A Facile Surface Preservation Strategy for the Lithium Anode for High-Performance Li–O2 Batteries

Protecting an anode from deterioration during charging/discharging has been seen as one of the key strategies in achieving high-performance lithium (Li)–O2 batteries and other Li–metal batteries with a high energy density. Here, we describe a facile approach to prevent the Li anode from dendritic growth and chemical corrosion by constructing a SiO2/GO hybrid thin layer on the surface. The uniform pore-preserving layer can conduct Li ions in the stripping/plating process, leading to an effective alleviation of the dendritic growth of Li by guiding the ion flux through the microstructure. Such a preservation technique significantly enhances the cell performance by enabling the Li–O2 cell to cycle up to 348 times at 1 A·g–1 with a capacity of 1000 mA·h·g–1, which is several times the cycles of cells with pristine Li (58 cycles), Li–GO (166 cycles), and Li–SiO2 (187 cycles). Moreover, the rate performance is improved, and the ultimate capacity of the cell is dramatically increased from 5400 to 25,200 mA·h·g–1. This facile technology is robust and conforms to the Li surface, which demonstrates its potential applications in developing future high-performance and long lifespan Li batteries in a cost-effective fashion.


■ INTRODUCTION
Aprotic Li−O 2 batteries (LOBs), a promising energy solution for automotive and aerospace engineering, have superior advantage because of their high energy densities, when compared with conventional lithium−/sodium−ion batteries (LIBs/SIBs). 1−8 Currently, the technical development in aprotic LOBs encounters substantial bottlenecks, including high charge overpotential, passivation of the cathode, and poor stability of the electrolyte and Li metal anode. 7 Significant efforts have been made to enhance the cyclic stability and other electrochemical performances of LOBs by promoting the sluggish kinetics of oxygen reduction and evolution using proper catalysts, 9 porous cathode materials, 10,11 redox mediators, 12,13 and stable electrolytes. 14,15 Despite its natural scarcity, problematic stability, and safety issues, 7, 8 Li metal is regarded as an ideal anode material for rechargeable batteries because of its superhigh specific energy (3860 mA·h·g −1 ), low redox potential (−3.04 V vs standard hydrogen electrode), and low mass density (0.534 g·cm −3 ). However, Li anodes encounter challenges such as the uneven stripping/plating of Li ions and the low Coulombic efficiency during charge/discharge cycles, which usually lead to the dendritic growth of Li to cause the short-circuiting and "dead Li" (i.e., broken Li branches) after the repeated charging/ discharging operation 16 and a rapid degradation in the battery cell. 17 The decaying in LOBs can be accelerated as the crossover oxidative species such as soluble oxygen reduction intermediates (O 2 − , LiO 2 − , etc.), H 2 O, and other reaction byproducts can accelerate Li loss, thus resulting in a declined cycle life. 18,19 One of the recent research interests has been focused on the protection of Li anodes because the dendritic deposition of lithium metal and dendrite formation along the solid− electrolyte interphase (SEI) are unveiled as key issues to determine the safety and performance of LOBs. 20,21 The strategies exercised for protecting Li anodes can be categorized in the following: (1) using poreless, air-impermeable, and waterproof separators to prevent O 2 , H 2 O, and other soluble species in the electrolyte from corroding the anode in LOBs; 22−24 however, the increase of the overall cell resistance remains under concern. (2) Optimizing the structure of SEI layers, where the lifespans of LOBs can be extended after treating the SEI layers on Li anodes chemically (including the reaction with CO 2 , N 2 , and F-and B-containing electrolyte additives) 25−29 or electrochemically. 30−32 (3) Generating ionically conductive, mechanically robust, and chemically stable protection layer(s) on Li anodes, and the commonly used materials for such protection layers include polymers (e.g., poly (1,4-dioxacyclohexane) 33 ), inorganic compounds (e.g., germanium or indium compounds 34−36 ), and composites (e.g., Al 2 O 3 /PVDF-HFP or AlF 3 /PEDOT-PEG 37,38 ).
