Introduction

In recent years, the rapid growth of the worldwide population has led to the increasing consumption of energy. At the same time, the consumption of fossil fuels has caused a significant increase in the emissions of carbon dioxide and other greenhouse gases, which has aggravated global climate change and environmental pollution. Because of the above-mentioned problems, it is very urgent to find a clean and environmentally friendly energy storage material to address environmental pollution and the energy crisis [1,2,3]. The United Nations Sustainable Development Goals were laid out in 2015, and refer to “ensuring access to affordable, reliable and sustainable modern energy for all”. This view fully illustrates the importance of sustainable energy to the production and life of people around the world. Reducing the cost of energy materials, improving the safety of materials, and selecting materials with abundant reserves are important guarantees for the development of sustainable energy materials [4,5,6]. Therefore, selecting sustainable energy materials with excellent physical and chemical properties, such as porous carbon materials, porous polymers, and metal–organic frameworks (MOFs), can lead to the development of energy storage devices with excellent performance according to the need and minimize pollution in the process [7,8,9,10,11,12].

MOFs are periodic network structures comprised of metal ions and organic ligands, which have developed rapidly in recent years [13,14,15,16]. After Yaghi’s group synthesized the first MOF in the 1990s, researchers prepared zero-, one-, two- and three-dimensional MOFs with different structures by changing the central metal and organic ligand, using solvothermal, microwave thermal, and electrochemical methods and so on [17,18,19,20,21,22]. At present, more than 20,000 cases of MOFs have been reported, which have the advantages of unique structure and easy control of porosity [23,24,25,26,27,28,29,30,31,32,33,34,35,36]. In addition, MOFs act as precursors to yield a range of carbon material heterogeneities, metal oxides, and metal sulfides through heat treatment [37,38,39]. In general, the structure of MOFs is well preserved in the pyrolysis process [40,41,42,43]. Therefore, both MOFs and their derived materials have the advantages of high surface area, ordered multistage porosity, and controlled synthesis structures, which are conducive to exhibiting excellent performance in the electrochemical field [44]. However, a single MOF material has difficulty meeting the requirements of practical applications [45,46,47]. Therefore, by combining MOFs with other functional materials, the advantages of the two materials are complemented. When appropriate polymers and necessary molding technologies are introduced into MOFs, polymers can not only serve as adhesives, and enhance the mechanical properties of composite materials, but also ensure the chemical stability of materials. The as-obtained composites retain the characteristics of uniform porosity and flexibility, which can effectively expand the application range [48,49,50,51].

Electrospinning is one way to better combine MOFs with polymers. This is a mature electrospray process that enables the large-scale preparation of micro-/nanofibers by adjusting the size of the needle. Since the sixteenth century, Gilberts claim that liquids have electrostatic attraction [52, 53]. Figure 1 demonstrates briefly the development of electrospinning [54,55,56,57,58,59]. After 500 years of development until 2000, the research focus of electrospinning technology changed from simple preparation technology to the principle and mechanism of electrospinning technology. Therefore, electrospun fiber can better regulate the size, structure, composition, morphology, and porosity of electrospun materials [60, 61].

Fig. 1
figure 1

Development of electrospinning briefly

In conclusion, the development of electrospinning in the early twenty-first century has the following characteristics: (1) plenty of polymers’ heterostructure has been obtained via electrospinning; (2) the morphology control ability of electrospun nanofibers is enhanced; (3) multicomponent inorganic composite fibers can be developed; (4) the research focus is on industrial production that can optimize the process [62,63,64,65,66]. Common methods for constructing fiber structures are as follows: (1) Electrospinning composite nanofibers form by adding the need material directly into the electrospinning solution [67,68,69]. (2) Special structure, such as core–shell, tube-in-tube, multicore cable-like, rice grain shape, helical, ribbon-like, necklace-like, multichannel tubular, nanowire-microtube, firecracker shape, and hollow fiber structures [70,71,72,73,74,75,76,77,78,79,80,81,82,83,84]. (3) Nanofiber composites with controllable fiber diameter and personal material properties were prepared [85, 86]. To further study the development trend of MOFs and electrospinning, the two keywords were searched. MOF material was first reported in 1977. The electrospinning technique was first reported by Xiong in 1997. The quantity and quality of research papers on electrospinning and MOFs have improved since the twenty-first century. Electrospinning and metal–organic frameworks both increased from only a few papers per year to 5583 and 7483 in 2022 (Fig. 2) [87,88,89,90,91]. In the reported literature, polymer nanofibers can become the most ideal framework for MOF materials because of their large specific surface area, high porosity, high mechanical strength, and good permeability. The electrospinning nanofiber framework can effectively inhibit the agglomeration of MOF particles and increase the utilization of active sites.

Fig. 2
figure 2

Number of published papers between 1998 and 2022.10, with the keyword “electrospinning and metal–organic frameworks”

In this paper, the latest research results of MOFs nanofiber composites materials and their derived composites materials are reviewed and applied to supercapacitors, lithium-ion batteries, sodium-ion batteries, potassium-ion batteries, and lithium–sulfur batteries (Fig. 3). It highlights that the structural design has a great influence on the electrochemical performance. The development prospects and challenges of MOF nanofibers have prospected. This provides more references for subsequent researchers in electrode design [92,93,94,95].

Fig. 3
figure 3

Schematic diagram of the main application fields of MOF nanofibers

Feasible Strategies and Preparation of MOF-Based Materials

Electrospinning is a way of making composite nanofiber membranes from polymers and MOFs. Under the action of high-voltage static electricity, different kinds of polymers and different solvents can form a specific electrospinning solution. There are usually two ways to prepare the MOF/polymer fiber membrane via electrospinning: direct electrospinning of the prepared MOFs and polymers; in situ growth of MOFs on the surface polymers nanofiber by electrospinning [96,97,98].

