Faculty of Metallurgy and Energy Engineering, National Engineering Research Center of Vacuum Metallurgy, Kunming University of Science and Technology, Kunming 650093, China
Received Date:
09 September 2024 Accepted Date:
22 October 2024 Revised Date:
03 October 2024 Available Online:
15 December 2025
Abstract:
Sodium-ion batteries (SIBs) are the promising rechargeable batteries in large-scale energy storage systems for their low cost, high safety, wide temperature range adaptability, environmental friendliness and excellent fast-charging capabilities. Significant research endeavors in SIBs have focused on the exploration of high-performance electrode materials and thorough investigation of their mechanisms. Na2FePO4F (NFPF) is one of potential cathode materials because of low cost, minimal volume strain and extended cycle performance. This review summarizes the crystal structure, sodium ion migration pathways, and synthesis methods of NFPF and discusses the effect of various strategies including hybridization with carbon materials, ion doping, morphology control and electrolyte optimization on its electrochemical performance. Additionally, the application of the NFPF in different batteries is summarized. Finally, the challenges and future directions of NFPF are proposed. This review is both timely and important for promoting the applications of cost-effective NFPF.
Energy is fundamental to human survival and development. The excessive extraction and utilization of traditional fossil fuels have resulted in energy crises and environmental challenges. Consequently, green and affordable renewable energy, such as solar, geothermal, wind, and tidal energy, has garnered widespread attention [1]. However, these sustainable energy sources are affected by factors such as location and weather, resulting in intermittency and instability. Therefore, it is necessary to convert these energies into electrical energy and store them in large-scale energy storage systems for continuous and stable power output [2–4]. Since Sony pioneered the commercialization of lithium-ion batteries (LIBs), they have become the predominant technology for mobile power supplies, electric vehicles, handheld electronic products and large-scale energy storage systems because of high energy capacity and volume density [5,6]. Whereas, the constrained availability of lithium resources, may impede the future large-scale implementation of LIBs in renewable energy storage field [7].
Beneficial from abundant resources of sodium and similar electrochemical properties to lithium, sodium-ion batteries (SIBs) are considered as the potential supplement for LIBs in large-scale energy storage. Since research on SIBs commenced at 1970s, cathode materials have been considered an important component in the development of SIBs [8]. Recently, layered transition metal oxides [9–11], polyanionic compounds [12,13], and Prussian blue analogs [14,15] have become the research focus of cathode materials. Although layered oxides can initially achieve a high specific capacity exceeding 190 mAh/g, their cycling performances are suboptimal, which capacity degradation is observed over long cycles. Moreover, these oxides undergo complex phase transitions during sodiation and desodiation processes, involving multistep charge-discharge reactions across a broad potential range, which poses challenges for applications [16]. Prussian blue analogs often contain lattice defects and coordinated water, which significantly reduce active sites and lead to electrolyte loss, resulting in unsatisfactory electrochemical performance [17]. Consequently, polyanionic compounds have emerged as extensively researched electrode materials because of structural stability and superior electrochemical performance.
Polyanionic compounds are composed of tetrahedral anionic units (XO4)n- or their derivatives (XmO3m+1)n- (where X = S, P, Si, As, Mo, or W) combined with strong covalently bonded MOx polyhedral (M represents a transition metal). In most polyanionic compounds, (XO4)n- not only allows for rapid ionic conduction within the open framework of selected working alkali ions during discharge, but also stabilizes the working redox potential of the transition metals [12]. Among various polyanionic compounds, NaFePO4 initially attracts research interest as a cathode material for SIBs because it is most similar to LiFePO4. NaFePO4 can be divided into two different phases: Olivine and maricite phases. Maricite NaFePO4 does not have tunnels for Na+ so it is considered to have no electrochemical activity [18]. While for the olivine phase, adjacent FeO6 units share corners with PO43+ which forms a one-dimensional Na+ channel along the b-axis [19,20]. Besides, the olivine NaFePO4 undergoes an irreversible phase transition above 480 ℃ and its crystal structure is prone to retain lithium due to the inadequacies in the synthesis method, resulting in the decay of capacity [21]. Consequently, as a member of polyanionic compounds, Na2FePO4F (NFPF) which has stable structure and two-dimensional layered structure is gaining more and more attention as a potential candidate on account of its low cost, significant energy density, minimal volume strain, and prolonged cycle life [22]. Moreover, the strong electronegativity of fluorine ions increases redox potentials of NFPF when compared to NaFePO4. NFPF crystallizes in orthorhombic structure and has a layered structure which allows Na+ to diffuse along the [001] and [100] directions. It exhibits two voltage plateaus at 2.9 and 3.1 V and its theoretical specific capacity is 124 mAh/g [23]. In addition, besides SIBs, NFPF can serve as cathode material in many other battery storage systems, such as LIBs, potassium-ion batteries (PIBs). Recent research efforts have made significant strides in understanding its charge and discharge mechanism. Consequently, it is crucial to present a comprehensive summary of the current literature to guide future design strategies and facilitate its practical application.
This review comprehensively analyses the crystal structure, sodium ion migration pathways, and synthesis methods of NFPF reported to date, and examines various strategies for enhancing electronic conductivity and improving its electrochemical performance. In addition, NFPF applied to other systems is summarized. Finally, the challenges associated with NFPF and the outline of future research directions are further discussed.
2.
Crystal structure and migration pathways of NFPF
Different from olivine or maricite NaFePO4, NFPF is a kind of layered fluorophosphate compounds that belongs to the orthorhombic system with the space group of Pbcn and exhibits a smaller volume change (3.7%) than LiFePO4 (6.7%) [23]. And similar to the most of polyanion-type electrode materials, its structure is composed of paired coplanar iron octahedra, each of which were coordinated to four oxygen ligands and two fluoride ligands (Fig. 1a) [24]. The Fe2O6F3 dimer octahedral units are connected to other paired iron octahedra along the a-axis through corner-sharing (via F2), while the iron octahedral chains are connected along the c-axis through corner-sharing of PO43- tetrahedra [25]. Depending on the connectivity of the framework, NFPF may exhibit a high potential due to the induced effects of the PO43- group and the strong electronegativity of fluorine ions [26]. Nevertheless, the insulating PO4 tetrahedral structure blocks electron transmission, causing electronic transport pathways to be obstructed during the charging and discharging process, leading to low electronic conductivity [27]. Moreover, the calculation results of partial density of states (PDOS) and band structure indicate that NFPF has an indirect band gap (~3.27 eV) which means its electronic conductivity is poor [28].
Figure 1
Figure 1.
(a) Orthorhombic NFPF structure with a space group of Pbcn. Reprinted with permission [28]. Copyright 2019, Elsevier. (b) Ex-situ XRD patterns of NFPF@C samples during the first charge/discharge process. Reprinted with permission [31]. Copyright 2024, Wiley-VCH. (c) Discharge mechanism for Na1+xFePO4F at 0.5 < x < 1. Spheres represent Na1+xFePO4F particles. Red regions show the NFPF phase, and green regions show the P21/b Na1.5FePO4F phase. Yellow arrows show the diffusion path. Reprinted with permission [34]. Copyright 2018, Wiley-VCH. (d) Na+ migration path along the a- and c-axis in NFPF; octahedral FeO4F2 and tetrahedral PO4 are represented by grey and yellow respectively. Na+ migration path, (e) along the a-axis, and (f) along the c-axis. Reprinted with permission [32]. Copyright 2013, Royal Society of Chemistry.
In a typical charge/discharge curve, NFPF exhibits two voltage plateaus at 2.9 and 3.1 V. Furthermore, on account of the monoelectronic reaction of Fe2+/Fe3+, the theoretical specific capacity of this material is calculated to be 124 mAh/g. According to their distinct oxygen coordination environments, sodium sites can be classified into two unique types, referred to as Na1 and Na2 [29]. For the Na1 site, they have different Na-Fe distances, which are 4.55 and 3.45 Å, while all Na-Fe distances at the Na2 site are about 3.33 Å. The angles of two Na-F-Fe bond at the Na1 site are close to 180°, while the angles for other bonds are around 90°. But at Na2 site, all Na-O/F-Fe bond angles are approximately 90°. Only one site in these two different electrochemical environments can be delocalized from the NFPF crystals and thus exhibit electrochemical activity [30].