Graphene and graphene oxide (GO) have been reported previously to offer protection to the Li anode by preventing from dendrite growth, corrosion, and strengthening the stability of SEI. 39−43 Li and co-workers 39 employed reduced GO to prevent Li dendrites by directing their growth. Zhao et al. 41 encapsulated a Li alloy with graphene sheets to prevent the corrosion from air. Gao and co-workers 43 designed a molecular level SEI by embedding polymeric Li salts with GO, which provides the SEI to have excellent stability and good passivation properties, which prevents the Li metal anode from dendrite growth and corrosion. Silica was also used for Li metal protection. Tang and co-workers applied silica nanoparticles (SiO 2 NPs) in a cross-linked polymer solid-state electrolyte for increasing the Li ionic conductivity by forming 3D ion transport paths. 44 Lin et al. employed a silica aerogel as a backbone for a polymer electrolyte for high Li ionic conductivity and modulus. 45 Herein, we propose an anode protection strategy by developing an artificial composite layer consisting of SiO 2 NPs and GO nanosheets on the Li metal (Li−SiO 2 /GO) anode. GO sheets act as a barrier to hinder the growth of Li dendrites and the deteriorative attack of oxygen species from the electrolyte; the SiO 2 NPs embedded in GO sheets can provide an effective interlayer and pores for Li ion diffusion by preventing the aggregation of GO sheets. With this designed protection, the LOBs with Li−SiO 2 /GO anodes exhibit superior cycling stability, rate performance, and ultimate capacity. We hope this anode protection technology will find future applications in next-generation sustainable energy solutions. Preparation of SiO 2 /GO Hybrids. GO was prepared using the modified Hummer's method. 46 Typically, 1.0 g of graphite, 0.5 g of NaNO 3 , and 23 mL of H 2 SO 4 were mixed and stirred at 0°C for 1 h. The mixture was heated to 35°C, and then 3.0 g of KMnO 4 was slowly added. After stirring for 7 h, another 3.0 g of KMnO 4 was added, and the mixture was kept stirring for the next 12 h. The mixture was cooled down in air, and then 400 mL of ice water was added under stirring. After that, 30 wt % H 2 O 2 was gradually added until the color of the mixture turned from brown to yellow. The solid content in the mixture was separated by centrifugation at 8000 rpm and then repeatedly rinsed with deionized water until the supernatant became neutral. The precipitate was mixed with deionized water under ultrasonic stirring for 3 h and then allowed to stand for 12 h. The GO suspension in the container was removed and used for further synthesis.

■ EXPERIMENTAL SECTION
SiO 2 NPs were ultrasonically dispersed in the GO suspension with a mass ratio of 1:2 (SiO 2 /GO) for 2 h; the precipitate was then removed and dried in a vacuum oven at 80°C for 24 h. The dried SiO 2 /GO powder was carefully ground and then dispersed in DME (5 mg·mL −1 ), where 0.5 mg of the slurry was repeatedly pipetted until it completely covered the surface of a Li plate, followed by a smoothening process using a fine blade, and after drying in a glove box at ambient temperature, the Li anode with protective SiO 2 /GO coating is obtained and denoted Li−SiO 2 /GO. The Li anodes coated with either monocomponent SiO 2 or GO were prepared as reference samples, denoted Li−SiO 2 and Li−GO, respectively. Various GO/SiO 2 mass ratios (0.5:1, 1:1, 2:1, 3:1, and 4:1) and loading amounts (0.1, 0.2, 0.6, 0.8, 1.0, and 1.5 mg) were prepared and constructed as a preserving layer on the Li anodes.
Battery Assembly and Testing. The batteries were assembled and built into CR2032 coin cells in an Ar-filled glove box (MIKROUNA, Super, H 2 O < 0.1 ppm, O 2 < 0.1 ppm) and tested using a standard battery testing system (CT-3008W-5V10mA, Neware Technology Limited).
LOBs were assembled using both Li anodes and MWNT cathodes in the CR2032 cells. The Li anodes employed in the experiment included the pristine Li plate and Li plates with different surface coatings. The cathode was prepared by spraying the MWNTs in ethanol slurry onto a carbon paper at a loading of 0.1 mg·cm −2 . To assemble the coin cells, an anode cap was first placed, and then a Li anode (either pristine Li or one of the anodes with surface coatings), a glass fiber separator (wetted with 100 μL of 1 M LiClO 4 /DMSO electrolyte), a MWNT cathode, and a cathode cap were placed in sequence and encapsulated. The cells were mounted separately in the holders of a home-made battery testing box filled with pure oxygen at 1.0 atm. The cyclic performance test was settled at a rate of 1 A·g −1 with a fixed capacity of 1000 mA·h·g −1 within the potential window from 2.0 to 4.5 V. The rate performance was measured at current densities of 2, 3, and 5 A·g −1 , and the capacity of 1000 mA·h·g −1 within the potential cutoff was from 2.0 to 4.5 V.