Structural Design MOFs’ Hybrid

MOFs’ powder materials are usually synthesized first. As long as the selected MOF does not react with the electrospinning solution, the original channel and structure of the MOFs material will not be damaged during the electrospinning process. Subsequently, parameters (the concentration of MOFs, applied voltage, the viscosity of the slurry, injector of diameter, and type of needle) could be adjusted to optimize the experimental conditions. This process enables MOFs to be directly embedded into the polymer nanofiber. Finally, a large number of MOFs polymer nanofibers with controllable structures are as-obtained [99, 100].

In Fig. 4a, Lu’s group prepared the membranes of UIO-66/PVA composite by rapid electrospinning technique [101]. PVA contained a hydroxyl group and the MOF contains carboxylic acid, which is converted to an ester bond through esterification, making MOF and PVA connected to form the fibrous network. The liquid electrolyte immersed into the electrospun MOF–PVA composite membranes (EMP). The anions of the electrolyte automatically adsorbed to the MOF and fixed in the MOF channels.

Fig. 4
figure 4

Schematic diagram of MOF directly added into the electrostatic spinning solution to prepare materials. a MOF–PVA composite membrane; bc the SEM of MOF–PVA; reprinted with permission from ref. [101], Copyright 2019, Wiley-VC; d ZIF-7/PAN; e TEM images of PAN-NF; f ZIF-7/PAN-NF; reprinted with permission from ref. [102], Copyright 2016, Wiley-VC; g the structure of different types of MOFs; h optical and scanning electron microscopy of composite materials; reprinted with permission from ref. [103], Copyright 2016, American Chemical Society

This process improved the efficient transport of lithium ions in the electrolyte, increased the Li-ion migration number and lithium-ion conductivity. In addition, the introduction of MOF particles reduced the decomposition of the electrolyte and promoted the electrode reaction kinetics (Fig. 4b). Figure 4c displays an SEM of EMP. The overall morphology was beaded fibrous, and the regular octahedron shape of the MOF distributed in the PVA nanofiber. Yin et al. produced ZIF-7/PAN nanofiber with the variation of the slurry fabrication process electrospinning (Fig. 4d). In this process, the raw materials of ZIF-7 directly put into the electrospinning solution. This method helped to control the size of the ZIF-7 much less than the diameter of the PAN nanofibers. In Fig. 4e, f, the TEM of a single PAN nanofiber owned a slippy surface and a uniform diameter of 200–300 nm. The ZIF-7/PAN composite demonstrated that ZIF-7 evenly was distributed in the fibers, and the fiber surface became more corrugated [102]. Wang and his team prepared a series of flexible composite nanofiber films by embedding four MOFs into three polymers (PAN, PVP, and PS) by electrospinning (Fig. 4g).

The composite membrane was used as a contaminant filter. Agglomeration was prevented by adjusting the optimal loading between the MOFs and the polymer. Pollutants interact with MOFs in different forms to achieve filtration. Figure 4g shows that the four MOFs have a unique spatial structure that can effectively adsorb different functional groups, enhancing the ability of the composites to capture different types of pollutants in air filters. Figure 4h shows the digital photos and corresponding SEM photos of different MOF on different electrospinning substrates [103]. Different Ni-BDC loadings of Ni-BDC/PAN nanofiber membranes were obtained, and the proton conduction property of the composite was studied by Liu’s group. The imidazole molecules added into one-dimensional MOF/polymer fibers significantly affected their proton conductivity. The Ni-BDC and imidazole put into the electrospinning solution to fabricate the composite film [104]. The structure of Ni-BDC promoted the movement of imidazole molecules.

Hybrids of MOF-Derived Nanostructures

MOF derivatives are mainly divided into porous carbon and composite materials. It has the following advantages in energy storage. Hollow and porous structures contribute abundant active sites; multi-dimensional derivative materials could provide more ways of electron transport; the derivative film materials can play their flexible and self-supporting characteristics. Therefore, researchers have made some achievements in MOF-derived materials in recent years [105,106,107,108,109].

The remarkable surface area, lightweight, super-long stability, and open porosity of carbon materials have been diffusely applied in Li+/Na+/K+ ion batteries and supercapacitors (SCs) [110,111,112,113,114,115]. At the same time, electrospun nanofiber materials form a series of carbon-based composites through annealing treatment, which has also become a hotspot of supercapacitor materials. Kim et al. designed the flexible membrane (Co/N-CNT@PCNF900) vis a facile method, which processed no need for extra reducing gases [116]. Co(NO3)2@PAN-NF membrane formed by adding Co(NO3)2 to polyacrylonitrile (PAN) by electrospinning. The as-prepared membrane put into a 2-methylimidazole solution and obtained ZIF-67@PAN-NF, which subsequently annealed to obtain Co/N-CNT@PCNF900 in Fig. 5a. The SEM of Co/N-CNT@PCNF900 nanofibers had an entwined N-CNT network structure, which was beneficial to increase the flexibility of the membrane. The characteristics of the flexible and porous carbon nanofibrous (Co/N-CNT@PCNF900 membrane) possessed uniform pore size distribution, and homogeneous metal atoms, which displayed remarkable electrocatalysis and SCs. The design strategy combined the advantages of MOF and electrospinning to form a low-cost and universal method. Sun reported a kind of nitrogen-doped nanofiber carbon (N-NFC) derived from MOF in Fig. 5b. MOF layers grew on the electrospun polymer nanofiber with controllable size and uniform morphology. The ions between polar groups on the polymer side chain and the metal promoted nucleation of MOF, and steric hindrance resulted in uniform growth of the MOF. After high-temperature annealing, these nanofibers transformed into N-NFC. The amount of the outer carbon shell could be regulated by the ratio between organic ligands and metal ions. A low concentration of an organic ligand could increase the size of ZIF-8, which also reduced the amount of ZIF-8 on the surface. N-NFC-8 exhibited the best concentration ratio, and the hollow carbon skeleton was evenly distributed on the surface [117]. The core-sheath nanofibers transformed into N-NFC during high-temperature pyrolysis. N-NFC structure has the following advantages: (1) The highly conductive one-dimensional nanofiber structure is the core; (2) the outer layer is a uniformly distributed hollow carbon shell; (3) there are a large number of micro/mesoporous pores in the structure; (4) uniform n-doping is achieved. Finally, it shows outstanding electrochemical performance.

