Peering into few-layer black phosphorus nanosheets: from preparation to battery applications

Few-layer black phosphorus (FLBP) nanosheets feature high charge storage capacity, large surface area, considerable mechanical flexibility, high carrier mobility and adjustable intrinsic band gap, allowing wide applications in electrochemical storage and conversion. This article presents a review on the preparation of FLBP materials and their applications in rechargeable batteries, including alkali metal batteries. Top-down and bottom-up synthetic strategies of FLBP are overviewed with examples of mechanical, liquid-phase and electrochemical exfoliation routes and chemical vapour and pulsed laser deposition. The electrochemical properties, structure-performance relationship and electrode mechanisms of FLBP are demonstrated as anode materials for storage of lithium, sodium and potassium ions. Issues including huge volume expansion and structural instability are discussed, along with solving strategies such as composing with nanostructured carbon, MXene, conductive polymer and transition metal oxides. Furthermore, the remaining challenges and future perspectives for the electrochemical use of FLBP are highlighted.

Recently, FLBP has been intensively investigated for electrochemical energy conversion and storage, particularly in rechargeable batteries [19][20][21].As an anode material, BP electrochemically reacts with Li, Na and K to form Li 3 P, Na 3 P and K 4 P 3 alloys, delivering theoretical specific capacity of 2596, 2596 and 1154 mAh g −1 , respectively.Additionally, the high electron conductivity, ionic diffusivity and surface area of FLBP aids the anchoring of polysulfides in the sulphur cathode to restrain the shuttling problems in lithium-sulfur batteries (LSBs).The last few years have witnessed progress in the electrochemical application of FLBP.Herein, we intend to provide an overview of this emerging field.We begin with a brief introduction of the preparation and structural properties of FLBP.Then, the latest advance in the electrode use of FLBP for rechargeable lithium-ion batteries (LIBs), sodium-ion batteries (SIBs), potassium-ion batteries (PIBs) and LSBs are sequentially presented with selected examples and discussion on the issues and solving strategies.Finally, the remaining challenges and future perspectives are highlighted in the field of applying FLBP in electrochemical energy devices.

Preparation of FLBP
Like in the case of other 2D materials, the synthetic strategies of FLBP can be generally divided into two categories of top-down and bottom-up methods, as schematically shown in figure 2. Top-down routes start with bulk BP precursor and take advantage of external physical or chemical driving forces to exfoliate BP layers in different media, which can serve as dispersion agents to prevent layer restacking.For the bottom-up approaches, phosphorus is often evaporated firstly and deposited on a desired substrate after condensation.Heating or pulse laser is usually adopted for the evaporation process.In this section, the preparation of FLBP is briefly summarized with different exfoliation and deposition approaches.

Mechanical exfoliation
The traditional mechanical exfoliation method is performed using scotch tape to obtain FLBP.In brief, bulk BP is adhered on a slice of scotch tape, then another slice of scotch tape is pasted on the other side of BP for shedding [27,28].By repeating the process several times, FLBP can be obtained and is transferred onto Si/SiO 2 substrate to remove the tape with organic solvents such as methanol, isopropyl alcohol (IPA) and acetone.By this method, Martel group prepared mono-, bi-and multilayer FLBP [22].For ultrathin layers, the lower intensity ratio of A g 1 /A g 2 was a signature of oxidation (figures 3(a)-(c)).In addition, the position of Raman peak was sensitive to the thickness of FLBP (figure 3(d)).Sha et al amended the traditional scotch tape mechanical exfoliation method by introducing metal (Au or Ag) onto Si/SiO 2 substrate (figure 3(e)) [29].As the binding energy between the metal and FLBP was higher than the interlayer van der Waals force in bulk BP, FLBP was easily peeled off from bulk BP.The FLBP exfoliated by Au-assisted method was 100 times as large as that exfoliated by traditional scotch tape method (figures 3(f)-(h)).The obtained FLBP featured with the lateral size of larger than 50 µm, hole mobility of 68.6 cm 2 V −1 s −1 and current on/off ratio of 2 × 10 5 .The mechanical exfoliation method favours preparation of FLBP with large size and high crystal quality, but falls short in low yield, limiting scale-up production and practical application.