The ex-situ XRD was investigated to understand the detailed phase transitions of NFPF during the charge and discharge processes, as illustrated in Fig. 1b [31]. NFPF experiences two transitions during charge or discharge processes. In the electrode reaction process, there is an intermediate phase Na1.5FePO4F belonging to the P21/b space group, in which the Na2 site of the Pbcn structure undergoes a division into two distinct sites, Na21 and Na22 [26]. When charged to around 3.0 V, NFPF releases 0.5 Na+ to form Na1.5FePO4F. Continuing to charge to around 3.2 V, Na1.5FePO4F releases 0.5 Na+ to form NaFePO4F. Upon discharge to around 3.0 V, NaFePO4F inserts 0.5 Na+ to form Na1.5FePO4F. When discharged to around 2.8 V, Na1.5FePO4F inserts 0.5 Na+ to form NFPF. The phase transition process is illustrated by Eqs. 1 and 2:
The layered structure allows Na+ to diffuse along the [001] and [100] directions. Tripathi [32] confirmed through neutron diffraction maximum-entropy method that the lowest energy path for NFPF involves parallel migration of Na along the a- and c-axis directions, with remote diffusion barriers of approximately 0.3 and 0.4 eV, respectively. The migration paths are illustrated in Figs. 1e and f. They also identified six unique Na-Na distances within the layered NFPF unit cell (labeled as N1 to N6), where migration along the c-axis follows the hopping sequence of N4-N3-N6, whereas migration along the a-axis takes place via the pathway of N3-N5 (Fig. 1d). These findings suggest, in contrast to NaFePO4, which only has 1D channel Na+ migration pathway along the b-axis [19,33], NFPF exhibits elevated Na migration rates within the channels of the a- and c-axis, enabling a 2D faster Na+ migration across the ac plane. And ions in NFPF is less likely to be blocked by anti-site defects, contrasting sharply with the substantial effect on Na+ migration observed in olivine materials characterized by one-dimensional diffusion [32].
The intermediate phase Na1+xFePO4F (composed of Na2FePO4F, Na1.5FePO4F or Na1FePO4F) also exhibits two-dimensional diffusion pathways along the a- and c-axis. So, Shinagawa et al. [34] believed that Na1+xFePO4F features intra-phase and inter-phase diffusion pathways simultaneously (Fig. 1c). When the current density is very low, Na+ diffuses through inter-phase diffusion to the Na1.5FePO4F phase and subsequently diffuses to the surface of the particles. However, the amount of Na+ diffusing from the boundaries will be minimal as there are only two interfaces between the NFPF and Na1.5FePO4F phases. At moderate rates, Na1+xFePO4F has more than two interfaces, allowing for a greater amount of Na+ diffusion from the interfaces outward. Na+ can only diffuse to the particle surface from the outermost layer interface through inter-phase diffusion pathways on phase interfaces. In other interfaces, Na+ not only needs to diffuse through inter-phase pathways but also through the a-axis diffusion pathways, namely intra-phase diffusion pathways to reach the surface.
3.
Synthetic processes of NFPF
The preparation method of SIBs cathode materials directly determines characteristics including composition, morphology, crystal and surface structure of NFPF, thereby decisively influencing electrochemical performance. Currently, the preparation of NFPF materials follows similar to that of other cathode materials, mainly involving solid-state method, spray drying methods, sol-gel methods, hydrothermal method, etc. The morphology of the materials obtained using the respective methods is shown in Fig. 2.
Figure 2
Figure 2.
Preparation schematic and corresponding SEM of NFPF samples prepared by different synthesis methods: (a, b) Solid state method. Reprinted with permission [35]. Copyright 2023, Elsevier. (c, d) Spray-drying. Reprinted with permission [36]. Copyright 2014, Elsevier. Reprinted with permission [53]. Copyright 2022, Wiley-VCH. (e, f) Sol-gel method. Reprinted with permission [39]. Copyright 2018, Elsevier. (g, h) Hydrothermal reaction. Reprinted with permission [40]. Copyright 2020, Elsevier. (i, j) Electrospinning method. Reprinted with permission [42]. Copyright 2019, Wiley-VCH.
The simple solid-state method is among the most widely utilized techniques for synthesizing NFPF cathode materials, known for its advantages of simplicity in operation and a short process. However, its drawback lies in the difficulty of controlling the primary particle size of the material, leading to uneven morphology. Gong and colleagues [35] synthesized NFPF/C through a simple solid-state method. Initially, NaF, NH4H2PO4, CH3COONa, FeC2O4⋅2H2O and C6H8O7⋅H2O were milled together and calcined. Subsequently, the above powder and glucose were re-milled and calcined again to obtain NFPF/C (Fig. 2a). Finally, the sizes of NFPF/C particles were ranging from 50 nm to 150 nm (Fig. 2b). And they found that the F elements will be lost from NFPF as the calcination temperature increases when the NFPF decomposes into Na3PO4, Fe3(PO4)2, and F2 at high temperatures. The spray drying method can solve the problem of heterogeneous particle and its schematic process is shown in Fig. 2c. It involves first refining the particle size, followed by a pyrolysis reaction of aerosol droplets. The precursor solution is atomized into small droplets under high-pressure airflow and high temperature, then rapidly dried. Therefore, compared to the solid-state method, materials synthesized through spray drying methods exhibit advantages such as high uniformity and controllable particle size. Magali [36] synthesized NFPF with particle size from a few microns to about 10 μm by spray drying method (Fig. 2d). Another method is sol-gel method. The materials produced by sol-gel technique take an advantage of high specific surface area and porous structure, thereby substantially increasing contact area of electrode and electrolyte [37]. Typically, the primary atoms in inorganic salts are chelated onto organic compounds (such as citric acid, oxalic acid), forming a metastable sol system. Subsequently, the solution is dried to yield a dry gel, characterized by a relatively uniform distribution of various metal atoms. Ultimately, high-temperature calcination can eliminate the organic macromolecules, resulting in a relatively uniform product and forming carbon layer on the material surface [38]. Synthesized via sol-gel method, the NFPF/C material is composed of small particles and their sizes are about 200–300 nm stacked together to form porous structure (Figs. 2e and f). This distinctive structure facilitates multi-channel diffusion pathways, ensures continuous ionic conduction, and offers stable structure. Remarkably, the capacity retention rate remains at 82.6% after 1000 cycles at 0.5 C [39]. However, this method requires substantial quantities of organic solvents and the materials usually have a high carbon content, which is not conducive to commercialization.
The powder obtained by the hydrothermal method tends to be small and uniform, slightly aggregated, and easy to obtain suitable crystal shapes. Depending on the solvent used, the morphology of materials may also vary. As shown in Figs. 2g and h, Xun et al. [40] synthesized NFPF nanorods at 180 ℃ for 12 h. These nanorods exhibit a uniform one-dimensional structure, characterized by an average diameter of 50 nm and lengths ranging from 200 nm to 500 nm. A rapid method for synthesizing NFPF is the solution combustion method. Sharma et al. [41] dissolved Fe(NO3)3⋅9H2O, NaH2PO4, NaF, and citric acid (C6H8O7) in deionized water. The precursor solution was subsequently heated to a temperature just above water's boiling point to facilitate dehydration, producing a dark brown combustion ash as an intermediate composite. Then, after granulating the ash, it was annealed at 600 ℃ for 1 min to rapidly form NFPF. Without any further optimization, a reversible specific capacity and excellent cycling stability could be achieved. However, the synthesis conditions of this method are challenging to control, and Fe2+ is prone to oxidation.