Fig. 5
figure 5

Preparation mechanism of MOF-derived carbon composites. a Co/N-CNT@PCNF membrane; reprinted with permission from ref. [116], Copyright 2021, American Chemical Society; b nitrogen-doped Nanocarbon (N-NFC); reprinted with permission from ref. [117], Copyright 2019, Royal Society of Chemistry

Recent Progress of MOF-Based Materials in Energy Storage

For the past few years, supercapacitors, lithium/sodium-ion batteries, play a significant role in energy storage [118,119,120,121]. The choice of electrode materials determine its electrochemical performance. Therefore, a great deal of researches have focused on exploring electrode materials with high efficiency, long cycle life, low cost, and high specific capacitance. However, MOF nanofibers contribute more electrochemistry active sites in the reaction, due to their aperture structure with larger surface area and adjustable aperture. In this section, research on MOF nanofibers and their derived materials progress and summarize their energy storage performance in Table 1.

Table 1 The MOF nanofiber and derived-MOFs’ materials for energy storage performance

Supercapacitors

The supercapacitor is a kind of energy storage device that can carry out high-power fast charging (within a few seconds) and ultralong cycle life. MOFs and their derived materials tend to have a large active surface area, reasonable pore size, and chemical stability. At the same time, the metal ions contained in MOFs can also provide more sites for pseudocapacitor energy storage. Therefore, MOF nanofibers and derived materials have attracted wide attention in the application of supercapacitors.

Wang demonstrated a simple strategy to prepare bimetallic MOFs on preoxidized polyacrylonitrile nanofibers (PPNFs). The metal ions in the structure provided pseudocapacitor energy storage for electrochemical reactions. First, polyacrylonitrile was used as a raw material to form fiber films by electrospinning and calcination to obtain PPNFs. Then, through solvothermal reactions, metal ions coordinated with terephthalic acid to obtain the M-Ni MOF in Fig. 6a [122]. In Fig. 6b, c, the SEM images of PPNF@Co–Ni MOF, and PPNFs are completely covered by a globular Co–Ni MOF with a diameter of 1.4–1.8 μm. However, when Zn ions are introduced, Zn-Ni MOF grew evenly on PPNFs, but the flower globules disappeared, and the surface of PPNFs was relatively smooth with only a few nanosheets. Meanwhile, the hierarchical flower-like nanosheets demonstrated a remarkable specific surface area. The introduction of Co into PPNF@Ni MOF improved redox activity, increased electronic conductivity, and exhibited the synergistic effect of different elements. Among them, PPNF@Co–Ni MOF showed remarkable electrochemical performance in supercapacitors. At the current density of 1 A g−1, the specific capacity delivered 1096.2 F g−1 (548.1 C g−1). Yoon reported an easily prepared and low-cost Mn@ZnO/CNF electrode material for supercapacitors, as shown in Fig. 6d [123]. Mn-doped ZIF-8 directly grown on polyacrylonitrile nanofibers. The composite fiber carbonized to form a dodecahedron Mn@ZnO with CNF as the core and a high surface area. Heat treatment carbonization of MOFs allowed the metal oxides to be well-distributed on the porous carbon framework. This structure provided more pathways for fleetly electron transfer and increased the active site of the Faraday reaction. Meanwhile, the introduction of carbon material increased the electroconductivity of the composite [124,125,126,127,128]. The SEM of Mn-ZnO showed the uniform growth of dodecahedron on the surface of polyacrylonitrile fiber (Fig. 6e) and the optical photo of flexibility supercapacitor devices, as shown in Fig. 6f.

Fig. 6
figure 6

Fabrication diagram of annealed electrospinning composite film electrode in supercapacitor. a M-Ni-MOF (M=Co, Zn, Cu, and Fe); b SEM of as-obtained materials PPNF@Co–Ni-MOF; c PPNF@Zn-Ni-MOF; MOF; reprinted with permission from ref. [122], Copyright 2020, American Chemical Society; d Mn@ZnO/CNFs; e SEM of Mn@ZnO/CNFs; f flexible electrode LED power supply; reprinted with permission from ref. [123], Copyright 2019, Elsevier

Xu’s group prepared the 3D macrostructures’ construction of nanoporous Co3O4, which was synthesized by the PAN/ZIF-67 nanofibers after annealing treatment in Fig. 7a. Co3O4 derived in situ from MOF and distributed uniformly in electrospun polyacrylonitrile nanofibers. After calcination, Co3O4 inherited the 3D pore structure from the PAN/ZIF-67 nanofibers. The 3D network increased the diffusion path of ions. The unique spatial structure and appropriate porosity enhanced the enhancing capacitance. The high capacitance of the Co3O4 electrode exhibits 970 F g−1 at 1 A g−1 [129]. Similarly, in 2021, Kim and his team used similar nanoparticles infused with tiny Co3O4 (NPs) [130]. After annealing, electrospinning PAN became the CNFs. Cobalt carbonate hydroxide nanohairs grown on CNFs by hydrothermal method and completely transformed into ZIF-67@CNFs. Under low temperatures and no reducing atmosphere, a 3D array Co3O4/N-CNTs@CNF is obtained. This design enabled MOF as template not only to serve as the source of CNTs, N-doping, and a small amount of Co3O4 NPs, but also to ensure a large specific surface area, high porosity, and uniform composition distribution in the composite structure. The multi-step structure construction and annealing were beneficial to the stability of the three-dimensional structure. In this design idea, through low temperature and non-reducing gas, and without the need for additives could design a composite structure with multiple components.