Liquid-phase exfoliation
Liquid-phase ultrasonic exfoliation has been widely used to strip 2D materials.This method usually consists of three steps: (a) deconcentration of bulk BP in water or organic solvent, (b) ultrasonication for stripping, and (c) refinement [30][31][32].In the second step, ultrasonic wave cavitates the bubbles and the pressure released by the bubble exfoliates the phosphorene layers held by van der Waals interaction.Sonication time directly affects the stripping.Liu group discussed the effect of sonication on the size and thickness of FLBP [33].The layer number and lateral size of FLBP was found to decrease with increasing sonication time (figures 4(a)-(d)).FLBP prepared at ultrasonic time of 60 min showed the optimized structure with thickness of 1.9-2.1 nm and horizontal dimension of 10-15 µm.Additonally, the type of solvent plays an important role in this exfoliation method.Hersam et al peeled off bulk BP in different organic solvents such as IPA, N-methyl-2-pyrrolidone (NMP), hexane, ethanol, dimethyl formamide, acetone and chloroform [34].The FLBP concentration increased with surface tension and boiling point of the solvent (figures 4(e) and (f)).Among seven solvents, NMP was optimized to obtain stable FLBP dispersion.The lateral size and the thickness of FLBP were controlled by the centrifugation speed.To decrease BP degradation, the prepared FLBP is needed to store under dark condition with relative humidity of ∼21.2% and average temperature of 27 • C. The exfoliated FLBP possessed a considerable hole mobility of 25.9 cm 2 V −1 s −1 and a current on/off (e) Illustration of the metal-assisted exfoliation process for FLBP.(f) Optical microscopy of exfoliated FLBP on Au layer.(g) FLBP exfoliated using the 'scotch-tape' method (left) and the Au-assisted method (right).(h) Total area of FLBP on ten different samples.Reproduced from [29] with permission of The Royal Society of Chemistry.ratio of 1.6 × 10 3 .Xie group also reported the use of oxygen-free water to strip bulk BP [35].Sonication for 8 h, the concentration of FLBP with thickness of 2 nm (about four layers) was estimated to be 0.02 mg ml −1 .Water-exfoliated FLBP could maintain its high singlet oxygen generation ability for catalysis and photodynamic applications.Furthermore, peeling agents include other solvents, such as dimethyl sulfoxide, N-cyclohexyl-2-pyrrolidone (CHP), GBL (γ-butyrolactone) and so on [15,36,37].
To augment the production yield and water stability of FLBP, Yu and his co-workers designed a basic NMP liquid exfoliation route to prepare FLBP with controllable dimension and thickness [13].The sonication process was conducted in a saturated NaOH/NMP solution.Zeta potential of the basic NMP-exfoliated FLBP (−30.9 mV) was lower than that of pure NMP-exfoliated FLBP (−19.7 mV), indicating higher stability of the former.The OH ions absorbed on the surface of FLBP resulted in negative charge and thus enhanced stability in water.The obtained FLBP showed a lateral size and thickness of about 670 and 5.3 ± 2.0 nm, respectively, when the dispersed FLBP solution was centrifuged at 12000 rpm (figure 4(g)).Increasing the centrifugation to 18000 rpm reduced the thickness and average diameter (figure 4(h)).
Compared to traditional organic solvents, ionic liquids (ILs) feature advantages such as high ionic conductivity, high thermal stability, excellent solubility non-volatility and solvent recyclability [39].Thus, ILs result in with high concentration and stability of peeled-off 2D materials such as graphite and MoS 2 nanosheets [40,41].Mu et al utilized nine ILs to produce FLBP through liquid exfoliation [38].Coulombic forces and π-π interactions between aromatic ILs cations and FLBP layers exerted significant roles in stripping of bulk BP and stabilization of FLBP nanosheets.The anions of ILs affected the concentrations of FLBP through the sequence of surface tension and hydrogen-bond ability, while the cations of ILs affected the concentration via changing cationic chain length.The concentration of FLBP attained as high as 0.95 mg ml −1 when sonicated in 1-hydroxyethyl-3-methylimidazolium trifluoromethansulfonate ([HOEMIM]-[TfO]), and the thickness of FLBP was approximately 3.58-8.90nm (figures 4(i)-(k)).The ILs-assisted liquid-phase sonication exfoliation benefits high concentration and stability but is costly.Using cheap, non-toxic solvents to improve the yield and achieve controllable thickness and horizontal dimension of FLBP deserves further studies.

Electrochemical exfoliation
Electrochemical exfoliation is a simple, rapid and environmentally friendly method to synthesize FLBP nanosheets.This process is usually performed at a selected working voltage in aqueous or organic electrolytes.In the common two-electrode system, bulk BP and Pt serve as the working and counter electrode, respectively, and inorganic/organic salt aqueous solution or deoxygenated propylene carbonate as electrolyte [42,43].Under electric field, cations such as Li + , Na + , K + , and tetraalkylammonium (alkyl = methyl, ethyl, or n-butyl) cations (referred as TMA + , TEA + , or TBA + , respectively) migrate inside BP to resist the van der Waals interactions between BP layers (figures 5(a) and (b)).Whether bulk BP can be successfully peeled off depends on the applied voltage and electrolyte.Feng et al studied the influence of electrolytes on the preparation of FLBP [44] and demonstrated that increasing the radii of intercalated cations (from Li + 0.09 nm to TBA + 0.83 nm) resulted in an increase of the yield from 5% to 78%, as shown in figure 5(c).A 150 s electrochemical exfoliation at −8 V generated FLBP with the thickness of 1.3-9.5 nm and the horizontal dimension of 2-20.6 µm (figures 5(d) and (e)).
In another example, Xie et al chose hexadecyltrimethylammonium chloride (CTAC) as intercalation agent to strip bulk BP in water [23].Under applied voltage, the CTA + intercalated in bulk BP and then decomposed to form trimethylamine bubbles.In the exfoliation process, the applied voltage, liquid temperature and concentration of CTAC were found to have large impact on the stripping of BP.By applying a negative potential (−30 V) for 20 s, exfoliation process could not take place at lower temperature of 50 • C.Under optimized condition of −30 V, 0.5 M CTAC and 50 • C, the acquired FLBP nanosheets had thickness of 5-8 nm (figures 5(f) and (g)).The adsorption of CTA + on the surface of FLBP nanaosheets contributed to improve the stability of FLBP, as shown in figure 5(h).Accordingly, uniform FLBP with controllable layer number can be synthesized by optimizing electrolyte, voltage and centrifugation in the electrochemical exfoliation process.Electrochemical exfoliation exhibits high peeling efficiency and yield.Effectively removing the residual electrolyte on the surface of FLBP and preventing the oxidation of FLBP are the problems that need to be solved for the industrialization development of this method.

