Scalable Clean Exfoliation of High‐Quality Few‐Layer Black Phosphorus for a Flexible Lithium Ion Battery

Few-layer black phosphorus (BP) nanosheets that are clean and of high quality, are efficiently produced by exfoliating bulk BP crystals, which are prepared by a scalable gas-phase catalytic transformation method in water. They are stable enough in water for further processing and applications. As an example, these BP nanosheets are combined with graphene to give high-performance flexible lithium-ion batteries. Atomically thin black phosphorus (BP) has recently attracted a great deal of interest because of its unique electronic and optical properties and a wide range of promising applications. It has a tunable direct bandgap bridging the energy gap between semimetallic graphene and various large-bandgap transition metal dichalcogenides (TMDCs),1, 2 a high carrier mobility,3 good current saturation characteristics,4 and in-plane anisotropic properties,5 which open up applications for it in electronics and optoelectronics such as high-performance radio-frequency6 and logic transistors,4 near- and mid-infrared photodetectors and modulators,7 mid-infrared polarizers and polarization sensors,8 and plasmonic devices.9 It has also been predicted to have negative Poisson's ratio,10 anomalous elastic properties,11 and high thermoelectric figure of merit.12 In addition, BP is a very promising high-specific-capacity electrode material for lithium ion batteries (LIBs) and sodium ion batteries.13, 14 Mechanical exfoliation of bulk BP was the first method to be developed and is currently still the main method of preparing atomically thin BP sheets that have high structural and electronic quality suitable for fundamental studies as well as electronic and optoelectronic devices, however, the yield is very low. For many important classes of applications, such as battery electrodes, printed electronics, and solar cells, large-scale production of defect-free atomically thin BP in a processable form such as liquid suspensions, inks, or dispersions is urgently required. Very recently, several groups have reported the production of few-layer BP nanosheets in dispersion form by liquid-phase exfoliation of bulk BP in appropriate solvents.15, 16 However, the solvents used for exfoliation have a high boiling point (N-methyl-2-pyrrolidone (NMP), ca. 204 °C; dimethyl formamide (DMF), ca. 153 °C; dimethyl sulfoxide (DMSO), ca. 189 °C; N-cyclohexyl-2-pyrrolidone (CHP), ca. 154 °C at 7 mm Hg), so they are difficult to remove when BP nanosheets are processed in films or composites, especially considering the low stability of BP nanosheets. Such solvent residuals cover the surface of BP nanosheets and inevitably limit the utilization of their intrinsic properties in applications. In addition, scalable synthesis of high-quality bulk BP with high efficiency is also challenging, although it is the prerequisite for large-scale production of BP nanosheets. Several methods have been developed for the production of bulk BP, including heating white phosphorus under high pressure,17 transforming white phosphorus in mercury18 or liquid bismuth,19 and transformation of red phosphorus using high-energy mechanical milling at ambient conditions.13 However, these methods either use toxic chemicals or complex apparatuses, or are time-consuming, or only give small BP crystals or BP nanoparticles. Compared to these methods, the recently developed mineralizer-assisted gas-phase transformation method shows a great potential to efficiently produce large-size BP crystals under simple and safe conditions.20, 21 Here, we report the synthesis of centimeter-size bulk BP crystals by an efficient mineralizer-assisted gas-phase transformation method and scalable clean production of few-layer BP nanosheets by exfoliating these BP crystals in water, utilizing the hydrophilic nature of BP. The bulk BP crystals show high purity and high quality, with a mobility of ca. 242 cm2 V−1 s−1 and current on/off ratio of ca. 5000 at room temperature for their few-layer counterparts made by mechanical exfoliation, and can be efficiently exfoliated in water to yield a few-layer BP nanosheet dispersion with high concentration. The BP nanosheets retain the high quality of the bulk crystals, have very high crystallinity, and are free of impurities and stable enough in water for further processing and applications. As an example, we demonstrate the use of these BP nanosheets for paper-like high-performance flexible LIB electrodes by combining them with highly conductive graphene sheets, which show a high specific capacity of 501 mAh g−1, excellent rate capability, and prolonged cycling performance at a current density of 500 mA g−1. The Experimental Section gives details of the synthesis of large-size high-quality BP crystals using the mineralizer-assisted gas-phase