Fig. 7
figure 7

Preparation process of MOF-derived nanostructures in supercapacitor. a Design idea of nanoporous Co3O4; reprinted with permission from ref. [129], Copyright 2021, Elsevier, b CNF@Ni-CAT; reprinted with permission from ref. [114], Copyright 2019, Royal Society of Chemistry; c CNT@HCNF-x electrode; reprinted with permission from ref. [131], Copyright 2022, Wiley-VC

Han reported the core–shell structure of CNF@Ni-CAT as a flexible membrane by electrospinning [114]. After the carbonization of 800 °C, the electrospinning of PAN converted to carbon nanofibers (CNFs). Ni-CAT nanorods grown on the surface to accelerate the transport between ions (Fig. 7b). This design strategy effectively improved the agglomeration problem of MOFs and increased the contact area between the electrode material and the electrolyte. The composite film material also had excellent flexibility, which made a certain contribution to the application of flexible electronic energy storage. Kim’s group prepared highly porous N-doped carbon nanotubes (N-CNTs) on hollow carbon nanofibers, owing to a high capacitance, as shown in Fig. 7c [131]. First, polyacrylonitrile (PAN) and polymethyl methacrylate (PMMA) electrospun to obtain the sheath–core nanofibers. Then, the ZIF-67 grew evenly on the sheath–core nanofibers. The ZIF-67/sheath–core nanofibers were converted into hollow carbon nanofiber composite porous N-doped carbon tubes (CNT@HCNF-1.5) by the one-step method under an optimal precursor ratio and appropriate heat treatment temperature. The preparation processes were simple and did not require reducing gas. CNT@HCNF-1.5 reached a capacitance approaching 712 F g−1.

Lithium-Ion Batteries

Lithium-ion batteries (LIBs) are the core component of traditional storage devices and electric vehicles because of their ultra-high energy density and admirable cycle life. However, LIBs still face many challenges. For example, in the process of Li-ion intercalation/deintercalation, the volume of the electrode material changes, resulting in capacity attenuation; ion diffusion and transport are slow in some electrode materials. Therefore, it is very important to design electrode materials carefully for lithium-ion energy storage. The porous structure of MOF and its derivatives buffered the volume change during the process of lithium/dilithium and improve cyclic stability. The integrated electrode design avoids using a binder and improves the overall conductivity of the electrode material [132,133,134,135,136].

Zhang and their partners proposed that MoS2 nanoplates were directly grown on the surfaces of N-doped porous carbon nanofibers (PCNFs) and obtained hierarchically from PCNF@MoS2 fibers by electrospinning. Figure 8a shows the synthesis process of PCNF@MoS2. High-length–diameter nanofibers of ES-PAN@ZIFs were prepared using PAN and nitrogen-rich ZIF-8 NPs by electrospinning. ES-PAN@ZIFs nanofibers annealed in an Ar atmosphere to get N-doped PCNF. By hydrothermal method, homogeneous MoS2 grown on the PCNFs. The well-designed PCNF@MoS2 could prevent Li+ from damaging the material structure and good reversibility. The capacity at 1 A g−1 current density was 1116.2 mAh g−1 after 450 cycles [137]. Zhang and his partners proposed an MOF-derived carbon necklace structure, which interconnected by heteroatoms and metal nanoparticles [138]. ZIF-67 and MIL-88 severally put into polyacrylonitrile (PAN) electrospinning solution to form MOF@PAN chain precursor films by electrostatic spinning, as shown in Fig. 8b. ZIF-67@PAN and MIL-88@PAN derived carbon necklace paper (CNPs) and indicated as ZCNP and MCNP by direct pyrolyzation of precursor film in the N2 atmosphere. During the pyrolysis process, the MOFs transformed into nitrogen-doped porous carbon materials, metal particles, which facilitated the storage of lithium ions. PAN fibers pyrolyzed to form a 3D conductive network, which formed a larger contact area, thus promoting the fast transfer of electrons in the electrochemical reaction. Meanwhile, this material had a self-supporting structure, which regarded as an anode of LIBs for performance testing without additional conductive agents and binders. The ZCNP and MCNP exhibited outstanding cyclic life and an impressive capacity of 1200 and 980 mAh g−1 in the current density of 200 mA g−1. Figure 8c shows the SEM of the ZIF-67-derived structure. After heat treatment, polyhedral nanocages embedded into the 3D carbon fiber conductive network. ZIF-67 could still maintain its polyhedral structure. The wrinkles on the surface of polyhedra related to the decomposition of organic groups in MOF and PAN. It displayed similar results derived from MIL-88@PAN in Fig. 8d. Tang et al. introduced 7 nm Mn3O4 into MOF-derived nitrogen-doped porous carbon fibers (NPCFs) via the sol–gel method and electrostatic spinning method [139]. The design idea was beneficial to reduce the agglomeration of Mn3O4 in composites. The open space structure can relieve the volume expansion during Li+ charge/discharge cycles. Rugged one-dimensional construction contributed to outstanding electrical conductivity. NPCFs provided a more effective path for rapid ion and electron transfer. The strongly coupled interaction between Mn3O4 and NPCFs occurs, and its electrochemical property was observably enhanced. The Mn3O4/NPCFS-1.4 flexible material can reach 1058 mAh g−1 at 50 mA g−1.

Fig. 8
figure 8

The scheme for the MOF-derived carbon fiber preparation diagram in lithium-ion battery. a PCNF@MoS2; reprinted with permission from ref. [137], Copyright 2019, Elsevier; b carbon necklace paper; SEM of derived from c MIL-88@PAN and d ZIF-67@PAN; reprinted with permission from ref. [138], Copyright 2017, Wiley-VC