Electrospinning is a typical technique for synthesizing self-supporting electrode materials with one-dimensional nanostructures. It can tightly attach the flexible membrane woven and from NFPF@C nanofibers directly onto the current collector to function as a binder-free positive electrode for SIBs. Wang and his colleagues [42] used the electrospinning process to achieve uniform encapsulation of NFPF nanoparticles (approximately 3.8 nm) within porous nitrogen-doped carbon nanofibers (Figs. 2i and j). This approach achieved a high reversible specific capacity, outstanding rate performance, as well as unprecedentedly high cycling stability for SIBs cathode materials.
However, many researchers have reported that the synthesis of NFPF currently faces issues with Fe3+ impurities or the loss of F which has a great influence on rate property [18,43,44]. As mentioned above, the NFPF will decompose into Na3PO4, Fe3(PO4)2, and F2 at high temperatures. Besides, Tang [45] reported that the diffraction peaks were observed in sample prepared with oxalic acid while samples synthesized with ascorbic acid, citric acid, and glucose were all pure. Li [46] found that the excess addition of citric acid was likely to form impurities and a low addition would lead to the crystallinity of the final product being unsatisfactory. Hence, it is crucial to pay attention to the calcination temperature, as well as the selection and amount of carbon source during synthesis processes.
Overall, although solid-state method is simple to operate, the final material is inhomogeneous and prone to agglomeration, which limit the electrochemical properties. Sol-gel and hydrothermal methods have many advantages that conventional solid-state method cannot replicate, such as homogeneous reactant mixing and small NFPF particle size, which are considered to be effective methods for achieving fine structure design and uniform elemental doping. In addition, although the composites synthesized by electrospinning have good multiplicity properties, the slow production speed of electrospinning limits its large-scale production and application [47]. For the issues with Fe3+ impurities and the loss of F, it is possible to address these problems by calcining in a reducing atmosphere, adjusting the calcination temperature, as well as the selection and amount of carbon source during synthesis reasonably.
4.
Modification strategies of NFPF
With excellent structural stability, simple 2D Na+ transfer pathway and relatively high operating voltage, NFPF exhibits significant promise for high-performance sodium storage. Nevertheless, the rate performance and reversible capacity of NFPF are limited by low inherent electronic conductivity and the restricted number of active sodium sites [48]. Therefore, enhancing electronic conductivity is essential for achieving superior electrochemical performance. To this end, numerous attempts and efforts have been made to modify NFPF mainly including hybridization with carbon materials, ion doping, morphology design. Through the application of one or more of these strategies, a notable enhancement in the electrochemical properties of NFPF is achieved. Summarized in Fig. 3 are several modification strategies, while a detailed overview of the electrochemical performance of NFPF is shown in Table 1 [24,28,31,35,37,39,40,42,44–46, 48–75].
Figure 3
Figure 3.
A schematic representation of the modification techniques of NFPF.
Hybridization with carbon materials is considered as an attractive way. Usually, hybridization with carbon materials will form carbon layers or carbon networks. Carbon layers have many advantages, including excellent electrical conductivity, exceptional chemical stability, low thermal expansion coefficient, and lightweight so that greatly improve electrochemical performance. Combining a carbon source with active materials results in the formation of an external conductive framework while simultaneously protecting the internal structure of NFPF. This process reduces direct contact between the active material and the electrolyte, thereby minimizing side reactions. The carbon layer possesses several notable advantages of (1) preventing the oxidation of Fe2+, (2) reducing particle size, (3) increasing electrical conductivity and (4) enhancing ionic diffusivity [76]. However, achieving a uniform carbon layer is not easy because of the significant influence of the preparation method. The migration of Na+ between the electrolyte and cathode materials is influenced by both the thickness and uniformity of the carbon layer on the outer layer of active material, highlighting the importance of precise control. Typical carbon materials are glucose, sucrose, ascorbic acid, and carbon nanotubes, among others.
4.1.1
Sucrose and glucose
Sucrose and glucose are common organic carbon sources, and typically there are two methods to coat them on the surface of materials. The first method involves polymerizing sucrose or glucose on the material surface through hydrothermal reaction, followed by high-temperature carbonization in an inert atmosphere to convert the polymer into carbon. The other method is to mix them with the material uniformly and dry. Subsequently, high-temperature carbonization in an inert atmosphere is also carried out to obtain a carbon coating. For example, using glucose as carbon source, uniform carbon layer was obtained by hydrothermal reaction and the structural characteristics of NFPF/C spheres were studied using HRTEM [49]. As illustrated in Fig. 4a, a carbon layer with an approximate thickness of 6 nm was detected on the surface of NFPF particles. The conductivity of NFPF could be considerably enhanced by the presence of a thin, evenly dispersed carbon layer. Hu et al. [50] prepared NFPF/C nanocomposites (NFPF/C-PG) using glucose as a carbon source via the solid-state method. The in-situ polymerization of PFA played a dual role: it prevented the aggregation of nanoparticles, thereby guaranteeing the uniform distribution of NFPF, and during the heating phase, the carbonization of PFA facilitated the development of a conductive carbon network. Therefore, the NFPF/C-PG samples exhibited superior electrochemical properties compared to those that did not use FA or glucose during synthesis process. The NFPF/C-PG demonstrated a stable charge-discharge platform, providing high discharge capacities and exhibited excellent cycling stability.
Figure 4
Figure 4.
(a) TEM of NFPF/C. Reprinted with permission [49]. Copyright 2017, Springer. (b) HRTEM of NFPF/C-650. Reprinted with permission [35]. Copyright 2023, Elsevier. (c) SEM images of Na2Fe0.6Mn0.4PO4F/C prepared with ascorbic acid. Reprinted with permission [45]. Copyright 2020, Frontiers Media S.A. (d) Schematic representation of NFPF-CB and NFPF-CNT growth process. (e) SEM micrographs of NFPF-20CNT. Reprinted with permission [79]. Copyright 2017, Springer. (f) HR-TEM images of NFPF@C@MCNTs. Reprinted with permission [53]. Copyright 2022, Wiley-VCH. (g) LbL nano-assembly process for the formation of NFPF/CNT multi-layered cathode. Reprinted with permission [54]. Copyright 2015, Elsevier. (h) TEM images NFPF@CNT&GN. Reprinted with permission [40]. Copyright 2020, Elsevier. (i) HRTEM images of 5T-NF@C. Reprinted with permission [31]. Copyright 2024, Wiley-VCH. (j) FESEM images of NFPF. Reprinted with permission [55]. Copyright 2015, Royal Society of Chemistry.
During the synthesis processes, the calcination temperature impacts both the development of the carbon layer on the surface of NFPF and the electrochemical characteristics of final composites. Gong et al. [35] researched the electrochemical properties of NFPF/C samples at different calcination temperatures using glucose as carbon source. According to the Raman spectra, the ID/IG increases and then decreases as calcination temperature increases, indicating that the temperature will influence the disorder degree of carbon layer. Calcined at 650 ℃, the high-purity NFPF/C has a stable fluor phosphorus structure alongside a carbon framework that is relatively disordered. This structure enhances ion diffusion kinetics. Furthermore, the significant disorder within the carbon layer of NFPF/C-650 facilitates greater Na absorption and improves the performance of battery (Fig. 4b). Moreover, the small size and dispersed particles not only aid the (de)intercalation of Na+ but also reduce the Na+ diffusion distance contributing to superior cycling performance and enhanced rate capacity. Finally, the NFPF/C-650 obtained the best electrochemical properties, with a specific capacity of 120 mAh/g at 0.1 C and the capacity retention was 76.86% after 200 cycles at 1 C.