To characterize the ionic conductivity of coating layers, Li|SS (stainless steel) cells were assembled and subjected to electrochemical impedance spectroscopy (EIS) analysis in an O 2 -free atmosphere. An anode cap, a Li plate, a GF separator (wetted with 100 μL of 1 M LiClO 4 /DMSO electrolyte), a SS plate, and a CR2032 cathode cap (without holes) were placed in sequence and encapsulated to construct a Li|SS cell. To investigate the stability of Li anodes with surface coatings in an O 2 atmosphere, a Li|SS foam cell was assembled; the approach used was similar to the Li|SS cell, except for using a CR2032 cathode cap with holes and a SS foam to replace the SS plate. The Li|SS cells and Li|SS foam cells with Li−GO, Li−SiO 2 , and Li−SiO 2 /GO were also assembled using the same approaches. To characterize the resistance of LOBs (the assembly approach was the same as LOBs) after cycling for certain times, the EIS test was carried out after introducing N 2 to remove any O 2 . All EIS plots were recorded on an electrochemical station (CHI 660, CH Instruments) at an open-circuit potential with 5 mV of ac amplitude in the frequency range from 0.1 Hz to 1 MHz.
Symmetric Li|Li cells were assembled to investigate the protection of Li anodes with different coatings in an O 2 -free atmosphere, where an anode cap, a Li plate, a glass fiber separator (wetted with 100 μL of 1 M LiClO 4 /DMSO electrolyte), another Li plate, and a CR2032 cathode cap (without holes) were placed in sequence and encapsulated. In the case of O 2 , the Li plate with a hole (diameter of 2 mm) and a CR2032 cathode cap with holes were used, and the assembly approach was similar to those without O 2 . The same approach was also applied to the assembly of symmetric Li|Li cells with Li−GO, Li−SiO 2 , and Li−SiO 2 / GO. The charge/discharge cycling was carried out with a time period fixed for 1 h at the current density of 0.1 mA·cm −2 .
Characterization. A field-emission scanning electron microscope (S-4800, Hitachi) and a transmission electron microscope (JEM-2100F) operating at 200 kV were employed to observe the morphology of the anodes and cathodes. An X-ray diffractometer (X'Pert PRO) equipped with a Cu Kα radiation source (λ = 1.54059 Å) was used to characterize the composition and structure of the coating and battery discharge products. An X-ray photoelectron spectrometer (ESCALAB 250Xi, Thermo Fisher Scientific) using Al Kα radiation was applied to characterize the components of coating layers and SEI. The electron conductivity of coating layers was measured using a high-temperature resistivity measurement system (RMS-1000I, Partulab Technologies). The specific surface area was determined from the N 2 adsorption/ desorption isotherms using a surface area analyzer (NOVA1200e, Quantachrome).

■ RESULTS AND DISCUSSION
Structural Identification of Protective Coating Layers. Figure 1a reveals a laminated structure for the SiO 2 /GO coating layer, that is, the SiO 2 NPs are intercalated among GO sheets, which effectively inhibits the overlapping aggregation of GO sheets and thereby produces nano-/mesopores for Li ion transportation. Two more cells with SiO 2 and GO individually coated anode were also assembled and tested for comparison; this is done to reveal the combined functions of both SiO 2 and GO when they are laminated together. Morphology information of SiO 2 and GO is shown in the TEM images in Figure S1.
The N 2 adsorption/desorption isotherms in Figure 1b demonstrate that both SiO 2 and SiO 2 /GO exhibit type IV nitrogen sorption isotherms, while GO presents a low porosity likely because of the restacking of the GO sheets. The Brunauer−Emmett−Teller (BET) method shows that the SiO 2 /GO coating delivers a specific surface area of up to 172.70 m 2 ·g −1 , SiO 2 presents a specific surface area of 146.25 m 2 · g −1 , and the specific surface area of GO is just 2.35 m 2 ·g −1 . The estimated average pore sizes are 5.42 nm with a pore volume of 1.9 × 10 −3 cm 3 ·g −1 for the GO layer, 6.27 nm with a pore volume of 176 × 10 −3 cm 3 ·g −1 for the SiO 2 layer, and 3.8 nm with the largest pore volume of 312 × 10 −3 cm 3 ·g −1 for the SiO 2 /GO layer (Figure 1c). The largest specific surface area of the SiO 2 / GO coating layer is caused by the intercalation of SiO 2 NPs among the GO sheets, and this is in good agreement with the TEM results in Figure 1a.
Next, we used X-ray diffraction (XRD) analysis to assess the structure of the coating layers (Figure 1d). For the GO coating, an intense and sharp peak is observed at 8.8°corresponding to the interplanar spacing of ca. 1.0 nm of GO sheets, 47 and a bump at 25.2°resulted from the oxidized graphite. The XRD pattern of SiO 2 presents a broad peak at 22.6°, assigned to the (101) facet of silica (JCPDS no. 82-1235). As for the SiO 2 /GO hybrid, a broad peak appears at about 22.6°, possibly attributed to the overlapped oxidized graphite and SiO 2 peaks, and the diffraction peak of GO at ca. 8.8°vanished, which indicates the absence of GO aggregation, in line with the TEM and BET analyses. We further applied X-ray photoelectron spectroscopy (XPS) to characterize the component of coating layers ( Figure S2). The C 1s spectrum of the GO layer shows four peaks at 284.8, 286.5, 287.6, and 290.0 eV, which are assigned to the C−C, C−OH, CO, and COOH groups, respectively ( Figure S2a). The Si 2p spectrum of the SiO 2 layer displays one peak at 103.5 eV attributed to SiO 2 ( Figure S2b). The C 1s and Si 2p spectra of the SiO 2 /GO layer exhibit similar peaks, indicating that the SiO 2 / GO composite contains the same component of GO and SiO 2 ( Figure S2c (Figure 2d), respectively. The standard EIS analysis is used to compare the effect of GO, SiO 2 , and SiO 2 /GO coatings on the Li ion conductivity in Li|SS cells versus pristine Li, which is listed in Figure 2c. 38 The Nyquist plot (Figure 2a,b) shows that the intercept of the real axis represents the impedance of Li + diffusion (R s ) from the electrolyte solution to lithium metal. 38,49 In the cell with pristine   Figure 3j,k shows a fluffy profile on the top layer mossy Li, which looks different from the pure Li metal; this is the evidence that the surface of these two anodes has been corroded. Those cells with SiO 2 (Figure 3h,l) and SiO 2 /GO (Figure 3i,m) coating layers have no obvious mossy Li, and cracks can be seen after cycling for 800 h; in addition, the cross-sectional image shows a compacted SEI. These results demonstrate that SiO 2 and SiO 2 /GO coating layers can facilitate stable SEI formation and regulate Li deposition.
The long-term cyclic stability of Li symmetric batteries with GO, SiO 2 , and SiO 2 /GO coating layers was also investigated in an O 2 atmosphere. As shown in Figure S4a Figure  S4g,h), and no obvious cracks are visible from the side view ( Figure S4k,l). The Li−SiO 2 /GO electrode shows an appearance with a smoother surface and a better evenly distributed cross section (Figure S4i,m).
LOB Performance Evaluation. The cyclic stability and rate performance of LOBs with the pristine Li, Li−GO, Li−SiO 2 , and Li−SiO 2 /GO anodes are assessed and compared (Figure 4).
The cell with pristine Li can only operate for 58 cycles ( Figure  4a). The discharge potential stays at about 2.72 V before the 15th cycle and then gradually dropped to 2.22 V at the 58th cycle. The charge potential keeps increasing from 4.10 V from the 1st cycle and ends up with ca. 4.24 V at the 58th cycle. The charge/discharge profiles of the cell with the Li−GO anode in Figure 4b suggest a better cyclic performance 166 times. The discharge plateau is maintained at 2.79 V within the initial 58 cycles, which then reduced down to 2.63 V at the 166th cycle. The charge plateau starts from 3.90 V at the 1st cycle and gradually increases to 4.42 V at the 166th cycle. The cell with the Li−SiO 2 anode can cycle 187 times, as displayed in Figure 4c, during which the discharge voltage is maintained at 2.76 V within 120 cycles and finally decreases to 2.64 V at the 187th cycle. The charge voltage is located at 3.83 V at the 1st cycle and approaches 4.41 V at the end.