Zhang’s group reported an assisted anion-exchange strategy for CoSnx@CPAN nanofibers, by coaxial electrospinning in Fig. 9a. Co(CH3COOH)2 and 2-methylimidazole stirred separately in a PAN solution. The two solutions transferred into two identical syringes through coaxial electrostatic spinning, and ZIF-67@PAN nanofiber membranes were formed. The composite nanofibers immersed in Na2SnO3 solution for anion exchange. Anion exchange required neither additional input of energy nor a complex experimental environment. This special design concept encapsulated well-dispersed nanoparticles in a three-dimensional hierarchy polymer nanofiber frame. CoSnx@CPAN enhanced the storage and diffusion of electrons/charges. The obtained CoSnx@CPAN and CoSn(OH)6@PAN showed remarkable performance in lithium-ion batteries [140]. Liang et al. presented a process via electrospinning Mn-MOF. After annealing, the yolk–shell MnOx contained carbon nanofibers in Fig. 9b [106]. As shown in Fig. 9c, Mn-BTC spheres and polyacrylonitrile (PAN) formed a fiber composite material with strings of beads using an electrostatic spinning structure. Increasing the concentration of Mn-BTC electrospinning solution can increase the number of spheres in the fiber composite. Subsequently, the Mn-BTC@PAN composite material was peroxidation treated in the air at 300 °C. The sieve shell structure of the Mn-BTC nanospheres was oxidized in the pre-oxidation step and showed in the TEM of Fig. 9d. Finally, it is further flamed in N2 at 600 °C for 2 h to form ysMnOx@NC nanofibers (Fig. 9e). The intermediate heating step can stabilize the structure of the nanofibers and facilitate the retention of the structure after carbonization. The yolk–shell of MnOx accepted volume expansion in lithium-ion migration processes and effectively prevented the pulverization of active substances, while the outer carbon fiber shell ensured the integrity of the structure during charge/discharge. The morphology and composition of yolk–shell MnOx composite nanofibers with different loading loads investigated and characterized. It proposed that the preparation mechanism of this structure might be Oswald curing and the Kirkendall process. After 1000 cycles, this structure still displayed excellent stability. The dual-buffer design can also present the process of lithium/dilithium by ex situ imaging. This synthesis strategy had certain versatility and can be used as a reference for preparing other transition metal oxides [141]. Kim et al. prepared SnO2–Co3O4 nanofibers with oxygen vacancies, as shown in Fig. 9f. ZIF-67, Sn precursor, and PVP mixed as an electrospinning solution for electrospinning. ZIF-67 uniformly distributed on the surface of PVP fiber. After calcination, ZIF-67 decomposed on the nanofibers to form the mesoporous site SnO2–Co3O4. Due to the introduction of ZIF-67, more oxygen vacancies formed as active sites for lithium storage, which improved the conductivity of the composites and facilitated the transport of ions and electrons. In addition, one-dimensional structures with high surface area and high porosity can show excellent electrochemical performance, with a reversible capacity of up to 1287 mA h−1 at the current density of 500 mA g−1, after 300 cycles [142].

Fig. 9
figure 9

Schematic diagram of MOF-derived composite fiber membrane in lithium-ion batteries. a CoSnx nanoparticles; reprinted with permission from ref. [140], Copyright 2020, Elsevier; b ysMnOx@NC; the TEM image of c Mn-BTC@PAN-2. d Pre-oxidized Mn-BTC@PAN-2; e ysMnOx@NC-2; reprinted with permission from ref. [106], Copyright 2019, Wiley-VC; f Mesoporous SnO2–Co3O4 nanofibers; reprinted with permission from ref. [142], Copyright 2018, American Chemical Society

Lithium–Sulfur Batteries

Lithium–sulfur battery demonstrates potential in the electrochemical application, owing to its advantages of splendid theoretical energy density (2600 Wh kg−1), low cost, and sufficient reserve. Unfortunately, the rapid dissolution of polysulfide results in decrease its capacity and cycle life. The key to solving those problems is to design effective structures to prevent dissolution. One feasible way to overcome the above shortcomings is to construct electrode materials with high conductivity and abundant porosity. MOF-derived fiber electrodes offer great potential in this regard.

Qiu and his partners designed ZIF-67 as a functional sandwich structure grown on carbon nanofibers (CNFs). The structure had a uniform microporous carbon skeleton and mesoporous ZIF-67 as the active site, which can effectively adsorb polysulfide and accelerate the movement of polysulfide. At the same time, the carbon nanofibers’ structure can promote the electroconductibility of the ZIF-67/CNFs’ material and supply a pathway for electron transfer, as shown in Fig. 10a. Therefore, a well-designed sandwich structure helped to inhibit the dissolution of polysulfide and reduce the shuttle effect utilizing physical blockade. Finally, the ZIF/CNFs displayed 1334 mAh g−1 and excellent cyclic life in 1 C. [38]. Yao et al. presented the 3D ivy-structured MoS2@N-CNFs via electrospinning. First, N-CNFs films obtained by electrospinning and annealing of PAN and ZIF8, and then, MoS2 nanosheets uniformly grown on the surface of N-CNFs fibers by the hydrothermal method in Fig. 10b [143]. The MoS2@N-CNFs composite membrane could be directly used as a cathode for lithium/polysulfide batteries. N-CNFs films with high conductivity can effectively reduce the internal resistance of the electrode as a lithium polysulfide. Meanwhile, the MoS2 nanosheet chemically combined with the collector. It had a high theoretical capacity and strong binding energy with polysulfide. The pyridine nitrogen mesopores in N-CNFs could be used to fix dissolved lithium to form polysulfide. The MoS2@N-CNFs cathode demonstrated a discharge capacity of 920.0 mAh g−1 at 0.2 C. DFT calculation showed that MoS2 nanosheet sulfide/metal–oxide nanosheet composite N-CNFs were a feasible strategy to promote the electrochemical behavior of lithium–sulfur batteries anode.