4.1.2
Ascorbic acid
Ascorbic acid not only functions as carbon source during the material synthesis process, but also, due to its excellent reducibility, can prevent the oxidation of Fe2+. Tang et al. [45] investigated the influence of different carbon sources, including oxalic acid, ascorbic acid, citric acid, and glucose, on the structural characteristics, morphology, and electrochemical performance of Na2Fe0.6Mn0.4PO4F/C materials. The material synthesized with ascorbic acid reveals a hollow spherical morphology characterized by a dense surface and a particle size distribution between 1 µm and 2 µm (Fig. 4c). The sample derived from ascorbic acid exhibits lower electrochemical polarization when compared to those produced from other carbon sources according to kinetic studies of the electrode reactions. This finding underscores its superior electrochemical performance, which includes a high discharge capacity of 95.1 mAh/g at 0.05 C, enhanced rate capability, and commendable cycling stability. Different from spray drying method, Kawabe et al. [51] coated ascorbic acid onto the NFPF particles through solid-state method, effectively inhibiting the growth of NFPF particles and forming nano-scale carbon-coated materials that exhibited an initial specific capacity of 110 mAh/g at a rate of 0.05 C. Deng [52] utilized VC as carbon source to synthesize porous NFPF-C. This porous structure could facilitate the diffusion of sodium within the NFPF phase, which could enhance electronic conductivity and accelerate sodium ion diffusion and finally contribute positively to the electrochemical performance. After 1000 cycles, conducted at a rate of 4 C, NFPF-C demonstrated a capacity retention of 85%, achieving a specific capacity of 117 mAh/g at a rate of 0.1 C.
4.1.3
CNT
Carbon nanotubes (CNTs) exhibit intrinsic physical and mechanical properties along with self-lubricating capabilities. Due to their unique features, carbon nanotubes can improve the distinctive characteristics of materials, such as conductivity and high-temperature stability, making them highly attractive as electrode coating materials. CNTs can be classified into single-walled carbon nanotubes (SWCNTs) or multi-walled carbon nanotubes (MWCNTs) on the basis of diameter formed by bending on the carbon nanotubes. The diameter range of SWCNTs is 1–2 nm, while MWCNTs have a diameter of <100 nm, formed by flipping carbon nanotube sheets [77]. Due to the higher preparation and purification costs of SWCNTs, they have not been widely used as coating materials [78]. Abdelfattah Mahmoud et al. [79] compared the effects of carbon black (CB) and CNT on the structure, morphology, and electrochemical performance of NFPF samples. When 20 wt% CNTs was added, the synthesized NFPF had fewer impurities, higher crystallinity of the carbon layer and formation process is shown in Fig. 4d. By incorporating a network of carbon nanotubes on the surface and within collapsed spheres, a uniform three-dimensional carbon nanotube network was formed (Fig. 4e). The morphological analysis demonstrated that the incorporation of carbon nanotubes positively influenced the particle size and aggregation behavior of the NFPF samples, resulting in enhanced electronic conductivity. Furthermore, the charge transfer resistance (Rct) between the electrolyte and active material was decreased. The NFPF@C@MCNTs synthesized by Cao et al. [53] had a distinctive morphology characterized by nanoscale primary particles and minute secondary particles. These particles were evenly and uniformly coated with MCNTs (Fig. 4f). It delivered a reversible specific capacity of 118.4 mAh/g at 0.1 C. The specific capacity remained at 56.4 mAh/g after 700 cycles with 97% capacity retention at 5 C. Yan [54] alternately attached NFPF and CNT to a copper current collector by physical adsorption (Fig. 4g). The obtained multilayer cathodes were greatly flexile without any binders such as PVDF and had a porous structure with even NFPF nanoparticles and CNT. The NFPF/CNT positive electrodes demonstrated excellent cycling stability, impressive rate performance, favorable reversibility, and rapid kinetic properties. This is likely attributed to their unique multi-layer porous structure, which provides ample electron/ion conduction pathways during sodiation and desodiation processes, enhances the adsorption capacity of NFPF nanoparticles, and prevents detachment from the CNT. Xun et al. [40] constructed a conductive network by combining rod-shaped NFPF precursors with CNT and two-dimensional graphene nanosheets (GN), significantly enhancing the electrochemical performance of NFPF. First, a conductive network was formed by integrating the active material with CNT and GN (Fig. 4h). Through the interconnection of various dimensions, numerous channels for the transport of electrons, ions, and Na+ were provided. Moreover, the fragmented CNT and GN minimized the electronic transport and Na+ diffusion distances. The conductive network of NFPF@CNT&GN maintained its stability over extended cycling periods, which significantly enhanced the long-term cycling performance.
4.1.4
PVP
Since PVP contains the nitrogen element, the generated carbon layer is generally doped with nitrogen. The nitrogen-doped carbon layer is apparently effective in increasing the electron density and conductivity of the material. This is mainly due to the fact that nitrogen has higher electronegativity and extra valence electrons than C. Nitrogen doping improves the conductivity and electrochemical activity of carbon materials by increasing their localized electron density [80,81]. By adjusting the amount of PVP with the material, the thickness of the final carbon layer generated will vary from several nanometers to dozens of nanometers. Moreover, the PVP does not only form a carbon layer but also a carbon network to connect the dispersed materials to constitute a conductive network in both 2D and 3D. Deng et al. [31] capped PVP on the surface of NFPF with a carbon coating thickness of about 2–3 nm. As shown in Fig. 4i, PVP has excellent film-forming properties, and the carbon generated can be evenly distributed over the NFPF particles, which is beneficial to the surface/interfacial conductivity of these particles, and effectively inhibits the side reaction between NFPF and electrolyte.
Besides, PEG [24], Super P [55], polytetrafluoroethylene [56], g-C3N4 [57], and rGO [58] etc., are also used as the carbon source or carbon material and showed good electrochemical performance. As shown in Fig. 4j, Markas [55] synthesized smaller and more uniform sample by high-energy ball milling which exhibited stable cycling performance. In conclusion, the current coating modification method applied to NFPF is mainly carbon coating, which can form a conductive network on the surface of the material to improve its conductivity and help reduce particle size, while the coating materials are mainly organic and inorganic materials. Organic materials as a carbon source, first of all, the organic materials need to be uniformly distributed on the surface of NFPF then undergo calcination in an inert atmosphere. The organic materials carbonize at high temperatures, forming amorphous materials, for enhancing the electrochemical properties of NFPF. On the other hand, inorganic materials such as CNTs have good conductivity, high mechanical strength, and structural stability. The coating process for these materials is simpler and does not require high-temperature treatment. However, inorganic carbon coatings are relatively costly. Nevertheless, to a certain extent, hybridization with carbon materials only improves the electrical conductivity of the material surface, yet it fails to improve the low localized electronic conductivity of material itself.
4.2
Ion doping
Ion doping is an effective way to improve NFPF with low intrinsic electron conductivity and limited active sodium sites. Ion doping can improve the microstructure by adjusting lattice spacing, creating defects to generate more sodium storage sites, and regulating redox platforms to achieve the purpose of stabilizing the crystal structure, improving the performance. According to density function theory (DFT), the band gap of NFPF decreases and the impurity states which is below the Fermi level are induced under the modification of metallic ions. The reduced band gap and formed impurity states result in that more active electrons are around Fe and they are more easily excited from the valence band to the conduction band, thereby improving the electronic conductivity of NFPF [31,59]. Meanwhile, the migration energy barrier is significantly decreased under the action of doping ions, indicating an enhanced Na+ diffusion kinetics which can significantly enhances electrochemical performance [60]. Up to now, Co3+ [28], Mn2+ [45], Li+ [61], Ni2+ [62], La3+ [63], V3+ [59], Mg2+ [64], Al3+ [74], etc. have been used as dopants of NFPF, which can be divided into two types according to the substitution sites: Na sites and Fe sites. The rate properties of different elements doped are summarized in Fig. 5.
Figure 5
Figure 5.
Rate performances of different dopant: (a) Co3+. Reprinted with permission [28]. Copyright 2019, Elsevier. (b) Al3+. Reprinted with permission [74]. Copyright 2021, Elsevier. (c) Mn2+. Reprinted with permission [45]. Copyright 2020, Frontiers Media S.A. (d) La3+. Reprinted with permission [63]. Copyright 2024, Elsevier. (e) V3+. Reprinted with permission [59]. Copyright 2021, Elsevier. (f) Mg2+ and Ni2+. Reprinted with permission [62]. Copyright 2022, Electrochemical Society of Japan. (g) Mg2+. Reprinted with permission [64]. Copyright 2024, Elsevier. (h) Ti4+. Reprinted with permission [31]. Copyright 2024, Wiley-VCH.