Solid-phase exfoliation
For solid-phase exfoliation of 2D materials, ball milling is often carried out with stripping agent.The generated shear force from milling breaks the weak van der Waals interactions between adjacent π-π stacked layers, similar to mechanical cleavage [45].Simultaneously, the stripping agent adsorbed on the surface of 2D materials counteracts van der Waals forces, facilitating the exfoliation process.This method has been utilized to peel off 2D materials such as graphite, BN and graphite fluoride [46][47][48].Common stripping agents include ammonia borane, melamine, alkali hydroxide, oxalic acid and urea [45,47,[49][50][51][52].Along this line, Yang et al ball-milled a mixture of bulk BP and LiOH at 250 rpm for 24 h to prepare FLBP with thickness of Surface functionalization of FLBP can be achieved simultaneously in ball-milling exfoliation of BP with stripping agents.For example, Cheng et al designed a green, fast and economic strategy to strip bulk BP [54].FLBP with a length of 150-400 nm and thickness of 2.15-4.87nm was synthesized by ball-milled bulk BP and urea at 150 rpm for 4 h (figures 6(a)-(c)).As a small polar molecule, urea was easily absorbed on the surface of bulk BP, compensating for van der Waals interactions between BP layers.X-ray photoelectron (XPS) and Fourier transform infrared (FTIR) spectra (figures 6(d)-(f)) revealed the functionalization of NH 2 groups on the edge of exfoliated FLBP, preventing the restacking and promoting the formation of nanosheets.The presence of functionalized FLBP would enhance electrochemical performance.Solid-phase exfoliation is a facile, fast, green and economical approach to obtain functionalized FLBP.However, the synthesis of FLBP with lateral dimension larger than 1 µm through this method still needs further experimental investigation.

Bottom-up methods
Chemical vapour deposition (CVD), in which gaseous phase is transferred to the solid form, is the most common bottom-up method to prepare 2D materials.For example, Wang et al directly prepared BP nanoribbons with the width of 400 nm, length of 60 µm and thickness of 20 nm from red phosphorus (RP) [55].Liu designed an efficient short-distance transport (SDT) growth approach to prepare high-quality BP with high yield of 98%, as shown in figures 7(a) and (b) [56].Their results also indicated that doping of BP by Sb or Te could improve the air stability of BP.In addition to CVD, pulsed laser deposition (PLD) can be processed at a lower temperature.As an example, Lau et al prepared amorphous FLBP with thickness of 2-10 nm at 150 • C [57].Interestingly, the band gap of FLBP decreased from 0.8 to 0.21-0.26eV with the increasing of thickness from 2 to 8 nm.The FLBP with a thickness of 2 nm showed high field-effect mobility of 14 cm 2 V −1 s −1 and moderate on/off current ratio of 100, much higher than other conventional elemental amorphous materials.FLBP equipped large size and surface area can synthesize through these two methods by optimizing the preparation conditions.But the technology, synthesis condition and cost limit the applications of these methods.
Besides the CVD and PLD techniques, liquid phase reactions are also adopted as bottom-up approaches to prepare FLBP.Hao et al prepared FLBP with the thickness of about 3 nm through hydrothermal method using RP and NH 4 F as raw materials at 200 • C for 16 h [58].In the hydrothermal process, the interaction between RP and NH 4 F and water promoted conversion of spherical RP to FLBP nanosheets, but the specific growth mechanism was not clear.Yan et al utilized a solvothermal reaction to produce FLBP by dispersing RP in ethanol, which was maintained at 400 • C in autoclave [26].This processing involved vapour-solid transformation generated from continuous vaporization-condensation procedure and consequent bottom-up assembly growth.The generated FLBP was holey with thickness of 0.5-4 nm (figures 7(c)-(e)), showing potential for electrochemical applications.Compared to CVD and PLD techniques, the complicacy and cost of the equipment are relatively decreased for liquid phase approach.
On the whole, the development of these bottom-up methods is still in the preliminary research stage.The safety of the experiment needs to be given full attention because the growth of FLBP usually starts from red phosphorus, restricting the practical applications.Hence, although bottom-up methods can be applied to synthesize FLBP (even monolayer phosphorene) with controllable layer number and large lateral dimension, are often demanding in experimental facility, conditions and cost control.Mechanical exfoliation method is conductive to the synthesis of FLBP with large size and high crystal quality, but is low in yield.Liquid-phase sonication exfoliation is the common and dominant top-down strategy to produce FLBP due to its simple apparatus and easy operation, whereas the yield and dimension control need to be further enhanced.Electrochemical exfoliation reveals high peeling efficiency and yield, which has been regarded as a potential technique in scale-up production and practical application.Solid-phase exfoliation is a facile, fast, green and economical approach to obtain FLBP without problem of residual solvent, but further research is needed to prepare FLBP with large lateral dimension.To sum up, top-down strategies are advantageous in simple and cost-effective processing.Functionalized FLBP is usually prepared by top-down methods to introduce functional groups during the preparation process.However, the lateral dimension and thickness are quite difficult to be controlled precisely.

