samples are shown in Figure 4. The first three cycles of MoS2_B1, -B2, and MoS2 NSs seem to be unstable, showing a clear change from the first to the second and third cycles. The voltage plateau of the first discharge curve was slightly reduced For lithiation: For delithiation: Finally, the solid electrolyte reaction at first cycles: As shown in Figure 3a, the bare MoS 2 materials show a cathodic peak in the first cycle at~0.76 V, which is the lithiation process to form Li x MoS 2 and the deep lithiation to form Mo and Li 2 S, as shown in Equations (1) and (2). The peak between 0.1-0.5 V could be due to the formation of the SEI layer (5). These results are consistent with previous reports on 1T MoS 2 in the first CV cycle [15,16]. From the second cycle, redox couple peaks were recorded at 1.05/1.72 V and 1.87/2.38 V, which are the reactions in Equations (2) and (3); and Equations (1) and (4), respectively. The CV curves of anodes MoS 2 _B1 and B2 were similar. In these two anodes, the first cycle shows cathodic peaks at~0.96 and 0.41 V, which correspond to the reactions (1) and (2), respectively. The anodic peaks were located at~1.7 and 2.3 V, which correspond to the reactions (3) and (4), respectively. SEI layer formation was recorded together with the peak of reaction (2) (2) and (5)). The peaks of the MoS 2 _B3 anode were positioned at lower potential compared to those of MoS 2 _B1 and -B2 electrodes, which were at~0.9 and 0.4 V vs. Li + /Li. This peak shift could be due to the formation of B 2 O 3 crystalline introducing a different interface to the electrolyte in comparison to the amorphous B 2 O 3 , which leads to the harder diffusion of Li in the first cycle. From the second cycle, the redox couple peaks were recorded at 0.82/1.73 V and 1.77/2.42 V, which correspond to the reactions (2) and (3); and Equations (1) and (4), respectively. The third cycle showed a similar curve to the second cycle, indicating the stable electrochemical reaction after the first cycle. Furthermore, the relative intensity of Mo's oxidation peak located at~1.73 V for the MoS 2 _B3 anode (Equation (3)) was significantly enhanced in comparison to those of MoS 2 _B1 and -B2 and bare MoS 2 NSs anodes. It is noted that the insertion of Li in MoS 2 at high potential is relative to the formation of a gel-like polymeric SEI layer due to the S dissolution in electrolyte [37]. The MoS 2 _B1, -B2, and bare MoS 2 NSs anodes show a high cathodic peak at~1.87 V after three cycles, which is higher than that located at 1.77 V of MoS 2 _B3 anode, indicating the higher amount of S was dissolved in electrolyte. Therefore, the MoS 2 _B3 anode has high amount of recovered MoS 2 NS, resulting in the high oxidation peak intensity of Mo to Mo 4+ . This could be due to the stability of crystalline B 2 O 3 allowing the insert/desertion of Li ions [38]. Moreover, the sulfur atoms have high electron affinity, thus, they could not pass through the B 2 O 3 lattice [39]. It indicates that the crystalline B 2 O 3 effectively protected the MoS 2 layer, preventing the loss of S atoms and the formation of gel-like polymeric SEI layer.
To further observe the effect of B 2 O 3 on the MoS 2 materials, the initial voltage profiles of B 2 O 3 decorated samples are shown in Figure 4. The first three cycles of MoS 2 _B1, -B2, and MoS 2 NSs seem to be unstable, showing a clear change from the first to the second and third cycles. The voltage plateau of the first discharge curve was slightly reduced from the MoS 2 NSs to the MoS 2 _B1, -B2, and -B3 samples, where two plateaus at 1.1/0.51 to 1.1/0.51, 1.1/0.50, and 1.1/0.48, respectively, are shown. This indicates that the B 2 O 3 crystals in the MoS 2 _B3 samples changed the lithium insertion potential. In the second and third cycles, the voltage plateaus were similar for the MoS 2 _B3 electrode, thereby indicating stable electrochemical properties from the second cycle. In addition, the initial discharge capacity of these anodes was high, but it reduced after each cycle owing to the formation of the SEI layer and degradation behavior in lithium ion batteries, such as cracks, sulfur dispersion in the electrolyte, and dendrite growth [40]. The initial discharge capacities for the bare MoS 2 NSs and MoS 2 _B1, -B2, and -B3 were 747.1, 717.7, 638.1, and 717.2 mAh·g −1 , respectively. The difference in the initial discharge capacities also depended on the formation of the SEI layer and the binding of B 2 O 3 to MoS 2 . The B 2 O 3 was reported as a low lithium ion storage capability [41]. Therefore, the increased amount of B 2 O 3 in MoS 2 led to the decreased charge/discharge capacities of MoS 2 anode materials. In particular, the charge/discharge capacities of the MoS 2 _B3 anode in the third cycle were 505.2/475.0 mAh·g −1 .  To evaluate the stability of the anode materials, cycling tests were performed at a current rate of 0.1 A·g −1 for 100 cycles, as illustrated in Figure 5a-d. Detailed comparison of specific capacities of as-prepared anode materials are also shown in Table 1. The MoS2 NSs showed stability for ~20 cycles, and its capacity was subsequently dramatically reduced and maintained at only ~109 mAh·g −1 at the 100th cycle (Figure 5a). The addition of B2O3 also resulted in very fast degradation, and the remaining capacity was ~125 mAh·g −1 and ~140 mAh·g −1 at the 100 th cycle in the MoS2_B1 and B2 anodes, respectively ( Figure  5b,c). The MoS2_B1 and B2 showed the enhancement of lattice spacing of MoS2, facilitating the insertion/desertion of Li ions. However, the amorphous B2O3 could not prevent the loss of S atoms. Therefore, MoS2_B1 and B2 anodes exhibited inferior stability to the pure MoS2 NS. In contrast, the crystalline B2O3 in the MoS2_B3 electrode showed a high capacity in the first cycle, and it demonstrated prolonged cycling stability for 100 cycles. As shown in Figure 5d, the capacity of MoS2_B3 at the 100th cycle was ~451 mAh·g −1 , which was 86.2% of the second cycle (~510 mAh·g −1 ) and ~62.9% of the first cycle. Therefore, it can be concluded that the enhancement of the redox reaction with the B2O3 crystals was effective in improving the stability of the MoS2 NSs. To evaluate the stability of the anode materials, cycling tests were performed at a current rate of 0.1 A·g −1 for 100 cycles, as illustrated in Figure 5a-d. Detailed comparison of specific capacities of as-prepared anode materials are also shown in Table 1. The MoS 2 NSs showed stability for~20 cycles, and its capacity was subsequently dramatically reduced and maintained at only~109 mAh·g −1 at the 100th cycle (Figure 5a). The addition of B 2 O 3 also resulted in very fast degradation, and the remaining capacity was~125 mAh·g −1 and 140 mAh·g −1 at the 100 th cycle in the MoS 2 _B1 and B2 anodes, respectively (Figure 5b,c). The MoS 2 _B1 and B2 showed the enhancement of lattice spacing of MoS 2 , facilitating the insertion/desertion of Li ions. However, the amorphous B 2 O 3 could not prevent the loss of S atoms. Therefore, MoS 2 _B1 and B2 anodes exhibited inferior stability to the pure MoS 2 NS. In contrast, the crystalline B 2 O 3 in the MoS 2 _B3 electrode showed a high capacity in the first cycle, and it demonstrated prolonged cycling stability for 100 cycles. As shown in Figure 5d, the capacity of MoS 2 _B3 at the 100th cycle was~451 mAh·g −1 , which was 86.2% of the second cycle (~510 mAh·g −1 ) and~62.9% of the first cycle. Therefore, it can be concluded that the enhancement of the redox reaction with the B 2 O 3 crystals was effective in improving the stability of the MoS 2 NSs.  The protective role of B2O3 was further confirmed via ex-situ XPS spectra, as sho in Figure 6. Both bare MoS2 NS and MoS2_B3 anodes were compared at the initial s and at 3.0 V 40 cycles. The initial state of MoS2 and B2O3 decorated samples presented same conditions of Mo 4+ and S 2− . However, after 40 cycles, bare MoS2 NSs showed sign cant change in Mo 4f peak. Mo 4+ peak intensity reduced, and the Mo 5+ and Mo 6+ pe appeared. Moreover, in S 2p spectrum, the S 2p peak split to S 2− peak at ~162 eV and a peak at ~163.6 eV, which might be related to the formation of polymeric SEI layer du the S dissolution or the unrecoverable Li2S [42,43], indicating unstable MoS2 NS ano On the other hand, the MoS2_B3 anode show a better stability, where the main peak of 4f assigned to Mo4+ was maintained with partial Mo 6+ peaks. It is noted that the Mo 6+ p might appear due to the sample preparation method as pointed out in previous rep [44][45][46]. The S 2p peak of MoS2_B3 showed a small change with S* 2− peak, which migh due to a partial loss of S to polymeric layer or unrecoverable Li2S. These results indi that B2O3 layer efficiently protected MoS2 layer, preventing the loss of S to electrolyte. small amount of S loss can be further improved after optimizing the B2O3 layers.  The protective role of B 2 O 3 was further confirmed via ex-situ XPS spectra, as shown in Figure 6. Both bare MoS 2 NS and MoS 2 _B3 anodes were compared at the initial state and at 3.0 V 40 cycles. The initial state of MoS 2 and B 2 O 3 decorated samples presented the same conditions of Mo 4+ and S 2− . However, after 40 cycles, bare MoS 2 NSs showed significant change in Mo 4f peak. Mo 4+ peak intensity reduced, and the Mo 5+ and Mo 6+ peaks appeared. Moreover, in S 2p spectrum, the S 2p peak split to S 2− peak at~162 eV and a S* 2− peak at~163.6 eV, which might be related to the formation of polymeric SEI layer due to the S dissolution or the unrecoverable Li 2 S [42,43], indicating unstable MoS 2 NS anode. On the other hand, the MoS 2 _B3 anode show a better stability, where the main peak of Mo 4f assigned to Mo4+ was maintained with partial Mo 6+ peaks. It is noted that the Mo 6+ peak might appear due to the sample preparation method as pointed out in previous reports [44][45][46]. The S 2p peak of MoS 2 _B3 showed a small change with S* 2− peak, which might be due to a partial loss of S to polymeric layer or unrecoverable Li 2 S. These results indicate that B 2 O 3 layer efficiently protected MoS 2 layer, preventing the loss of S to electrolyte. The small amount of S loss can be further improved after optimizing the B 2 O 3 layers. The electrical properties of the anode materials were evaluated using EIS measurements, as shown in Figure 7a. The equivalent circuit (modified Randle's model) contained series resistance (RS), charge-transfer resistance (RCT), SEI layer resistance (RSEI), a diffusion Warburg impedance element, and constant phase elements (CPE1 and CPE2). The extracted RCTs of the MoS2 NS and MoS2_B1, -B2, and -B3 samples were 150.3, 118.4, 118.2, and 148.6 , respectively. The addition of boron oxide did not significantly affect the resistance of the anode material. MoS2_B1 and -B2 showed reduced resistance. Then, the resistance increased in MoS2_B3 when B2O3 formed crystallinity owing to the low conductivity of B2O3. However, this resistance was still lower than that of the bare MoS2. As a 2D layered structure material, the conductivity of MoS2 decreases when the number of layers is reduced. In addition, the presence of B2O3 NPs may prevent the restacking of MoS2 NSs and the NS material from forming a bulk structure, thereby enhancing the conductivity of the anode material. The rate performance of MoS2_B3 is shown in Figure 7b. An increase in the current rate led to a decrease in capacity. At 1.0 A·g −1 , the capacity was maintained at ~155 mAh·g −1 . Nevertheless, the MoS2_B3 electrode can be recovered to almost 98% when decreasing the current rate to 0.1 A·g −1 , thus illustrating a highly reversible rate performance. The electrical properties of the anode materials were evaluated using EIS measurements, as shown in Figure 7a. The equivalent circuit (modified Randle's model) contained series resistance (R S ), charge-transfer resistance (R CT ), SEI layer resistance (R SEI ), a diffusion Warburg impedance element, and constant phase elements (CPE1 and CPE2). The extracted R CT s of the MoS 2 NS and MoS 2 _B1, -B2, and -B3 samples were 150.3, 118.4, 118.2, and 148.6 Ω, respectively. The addition of boron oxide did not significantly affect the resistance of the anode material. MoS 2 _B1 and -B2 showed reduced resistance. Then, the resistance increased in MoS 2 _B3 when B 2 O 3 formed crystallinity owing to the low conductivity of B 2 O 3 . However, this resistance was still lower than that of the bare MoS 2 . As a 2D layered structure material, the conductivity of MoS 2 decreases when the number of layers is reduced. In addition, the presence of B 2 O 3 NPs may prevent the restacking of MoS 2 NSs and the NS material from forming a bulk structure, thereby enhancing the conductivity of the anode material. The rate performance of MoS 2 _B3 is shown in Figure 7b. An increase in the current rate led to a decrease in capacity. At 1.0 A·g −1 , the capacity was maintained at~155 mAh·g −1 . Nevertheless, the MoS 2 _B3 electrode can be recovered to almost 98% when decreasing the current rate to 0.1 A·g −1 , thus illustrating a highly reversible rate performance. The recent works on the modification of MoS2 are shown in Table 2. The initial discharge capacities were high above 800 or even 1200 or 1400 mAh·g −1 . This might be due to the contribution to the capacity of the modified materials. From our method, B2O3 does not mainly contribute to the capacity, but effectively protects the MoS2 layer, which maintains the highly stable capacity. Moreover, the sintering method and the utilization of the The recent works on the modification of MoS 2 are shown in Table 2. The initial discharge capacities were high above 800 or even 1200 or 1400 mAh·g −1 . This might be due to the contribution to the capacity of the modified materials. From our method, B 2 O 3 does not mainly contribute to the capacity, but effectively protects the MoS 2 layer, which maintains the highly stable capacity. Moreover, the sintering method and the utilization of the boric acid are cost-effective ways. Therefore, it can be readily scale-up to the industrial size. We also believe that the use of carbon-based and co-active materials can further improve the electrochemical performance of the materials presented in this study for lithium ion storage application.

Conclusions
In summary, B 2 O 3 NP-decorated MoS 2 NSs were successfully fabricated via a facile chemical exfoliation and sintering process. The XRD, SEM, and TEM measurements confirmed that the crystalline B 2 O 3 could be formed at high boric acid content of over 30 wt%. The presence of B 2 O 3 created the lattice spacing of~7.37 Å in MoS 2 NS. Crystal B 2 O 3 formed with a lattice spacing of~2.58 Å, improving the redox reaction in the conversion of MoS 2 during the cycling process. The high intensity of Mo oxidation peak and the lower potential of lithium insertion into MoS 2 indicated B 2 O 3 layer played a role as a protective layer, preventing the dissolution of S atoms into electrolyte. The bare MoS2 material and amorphous B 2 O 3 in MoS 2 _B1 and -B2 anodes showed fast degradation after 20-40 cycles due to the loss of sulfur into the electrolyte. Meanwhile, the MoS 2 _B3 electrode with protectable crystalline B 2 O 3 layer demonstrated a stable capacity of~500 mAh·g −1 and a high-capacity retention of~86.2% after 100 cycles. These results suggest that B 2 O 3 NP decorations on anode materials could be a potential approach for high-stability anodes in lithium storage applications.
Author Contributions: T.P.N.: Conceptualization, methodology, validation, visualization, writing, review, and editing. I.T.K.: project administration, funding acquisition, review, and editing. All authors have read and agreed to the published version of the manuscript.