Wang et al. [61] explored the electrochemical performance of Na2−xLixFePO4F/C (0 ≤ x ≤ 2). They found that the space group of Na1.5Li0.5FePO4F changed to Pnma, while NFPF belongs to Pbcn. Moreover, the discharge capacity of Na2−xLixFePO4F/C gradually increased and reached the best capacity of 122.9 mAh/g when the x was equal to 0.5 (Na1.5Li0.5FePO4F). The results indicated that the addition of Li+ will change the crystal structure and thereby influence the specific capacity but the excessive Li+ is disadvantageous for enhancing the rate capacity.
4.2.2
Fe sites
The research on doping in Fe sites has garnered greater attention. For inactive elements, such as V3+, Ca2+, Mg2+, the dopants primarily help stabilize crystal structure and reduce Na+ diffusion barrier. When transition metal ions are doped, they can provide extra redox couple and increase the operating voltage. Dong et al. [59] synergistically improved the low conductivity of NFPF through V3+ doping (Na2Fe0.98V0.02PO4F, Na2Fe0.95V0.05PO4F and Na2Fe0.90V0.10PO4F) and carbon coating. Benefiting from enhanced Na+ diffusion kinetics and improved surface and bulk electronic conductivity, the Na2Fe0.95V0.05PO4F@C electrode among the three samples demonstrated the highest initial discharge capacity of 110.1 mAh/g at 0.1 C. Additionally, it showed a high-rate reversible capacity of 78.3 mAh/g at 10 C and maintained a long-term capacity retention of 83.8% after 600 cycles. They calculated the local density of states (LDOS) and verified that the defect states caused by V-doped are below the Fermi level which can be fully occupied by electrons. Thus, the V-doped material can form intrinsic electrons to improve electronic conductivity and enhances the stability of the original phase, making it more stable during the electrochemical process.
The active element intercalation into NFPF also obtains much attention. Yoshiteru and colleagues [65] discovered that the addition of Mn would provide extra voltage platforms. The specific capacity of NFPF at 6.2 mA/g was 100–110 mAh/g, with clear voltage platforms at 3.06 V and 2.91 V, while Na2Fe0.5Mn0.5PO4F had a discharge specific capacity of 110 mAh/g at 6.2 mA/g with three voltage platforms at 3.36 V, 3.04 V, and 2.86 V. The additional voltage platforms were likely attributed to the Mn2+/Mn3+ redox couple. However, Xie et al. [66] also synthesized a three-dimensional (3D) tunnel-structured Na2Fe0.5Mn0.5PO4F, which discharge specific capacity at 12.4 mA/g was 107 mAh/g, achieving approximately 86% of the theoretical capacity. The study of the K-edge XANES spectral behavior of iron and manganese indicated that the rate characteristic of the Na2Fe0.5Mn0.5PO4F electrode is primarily limited by the slow kinetics of the Mn3+/Mn2+ redox coupling.
As mentioned, there are two distinct sodium sites in crystallography, referred to as Na1 and Na2. Na2 is more readily extracted from the layered structure, which signifies its electrochemical activity, whereas Na1 remains inert. The disparity in electrochemical activity of two Na sites primarily stems from the varying number of bonds formed with semi-stable oxygen atoms [48,82]. Through the process of element doping, it is feasible to activate a portion of Na1, thereby improving the electrochemical properties of NFPF. Liu and colleagues [64] put forward a method to dope magnesium with d0 orbitals to increase intrinsic electronic conductivity and activate Na sites. In contrast to the 3d orbitals of Fe, the incorporation of d0 orbitals from Mg predominantly contributes p and s orbitals within the Mg-O bonds. This incorporation modifies the electronic states of the oxygen atoms, consequently leading to the "barrel effect" (Figs. 6a and b). This modification reduces the energy barrier for the transition from Na1 to Na2, while simultaneously creating additional Na+ migration pathways for the transition from Na2 to Na2. In comparison to NFPF (which exhibits a barrier of 1.02 eV), the influence of doped Mg on the migration between Na1 sites is relatively minor. However, it significantly enhances the transition from Na1 to Na2 (which features a barrier of 0.66 eV), illustrated in Figs. 6c and d. Furthermore, the doped Mg optimizes the electronic structure of NFPF, resulting in a reduced band gap. The NFMPF cathode demonstrates exceptional electrochemical performance, featuring an excellent capacity of 121.4 mAh/g and commendable cycling stability (Fig. 6e). Huang et al. [48] synthesized Na2Fe0.95Cu0.05PO4F/C with specific capacity of 74 mAh/g at 20 C and 119 mAh/g at 0.1 C. As illustrated in Fig. 6g, the 23Na solid-state nuclear magnetic resonance (ss-NMR) spectra for Na2Fe0.95Cu0.05PO4F/C indicated that the occupation of inactive Na1 of Cu-doped NFPF was the lowest which meant that Cu2+ activate Na1 site and reduce band gap, thereby enhancing the rate performance. As shown in Figs. 6f and h, the bonding environment around Na changed and the average bond length of Na-O/F increased after Cu doping indicates a weaker bonding environment around Na. The substitution of Cu2+ caused an elongation of the Na1-O/F bond length, weakened the influence of the undercoordinated O2 around Na1, and lowered the desodiation potential of the Na site, thus promoting the convention of inactive Na1 to the Na3 site (Figs. 6i and j). Furthermore, the bandgap was greatly reduced, thereby improving the intrinsic electronic conductivity. Comparison in the rate capabilities and the cycle performance is shown in Figs. 6k and l, indicating that FeCu-0.05 had a better cycling stability.
Figure 6
Figure 6.
(a) The differential charge of NFMPF. (b) The schematic illustration of the working mechanism by Mg2+ doping. (c) The energy barrier from Na2 → Na2 site of NFPF. (d) The energy barrier from Na1 to Na2 site of NFMPF when the migration site and Mg increased. (e) The cycle performance at 20 C for NFPF and NFMPF samples. Reprinted with permission [64]. Copyright 2024, Elsevier. (f) The local environment of the Na1 and Na2 sites in NFPF. (g) The occupation of Na1 calculated by refined results from XRD patterns and the corresponding specific capacity of different samples (at 0.1 C). (h) Depiction of the Na3 environment deriving from Na1 in FeCu-0.05. (i) 23Na ss-NMR spectra at 60 kHz of NFPF and FeCu-0.05. (j) The calculated de-intercalation potential of Na sites in different samples. (k) Comparison in the rate capabilities between FeCu-0.05 and the literature data for NFPF cathodes. (l) Cycling performance at 0.5 C. Reprinted with permission [48]. Copyright 2023, Wiley-VCH.
In conclusion, the strategy of ion doping proves to be effective in enhancing electrochemical properties, primarily due to the increased intrinsic electron conductivity and the activation of the inactive Na1 site. But as shown in above discussions, rate performance of NFPF will be limited by the excessive electrochemical inactive element doping or slow kinetics of active element. Therefore, researchers should comprehensively consider the element type, amount, doping site to promote the electrochemical properties effectively.