Fig. 10
figure 10

Schematic illustration of the as-obtained in the lithium–sulfur battery. a ZIF-67/CNFs; reprinted with permission from ref. [38], Copyright 2021, Elsevier; b MoS2@N-CNFs; reprinted with permission from ref. [143], Copyright 2019, Elsevier

Deng et al. fabricated high sulfur load CoS2–SPAN–CNT electrodes (Fig. 11a) [144]. First, PAN, CNT, and MeIM electrospun to obtain PAN–CNT–MeIM films. Adding into the salt solution containing Co at room temperature, ZIF-67@PAN–CNT film is obtained. After curing, the SPAN–CNT decorated with CoS2 film, which had a high sulfur load of 4.6 mg cm−2 (only 90 mm thick). As a result, the CoS2–SPAN–CNT electrode leaded to a capacity of 1240 mA h g−1 at 0.2 C. The modification of ZIF-67 in the structure helped to restrain the expansion of the fiber and the thickening of the membrane. Meanwhile, the introduction of CoS2 accelerated the redox kinetics of sulfur. According to the SEM images of the PAN–CNT–Meim membrane shown in Fig. 11b, the average fiber diameter of the PAN–CNT–Meim was about 430 nm. After modification of ZIF-67, the average diameter of the ZIF-67@PAN–CNT membrane increased to 714 nm (Fig. 11c). The resulting vulcanized CoS2 fixed on the surface of the SPAN–CNT film (Fig. 11d). Liu’s research groups also reported and synthesized the MOFs modified gel F-doped poly-m-phenyleneisophthalamide (PMIA) nanofiber membranes. PMIA of ZIF-67 and Cu-BTC is combined by blending electrospinning, as shown in Fig. 11e [145]. The electrochemical performance and safety of Li–S cells improved by the modified fiber film. F-doping is mainly contributed to the formation of a gel state and reduced the average fiber diameter of the prepared membrane. These membrane materials had smaller pore sizes, higher porosity, and excellent ionic conductivity. MOF doping enhanced the chemical adsorption of lithium polysulfide and showed a good catalytic effect in the redox reaction. In terms of the adsorption of poly lithium sulfide, the F-doped and MOFs co-doped membrane was beneficial to physically capture poly lithium sulfide, reduced the lithium dendrites, and helped to achieve uniform lithium deposition on the surface of lithium electrode during the process of charging and discharging. Both F-ZIF-67 and F-Cu-BTC-doped modified films demonstrated excellent electrochemical performance, with discharge capacities up to 1267.5 and 1272.2 mAh g−1 at 0.5 C. It is possible to functionalize MOFs and effectively improve the performance of lithium–sulfur batteries.

Fig. 11
figure 11

Schematic diagram of the as-prepared composites and the morphology characterization in the lithium–sulfur battery. a Process flow diagram of CoS2–SPAN–CNT electrodes; SEM images of b PAN–CNT–MeIM, c ZIF-67@PAN–CNT; d CoS2–SPAN–CNT; reprinted with permission from ref. [144], Copyright 2020, Royal Society of Chemistry; e the schematic illustration of the MOFs nanoparticles modified PMIA membrane; reprinted with permission from ref. [145], Copyright 2020, Elsevier

Sodium-Ion Batteries

Sodium-ion batteries (SIBs) have appealed much attention, because harmless properties, low cost, rich sodium resources, and electrochemical storage mechanism are similar to LIBs. However, sodium-ion have more radius than lithium-ion, which seriously affected the poor circulation and low kinetics of the electrode during the process of sodium/modification. The design and development of unique materials with cavity heterostructure could reduce the volume expansion and increase the electron transport path in the cycle process. MOFs are an ideal template for creating hollow structures. In this summary, MOFs have been used as a template to reasonably design the structure and apply it to the storage of sodium-ion batteries through electrospinning [146,147,148,149,150].

Ma et al. demonstrated hollow heterostructures (CoFe)Se2 three-dimensional (3D) carbon frameworks (CNS) via electrospinning and selenization treatment. The preparation route is shown in Fig. 12a. ZIF-67 nanocube was a self-sacrificing template. ZIF-67 and [Fe(CN)6]3− are transformed into layered hollow ZIF-67/Co-Fe PBA in the cation-exchange method. A hollow ZIF-67/Co-Fe PBA grown on one-dimensional polyacrylonitrile (PAN) nanofiber framework and Co3O4/Fe3O4@CNS obtained by annealing in an Ar atmosphere. After the selenization process, it converted into nano chains (CoFe)Se2@CNS composite materials [151]. The composite displayed remarkable cycling life and rate property in SIBs. After 3650 cycles and the current density of 5 A g−1, the capacity retention rate was still 98.6%. Changing the current density from 0.1 to 8 A g−1, the capacity decreased from 262.9 to 218.2 mA h g−1. The reasons are as follows: (1) The ingeniously designed hollow heterostructure coupled with the carbon skeleton. The hollow internal heterostructure provided enough buffer space for the movement of sodium ions, efficiently liberated the mechanical stress attributing to the volume variation of the material itself, enhanced the transport of electrolyte ions, and shortens the distance of ion transmission. (2) The anisotropy of electron transport enhanced by the three-dimensional network to accelerate the reaction kinetics of sodium ions. (3) The interesting internal electric field at the heterojunction derived from the combination of N-type CoSe2 and P-type FeSe2, which created a built-in electric field at the heterojunction interface, accelerating Na+ migration and adding more active sites. (4) The coupling effect of heterogeneous structure formation reduced the pulverization of materials in the cycling process, thus forming a stable SEI film [152, 153]. Li et al. fabricated the hierarchical NiCo2O4/NiO/carbon nanofibers via an electrospinning and annealing strategy (Fig. 12b) [154]. Co salt and Ni salt mixed into PAN and PVP to get 1D polymer (Co/Ni) fibers, which immersed in 2-MeIM and obtained the polymer (Co/Ni)@Ni/Co-MOF. After the carbonization and oxidation, hierarchical NiCo2O4/NiO/carbon nanofibers demonstrated a sodium-storage capacity of 210 mAh g−1 at 100 mA g−1. The one-dimensional carbonized nanofibers composite film inlaid with NiCo2O4 and NiO conductive particle and intertwined with each other to form a three-dimensional structure. It could effectively shorten the ion diffusion, and accelerated the diffusion kinetics along its one-dimensional form of high-speed electron mobility. The evenly dispersed particles increased the electrolyte exposure surface, which improved the material utilization rate.