Interestingly, the Li−SiO 2 /GO anode (Figure 4d) presents the best battery performance with a significantly extended lifespan of the LOB cell to 348 cycles. The discharge voltage remains above 2.70 V for 230 cycles and then declines to 2.43 V at the 348th cycle. The charge voltage begins with just 3.03 V at the 1st cycle, which increases to 3.68 V at the 58th cycle, 3.98 V at the 230th cycle, 4.12 V at the 290th cycle, and finishes as 4.30 V at the 348th cycle. The LOB with the Li−SiO 2 /GO anode also renders a better rate performance than the others (Figure 4e  The LOB performance can be affected by the SiO 2 /GO mass ratio and loading amount. As shown in Figure S6a, the cycle numbers of the Li anodes with SiO 2 /GO coatings are larger than those with sole GO or SiO 2 coating; the best performance appears at the SiO 2 /GO mass ratio of 1/2. In Figure S6b, we obtain the cycle numbers of 160, 180, 348, 215, 192, 176, and 159 for the Li−SiO 2 /GO anodes with a SiO 2 /GO weight of 0.1, 0.2, 0.5, 0.6, 0.8, 1.0, and 1.5 mg, respectively, with the SiO 2 /GO mass ratio of 1/2. An optimized loading is determined to be 0.5 mg of the coating materials applied on the Li anode. The TEM images of SiO 2 /GO with different mass ratios demonstrate that SiO 2 NPs severely aggregate when the GO/SiO 2 ratio is 0.5/1 ( Figure S7a) and 1/1 (Figure S7b), and no SiO 2 NPs filling in some areas of GO layers and multilayered covering of GO are also observed when the GO/SiO 2 ratio is 3/1( Figure S7c) and 4/1( Figure S7d). The side-view SEM images of Li−SiO 2 /GO with different loading amounts ( Figure S8) reveal a smooth and compact morphology for the SiO 2 /GO layers when the loading amount is less than 0.6 mg, and a rough and cracked morphology when the loading mass exceeds 1.0 mg, which is in agreement with the above battery testing results.
Structural Examination and Failure Analysis for Variant Li Anodes after Cycling. SEI evolution of Li anodes is investigated by XPS ( Figure 5, Table S1). For the pristine Li anode (Figure 5a, top 3 spectra), the C 1s spectrum shows three peaks at 284.8, 286.8, and 288.8 eV; the Li 1s spectrum exhibits a peak at 55.4 eV; and the O 1s spectrum displays two peaks at 531.7 eV and 533.1 eV; these indicate the existence of RCOCOOLi after soaking in the PC electrolyte for 48 h. As  (Figure 5d, bottom), the C 1s spectrum shows four peaks at 284.8, 285.4, 288.8, and 289.9 eV; Li 1s and O 1s exhibit one peak at 55.4 and 532.6 eV, respectively. C 1s of 289.9 eV is assigned to Li 2 CO 3 , possibly because the oxygen groups of GO participate in the formation of SEI. These results imply that after 10 cycles, RCOOLi and Li 2 CO 3 are the main components of the SEI, and corrosion of the Li−SiO 2 /GO anode by H 2 O is inhibited by the SiO 2 /GO coating layer.
Morphological evolution after charge/discharge cycling reveals the role of surface coating in preserving the Li anode. Figure 6a show the pristine Li anode and the pulverized anode after 58 cycles, which originally possesses a thickness of 348 μm (Figure 6e), suggesting that the battery failure is associated with the fast consumption of Li. The Li−GO anode presents a rough surface at the 58th cycle (Figure 6b), where Li metal still retains a thickness of 172 μm (Figure 6f). The Li−SiO 2 anode also shows a coarsened and scraggy surface after cycling 58 times (Figure 6c), leaving the Li metal with a thickness of about 186 μm ( Figure 6g). As for the Li−SiO 2 /GO anode, a smooth and compact surface is retained after 58 cycles (Figure 6d) with the average thickness of metallic Li for up to 264 μm (Figure 6h), It has been reported that highly active cross-over oxygen reduction intermediates (O 2 − and LiO 2 − ) attack the Li anode and cause corrosion; they also react with the electrolyte and produce water molecules which are harmful to the Li metal. 15,18,23,31 Sun et al. revealed that the side reaction and the volume evolution of the Li anode also caused damage to the cathode using neutron tomography. 52 Our previous work indicates that the protection of the Li anode by creating a stable SEI can also alleviate the passivation of the cathode. 31 Here, Figure 7 shows that when the Li anode is protected with an artificial layer, such as GO, SiO 2 , and SiO 2 /GO, the accumulation of solid products on the cathode is reduced.
Accumulation of solid discharge products can result in the passivation of the MWNT cathode, thus leading to battery failure, which can be traced from the SEM image of the pristine MWNT cathode in Figure 7a. For the cell with a pristine Li anode (Figure 7b), the MWNT cathode is totally covered with large blocks of discharge products at the 58th cycle, representing the passivation of the battery cathode. As for the cells with Li− GO (Figure 7c Figure S10a, and the peak at 32.5°can be attributed to the (101) facet of the LiOH crystal (JCPDS no. 85-0736). Correspondingly, the MWNT cathode in the cell with the Li−SiO 2 /GO anode also presents one peak at 32.5°, which is attributed to the (101) facet of LiOH (JCPDS no. 85-0736) at the 58th cycle ( Figure S10b), and the intensity is much lower than that of the cell with pure Li, indicating that less solid products are accumulated, which is confirmed by the SEM image of the cathode (Figure 7e). At the 348th cycle ( Figure S10c), the intensity of the three identifying peaks (at 20.4, 35.7, and 51.4°attributed to the (001), (110), and (200) facets of LiOH) increased; the process is much slower than that of the MWNT cathode from the cell with pristine Li. The presence of LiOH on both the cathode and the anode indicates the irreversible consumption of Li. The application of a coating layer effectively reduces the Li loss that normally arose from the dendritic growth, volume change, and corrosion during the stripping/plating process, which subsequently reduces the polarization and alleviates the cathode passivation.
Further investigation using electrochemical impedance proves the protection effect of the artificial layer. After cycling 58 times, N 2 was introduced into LOBs to remove O 2 for 30 min, and then the resistance of LOBs was characterized by EIS. Figure S11 presents the Nyquist plots of the LOBs with pristine Li, Li−GO, Li−SiO 2 , and Li−SiO 2 /GO as anodes and the equivalent circuit. Solution resistance (R s ), constant phase element 1 (CPE 1 ), and charge-transfer resistance (R 1 ) at the high-frequency region are assigned to the impedance at the Li/electrolyte interface; the constant phase element (CPE 2 ) and charge-transfer resistance (R 2 ) at the medium frequency region contributed to the impedance at the solid product/cathode interface. Values of resistance are shown in Table S2. The   confirmed that the protection coatings led to a more stable Li anode and a less passivated cathode. We considered that the nano-/mesopores introduced by SiO 2 NPs in the coating layer not only facilitate the diffusion of Li ions but also act as a guide for the uniform Li ion flux in the stripping/ plating process, which reduces the risk of localized Li dissolution and suppresses the dendritic growth of Li. 53−58 In practical LOBs, the introduction of the oxygen reduction reaction brings soluble intermediates (such as O 2 − and LiO 2 − ) and the decomposed byproducts of the electrolyte. 40,58 These may immigrate through the GF separator and cause serious chemical corrosion of Li anodes. Here, the incorporation of SiO 2 and GO manifests a synergistic effect of the barrier effect, the ionic conductance, and the Li ion flux guidance. The SiO 2 /GO layer in this research provides the best protection for the Li anode, leading to the most stable LOB that has never been reported elsewhere.

■ CONCLUSIONS
In summary, we describe a facile strategy to preserve the Li anode by coating a SiO 2 /GO composite layer, so that the dendritic growth and chemical corrosion of the Li anode during the electrochemical activities can be largely minimized or prevented. The structural composite layer with a large amount of nanopores resulted from the intercalation of SiO 2 NPs among the GO sheets, yielding an enhanced transportation of Li ions. The resultant LOB with the Li−SiO 2 /GO anode can reach more than 348 cycles at 1 A·g −1 with a capacity of 1000 mA·h·g −1 , which is several times the cells with the pristine Li (58 cycles), Li−GO (166 cycles), and Li−SiO 2 (187 cycles) anodes. The rate performance and ultimate capacity of the LOB with the Li− SiO 2 /GO anode are also significantly improved. We hope that this low-cost coating strategy find applications in future LIB technologies.

* sı Supporting Information
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.0c08355. TEM images, ultimate capacity curves of LOBs, cyclic performance comparison of LOBs, SEM images, XRD patterns, Nyquist plots, XPS investigation of SEI evolution of LOBs, and EIS analysis of LOBs (PDF)