4.3
Morphology control
The electrochemical properties are also influenced by morphology and structure of the material, which can be maximized by modulating the internal structure and surface morphology of the materials. There are a large number of studies aimed at increasing contact area between the electrolyte and the electrode to accelerate Na+ migration and maintain a stable cathode electrolyte interphase (CEI). Strategies mainly focus on changing the microstructure such as the creation of porous structures or the fabrication of nanoscale electrode materials. The morphology of NFPF is significantly influenced by the synthesis methods and conditions. The effect of different solvents on the morphology of NFPF was studied by Liu [44]. During the early phase of the hydrothermal reaction, small particles are drawn to one another and combine freely due to electrostatic forces, resulting in the formation of micrometer-sized spheres. The higher molecular weight of ethylene glycol compared to water leads to an increase in the viscosity of solution, which in turn creates resistance at spatial sites. Consequently, this heightened resistance to ion diffusion impedes the movement of particles and affects the growth kinetics. Therefore, compared to water, the attainment of uniform small spherical structures was more effectively facilitated when utilizing 30% EG as solvent. Furthermore, when polyethylene glycol (PEG) serves as the solvent in the hydrothermal process, shorter chains formed because of its elevated molecular weight and long-chain structure, which are influenced by thermal decomposition and shear forces. These shorter polymer chains adhere to the surfaces of the grains, acting as a flexible template. This mechanism not only mitigates particle agglomeration and restricts particle growth but also results in the development of an in-situ carbon coating after calcination. However, too high a viscosity of the polyethylene glycol solution may lead to undesirable consequences such as uneven local reaction and particle agglomeration. Consequently, the particles of the NFFF-EG sample exhibited the smallest size about 800 nm, characterized by effective dispersion and uniform distribution. Also, it showed the best electrochemical properties among three samples which provided a discharge specific capacities with 112.3 mAh/g at 0.1 C. This confirmed that the different solvent properties in the hydrothermal reaction led to different morphology. Ling [67] investigated how pH of the precursor solution controlled the morphology of NFPF powders. The aspect ratios of the products gradually increased during the pH increase from 8.5 to 11.5. This observation suggests that a higher pH facilitates the morphological transformation of NFPF from coarse rod-like crystals to elongated needle-like crystals. Notably, the particle size of NFPF-10.5/C was measured to be <100 nm, while its surface area increased to 126.2 m2/g, at a pH of 10.5. Electrochemical measurements indicated that the NFPF-10.5/C sample has a high specific discharge capacity of 114.3 mAh/g at 0.1 C. Pre-treatment of raw materials is also a useful way in controlling morphology of the final product. Some researchers first pretreated the precursor FePO4 to obtain rGO/FePO4, and then synthesized NFPF/C/rGO with a specific capacity of 118 mAh/g at 0.1 C and, after 100 cycles, capacity retention is 92.2% at 1 C [46]. This research indicated that pre-modification did not affect on the NFPF material formation.
Porous structure can also enhance electrochemical performance. The porous structure can primarily facilitate greater contact between the material and the electrolyte, thereby optimizing redox reaction. However, it may also lead to increased electrolyte consumption. As illustrated in Figs. 7a and b, Langrock [68] synthesized nanosized porous hollow C/NFPF by template-free ultrasonic spraying, achieving a specific capacity at 0.1 C of 89 mAh/g, capacity retention of 33% at 9 C, and still delivering 60 mAh/g after long cycles at 1 C. Ling [49] prepared hollow NFPF/C microspheres using SDS templates. As the hydrothermal time extended, the structure of the microspheres experienced three stages. The schematic illustration for the formation process and rate performance is shown in Figs. 7c and d, the final NFPF/C composite material showed outstanding electrochemical properties as a cathode material, demonstrating remarkable cycle stability and rate capability (97.5% capacity retention, equivalent to 116.9 mAh/g). It can be attributed to the special double-shell hollow structure of NFPF/C, which allows electrochemical reactions to occur on both internal and external surfaces of the material shells, effectively increasing the specific surface area and accelerating electrochemical reaction. Additionally, the nanopores on the surface of secondary particles provide accommodation space for the volume changes caused by sodium ion deintercalation, further improving cycle stability. Zhang [69] synthesized mesoporous NFPF@C composite material, achieving a high reversible capacity, splendid rate performance, and excellent cycling stability. Compared to the uncoated material, the microspheres NFPF@C material appeared more symmetrical redox peaks and pronounced voltage plateau, indicating that the electrochemical reversibility is enhanced by mesoporous structure and appropriate carbon layer. Li [37] synthesized NFPF/biochar nano composite hollow microspheres using yeast cell biological templates. The schematic diagram of its formation mechanism and rate performance is shown in Figs. 7e and f. This electrode features an excellent initial capacity, good cyclic stability, and high-rate capability, which is obtained from yeast cell templates and offers an enhanced specific surface, thus minimizing the diffusion distance for Na+. Additionally, the mesoporous carbon nanonetwork establishes a 3D structure that generates extra active sites, consequently promoting electron transport.
Figure 7
Figure 7.
(a) Schematic illustration of the formation process of hollow C/NFPF spheres. (b) Rate capability of C/NFPF. Reprinted with permission [68]. Copyright 2012, Elsevier. (c) Schematic illustration for the formation process of double-shelled hollow NFPF/C spheres during the solvothermal reaction. (d) Rate performance of double-shelled hollow NFPF/C spheres. Reprinted with permission [49]. Copyright 2017, Springer US. (e) Schematic of the proposed formation mechanism of NFPF/BC hollow microspheres. (f) Rate performance of the NFPF/BC-2 sample at various rates. Reprinted with permission [37]. Copyright 2021, Wiley-VCH.
In short, the nanoscale or porous NFPF is beneficial to achieve high-rate capability. Nevertheless, the reduction of particle size may lead to a decrease in volumetric energy density, which will hinder its practical application. And the morphology of NFPF is influenced by many factors, such as synthesis method, hydrothermal temperate, solvent, the pH of solution, leading to control the size and morphology difficultly. Thus, it is necessary to analyze the fundamental theory and other aided methods to develop NFPF with diverse structures. When designing the size and shape of NFPF, various factors should be considered comprehensively.
In addition to applying these modification strategies individually, combining them is considered an effective approach to making NFPF commercially viable. For example, Li et al. [74] connected Al with terephthalic acid ligands to form a 3D structure, and subsequently synthesized a NFPF-Al/MC which had uniform porous carbon nanonetwork structure, where some of the Fe2+ in the NFPF crystal were replaced by Al3+. Since the valence electron count of Al3+ is higher than Fe2+, this substitution introduces vacancies that can accelerate Na+ diffusion and enhance electronic conductivity. Additionally, the carbonization of terephthalic acid forms continuous conductive carbon coating network on the particle surface, significantly increasing the contact area between the electrode and the electrolyte. At 0.1 C, the specific capacity during the initial cycle measured 115.2 mAh/g. The specific capacity achieved an impressive 62.3 mAh/g, demonstrating a capacity retention of 70.3% after undergoing 500 cycles when the rate increased to 5 C. Deng and colleagues [31] synthesized titanium-doped orthorhombic NFPF with varying proportions by solid-state methods and coated with polyvinylpyrrolidone (PVP). Titanium doping enhanced the intrinsic conductivity of NFPF, accelerated the diffusion of Na+, and created more storage sites. The uniformly coated carbon enhanced both the surface and interface conductivity while mitigating side reactions. Through the synergistic influence of carbon coating and titanium doping, 5T-NF@C demonstrated remarkable electrochemical properties. It achieved a capacity of 108.4 mAh/g at 0.2 C and retained a substantial capacity of 80.0 mAh/g even under a high current density of 10 C. Furthermore, after 2000 cycles at 10 C, it exhibited a capacity retention rate of 81.8%. Guo [62] prepared a carbon-coated NFPF cathode material co-doped with magnesium and nickel. The Mg2+ and Ni2+ doping in NFPF, along with the carbon network coated, partially substitute for Mg2+ and Ni2+, which can suppress grain growth during the sintering process, enhance the conductivity of material, stabilize the overall structure, improve Na+ diffusion, and maintain the crystal structure, contributing to a significant increase in electrical conductivity and improvement of cycling performance and rate capability. In contrast to NFPF/C and Na2Fe0.85Mg0.15PO4F/C, the discharge capacity of Na2Fe0.8Mg0.15Ni0.05PO4F/C remains at 91.3% after 50 cycles conducted at 0.1 C. Moreover, when tested at 5 C, this material achieves a discharge specific capacity of 53 mAh/g.