transformation method. Figure 1a shows a 3 mm-sized BP crystal, and BP crystals as large as 6 mm can also be obtained (Figure S1, Supporting Information). Energy dispersive X-ray spectroscopy (EDS) measurements show that the sample is composed of only P element without any other elements being observed (Figure S2, Supporting Information). Atomic absorption spectroscopy (AAS) and inductively coupled plasma atomic emission spectrometry (ICP-AES) measurements further indicate that the purity of the BP crystal is higher than 99.8 at%, and therefore it can be directly used without purification. X-ray diffraction (XRD) measurements confirm that the samples are BP crystals with high crystallinity (Figure 1b). As reported previously, both good crystallinity and high purity are very important for keeping BP stable in air.17 Therefore, our BP crystals show a high thermal stability; they can endure temperatures of as high as ca. 400 °C in air (Figure S3, Supporting Information). To further confirm the high quality of the BP crystals, we fabricated monolayer to few-layer BP flakes by mechanical exfoliation and measured their Raman spectra and electronic transport properties (Figure 1c–f). Atomic force microscopy (AFM) measurements show that each BP flake has uniform thickness and a smooth surface (Figure 1d), indicating that the bulk BP crystals have a well-defined layered structure and can be easily exfoliated. As shown in Figure 1e, all the BP flakes show three Raman peaks between 300 and 500 cm−1: (365 cm−1), B2g (442 cm−1), and (470 cm−1).22 Note that these peaks have small widths of ca. 1.9, 3.1, and 2.5 cm−1, respectively, indicating the high quality of the samples. As previously reported,23 the intensity ratio of the peak to Si peak (ca. 521 cm−1) increases linearly with the thickness for samples less than 15 nm thick (Figure 1e, and Figures S4, S5, Supporting Information), which allows the thickness of the BP flakes to be identified. Figure 1f shows the room temperature transport property of a 10 nm-thick few-layer BP-based field-effect transistor (FET) with a back gate on heavily doped silicon wafer covered with 285 nm SiO2. The extracted field-effect hole mobility can reach 242 cm2 V−1 s−1 with an on/off current ratio of ca. 5 × 103. It is worth noting that this transport property is comparable to those of BP flakes with the same thickness reported so far,3, 4 further confirming the high quality of the bulk BP crystals prepared by the scalable mineralizer-assisted gas-phase transformation method. Similar to graphene oxide, BP sheets are highly hydrophilic, which allows good dispersion in water without the use of any surfactant or solvents such as DMF, NMP, DMSO, or CHP. Therefore, we directly exfoliated bulk BP crystals in water by sonication, which gives a homogeneous BP nanosheet dispersion (Figure 2a).The color of the dispersion changes from light yellow to light black as the concentration is increased. The Tyndall effect was observed in the BP nanosheet dispersion (inset of Figure 2b), indicating the colloidal nature of the dispersion. We further used UV-vis absorption spectroscopy to characterize the dispersion (Figure S6, Supporting Information). It is worth noting that the absorbance divided by cell length (A/l) at wavelength λ = 684 nm shows a perfect linear relationship with the concentration (C) (Figure 2b). This behavior is consistent with the Lambert–Beer law, A/l = αC, yielding an absorption coefficient of α = 2767 L g−1 m−1. Such consistence suggests the good dispersion of BP nanosheets in water. It is well recognized that high-concentration dispersions are highly desirable for various practical applications. We systematically studied the influence of sonication power (Ps), initial BP concentration (Ci), and sonication time (ts) on the concentration of the dispersed BP nanosheets in water. The speed of centrifugation was fixed at 2500 revolutions per minute (rpm). As shown in Figure 2c–e, the concentration of dispersed BP nanosheets increases linearly with increasing Ps and ts. Considering that high sonication power might severely break the P–P bonds, a low Ps no more than 380 W was used in our exfoliation experiments. Compared to other parameters, Ci has a much stronger influence on the concentration of BP nanosheets. The concentration increases more than 17 times when Ci increases from 1 to 6 mg mL−1. BP nanosheet dispersion with a concentration as high as ca. 0.4 mg mL−1 can be obtained for Ci= 6 mg mL−1, Ps = 380 W, and ts = 300 min, which corresponds to a yield of 6.7 wt%. The yield can be further improved to ca. 30 wt% by recycling the sediment 4 times. Figure 2f shows 1 L BP nanosheet dispersion with a concentration of 0.17 mg mL−1. AFM measurements (Figure S7, Supporting Information) indicate that the mean thickness of BP nanosheets is ca. 9.4 nm (corresponding to ca. 18 layers) for a centrifugation speed of 2500 rpm (Figure 2g), while it is reduced to ca. 5.2 nm (corresponding to ca. 10 layers) when the centrifugation speed is increased to 5000 rpm (Figure 2h). We used X-ray photoelectron spectroscopy (XPS) and XRD to characterize the chemical composition and structure of the original bulk BP crystals and the BP nanosheets obtained. Both samples show a strong BP XPS peak at ca. 130 eV with a very tiny oxidized phosphorus (POx) sub-band (ca. 134 eV), indicating that no strong oxidation occurred during the exfoliation process in water (Figure S8, Supporting Information). The POx sub-band is attributed to weak oxidation of the BP surface in the presence of oxygen. Moreover, both samples show sharp XRD peaks, indicating that the BP nanosheets retain the high crystallinity of the bulk BP crystals (Figure S9, Supporting Information). In addition, it is worth noting that the BP nanosheets show some additional small XRD peaks that are absent in the bulk BP crystals (Figure S9). This suggests that the BP nanosheets are randomly distributed without preferential orientation when they are exfoliated from bulk BP crystals. We further used transmission electron microscopy (TEM) to characterize the detailed crystal structure of the BP nanosheets obtained. As shown in Figure 3a and Figure S10 (Supporting Information), the BP nanosheets are very thin and ca. 200 nm in lateral size (Figure 3a). The lattice constants extracted from the diffraction pattern (inset of Figure 3a) are 3.36 Å and 4.43 Å, which are consistent with those of bulk BP.24 Figure 3b shows a scanning TEM (STEM) high angle annular dark field (HAADF) image of two BP nanosheets. The corresponding EDS elemental mapping (Figure 3c) and EDS spectrum (Figure S11, Supporting Information) show that the nanosheets are composed of uniform P element without obvious O element being detected. High-resolution TEM (HRTEM) images show that the nanosheets are very clean, without visible impurities or defects, and have perfect orthogonally symmetric structure (Figure 3d). These results indicate that the exfoliated BP nanosheets produced by sonication in water retain the high quality of the original bulk BP crystals, which is consistent with the XPS and XRD measurements. Previous studies have shown that the exfoliated BP nanosheets degrade under ambient conditions, reacting in the presence of water or oxygen.25 In particular, it was found that water is predominantly responsible for the degradation process of BP nanosheets. Considering the use of water during our exfoliation process, it is necessary to estimate the stability of our BP nanosheets before further processing and applications. To do this, 1 L of BP nanosheet dispersion was sealed in a glass bottle and the absorbance at λ = 684 nm was recorded as a function of time. It is worth noting that only ca. 10 wt% of the BP nanosheets was dissolved after one week, and still ca. 60 wt% of the nanosheets was left even after eight weeks (Figure S12, Supporting Infomration). These results indicate that our BP nanosheets are stable enough in water for further processing and applications, which is different from the previous observations.25 We suggest that the high stability of our BP nanosheets can be attributed to their intrinsic high crystallinity and high purity, as shown above, as well as the oxygen-isolated measurement conditions. It can be clearly seen from Figure S13 (Supporting Information) that the presence of oxygen can greatly enhance the degradation of BP nanosheets immersed in water. These experimental observations are consistent with recent theoretical calculations,26 which show that water prefers to attach to the surface of BP through hydrogen but does not interact directly with the pristine lattice of BP; however, water will interact with BP once it has been oxidized. Therefore, the BP nanosheet dispersion should be stored in oxygen-isolated containers such as sealed bottles after preparation. The mass production of high-quality clean BP nanosheets opens up the possibility of applications beyond electronic and optoelectronic devices. It is well known that BP is an attractive anode material with an ultrahigh theoretical specific capacity of 2596 mAh g−1 (about 7 times that of the commonly used graphite anode material), which, however, faces many challenges, such as rapid capacity decay owing to the low electrical conductivity and severe volume variation during the lithiation/delithiation processes.13 Considering the excellent electrical conductivity of graphene and the two-dimensional flexible characteristic of both graphene and BP nanosheets,27, 28 we designed layer-structured flexible BP nanosheet–graphene (BP-G) hybrid paper as an anode, which not only solves the above problems but also provides a great potential for use in flexible energy storage devices. In addition, it is worth noting that no inactive metallic current collector, other binders, and conductive additives are used in this hybrid anode, therefore, it is also expected that the energy density of the whole electrode can be greatly improved. The BP-G hybrid paper was fabricated by vacuum filtration of a mixed dispersion of graphene sheets and few-layer BP nanosheets (see the Experimental Section). The graphene sheets, produced by an intercalation–exfoliation method, have a thickness of less than 10 layers and a lateral size of 1–5 μm.28 Importantly, these graphene sheets are almost free of oxygen functional groups and defects, and have a very high electrical conductivity of ca. 1000 S cm−1.28 As shown in Figure S14 (Supporting Information), the electrical conductivity of the hybrid electrode increases with the loading amount of graphene sheets. On the other hand, to make full use of the high specific capacity of BP and increase the mass specific/volumetric capacity of the LIBs, the amount of graphene sheets should be kept as low as possible on the condition that the conductivity is high enough. Considering the balance of the electrical conductivity and specific capacity of the hybrid electrode, therefore, 20 wt% graphene sheets and 80 wt% BP nanosheets were used in our experiments to fabricate the BP-G hybrid electrode. As shown in Figure 4a, the BP-G hybrid paper is mechanically robust and can be bent through nearly 180° without breaking, indicating its great potential for use as an electrode in future flexible energy storage devices.29 Both the top-view (Figure 4b) and cross-sectional view SEM images (Figure 4c) of the paper show that all the graphene and BP sheets are closely contacted in a face-to-face manner because of their two-dimensional structure, which can be further confirmed by TEM observations and EDS elemental mapping (Figure 4d–f). Such good contact together with the excellent electrical conductivity of graphene sheets enables a rapid electron transport between BP nanosheets. In addition, the very small size of BP nanosheets and the hierarchical open porous structure of BP-G hybrid papers lead to a short diffusion length of lithium ions and large electrode–electrolyte contact area. The big difference in lateral size (about 10 times difference) implies that the small BP nanosheets can be wrapped by large graphene sheets in the hybrid paper, and therefore, their volume expansion can be accommodated. As a result, the BP-G hybrid paper is expected to have a greatly improved rate capability and cycling stability compared to the BP nanosheets themselves. We also tried to fabricate flexible BP nanoparticle–G hybrid paper (Supporting Information), however, the hybrid paper obtained is very brittle and could hardly be peeled off from the filter membrane because the BP nanoparticles are sparsely dispersed on the surface of graphene without intimate contact (Figure S15, Supporting Information). It readily broke into small pieces during attempts to peel it off (Figure S15). This further demonstrates the advantages of combining different two-dimensional materials together to form a sheet-on-sheet structure for flexible electrochemical energy storage applications. The electrochemical lithium storage behaviors of the BP nanosheet, G paper, and BP-G hybrid paper anodes were examined by discharge/charge voltage profiles at a constant current of 100 mA g−1 as shown in Figure 4g. In the second discharge curve, two typical voltage plateaus are found around 0.63 and 0.25 V for the BP-G hybrid paper electrode, which are ascribed to the transformation from elemental P to LixP and finally to Li3P.13 In the charge curve, two plateaus, located at about 1.35 and 1.55 V, are found, which respectively originate from the reverse reaction for the formation of LixP and P from Li3P.13 In contrast, only one short plateau appears at around 0.25 V for the BP nanosheet electrode, and a declined slope below 0.5 V for the graphene paper electrode. Moreover, the voltage gap between discharge and charge potentials for BP and G paper electrodes are much larger than for the BP-G hybrid paper electrode, exhibiting higher overpotential. We characterized the BP nanosheets after the first cycle, and found that they had changed from crystalline state to amorphous state (Figures S16 and S17, Supporting Information), which are similar to those observed in red phosphorus–graphene composites.30, 31 However, it is worth noting that the amorphous BP is still well dispersed and adhered to the graphene without noticeable disconnection. These results confirm that the excellent electrical conductivity of graphene and the good contact between graphene and BP sheets significantly promote the reaction kinetics of the electrode, which can also be verified by the following rate capability tests. As shown in Figure 4h, the BP-G hybrid paper electrode delivers a high specific capacity of 920 mAh g−1 at a current density of 100 mA g−1, based on the total mass of the electrode, which is much higher than those of BP nanosheets (180 mAh g−1) and G paper electrodes (435 mAh g−1) based on the total mass of the electrode and the theoretical capacity of graphite (372 mAh g−1).32, 33 Moreover, this capacity is also much higher than the total sum (231 mAh g−1) of the individual capacity contributions of BP nanosheets (180 × 80% = 144 mAh g−1) and graphene (435 × 20% = 87 mAh g−1) based on their mass ratio in the hybrid, suggesting a synergistic effect between graphene and BP nanosheets that improves the electrochemical performance (Figure S18, Supporting Information). The volumetric capacity is another important concern for practical application of a BP nanosheet-based anode. After the binder and metal current collector had been replaced with lightweight graphene sheets, the volumetric capacity and volumetric energy density of the BP-G hybrid paper reached 1030 mAh cm−3 and 1571 Wh L−1, respectively, which are comparable or even superior to the other reported values for high volumetric capacity lithium battery anodes.34-36 When the discharge/charge current density is increased to 500 mA g−1 or even 2500 mA g−1, the BP-G paper electrode still shows high specific capacities of 501 and 141 mAh g−1, which are comparable to those previously reported,37, 38 implying the good reversibility of the BP-G hybrid paper electrode at high current density. In contrast, the G paper electrode drops dramatically to below 200 mAh g−1 and the BP nanosheet electrode fails to function at a current density of 500 mA g−1. Moreover, when the current density is changed from 2500 mA g−1 back to 500 mA g−1, the BP-G electrode recovers its capacity of 502 mAh g−1, indicating the good rate capability and structural stability. More importantly, the BP-G hybrid paper electrode shows excellent prolonged cycle performance at moderate current density. When cycled at 500 mA g−1 over 500 cycles, it could still deliver a capacity of 402 mAh g−1 with an average Coulombic efficiency approaching 100% (Figure 4i). This means that the capacity retention is 80.2% and the capacity decay rate is as low as 0.04% per cycle, which makes BP-G hybrid paper attractive and promising as a high capacity/energy anode material with long cycle life. Based on the above structure analyses, the excellent cycling stability and rate capability of BP-G hybrid paper electrode can be attributed to the good contact between graphene and BP nanosheets, favorable charge-transport pathway, short diffusion length of lithium ions, large electrode–electrolyte contact area, and the effective confinement of BP by graphene wrapping. In conclusion, using a scalable mineralizer-assisted gas-phase transformation method, we have prepared centimeter-size high-quality BP crystals that show a mobility of ca. 242 cm2 V−1 s−1 and current on/off ratio of ca. 5000 at room temperature for few-layer samples obtained by mechanical exfoliation. With these high-quality BP crystals as starting materials, we demonstrated the scalable clean production of BP nanosheets by liquid exfoliation in water, utilizing the hydrophilic nature of BP. It was found that the BP crystals can be efficiently exfoliated in water to yield a few-layer BP nanosheet dispersion with a high concentration. Moreover, the BP nanosheets retain the high quality of their bulk crystals, and are free of impurities and stable enough in water for further processing and applications. As an example, we demonstrated the use of these BP nanosheets for high-performance flexible paper-like LIB electrodes by combining them with highly conductive graphene sheets, which shows a high specific capacity of 501 mAh g−1, excellent rate capability, and prolonged cycling performance at a current density of 500 mA g−1. Synthesis of Large-Size BP Crystals: Red phosphorus (900 mg), AuSn alloy (360 mg), and SnI4 (18 mg) were first sealed in a quartz ampoule (13 cm in length and 15 mm in diameter) evacuated to a pressure lower than 10−3 mbar. Then the sealed ampoule was placed horizontally in the reaction zone of a tube furnace (Lindberg Blue M (TF55035KC-1)) and heated to 650 °C within 1 h. After being kept at 650 °C for 24 h, the ampoule was cooled to 500 °C at a rate of 40 °C h−1, and then cooled to room temperature after being held at 500 °C for at least 30 min. During the above synthesis process, large BP crystals (about 860 mg) were formed on the cold end of the ampoule. By prolonging the reaction time we could achieve almost full conversion. Finally, the large BP crystals were picked out and washed with toluene to remove the residual mineralizer, followed by water and acetone washing. Preparation of Few-Layer BP Nanosheet Dispersion: The BP crystals obtained