FLBP for LIBs
As an anode for LIBs, the theoretical capacity of BP is as high as 2596 mAh g −1 due to the low molar weight and three-electron alloying-dealloying mechanism corresponding to the formation of multiple phases including LiP, Li 2 P, and Li 3 P [59,60].This capacity is significantly higher than the theoretical capacity of commercial graphite anode (372 mAh g −1 ).Moreover, among the allotropes of phosphorus, BP exhibits superior electric conductivity of 300 S m −1 , large layered structure with layer spacing of 5.3 Å and moderate density of 2.69 g cm −3 , facilitating high rate electrochemical performance [61].The average potential for Li-P alloy is about 0.7 V, which is suitable for high potential battery.During the lithiation process, Li + intercalates into BP layers along zigzag direction coupled with expansion along armchair direction [8].The severe volume expansion (300%) gives rise to inferior cycling performance [60,61].Strategies proposed to solve these issues include combining BP with other conductive materials and preparing FLBP to accommodate volumetric strain [20,[62][63][64].Compared with bulk BP, FLBP features adjustable band gap, high carrier mobility, large surface area and short ion diffusion path.The Li + diffusivity of monolayer phosphorene is 10 4 times that of graphene [65].Diffusion energy barrier along zigzag direction for phosphorene is 0.18 eV, much smaller than that of graphene (0.327 eV) [66].In addition, monolayer phosphorene experiences a small volume change of 0.2% during discharge-charge process [67].Because of these advantages, FLBP has been served as anode material for LIBs.
For lithium storage application, Chen et al prepared holey FLBP with lateral size of 1.0 µm and thickness of around 0.5-4 nm (1-8 layers) by sublimation-induced transformation method [26].This electrode exhibited high initial discharge and charge capacities of 2696 and 1969 mAh g −1 at 0.2 A g −1 , maintaining 62.4% of initial discharge capacity over 100 cycles.When the current density increased from 0.5 to 20 A g −1 , the discharge capacity decreased from 1600 to 630 mAh g −1 .Moderate cycling and rate performance were ascribed to the unique construction of the holey FLBP.Li-ions can be stored in the large nanopores to react with the neighboring layers of phosphorus, showing fast ion diffusion and electron transfer.In another example, FLBP was synthesized by a hydrothermal method using red phosphorus as raw material and NH 4 F as peeling agent [58].The 1st and 100th discharge capacity was 789 and 432 mAh g −1 , respectively.Additonally, Bonaccorso group obtained FLBP with an average horizontal dimension and thickness of 30 and 7 nm by liquid-phase exfoliation in low boiling point solvent (acetone) and compared the electrochemical performance of FLBP anodes synthesized from exfoliation in different solvents [68].As shown in figure 8(a), FLBP acetone delivered specific capacities of 382 and 345 mAh g −1 at 0.5 and 1 A g −1 , higher than those of FLBP CHP (below 200 mAh g −1 ).Additionally, Raman and energy-dispersive x-ray (EDX) elemental mapping revealed that FLBP acetone flakes homogeneously distributed on the Cu substrate due to the fast solvent removal (figures 8(b)-(g)).
Moreover, Lee and co-workers studied the effect of FLBP layers number on electrochemical performance [20].The layer number and lateral size were controlled by exfoliating at different solvents.The change of the layer number presented the following trend of FLBP hexane (3-5) < FLBP 2-propanol (8-10) ≈ FLBP methanol (8)(9)(10), while the lateral size depicting FLBP methanol (<2 µm) < FLBP hexane (2-4 µm) < FLBP 2-propanol (>5 µm).Specifically, FLBP hexane , FLBP methanol and FLBP 2-propanol showed stable cycling performance and capacities of 404.4,340.4 and 307.4 mAh g −1 after 200 cycles at 1 A g −1 .Among these samples, FLBP hexane exhibited the highest capacity at high rate.EIS curves demonstrated that thinner FLBP was in favour of Li + diffusion and charge transfer while larger lateral size could enhance the charge transfer despite increasing the Li + diffusion path.Hence, the optimization of aspect ratio contributes to enhanced electrochemical properties of FLBP.
Neat FLBP showed cycling stability but suffered from poor capacity and low initial Coulombic efficiency (40.2%) due to the formation of solid-electrolyte interphase (SEI) film on the surface of the nanosheets.Engineering the structures of FLBP with conductive carbon is effective to enhance electrochemical performance.Ren et al described a simple vacuum filtration method to obtain flexible FLBP-graphene paper electrode [69].The large lateral size difference (about ten times) between small FLBP and large graphene revealed that FLBP could be wrapped by graphene in the freestanding paper.This structure was beneficial to relieve the volume expansion during the discharge-charge process.Furthermore, the open hierarchical porous structure of the hybrid paper and the small size of FLBP brought out a short Li + diffusion path and large electrode-electrolyte contact area.As a result, this hybrid paper exhibited capacities of 920, 501 and 141 mAh g −1 at 0.1, 0.5 and 2.5 A g −1 , respectively.The electrode maintained 402 mAh g −1 over 500 cycles at 0.5 A g −1 .Without binder and metal current collector, FLBP-graphene paper electrode showed volumetric capacity and volumetric energy as high as 1030 mAh cm −3 and 1571 Wh L −1 , respectively.Using similar membrane technology, sandwich-structure G-BPGO thin film, which consisted of rGO-BPGO-rGO layers, was deposited on LIBs separator [70].The two layers of reduced graphene oxide (rGO) located on both sides of BPGO allowed a rapid electron transport through the hybrid film and protected BPGO from oxidation.This film electrode exhibited high initial discharge capacity of 2587 mAh g −1 with Coulombic efficiency of 71% and retained capacity of 1401 mAh g −1 over 200 cycles at 0.1 A g −1 , together with good rate capability of 656 mAh g −1 at 1 A g −1 .
To improve the stability of FLBP, Yan group prepared densely packed FLBP-graphene oxide paper (PG-SPS) by a simple and easily scalable spark plasma sintering (SPS) process [36].AFM and TEM images (figures 8(h) and (i)) showed average thickness of 3.1-4.3nm, corresponding to 7 ± 1 layers of phosphorene.The horizontal dimension of FLBP was about 50-1000 nm.Compared with pure FLBP and loosely stacked PG paper without SPS treatment, PG-SPS paper (figure 8(j)) could maintain excellent air stability in 60 d owing to effectively restrained permeation of H 2 O and O 2 in the hybrid paper.On account of reduced exposed surface and improved electrical connection between FLBP and rGO, PG-SPS depicted higher initial Coulombic efficiency of 60.2% than pure FLBP (11.5%) and PG paper (34.3%, figure 8(k)).This stable hybrid paper also delivered high gravimetric capacity (1306.7 mAh g −1 ) and volumetric capacity (256.4 mAh cm −3 ) at 0.1 A g −1 , superior cycling stability (91.9% retention over 800 cycles at 10 A g −1 ), and rate capacity (415 mAh g −1 at 10 A g −1 , figure 8(l)).
Conductive polymer also plays an important role in the morphology and structure of FLBP-based composites.For example, polyaniline (PANI) is a good coating layer candidate due to its high conductivity and easy preparation.Duan and co-workers fabricated PANI encapsulated BP-graphite by ball milling a mixture of BP and graphite and then in-situ polymerization of PANI [71].The thin PANI gel coating swollen by electrolytes was beneficial to generate stable SEI film, which restrained the continuous accumulation of poorly conductive Li fluorides and carbonates.In addition, the swollen PANI induced Li + and protons doping in the polymer matrix and absorbed corrosive hydrogen fluoride, which promoted charge transfer throughout the electrode and electrode-electrolyte interface (figure 9(a)).Hence, this electrode exhibited remarkable rate and cycling ability, including a high initial charge capacity of 1650 mAh g −1 at 0.26 A g −1 (figure 9(b)), reversible capacities of 440 mAh g −1 at high current densities of 13 A g −1 over 2000 cycles (figure 9(c)).At loading mass of 3.6 mg cm −2 , the capacity maintained at 750 mAh g −1 at 0.52 A g −1 .
Coating transition metal compounds is another strategy to enhance the electrochemical performance of FLBP.Metallic Ni 2 P is regarded as high-performance anode for LIBs due to its high activity and conductivity.Heterostructures of FLBP and Ni 2 P nanocrystal equip superiorities.On the one hand, Ni 2 P provides electrons to FLBP and effectively adjusts charge carrier concentration, while FLBP supplies diffusion paths for charge carriers with less interface scatting.These features are responsible to improve electrical conductivity.On the other hand, the deposition of Ni 2 P nanocrystal on the surface of FLBP can not only prevent restacking of FLBP but also suppress agglomeration of Ni 2 P nanocrystal, providing abundant active sites for electrochemical reaction.Based on these advantages, Yan group synthesized Ni 2 P@FLBP by coating Ni 2 P nanocrystal with approximate size of 5 nm on the surface of FLBP using solvothermal method [73].Compared with pure BP, Ni 2 P@FLBP possessed improved electrical conductivity (2.12 × 10 2 vs 6.25 × 10 4 S m -1 ), higher charge carrier concentration (1.25 × 10 17 vs 1.37 × 10 20 cm -3 ) and higher Li + diffusion coefficient (1.14 × 10 -14 vs 8.02 × 10 -13 cm 2 s -1 ).As a result, Ni 2 P@FLBP delivered superior capacities of 1196.3 and 602.8 mAh g −1 at 0.1 and 2 A g −1 , respectively, which were significantly higher than those of pure BP (290.9 and 68.6 mAh g −1 ).
As TiO 2 shows zero-strain characteristics and transfers electronic conductor after lithiation process, coating TiO 2 can effectively enhance the electrochemical properties of FLBP.For example, Wang prepared BP@TiO 2 composite electrode by coating TiO 2 nanolayers (thickness of about 50 nm) on BP electrode using vacuum evaporation method [74].This buffer layer prevented the connection between BP particles and electrolyte, lessening the in situ formed SEI layer on the BP electrode.The lithiation of the titanium layer also offered a stable Li x Ti y O z interface for accelerating Li + diffusion and electron transfer.As a result, the composite electrode delivered a capacity of 1049 mAh g −1 and Coulombic efficiency of 99% over 100 cycles at 0.2 C.
Metal-organic frameworks (MOFs) features porous stable structure and has been investigated as a potential candidate to solve the problems such as severe volume change and slow ion transport kinetics of FLBP.Zhou and his co-workers designed and prepared 2D FLBP/NiCo MOF hybrid using a facile method in a mixed solution of Ni 2+ , Co 2+ and benzenedicarboxylic acid (BDC) [72].The carboxylate groups in BDC anion not only chelated with metal ions (Ni 2+ , Co 2+ ) but also bonded with BP, transforming into a stable composite structure.The FLBP/NiCo MOF composite had large specific surface area for charge transport and could buffer the volume expansion during discharge-charge process because of 2D porous structure (figure 9(d)), which resulted in excellent cycling stability (853 mAh g −1 at 0.5 A g −1 over 200 cycles) and superior rate performance (398 mAh g −1 at 5 A g −1 over 1000 cycles), as shown in figures 9(e) and (f).
The electrochemical performances of FLBP and FLBP-based materials for LIBs are summarized in table 1. Coating with other materials such as carbon, conductive polymer, transition metal oxide, transition metal phosphates, and MOF can effectively increase the capacity and cycling properties of FLBP.However, the initial Coulombic efficiency remains low, leaving much room for further development of BP as a negative electrode material for LIBs.

FLBP for SIBs
SIBs have received much attention owing to the abundant resources and low cost of sodium.Furthermore, Al foil can be used as anode current collector to replace more expensive Cu foil because Na does not react with Al under a relatively low potential.Nevertheless, the larger ionic radius (1.02 Å) and higher molar mass (23 g mol −1 ) of Na result in sluggish kinetics of sodium storage in many host electrode materials, leading to lower energy density and unsatisfactory electrochemical performance [77,78].Advanced electrode materials are desirable for the development of SIBs.FLBP is one of the most promising anode materials for SIBs owing to its large interlayer spacing (5.3 Å), which far exceeds than that of graphite (3.4 Å).
In-situ TEM experiment was carried out to investigate the sodium storage mechanism of BP [21,79].Channels along x-and y-axes of BP were 3.08 and 1.16 Å (figure 10(a)), respectively.So firstly, Na + was inserted along the x-axis-oriented channels between the phosphorene layers to generate Na 0.17 P, while the distance along z-axis was also increased.This intercalation process imposed no apparent change on the structure of BP, thus delivering high cyclic reversibility in high potential range of 0.54-1.5 V. Subsequently, the alloy reaction occurred below 0.54 V, resulting in the formation of Na 3 P (corresponding theoretical capacity 2596 mAh g −1 ).This process led to a huge volume expansion of almost 500%, bringing out structure fracturing and significant capacity loss.From in-situ TEM images (figures 10(b)-(d)), the x-, y-, z-axes expansion was 0%, 92% and 160%, respectively.As a consequence, reducing the horizontal size and thickness of BP layers was beneficial to address the expansion stress along y-and z-axis directions.In this sense, electrode consisted of FLBP was appropriate anode material for SIBs.Ji group prepared FLBP with 2-11 layers by electrochemical cationic intercalation method [80].The thickness of FLBP could be controlled by the applied potential.The obtained FLBP delivered 1878.8,1108.5 and 321 mAh g −1 at 0.1, 0.5 and 2.5 A g −1 , respectively.
Small FLBP nanosheets sandwiched in large graphene sheets can offer many advantages.For example, the thin FLBP shortens the transport path of Na + and electrons to achieve favourable rate performance.Graphene layers on both sides of FLBP provide buffer regions to acclimatize anisotropic expansion for good cycling stability and the electrically conductive graphene accelerates the transfer of electrons from phosphorene to the current collector.In this regard, Cui group designed FLBP sandwiched between graphene layers with optimized phosphorene loading of 48 wt% [21].This hybrid electrode showed high charge capacity of 2440 mAh g −1 at 0.05 A g −1 and 83% capacity retention over 100 cycles (figure 10(e)).Even at high current densities of 12 and 26 A g −1 , the specific capacity attained 1200 and 645 mAh g −1 (figure 10(f)), respectively.Interestingly, the channel between graphene and phosphorene could be broadened to store more sodium ions when utilizing the composite of rGO and 4-nitrobenzenediazonium (4-NBD) modified BP [82].Additionally, electrophoretic deposition and chemical activation were adopted to prepare sandwich-structure
MXene, a new family of 2D materials with a formula of M n+1 X n T x (M=Ti, Nb, Mo, V, etc.; X=C, N; T=O, F, OH), possesses superior conductivity and steerable flexible interlayer spacing, which benefit to increase charge transport and electrochemical redox kinetics [87][88][89].As a most common MXene material, Ti 3 C 2 has a monolayer conductivity of 6.76 × 10 5 S m −1 .The diffusion barriers of Na + in Ti 3 C 2 with functional groups of F, O and OH are 0.19, 0.2 and 0.013 eV [90], respectively, indicating good Na + diffusion kinetics.Few-layer Ti 3 C 2 also exhibits considerable Na storage capacity of 267 mAh g −1 .Combining Ti 3 C 2 nanosheets with FLBP shows the following advantages.(a) The coupling of Ti 3 C 2 with FLBP can buffer the volume expansion and prevent the pulverization and aggregation of FLBP in cycling.(b) Ti 3 C 2 facilitates the transfer of both sodium ions and electrons.(c) The similar layered structures of Ti 3 C 2 and FLBP can form heterostructure that consisted of face-to-face contact, providing more ionic diffusion channels.As an example, Wang designed poly(diallyl dimethyl ammoniumchloride) (PDDA)-BP/Ti 3 C 2 composite as anode for SIBs [90].Modifying BP with PDDA could enhance the electronegativity of BP and avoid the oxidization of BP in water.This composite showed a high charge capacity of 1780 mAh g −1 at 0.1 A g −1 and respectable cycling stability of 658 mAh g −1 at 1 A g −1 after 2000 cycles.Furthermore, in-depth XPS anlaysis demonstrated that Ti 3 C 2 with F group exhibited stable cycling properties and enhanced Coulombic efficiency of FLBP/MXene owing to the formation of a fluorine-rich SEI film [91].
The electrochemical performances of FLBP and FLBP-based materials for SIBs are summarized in table 2. Coating with other materials such as carbon, conductive polymer and MXene can effectively strengthen the properties of FLBP.Compared to LIBs, the same problem of low initial Coulombic efficiency also exists in the application of FLBP in SIBs, which calls for more research efforts such as surface modification and electrolyte formulation.

FLBP for PIBs
Having close crust abundance to sodium, potassium-based batteries are attracting increasing interest [92][93][94][95][96][97].The redox potential of potassium deposition/plating is −2.88 V vs the standard hydrogen electrode, and graphite can be used as anode for PIBs.As K + has a smaller Stokes radius (3.6 Å) than that of Na + (4.6 Å) and Li + (4.8 Å), the lower desolvation energy gives rise to superior ion mobility and conductivity in the electrolyte.Nonetheless, the potassiation/depotassiation process often severely destroies the structure of the anode materials owing to the larger ion radius (1.38 Å) of K, leading to limited rate and cycling performance [98].It is desirable to exploit advanced host materials for fast potassium storage with high capacity and stable cycling properties.
The large interlayer distance (5.3 Å) of BP favours K + storage and transportation.An alloying mechanism to generate K 4 P 3 corresponds to a theoretical capacity of 1154 mAh g −1 [99].The alloying/dealloying process gives rise to large volume changes (∼411.78%).Koratkar utilized rGO as buffer layers to accommodate the huge volume expansion [99].The electrode exhibited reversible capacities of 710 and 230 mAh g −1 at current density of 0.1 and 1.2 A g −1 , respectively.At rate of 0.5 A g −1 , a capacity of 230 mAh g −1 with 0.16% capacity decay per cycle was observed over 300 cycles.To increase the long cycling performance, interstratification-assembled FLBP/V 2 CT x hybrid anode with 2D layer structure was designed via van der Waals interactions (figure 11(a)) [100].Characterizations from SEM, TEM and AFM (figures 11(b)-(g)) evidenced typical layer-by-layer stacking structure with thickness below 10 nm and location of FLBP layer between two ex-V 2 CT x nanosheets.The synergistic effect between FLBP and V 2 CT x supplied interlayer space for K + storage and 3D interconnected conductive network for fast K + transfer.

68
.7% [90] 658 mAh g −1 after 2000 cycles at 1 A g −1 Such hybrid structure also alleviated the volume change due to phase transformation upon cycling.Therefore, the composite electrode showed high charge capacity of 673 mAh g −1 at 0.1 A g −1 and superior cycling stability, with the capacity retention of 91% and 86% after 3000 cycles at 0.2 and 2 A g −1 , respectively (figures 11(h) and (i)).From table 3, it is clearly that the performance of FLBP as anode material for PIBs is far from satisfactory as there remains large gap between reversible capacity and theoretical value.Moreover, there are few studies on FLBP and FLBP-based materials as anodes for PIBs.

FLBP for LSBs
LSBs are attractive for their high energy density of 2600 Wh kg −1 and low cost of sulfur cathode.However, the shuttle effect of polysulfide, the solution of sulfur in electrolyte and the low electrical and ionic conductivity of sulfur lead to severe capacity loss, limiting the application of LSBs.To tackle these issues, phosphorene was introduced as highly effective polysulfide immobilizer and electrocatalyst.(i) Long-term cyclability of BPE@V2CTx.
Reproduced from [100] with permission of The Royal Society of Chemistry.carbon-based materials.In addition, the electrical conductivity of FLBP was 450 S m −1 , higher than that of metal oxides and carbon nitride.Hence, FLBP was regarded as a potential polysulfide immobilizer and electrocatalyst in LSBs.The edges of BP quantum dots also offer more catalytically active sites to effectively adsorb and catalyse polysulfide [102].Therefore, porous carbon/BP quantum dots exhibited a low rate of capacity loss (0.027% per cycle over 1000 cycles) as well as high capacity of 1266 and 784 mAh g −1 at 0.1 and 4 C, respectively.Furthermore, FLBP with high electronic conductivity and Li + diffusivity was deposited on the commercial Celgard separator to restrain polysulfides diffusion [103].During the discharge process, FLBP interacted with polysulfides through bonding with both Li and S atoms.The FLBP-modified electrode delivered a capacity retention of 86% after 100 cycles at 0.4 A g −1 , outperforming the graphene-modified cell (66%).

Summary and outlook
In summary, this article overviews the preparation of FLBP-based materials and their applications for rechargeable batteries.As shown in figure 13, the synthetic methods of FLBP involve mechanical exfoliation, liquid-phase ultrasonic exfoliation, electrochemical exfoliation, solid-phase exfoliation and bottom-up routes such as vapour deposition.The intrinsic electron conductivity, charge carrier mobility and large interlayer space of FLBP render considerable charge storage properties for lithium, sodium and potassium batteries.Additionally, the polysulfide anchoring and catalytic ability of FLBP offer advantages to improve lithium-sulfur cells.Despite solid progress in this field during the past decade, the electrode use of FLBP is still in a fledging stage and the potentials need to be further released by surmounting challenges.
Firstly, it remains challenging to synthesize FLBP with controllable layer number and uniform morphology in mass production.The present preparation, either top-down exfoliation routes or bottom-up deposition/solvothermal techniques, is often limited to laboratory level rather than industrial scale.For these synthetic methods, further study is required for optimization of experimental parameters such as temperature, substrate effect and selection of solvent and phosphorus source.New routes processing at low temperature and low energy consumption are highly desirable.Secondly, the problematic long-term stability of FLBP during storage calls for passivation and surface-protection techniques.Thirdly, the residual organic molecules on the surface of exfoliated FLBP nanosheets for stabilization and preventing layer restacking may affect electrochemical performance.It deserves further research on the relationship between residual functional groups and electrode properties.
Also, as anode for chargeable batteries, FLBP suffers from low Coulombic efficiency and limited rate performance, which may be attributed to decomposition of electrolyte, formation of SEI film, irreversible reaction of M + (Li + , Na + , K + ) on the facets/edges and defective sites, and slow reaction kinetics.For full battery, the low Coulombic efficiency of anode contributes to wastage of M + (Li + , Na + , K + ) furnished by cathode material, leading to rapid loss of deliverable capacity.The assembly of FLBP into the large-sized conductive materials such as graphene and MXene can reduce the side reactions and improve the packing density and rate performance.A protective and buffer layers composed of conductive polymer, transition metal oxide, transition metal phosphates or MOF on the edge of FLBP can also effectively decrease the side reactions and alleviate volume expansion to get good electrochemical performance.However, side reactions of FLBP with different electrolytes are not clear.Effects of thickness, lateral size, exposed facets/edges, and defects on the charge storage properties (intercalation and alloying reactions) remain not well understood and thus require more mechanistic investigations.In addition to electrode material, metal salts, electrolyte additives and binders also have an impact on the electrochemical property especially the Coulombic efficiency, which need to be further explored.Furthermore, full battery with FLBP as anode has been less reported owing to the huge capacity difference between FLBP and current prevailing cathode materials.Optimizing the proper anode/cathode mass or volume ratio is the key to the commercialization of FLBP-based full battery.
Last but not the least, along with the vigorous development of foldable monitor and wearable electronic devices, flexible energy storage devices are becoming a research hotspot.In this direction, it is important to explore FLBP-based rechargeable batteries that offer flexibility and high performance.The layer structure and large surface area of FLBP are beneficial to building self-supporting electrodes, which can be assembled into flexible paper-like rechargeable batteries.It deserves further research on the electrochemical performance of FLBP-based flexible rechargeable batteries in various folded states.
Although FLBP presents tremendous potential applications in electrochemical storage and conversion due to its intrinsic properties, there are still many challenges to render it marketable.As a consequence, the comprehensive understanding of rational design of hybrid nano/micro structure, characterization techniques, structure-performance relationship and electrode mechanisms should be enhanced to further improve the electrochemical properties of FLBP.With the continuous efforts, rapid development of FLBP-based rechargeable batteries can be anticipated in energy storage applications.

Figure 3 .
Figure 3. (a), (b) AFM images of FLBP after exfoliation on a SiO2/Si substrate.(c) Raman spectra of sample with the thickness of 5 nm measured in air at 24, 48, 96 and 120 min after exfoliation.(d) Raman spectra of different layered FLBP and bulk BP at 300 K. [22] Reprinted by permission from Springer Nature Customer Service Centre GmbH: Springer Nature, Nature Materials.(e) Illustration of the metal-assisted exfoliation process for FLBP.(f) Optical microscopy of exfoliated FLBP on Au layer.(g) FLBP exfoliated using the 'scotch-tape' method (left) and the Au-assisted method (right).(h) Total area of FLBP on ten different samples.Reproduced from [29] with permission of The Royal Society of Chemistry.

Figure 5 .
Figure 5. (a), (b) Illustration of the exfoliation procedure and reaction cell.(c) The variation of production yield using different intercalating cations.(d) SEM and TEM images of FLBP from electrochemical exfoliation.(e) AFM images of FLBP and corresponding statistic thickness from the height profile.[44] John Wiley & Sons.© 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.(f) TEM and (g) SEM images of FLBP.(h) Storage time-dependent relative absorbance (A/A0) of FLBP dispersion under the anoxic condition and normal condition.Reprinted with permission from [23].Copyright (2019), American Chemical Society.

Figure 7 .
Figure 7. (a) Schemes of uniform temperature SDT strategy to grow BP.(b) A summary of literature reports on the growth yield and purity of BP grown by different methods.Reprinted from [56].Copyright (2020), with permission from Elsevier.(c) TEM image of the holey FLBP.(d) AFM image and (e) the corresponding heights profiles.[26] John Wiley & Sons.© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Figure 10 .
Figure 10.(a) Schematic of sodiation in BP during discharge process.(b)-(d) Evolution of TEM images of sodiation in BP.(e) Charge capacities of phosphorene/graphene electrodes at 0.05 A g −1 .(f) Volumetric and mass capacities at different current densities.Reprinted by permission from Springer Nature Customer Service Centre GmbH: Nature, Nature Nanotechnology [21] A phosphorene-graphene hybrid material as a high-capacity anode for sodium-ion batteries, Sun J, Lee H, Pasta M, Yuan H, Zheng G, Sun Y, Li Y and Cui Y, © 2015.(g) Synthesis of the BP/rGO.(h) Planar and (i) cross-sectional SEM images of BP/rGO film.(j) Discharge-charge profiles for different rates.(k) Cycling performance of BP/rGO at 1 and 40 A g −1 .Reprinted with permission from [81].Copyright (2018) American Chemical Society.

Figure 12 .
Figure 12.(a) Schematic of FLP-CNF as the host for lithium polysulphide catholyte.(b) SEM of FLP-CNF.EDX mapping of (c) carbon and (d) phosphorus.(e) Rate performance.(f) Discharge-charge profiles at 0.2 C. (g) Cycling performance and Coulombic efficiency.Inset shows utilization of sulfur at 1 C. (h) Structure model and charge density distribution.Red, green, and violet spheres represent S, Li, and P, respectively.(i) Binding energy comparison for lithium polysulfides with phosphorene and carbon.[101] John Wiley & Sons.© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Figure 13 .
Figure 13.Overview of preparation and battery applications of FLBP.

Table 1 .
A comparison of the electrochemical properties of FLBP anodes in LIBs.

Table 2 .
A comparison of the electrochemical properties of FLBP anodes in SIBs.

Table 3 .
A comparison of the electrochemical properties of FLBP anodes in PIBs.