4.4
Electrolyte optimization
The electrode and electrolyte will react and form a CEI film on the cathode surface during the first cycle which has a great influence on the energy density and interface stability [83,84]. Although its formation may cause the loss of reversible capacity at first cycle, CEI prevents the dissolution of active materials and inhibit further oxidation of the electrolyte components [85]. Electrolyte optimization with the additives has been consider as the most practical method to form a stable CEI [86]. Lei [70] reported an irreversible cathodic peak appeared around 1.1 V of CV curve in the first cycle, which subsequently disappeared in the following cycles. They speculated that this is due to the interaction between some oxygen functional groups and Na, as well as the formation of a strong and stable CEI film. Kacemi and colleagues [57] added 5% FEC to 1 mol/L NaClO4 in PC electrolyte. They found that the addition of FEC helped prevent electrolyte decomposition, provide reversible a charge-discharge cycle and improve the electrochemical performance.
5.
Applications of NFPF
NFPF can be used not only in SIBs but also as cathode materials for LIBs [87,88], PIBs [73], aqueous sodium-ion batteries [89] and calcium-ion batteries [90].
When using lithium metal as the counter electrode, lithium ions preferentially replace Na2 sites in the crystal structure. After two cycles, the ion exchange between Na2 and lithium ions is completed and forms a new phase, NaLiFePO4F [91]. The NFPF/CNTs composite material, prepared by Brisbois, exhibits good electrochemical performance in 1 mol/L LiPF6 in EC/DMC (1:1, v/v), with a discharge capacity of 104 mAh/g at 0.1 C and 90 mAh/g at 1 C [43]. Hu et al. [92] synthesized NFPF/C composites with uniform particle distribution and lower charge transfer resistance using solid-state method. As shown in Fig. 8a, the charge/discharge curves of NFPF/C composite at the different rates showed a sloping curve characteristic rather than two well-defined voltage plateaus. This difference arises from the fact that in lithium-ion batteries, NFPF follows a solid solution mechanism during the charge and discharge process rather than forming intermediate phases [26]. The NFPF applied to LIBs exhibits enhanced electrochemical characteristic attributing to nanoscale particle and uniform carbon layer as nanostructures facilitated an increased interface contact between the electrode and the electrolyte, effectively reducing the transport distances for both electrons and Li+. Lu et al. [93] utilized sugarcane bagasse as a carbon source and the NFPF@C sample showed a specific capacity about 93 mAh/g but experienced capacity decay during cycling. Similarly, Ai et al. [94] employed as pre-treated bagasse a carbon source to synthesize NFPF@C composites. Under conditions of 10% waste sugarcane bagasse content and high temperature calcination, the composites demonstrated good crystallinity. At a rate of 0.1 C, the specific capacity of NFPF@C was 128 mAh/g at first cycle, representing a threefold increase compared to the uncoated NFPF material.
Figure 8
Figure 8.
(a) Charge/discharge curves at different rates. Reprinted with permission [92]. Copyright 2019, Central South University. (b) Charge/discharge curves extracted from the rate capability experiment at C/15, C/10, C/5, and 1 C rates. Reprinted with permission [73]. Copyright 2023, Elsevier. (c) Nyquist plot of NFPF and NFPF/CNT with its corresponding fitted curve. (d) Galvanostatic potential-capacity profiles (vs. both SHE and Na/Na+) of combustion synthesized NFPF half-cell in aqueous electrolyte. (e) Schematic presentation of full cell aqueous battery comprising of NFPF fluorophosphate cathode and NaTi2(PO4)3 NASICON anode with a potential difference of 1 V (vs. SHE). (f) Galvanostatic potential-capacity profiles (vs. NaTi2(PO4)3) of aqueous full cell comprising NFPF cathode and NaTi2(PO4)3 NASICON anode. Reprinted with permission [89]. John Wiley and Sons Ltd.
Besides Li+/Na+ dual-ion batteries, the potential of NFPF in KIBs has also been investigated. Bodart et al. [73] assessed the performance of NFPF/CNT composites in potassium-ion batteries, demonstrating a consistent capacity of 114 mAh/g at a C/15 rate within a voltage range of 2–4.2 V. The average discharge capacities recorded at cycling rates of C/10, C/5, and 1 C were 82, 68, 45, and 80 mAh/g, respectively. Notably, Rct significantly decreased due to the incorporation of CNT, dropping from 234 Ω to 64 Ω (Fig. 8c), thereby validating the superior electronic conductivity of the NFPF/CNT composite material (Fig. 8b).
Although aqueous batteries have a lower energy density, compared to the organic electrolytes currently used, aqueous electrolytes offer faster ion dynamics, higher safety, lower costs, and minimal pollution. These characteristics allow aqueous batteries to be applied more broadly in areas where cost is prioritized over energy density [95]. Since cost and safety are crucial metrics for batteries, identifying electrode materials that are abundant, easy to synthesize, and stable has become a primary objective in the development of aqueous batteries. As shown in Fig. 8d, Sharma [89] applied NFPF in aqueous sodium-ion batteries, achieving discharge high-rate capability and good cycling stability. Additionally, as shown in Fig. 8e, the author reported a discharge capacity of 85 mAh/g for full battery using NFPF as the positive electrode and NASICON-type NaTi2(PO4)3 as the negative electrode, with a battery potential of 0.7 V (vs. SHE) and the energy density is 30 Wh/kg (Fig. 8f).
Furthermore, in calcium-ion batteries, the calcium-doped NFPF demonstrated an initial capacity of 60 mAh/g at 10 mA/g. After undergoing 50 cycles, the capacity increased to 80 mAh/g, which can likely be linked to the smaller particle size caused by electrochemical strain during the cycling process, which consequently leads to an increased surface area and enhanced utilization of the material [90].
6.
Conclusion
Iron-based phosphate NFPF is considered as an optimal material for SIBs because of its low cost, abundant resources, and small volume change. This article presents research progress about crystal structure, channels for Na+ transportation and synthetic methods of NFPF. It is a kind of layered fluorophosphate compounds which allows Na+ diffuse along the a- and c-axis direction and features intra-phase and inter-phase Na+ diffusion pathways. Besides, this review provides an in-depth analysis of modification methods to increase the electrochemical capacity of NFPF from the perspectives of improving material conductivity and increasing active Na sites. Element doping and hybridization with carbon materials are considered to be an effective means to enhance electrochemical performance. Hybridization with carbon materials can effectively improve the conductivity of NFPF and reduce particle size, while element doping helps enhance the electronic conductivity and activate inert Na1, thereby improving electrochemical performance. For morphology control, the nanoscale or porous structure are beneficial to achieve high-rate capability. Electrolyte optimization helps the electrode to form a stable CEI film, thereby improving cycle stability. Additionally, the application of NFPF in other systems including LIBs, PIBs, aqueous SIBs and calcium-ion batteries is discussed. Despite the considerable advancements achieved in previous studies, obstacles continue to persist in its progression:
(1) Suitable dopants and sites. Although numerous efforts in Fe site doping have already been made to enhance the structural stability and electrochemical performance of NFPF, there are still many undeveloped transitions metal ions. Additionally, studies on doping at sodium and anion sites remain comparatively limited. But finding the NFPF cathode materials with “highest” electrochemical performance through various possible combinations is a time-consuming and labor-intensive task for researchers. Therefore, combing machine learning with theoretical calculations to find suitable materials is probably a promising research direction to obtain the high-performance NFPF materials.
(2) Modification design. Morphological design and the integration with carbon materials can significantly enhance electrochemical performance. The conductivity, reaction activity, and charge storage capacity can be improved by optimizing the morphology of materials, leading to superior electrochemical performance. On the other hand, carbon materials are often used to enhance electrochemical performance due to their excellent conductivity and chemical stability. Therefore, there is an urgent need for researching and developing more efficient and cost-effective preparation technologies to promote the practical application of these innovative strategies.