Fig. 12
figure 12

Schematic illustration of the MOF-derived composite structure in sodium-ion batteries. a (CoFe)Se2@CNS; reprinted with permission from ref. [151], Copyright 2020, Royal Society of Chemistry; b NiCo2O4/NiO/carbon nanofibers; reprinted with permission from ref. [154], Copyright 2019, Elsevier

Jiang’s group designed the hybrid of Fe7S8/N-doped carbon nanofibers (Fe7S8/N-CNFs) by electrospinning [155]. MIL-88A-Fe/PAN composite membrane obtained by directly adding MIL-88A-Fe to PAN for electrospinning (Fig. 13a). However, Fe7S8/N-CNFs demonstrated a high capacity of 649.9 mAh g−1 at 0.2 A g−1. The excellent electrochemical performance attributed to: the average diameter of Fe7S8 nanoparticles was 17 nm, which shorten the Na+ diffusion path; the one-dimensional carbon network and three-dimensional carbon nanocage structure combined to enhance the conductivity of the composite; Fe7S8 was uniformly dispersed on the surface of the carbon fiber, which could effectively slow the agglomeration and volume expansion of Fe7S8 during the electrochemical reaction. It also provided more active sites for Na+ adsorption. Zhao et al. constructed a carbon-coated metal sulfur selenide solid-solution CoSSe nanosheet on carbon nanofiber (CNF@CoSSe@C) by electrospinning [156]. The electrospun polyacrylonitrile (PAN) nanofiber membrane was used as the precursor, and then, the ZIF-67 nanosheet was embedded in the PAN by immersion. PAN@ZIF-67 was carbonized, vulcanized, and selenized to obtain a carbon-coated CoSSe nanosheet, grown on one-dimensional PAN (Fig. 13b). This structure formed a three-dimensional network structure of PAN as a branch and CoSSe@C as a leaf. CNF@CoSSe@C had a capacity of 479.4 mA h g−1 at 1 A g−1 as the negative electrode of the SIBs. This carefully designed structure was conducive to enhancing the conductivity and mechanical strength of materials, forming more abundant electrochemical active sites, and the three-dimensional structure provided more channels for ion transport. DFT showed that this structure had the lowest migration barrier energy for sodium ions and accelerates the diffusion of Na+.

Fig. 13
figure 13

Preparation process diagram of the as-prepared electrode in sodium-ion batteries. a Fe7S8/N-CNFs; reprinted with permission from ref. [155], Copyright 2021, Wiley-VC and b CNT@CoSSe@C; reprinted with permission from ref. [156], Copyright 2022, Royal Society of Chemistry

Potassium-Ion Batteries

Potassium-ion batteries (PIBs) have become a new type of energy storage device and attract much attention because of their low redox potential, abundant reserves, and low cost. The combination of MOF and electrospinning to prepare potassium-ion battery electrode materials can effectively shorten the ion diffusion path. MOFs as a template could provide a stable structural framework for its derivative materials to alleviate volume expansion and improves conductivity.

Zhang et al. fabricated Co0.85Se@CNFs nanoboxes by embedding Co-MOF into PAN for carbonization and selenization (Fig. 14a). The Co0.85Se@CNFs flexible films had a high active material load [157]. In addition, the composite nanobox structure had a large internal space, which was convenient for K+ diffusion and transmission, and slowed down the deformation of Co0.85Se in the electrochemical reaction. Using carbon encapsulation, Co0.85Se restricted to a certain area to prevent material crushing. Thus, after 100 cycles, this nanocomposite film displayed a capacity of 353 mAh g−1 at 0.2 A g−1, and with the current density increasing to 5 A g−1 a specific capacity of 166 mAh g−1 in PIBs. Park’s group demonstrated the preparation process of ZnSe@PCNF in Fig. 14b [158]. First, a diameter of 100 nm of ZIF-8/PAN as-obtained by the electrostatic spinning method and N-doped carbon film obtained by annealing treatment. After the selenization treatment, the ZnSe@N-doped carbon (ZnSe@NC) material formed and displayed an excellent capacity of 270 mA h g−1 at 0.5 A g−1. Using MOFs as precursors, a simple process uniformly confined metal selenide nanoparticles to carbon fibers. In addition, metal Zn transformed into ZnSe with excellent potassium-ion storage capacity through the selenization process. This structure provided enough active sites for K+ to store energy and shorten the diffusion path of ions. N-doped carbon materials helped to restrain the volume expansion of ZnSe nanocrystals in electrochemical reactions and increased their high conductivity.

Fig. 14
figure 14

The synthetic process of MOF-derived nanostructure in potassium-ion batteries. a Co0.85Se@CNFs; reprinted with permission from ref. [157], Copyright 2020, Elsevier; b ZnSe@PCNF composite; reprinted with permission from ref. [158], Copyright 2021, Elsevier

Lu’s group grew MoP ultrafine nanoparticles on N, P co-doped carbon nanofibers (MoP@NPCNFs) by electrospinning (Fig. 15a) [159]. N, P co-doped carbon nanofibers obtained through the carbonization of ZIF-8 and PAN electrospinning. After being immersed in ammonia molybdate and phosphorization process, MoP@NPCNFs carbon nanofibers are prepared. It displayed a noteworthy capacity of 320 mAh g−1 at 100 mA g−1, and 220 mAh g−1 at 2 A g−1. Thus, the strong interaction between MoP particles and NPCNFs effectively alleviated electrode fragmentation and agglomeration; the three-dimensional network structure increased the contact area between the material and the electrolyte, accelerated the reaction kinetics, and had excellent rate performance; Flexible membrane electrodes can be used in flexible energy storage devices. Wang’s group prepared the CoZn nanoparticle structure and mosaicked it in hollow carbon tubes (CoZn@HCT) [160]. By the coaxial electrospinning, co-doped ZIF-8/polymethyl methacrylate (PMMA) and polyacrylonitrile (PAN) were the inner axes and the outer layer (Fig. 15b). After annealing, CoZn nanoparticles coated in N-doped hollow carbon tubes. A cavity existed in the CoZn@HCT structure to accommodate the volume change of K metal during the cycle. The adsorption energy of CoZn and K+ was the lowest by DFT calculation, which improved the interaction between K and the electrode. The conductive CoZn@HCT skeleton can effectively reduce the local current density and adjust the uniform distribution of ions.

Fig. 15
figure 15

Preparation process diagram of MOF-derived nanostructure in potassium-ion batteries. a MoP@NPCNFs; reprinted with permission from ref. [159], Copyright 2020, Wiley-VC; b CoZn@HCT composite; reprinted with permission from ref. [160], Copyright 2021, Elsevier

Conclusion and Future Perspectives

MOFs and electrospinning are combined to give full play to the superiorities of both to flexibly design composite nanomaterials with controllable pore size and structure. Noteworthy, these kind of composite nanofibers have plenty of advantages on open pore volume and well-proportioned distribution of metal active sites. The one-dimensional fiber and three-dimensional carbon skeleton facilitate ion storage and transportation, which contribute to boosting the conductivity of the as-obtained composite. The well-designed composite nanomaterials are diffusely applied to energy storage. Although the combination of MOFs and electrospinning in the field of energy storage has made some research progress, it has both advantages and challenges in the main preparation methods.

  1. (1)

    The as-obtained MOF was first put into the electrospinning solution, and then, electrospinning was carried out to prepare the composite material. Noteworthy, the materials synthesized by this preparation route have the advantages of simple synthesis, excellent mechanical properties of composite films, and adjustable structures and functions. However, they also has certain limitations. (i) The presence of MOFs in composite films. MOFs have a significant effect on the conductivity of electrospinning precursors, and the addition of too many MOF crystals has a significant effect on conductivity. (ii) The size of the MOFs themselves needs to match the size of the electrospun fiber diameter, and the size of the MOFs is too large, which can easily lead to an uneven spinning effect.

  2. (2)

    MOF were grown on the electrospun polymer film or carbon-based composite films by hydrothermal or other methods. Compared with the previous direct blending electrospinning method, MOFs were grown directly on the surface of the polymer membrane, and the pores of the MOFs were not covered by polymer fibers. However, the limitations lie in the following: (i) The solvent of the electrospun solution easily affects the nucleation and growth process of the MOF. The selected electrospinning solution must not react with or corrode MOFs, to ensure that the structure of MOFs is not destroyed. High temperature tends to make MOFs undergo a phase transition. (ii) Long reaction times and excessive temperatures may affect the stability of the polymer film.

  3. (3)

    The MOF polymer composite film was annealed to obtain the derived-MOF composite fiber material. (i) After heat treatment, some metal ions of MOFs remain in the structure, resulting in an increase in the overall mass of the electrode material. (ii) Heteroatoms in organic ligands are easy to directly sublimate due to high temperature, so doping is reduced. Therefore, in subsequent studies, it is necessary to further explore in detail the time and temperature of MOF structure carbonization to maximize the retention of heteroatoms. (iii) The mechanical properties of some composite films are partially reduced after heat treatment.

In addition to solving the above problems, there are several ways to broaden the application of MOFs and classical spinning composites.

  1. (1)

    Development of new composite materials: More than 20,000 types of MOFs have been reported. Different MOFs have different pore structures and can accommodate electrolyte ions of different sizes in electrochemical reactions. The composite material is composed of different types of MOF fibers, which enriches the diversity of electrode materials. It is worth noting that some MOFs have the disadvantages of poor conductivity and low crystallinity [161, 162]. However, with the advent of DFT meters, MD simulations, machine learning, and the use of emerging computer techniques, the relationship between the structure of different MOFs and electrons can be predicted. These ways play a screening role in electrode materials. Through theoretical simulations, both orbital interactions of orbital overlap between metal ions and organic ligands and π–π superposition between efficient ligands and ligands enhance the conductivity of MOFs. Moreover, through the reasonable selection of organic ligands and metal ions, MOFs have as a small molecular weight and many redox active sites as possible, and finally obtain satisfactory capacity.

  2. (2)

    Increase of electrochemical capacity: Through material design, the electrochemical performance of MOF fiber membranes can be improved. (i) By changing the structure and shape of the nanofibers (special nanostructures, such as hollow structures, core–shell structures, brush-like structures), more energy storage active sites are added to facilitate the transport of electrons. (ii) Through interface engineering, defect construction, and electronic structure regulation, more interfaces are provided for electrochemical reactions.

  3. (3)

    Stability of electrode materials: The stability of MOFs in water, chemical solvents, and heat is still not satisfactory. As the pore size increases, the chemical and thermal stability usually deteriorate. During the electrochemical process of MOFs, whether there is fragmentation and phase transition is a question that needs to be considered in follow-up research. In situ characterization tests can be used to record whether the structure of the electrode material changes during the electrochemical reaction, proving the electrochemical reaction mechanism effectively. Similarly, precise control of the preparation of MOF derivative structure must be carried out. Therefore, the products produced after cyclic testing can be characterized in subsequent studies.

  4. (4)

    Optimizing the material configuration: MOFs, derived-MOFs materials, and electrospun carbon fiber structures have different energy storage mechanisms. At the initial stage of material design, the ratio of composite materials is optimized to maximize the electrochemical performance of composite materials. Moreover, usually electrospinning precursor solutions contain organic components, and composite materials containing organic components are directly used, and energy storage devices are polluted in the process of later recycling and processing. However, although no organic solution technically of melt electrospinning can avoid these problems, high system viscosity has an impact on nanofiber production.

  5. (5)

    Optimize production costs: Because of the high price of organic ligands, composite membrane materials have not been industrialized. Therefore, reducing the cost is also a big test of the application of composite membrane materials. Finally, industrializing electrospinning remains a challenge. At present, most electrospinning technologies use single-needle electrospinning. However, it focuses on the research and use of various multi-needle/multi-jet electrospinning methods, which is convenient for the rapid and large-scale preparation of complex fibers [163,164,165].

By solving the above problems, MOF composite fibers can achieve better performances in more fields. To sum up, MOFs and their derived materials are used to reasonably design and construct functional heterogeneous fiber structures through electrostatic spinning, and they may have uses in the energy storage. Finally, it is hoped that MOF and its derived nanofibers will be applied industrially from the laboratory as soon as possible.