were first ground to BP powders, and then the powders were dispersed in deionized water (20 mL) with an initial concentration of 1–10 mg mL−1 by tip sonication (Scientz-IID ultrasonic homogenizer, P0 = 950 W, output power: 0.1P0–0.5P0) for 30–300 min. After the dispersion had settled for 12 h, the supernatant was decanted and then centrifuged at 1500–5000 rpm for 30 min (TGL-16C, Shanghai Anting Scientific Instrument Factory). Finally, the resulting BP nanosheet dispersion (supernatant) was collected for further structural characterization and LIB applications. Structural Characterizations: The morphology and crystallinity of the large-size BP crystals were characterized by SEM/EDS (Nova NanoSEM 430, 10 kV/5 kV) and XRD (D-MAX/2400 using Cu Kα radiation). The purity of the bulk BP crystals was characterized by AAS (Z2300, Hitachi) and ICP-AES (PE8300, PerkinElmer). The structure of the few-layer BP flakes made by mechanical exfoliation was characterized by optical microscopy using a Nikon ECLIPSE LV100D and Raman spectroscopy using a LabRAM HR800 (632.8 nm He-Ne laser, spot size ca. 1 μm2, 100× objective lens). The thermal stability was measured from 30 °C to 600 °C at a heating rate of 10 °C min−1 in air or Ar using a Netzsch-STA 449C, and the electrical properties were measured under high vacuum (10−5 torr) at room temperature using a Keithley 4200 semiconductor parameter analyzer. The absorption of BP nanosheet dispersions was measured using a UV-vis absorption spectrometer (JACSO V-550). The thicknesses of the few-layer BP nanosheets exfoliated in water were measured using AFM (Nanoscope IIIa). The detailed chemical compositions of bulk BP crystals and BP nanosheets were characterized by XPS with an ESCALAB250 (150 W, spot size 500 mm) using Al Kα radiation; all spectra were calibrated to the binding energy of adventitious carbon (284.8 eV). Their detailed structure was characterized by TEM (JEOL JEM 2010, 200 kV), STEM (FEI Tecnai F30, 300 kV), and HRTEM (FEI Titan3 G2 60–300 S/TEM, fitted with two CEOS Cs aberration correctors and monochromator, 60 kV). AFM samples were prepared by dropping the BP nanosheet dispersion onto SiO2/Si substrates followed by drying on a hot plate and acetone washing. TEM samples were prepared by directly dropping the BP nanosheet dispersion onto carbon grids (230 mesh) followed by acetone washing. Fabrication of BP-G Hybrid Paper and G Paper and Electrochemical Measurements: The electrochemical properties of BP nanosheets, G paper, and BP-G hybrid paper as anode materials in half cells were evaluated by a galvanostatic charge/discharge technique. The BP-G hybrid paper and G paper were fabricated by vacuum filtration. The few-layer BP nanosheet dispersion was first filtered, washed with ethanol, and dried overnight at 60 °C in vacuum. Then the mixture of the obtained BP nanosheet powder (80 wt%) and few-layer graphene powder made by the intercalation–exfoliation method (20 wt%) or pure few-layer graphene powder was dispersed in NMP, followed by tip-sonication (190 W, 60 min) and vacuum filtration. After being washed with ethanol and drying overnight at 100 °C in vacuum, the BP-G hybrid paper or G paper was peeled off from the filter membrane. The obtained BP-G hybrid paper or G paper was directly used as the anode, and lithium-metal foil was used as the counter and reference electrodes, which were separated by Celgard 2400 separator. The mass loading of BP-G paper or G paper was 1–1.5 mg cm−2, and the capacity was calculated based on the total mass of the electrode. The BP nanosheet electrode was prepared by mixing 70 wt% BP powder with 20 wt% conductive carbon black (super P) as a conducting agent and 10 wt% polyvinylidene fluoride dissolved in NMP as a binder to form a slurry, which was then coated onto a copper foil and dried under vacuum at 100 °C for 12 h. The foil was shaped into a circular disk with a diameter of 12 mm and finally dried in a vacuum oven at 120 °C for 6 h. The total mass of the electrode was about 1.3–1.7 mg, and the geometrical area of the electrode was 1.13 cm2. Coin cells (size 2032) were assembled in an argon-filled glove box with the BP nanosheet electrode or G paper electrode or BP-G hybrid paper electrode, and a mixture of 1 M LiPF6 in ethylene carbonate, dimethyl carbonate, and ethylmethyl carbonate (1:1:1 vol) as the electrolyte. Charge/discharge measurements were carried out galvanostatically at various current densities over a voltage range of 0.001 to 3 V (vs. Li+/Li0) using a battery testing system (LAND CT2001A). L.C. and G.M.Z. contributed equally to this work. This work was supported by the National Natural Science Foundation of China (Nos. 51325205, 51290273, 51221264, and 51172240) and the Chinese Academy of Sciences (Nos. KGZD-EW-303–1 and KGZD-EW-T06). As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.