(3) Mechanism investigation. It is crucial to go deep into the material structure and reaction processes. Detailed studies on the Na+ storage mechanism of NFPF have been conducted using various characterization techniques, including in-situ X-ray diffraction (in-situ XRD) and solid-state nuclear magnetic resonance (ss-NMR). However, advanced material characterization methods, such as synchrotron X-ray diffraction (SXRD), X-ray absorption spectroscopy, and neutron diffraction are still required to accurately understand the reaction mechanisms of NFPF. These tests are characterized by high resolution and non-destructive capabilities, enabling researchers to reveal the structure of the material, thus informing strategies for optimizing NFPF performance.
(4) Electrolyte. Electrolytes are critical components of batteries, directly influencing their electrochemical performance and safety as they will react with the cathode materials to form a cathode electrolyte interphase. Currently, the commonly used electrolytes for NFPF are ether or ester electrolytes. Electrolyte additives such as fluoroethylene carbonate (FEC) play a positive role in carbonate solutions containing sodium salts. To further optimize the performance of NFPF, it is crucial to study the composition, concentration, and additives of the electrolyte in depth. Different electrolyte formulations can result in significant variations in battery cycle stability and rate performance. Therefore, systematically exploring the impact of these factors on NFPF battery performance not only provides a theoretical basis for battery design and optimization but also enhances their practical application performance, ensuring reliability and safety under various operating conditions.
(5) Full cell. Current research primarily focuses on the electrochemical performance of half-cells. However, in practical applications, testing only half-cells is insufficient to comprehensively evaluate the practicality of material. Therefore, conducting a comprehensive performance assessment of NFPF in full cell is important to better understand its potential and limitations in practical applications. This provides a more solid theoretical foundation and practical guidance for the future development of battery technologies.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
This work was supported by National Natural Science Foundation of China (No. 52064031), Natural Science Foundation of Yunnan Province (Nos. 202301BE070001–014, 202301AT070150, 202402AB080001), and the Analysis and Testing Foundation of Kunming University of Science and Technology (No. 2022T20210182).
[1]
M. Höök, X. Tang, Energy Policy 52 (2013) 797–809.
[2]
X. Deng, J. Li, L. Ma, et al., Mater. Chem. Front. 3 (2019) 2221–2245. doi: 10.1039/c9qm00425d
[3]
A. Noori, M.F. El-Kady, M.S. Rahmanifar, et al., Chem. Soc. Rev. 48 (2019) 1272–1341. doi: 10.1039/c8cs00581h
[4]
Y. Jiang, Z. Zhang, H. Liao, et al., ACS Nano 18 (2024) 7796–7824. doi: 10.1021/acsnano.4c00613
H. Hu, Y. Wang, Y. Huang, et al., J. Cent. South Univ. 26 (2019) 1521–1529. doi: 10.1007/s11771-019-4108-5
[93]
L. Mingjun, G. Rongting, L. Wei, et al., J. Electroanal. Chem. 922 (2022) 116772.
[94]
A. Huiting, L. Wei, L. Zheng, et al., J. Alloys Compd. 854 (2021) 157156.
[95]
Y. Niu, Y. Zhang, M. Xu, J. Mater. Chem. A 7 (2019) 15006–15025. doi: 10.1039/c9ta04274a
Figure 1
(a) Orthorhombic NFPF structure with a space group of Pbcn. Reprinted with permission [28]. Copyright 2019, Elsevier. (b) Ex-situ XRD patterns of NFPF@C samples during the first charge/discharge process. Reprinted with permission [31]. Copyright 2024, Wiley-VCH. (c) Discharge mechanism for Na1+xFePO4F at 0.5 < x < 1. Spheres represent Na1+xFePO4F particles. Red regions show the NFPF phase, and green regions show the P21/b Na1.5FePO4F phase. Yellow arrows show the diffusion path. Reprinted with permission [34]. Copyright 2018, Wiley-VCH. (d) Na+ migration path along the a- and c-axis in NFPF; octahedral FeO4F2 and tetrahedral PO4 are represented by grey and yellow respectively. Na+ migration path, (e) along the a-axis, and (f) along the c-axis. Reprinted with permission [32]. Copyright 2013, Royal Society of Chemistry.
Figure 4
(a) TEM of NFPF/C. Reprinted with permission [49]. Copyright 2017, Springer. (b) HRTEM of NFPF/C-650. Reprinted with permission [35]. Copyright 2023, Elsevier. (c) SEM images of Na2Fe0.6Mn0.4PO4F/C prepared with ascorbic acid. Reprinted with permission [45]. Copyright 2020, Frontiers Media S.A. (d) Schematic representation of NFPF-CB and NFPF-CNT growth process. (e) SEM micrographs of NFPF-20CNT. Reprinted with permission [79]. Copyright 2017, Springer. (f) HR-TEM images of NFPF@C@MCNTs. Reprinted with permission [53]. Copyright 2022, Wiley-VCH. (g) LbL nano-assembly process for the formation of NFPF/CNT multi-layered cathode. Reprinted with permission [54]. Copyright 2015, Elsevier. (h) TEM images NFPF@CNT&GN. Reprinted with permission [40]. Copyright 2020, Elsevier. (i) HRTEM images of 5T-NF@C. Reprinted with permission [31]. Copyright 2024, Wiley-VCH. (j) FESEM images of NFPF. Reprinted with permission [55]. Copyright 2015, Royal Society of Chemistry.
Figure 6
(a) The differential charge of NFMPF. (b) The schematic illustration of the working mechanism by Mg2+ doping. (c) The energy barrier from Na2 → Na2 site of NFPF. (d) The energy barrier from Na1 to Na2 site of NFMPF when the migration site and Mg increased. (e) The cycle performance at 20 C for NFPF and NFMPF samples. Reprinted with permission [64]. Copyright 2024, Elsevier. (f) The local environment of the Na1 and Na2 sites in NFPF. (g) The occupation of Na1 calculated by refined results from XRD patterns and the corresponding specific capacity of different samples (at 0.1 C). (h) Depiction of the Na3 environment deriving from Na1 in FeCu-0.05. (i) 23Na ss-NMR spectra at 60 kHz of NFPF and FeCu-0.05. (j) The calculated de-intercalation potential of Na sites in different samples. (k) Comparison in the rate capabilities between FeCu-0.05 and the literature data for NFPF cathodes. (l) Cycling performance at 0.5 C. Reprinted with permission [48]. Copyright 2023, Wiley-VCH.
Figure 7
(a) Schematic illustration of the formation process of hollow C/NFPF spheres. (b) Rate capability of C/NFPF. Reprinted with permission [68]. Copyright 2012, Elsevier. (c) Schematic illustration for the formation process of double-shelled hollow NFPF/C spheres during the solvothermal reaction. (d) Rate performance of double-shelled hollow NFPF/C spheres. Reprinted with permission [49]. Copyright 2017, Springer US. (e) Schematic of the proposed formation mechanism of NFPF/BC hollow microspheres. (f) Rate performance of the NFPF/BC-2 sample at various rates. Reprinted with permission [37]. Copyright 2021, Wiley-VCH.
Figure 8
(a) Charge/discharge curves at different rates. Reprinted with permission [92]. Copyright 2019, Central South University. (b) Charge/discharge curves extracted from the rate capability experiment at C/15, C/10, C/5, and 1 C rates. Reprinted with permission [73]. Copyright 2023, Elsevier. (c) Nyquist plot of NFPF and NFPF/CNT with its corresponding fitted curve. (d) Galvanostatic potential-capacity profiles (vs. both SHE and Na/Na+) of combustion synthesized NFPF half-cell in aqueous electrolyte. (e) Schematic presentation of full cell aqueous battery comprising of NFPF fluorophosphate cathode and NaTi2(PO4)3 NASICON anode with a potential difference of 1 V (vs. SHE). (f) Galvanostatic potential-capacity profiles (vs. NaTi2(PO4)3) of aqueous full cell comprising NFPF cathode and NaTi2(PO4)3 NASICON anode. Reprinted with permission [89]. John Wiley and Sons Ltd.
DownLoad:
下载:
