Engineering strategies of metal‐organic frameworks toward advanced batteries

Metal‐organic frameworks (MOFs) integrate several advantages such as adjustable pore sizes, large specific surface areas, controllable geometrical morphology, and feasible surface modification. Benefiting from these appealing merits, MOFs have recently been extensively explored in the field of advanced secondary batteries. However, a systematic summarization of the specific functional units that these materials can act as in batteries as well as their related design strategies to underline their functions has not been perceived to date. Motivated by this point, this review dedicates to the elucidation of diverse functions of MOFs for batteries, which involve the electrodes, separators, interface modifiers, and electrolytes. Particularly, the main engineering strategies based on the physical and chemical features to enable their enhanced performance have been highlighted for the individual functions. In addition, perspectives and possible research questions in the future development of these materials have also been outlined. This review captures such progress ranging from fundamental understanding and optimized protocols to multidirectional applications of MOF‐based materials in advanced secondary batteries.


| INTRODUCTION
The booming growth of portable electronic devices and electric vehicles is calling for high-performance energy storage media. 1 Currently, rechargeable batteries are the most popular energy storage systems owing to their favorable Coulombic efficiencies, alleviated self-discharge traits, and changeable energy densities, however, it is still a formidable challenge to concurrently realize the long cycle stability, fast power capability, and high energy density of batteries. [2][3][4] Since the performance of battery systems is highly dependent on their crucial components such as electrodes, separators, interface modifiers, and electrolytes, enormous efforts have been paid by research communities to the pursuit of adequate multifunctional materials for boosting the performance of these components. [5][6][7][8][9] Metal-organic frameworks (MOFs) were first defined in the 1990s by Yaghi′s group. 10 They are constructed by the coordination of the metal ions with organic ligands, 11,12 representing a unique class of organic-inorganic hybrid porous materials. Thanks to the adjustable pore sizes, large specific surface areas, controllable dimensions, easy functionalization, as well as green and safe preparation processes, 13 MOFs have been widely utilized as functional materials in various fields including hydrogen storage, 14,15 chemical sensor, 16,17 and drug delivery. 18,19 Recently, many researchers have deployed MOF-based materials, including pristine MOFs, MOF-involved composites, and MOF-derived inorganic materials, in energy storage devices (e.g., supercapacitors, [20][21][22] lithium-ion batteries [LIBs], [23][24][25][26] sodium-ion batteries [SIBs], [27][28][29][30] lithiumsulfur batteries [LSBs], 31,32 potassium-ion batteries [KIBs], 33,34 zinc-ion batteries [ZIBs], [35][36][37][38] fuel cells 39,40 ) and other fields. 41,42 Among them, the investigations on MOF-derived materials that are composed of carbonencapsulated transition metal oxides, phosphides, and chalcogenides, have achieved huge progress in energy storage applications. Compared to MOFs themselves, MOF derivatives usually present higher electrical conductivities, which are beneficial for charge transfer during the electrochemical reaction and thereby display remarkable performance. [43][44][45] Nevertheless, the posttreatment (e.g., high-temperature annealing or solvothermal reaction) toward MOFs to obtain corresponding MOF-derivatives unavoidably leads to the severe loss of pores and the significant decrease of surface areas, which degrade the intrinsic advantages of MOFs and limit some of their critical applications such as separators, interface modifiers, and electrolytes. For the pristine MOFs, the porous nature enables the facile electrolyte penetration and ion transportation, as well as the exposure of a large number of active sites within the structures, showing great potential for their application in batteries. In recent years, the research of MOFs in various applications such as electrode materials, electrolytes, separators, and interface modifiers has been reported successively. Nevertheless, like a double-edged sword, many pristine MOFs also suffer from poor electrical conductivity and structural instability, which hinder their practical utilization. Therefore, advanced engineering strategies for MOF materials have become a hot research area, where opportunities and challenges coexist.
On the basis of the above perspective, the up-to-date progress of the MOFs rather than their derivatives applied in battery systems is presented and discussed in this contribution. In the first part, we aim to introduce the advance of MOFs as electrodes as well as the strategies to boost their electrochemical performance. Particularly, the engineering strategies such as morphology and structural modification, elemental doping, design of conductive MOFs, and construction of MOF-based composites are emphasized. Following this part, the progress of MOFs for separators has been discussed. Remarkably, three approaches on how to effectively use the MOFs to optimize the performance of separators are highlighted, which are the design of MOFs as free-standing separators, coating pristine MOFs onto the conventional separators, and coating MOFs-based composites onto the conventional separators, respectively. Inspired by the successive progress achieved by using MOFs for electrodes and separators, the use of MOFs in interface modifiers and electrolytes is subsequently described. In terms of the former one, MOFs can be coated onto the anodes in a pristine and composite manner, both of which can improve the anode property. Regarding the latter one, three delicate methods including the creation of the MOFs host, fabrication of ionic liquids (ILs)-laden MOFs hybrids, and serving as the additives for the electrolytes with improved properties are discussed. The above-related regulation strategies have been schematically displayed in Figure 1. Finally, perspectives and research questions for future development are proposed. We hope this review could stimulate continuous innovations for designing advanced MOFs with more stable structures and better performance in various energy storage devices.

| ELECTRODE MATERIALS
MOFs have been used as electrode materials in batteries for a long time. As early as 2006, Li et al. 46 synthesized Zn 4 O (1,3,5-benzenetribenzoate) 2 and named it as MOF-177. When applied in LIBs, they found that MOF-177 electrodes exhibited a relatively high capacity, which marked the beginning of research on MOFs application in electrode materials. Following this observation, Prussian blue and its analogs, as one of the most typical MOFs, have also been further demonstrated to be able to show certain electrochemical performance in both LIBs 47 and SIBs. 48 The reason why MOFs can be served as electrode materials can be due to the following aspects. Firstly, abundant redox active sites in MOFs provide multielectron electrochemical reactions and exhibit high theoretical capacities. Secondly, the abundant micropores, mesopores, and regular channels in MOFs are beneficial to faster ion transfer, electrolyte permeation, and volume expansion laxation during the electrochemical processes, enabling long cycling stability and excellent rate capability. Besides, most defects in metal centers could also accelerate the charge transfer inside the frameworks. 26,49 For instance, An et al. 50 synthesized a flexible and wavy layered nickel-based MOF (Ni-MOF). The two-dimensional (2D) layered structure allowed the Li + insertion/extraction, while the organic ligands, 3,3′,5,5′-tetramethyl-4,4′-biphenyldiazepine (H 2 Me 4 bpz), were conducive to enhancing the flexibility and stability of the ripple-like structure. This Ni-MOF anode displayed a high capacity of 320 mAh g −1 at 50 mA g −1 . Inspired by this finding, Hu et al. 51 developed Mn-1,4-benzenedicarboxylate (Mn-MOF) framework as an anode material for LIBs. The Mn-MOF electrode delivered a high reversible capacity of 974 mAh g −1 after 100 cycles at 100 mAh g −1 , exhibiting good lithium storage properties. Further, just in the last 2 years, Mutahir et al. 52 designed a pillar layered MOF, Co (BDC)TED 0.5 , consisting of an aromatic organic ligand terephthalic acid (1,4-H 2 BDC) and pillar ligand triethylenediamine (TED). When investigated as an anode material in LIBs, the pillar layered framework not only provided the space for Li + diffusion but also mitigated the volume expansion during cycling, which exhibited a high capacity of 614 mAh g −1 at 0.2 A g −1 . It can be seen that MOFs are one of the most ideal electrode materials for batteries.

| Electrochemical mechanisms
The electrochemical mechanisms of MOFs as electrode materials are based on the intercalation/deintercalation reactions or conversion reactions.
As cathode materials, most MOFs undergo the reversible intercalation/deintercalation mechanism during the redox processes. Typically, MIL-101(Fe) 53 was investigated as cathode material of LIBs due to the existence of variable-valence metallic iron ions in the structure (Figure 2A). Fe 3+ /Fe 2+ redox peaks were observed in cyclic voltammogram (CV) curves ( Figure 2B) which served as electroactive sites for Li + storage. Prussian blue and its analogs are a type of promising category as cathode materials for LIBs, which deliver high specific capacities via a reversible twoelectron electrochemical reaction. Interestingly, the same reaction mechanism above for MOFs could also be observed in other battery systems. For example, Zhang et al. 54 Figure 2C). During the charging/discharging processes, the K + intercalated/deintercalated in the framework, and the Fe 3+ /Fe 2+ was the redox couple ( Figure 2D,E).
When used as anode materials, MOFs containing the redox-active ligands can provide insertion sites, and also can undergo reversible intercalation/deintercalation processes. Hu et al. 50 synthesized Mn-1,4-benzenedicarboxylate (Mn-1,4-BDC) MOF as the anode in LIBs. As shown in Figure 2F, during the charging process, Li ions reacted with the carboxylic groups and aromatic rings, providing a high theoretical capacity of 979 mAh g −1 ( Figure 2G,H). Similarly, Yin et al. 56 synthesized a bimetallic Ni-Mn-MOF using 3,3′,4,4′-biphenyltetracarboxylic acid (H 4 BPTC) as the organic ligands and applied it as the anode in LIBs. During the discharging process, Li + ions were adsorbed on the carboxyl groups to form Li-O bonds and six Li + ions were Reproduced with permission. 55 Copyright 2020, Chemistry Europe. CV, cyclic voltammogram; Mn-1,4-BDC, Mn-1,4-benzenedicarboxylate; MOF, metal-organic framework. embedded in the benzene rings. The units of this MOF structure have a total number of 16 Li + active sites, delivering a high specific capacity. In addition to the intercalation/deintercalation processes, researchers have also found certain conversion-type MOF anodes, in which the reversible transformation between MOFs and the corresponding metallic nanoparticles or lithium-metal alloys is expected. Moreover, it has been found that the metal ions can also be adsorbed to form M-S bonds and provide a high capacity. Recently, Wang et al. 55 , respectively. As shown in Figure 2I, The S-S bonds broke and combined with Li + to generate S-Li bonds during the discharging process. Meanwhile, the high-valence metal ions lost electrons and transformed into low-valence metal ions. The electrochemical reactions were highly reversible during the charging process.
Based on the above-illustrated examples, it is clear that MOFs have the ability to show great potential for application as electrode materials. However, many pristine MOFs suffer from inferior structural stability during the electrochemical process, resulting in poor cycling stability. On the other hand, the poor electrical conductivity of pristine MOFs displays large charge transfer resistances and limits the charge transfer kinetics. These shortcomings have seriously impacted the direct application of MOFs in batteries. To solve these problems, various regulation strategies, including morphology and structure modification, elemental doping, design of conductive MOF, and construction of MOFbased composites, have been put forward.

| Morphology and structure modification
Optimizing the morphologies and structures of electrode materials is beneficial to shorten the diffusion distance of electrons and ions, and to promote electrolyte infiltration, thus enhancing the electrochemical properties. Inspired by the synthesis process of MOFs, two entry points, that is, the mode of heating and the type of solvents, could be considered to realize the optimization of the morphologies and structures.
The formation of MOF particles includes the nucleation and growth processes of the crystals. Normally, rapid heating could guide nanoparticle formation by promoting the nucleation rate and improving the reversibility of the coordination bonds, which is hard to achieve via traditional hydrothermal methods. Current technologies that enable rapid heating mainly include microwave heating mode and ultrasonic heating mode. Microwave heating can heat quickly and uniformly. The crystal nucleus forms rapidly and grows into nanoparticles with uniform size. By controlling the microwave reaction conditions, the morphologies and structures of MOFs can be adjusted. On the basis of this point, Guo et al. 57 investigated the effect of microwave irradiation times on the morphological evolution of Fe-based MOFs. By controlling the irradiation time, the porosity of the material can be regulated, and finally, mesoporous yolkshell MOF octahedrons can be obtained, which performed outstanding lithium storage performance. Similarly, Desai et al. 24 synthesized a 3D Li-based 2-nitro terephthalate (Li-NTA) by microwave heating and applied it as the anode of LIBs, which achieved excellent electrochemical performance. More encouragingly, Skoda et al. 58 recently introduced a simple microwave-assisted synthesis to obtain the biphenyl-4,4′dicarboxylate-based cobalt MOF (Co-Bpdc) ( Figure 3A) and applied it as anode material in SIBs. At 0.2 C, the batteries exhibited a high initial capacity of 218 mAh g −1 . The capacity remained at 71.6% after 100 cycles and the Coulombic efficiency was nearly 100% ( Figure 3B). Ultrasonic heating can generate a large number of nanoscale local hot spots with instantaneous high temperature and high pressure, promoting nucleation and limiting the growth of the particle sizes. Li et al. 59 prepared Co (II) terephthalate-based MOF (u-CoOHtp) nanosheets with oxygen vacancies by a facile ultrasonic method ( Figure 3C). The u-CoOHtp showed a micronscale layered crystal structure, while the bulk MOF Co 2 (OH) 2 tp (b-CoOHtp) obtained by the conventional heating method was composed of densely stacked and randomly assembled ultrathin nanosheets. The reversible capacity of u-CoOHtp reached 372 mAh g −1 after 50 cycles with Coulombic efficiency approaching 100% at 50 mA g −1 ( Figure 3D). However, the b-CoOHtp delivered a reversible capacity of only 119 mAh g −1 after 50 cycles at 50 mA g −1 . The electrochemical property of u-Co-MOFs was significantly superior to the b-Co-MOFs.
On the other hand, similar to the hydrothermal/ solvothermal methods for MOFs synthesis, the solvents play an important role in constructing the MOFs structures. Yin et al. 56   porosity and specific surface area ( Figure 3E). While prepared with ethylene glycol, the Ni-Mn-BPTC-g-MOF exhibited irregular spheres that the complexation of organic ligands and the presence of ethylene glycol blocked the pore channels of the MOF, resulting in more inferior electrochemical properties than Ni-Mn-BPTC-e-MOF. The better Ni-Mn-BPTC-e-MOF showed a high initial discharged capacity of 1380 mAh g −1 at 100 mA g −1 .
After 200 cycles, the capacity still maintained 1269 mAh g −1 with 92% capacity retention and the Coulombic efficiency was nearly 100% ( Figure 3F).

| Elemental doping
Most pristine MOFs suffer from inferior intrinsic electrical conductivity and structural instability during the charging and discharging processes. Elemental doping is believed as one of the most significant and effective strategies, which can engineer intrinsic electrical conductivity, defects, crystal structures, and redox plateaus of materials to improve their electrochemical performance. In addition, elemental doping can well preserve the original morphological characteristics of the pristine MOFs.
Effective metal cation doping can usually suppress the crystalline phases and structure changes of MOFs during the redox processes. Generally, doping elements with a larger ionic radius in the metal sites can usually induce the expansion of the lattice frameworks, thus expanding the ion migration channels. stability. Zhu et al. 61 prepared aluminum (Al)-doped cobaltnickel double hydroxides (Al-CoNiDH) on carbon clothes ( Figure 4A). Al-doping relieved the crystal phase and structural changes during the electrochemical processes. When applied as anode materials for Ni-Zn batteries, the Al-CoNiDH retained a high reversible specific capacity of 160 mAh g −1 at 15 mA cm −2 after 5000 cycles ( Figure 4B). While undoped cobalt-nickel double hydroxide (Al-CoNiDH-0%) only retained a capacity of 55 mAh g −1 . To improve the reaction kinetics, the metal ions with larger ionic radii were doped into the structures to expand ion migration channels. Li et al. 62 partially replaced Ni and Co in NiCo-MOF with cerium (Ce), which stretched the framework ( Figure 4C) and provided large channels for faster ion migration and enhanced the rate capability. Additionally, the substitution of Ni/Co by Ce ensured structural stability during the redox reactions, thus improving the cycling stability. When the Ce dopant amount was 1% (NiCo-MOF-1%), the rate performance enhanced dramatically. The charge and discharge profiles of the NiCo-MOF and NiCo-MOF-1% were measured at current densities ranging from 2 to 20 A g −1 ( Figure 4D,E). At 2 A g −1 , the capacity of NiCo-MOF-1% was 286 mAh g −1 , maintaining 93% capacity at 20 A g −1 , while the NiCo-MOF only delivered a capacity of 215 mAh g −1 at 2 A g −1, and the capacity only retained 86% at 20 A g −1 .
Anion doping could also significantly improve the electrochemical properties of pristine MOFs. Fluorine (F) doping can increase the crystallinity of microporous materials and is beneficial to the formation of high crystalline MOFs phases. 64 On the other hand, F doping can enhance ions insertion/extraction reversibility and energy storage performance. 63 He et al. 63 chose F to partially replace the BDC linkers and synthesized the F-doped Mn-MOF. F has high electronegativity and strong interaction effect with Li + than the oxygen element. More importantly, the formed M-F bonds can promote the Li + transfer rate at the active material interfaces and protect the surface of the material from hygrogen fluoride attacking ( Figure 4F). The F-doped Mn-MOF anode in LIBs demonstrated a high reversible capacity of 2700 mAh g −1 at 100 mA g −1 and maintained 927 mAh g −1 with an excellent capacity retention of 78.5% after 100 cycles ( Figure 4G). Meanwhile, the F-doped Mn-MOF had a much smaller charge-transfer resistance as shown in impedance plots ( Figure 4H). Similarly, Wei et al. 65 synthesized a hollow-structured F-doped Co-MOF material. When applied as the anode material of LIBs, it performed an outstanding reversible capacity, excellent rate capability, and long cycling performance.

| Design of conductive MOFs
In practical applications, the electrochemical performance of pristine MOFs is greatly limited by their poor electrical conductivity. Researchers have explored the reasons and adjusted the MOFs to improve their electrical conductivity. Pathak et al. 66 found that the poor electrical conductivity mainly originated from the low overlaps between the d orbitals of metal ions and the p orbitals of oxygen. They demonstrated a strategic design in which the orbitals overlaps increased by replacing O atoms with S atoms, which formed metalsulfur (M-S) bonds and M-S planes integrated within the MOFs ( Figure 5A,B). This conductive MOF achieves high electrical conductivity of 10.96 S cm −1 . Through reasonable molecular structural design, the electrical conductivity of pristine MOF materials can be improved effectively. Metal-catecholate (M-CATs), composed of metal ions and 2,3,6,7,10,11-hexahydroxytriphenylene (HHTP) ligands, are a class of conductive MOFs. The high electrical conductivity mainly lowered the interactions between O bonds and p-p orbitals, where metal nodes and organic linkers serving as charge carriers enabled fully charged delocalization. Guo et al. 67 constructed 1D Ni-CAT nanorods as the anode materials of LIBs, which not only exhibited high Li + diffusion kinetic (D Li is 10 −9 -10 −10 cm 2 s −1 ) ( Figure 5C) but also delivered a high discharged capacity of 889 mAh g −1 at 0.2 A g −1 as well as excellent cycling stability ( Figure 5D). In addition, introducing conductive organic groups into the MOFs structures is also an effective strategy to improve the electrical conductivity. Weng et al. 68 choose tetrathiafulvalene derivatives (TTFs) to partly replace benzene ring linkers in the zinc-MOF (Zn-MOF) and obtained the 3D TTFs-Zn-MOF by controlling the contents of TTFs. TTFs are a kind of sulfur-rich conjugated organic molecules with 14 π-electrons and possess better electrical conductivity than benzene rings in this Zn-MOF ( Figure 5E). The 3D TTF-Zn-MOF anodes showed long-term cycling stability in LIBs, remaining a specific discharge capacity of 1117.4 mAh g −1 after 150 cycles at 200 mA g −1 ( Figure 5F).

| Construction of MOF-based composites
MOF-based composites with suitable structures and adjustable composition can incorporate the superiority of the pristine MOFs and other functional materials, by which the aforementioned issues of pristine MOFs as electrode materials can be effectively tackled. Generally, two situations should be considered for hybrid secondary materials with MOF to construct the composites as electrode materials. One is that the MOFs function as the real active electrodes, then they are used to hybrid with conductive materials, like carbon materials and conductive polymers, and so on, by which the electrical conductivity and electrochemical performance of pristine MOFs can be improved. In the second situation, MOFs function as the host with inert electrochemical performance, then their rich porosity and large specific surface areas can be utilized, by which they serve as the host to composite with or load active materials to construct the composites and to alleviate the volume expansion of the incorporated materials during electrochemical processes. The low electrical conductivity of pristine MOFs results in their poor electrochemical performance. To improve the electrical conductivity, pristine MOFs as the anode materials, are composited with conductive materials, such as carbon or conductive polymers. Carbon materials, such as reduced graphene (rGO) and carbon nanotubes (CNTs), have the advantages of 1D or 2D structures with high electrical conductivity and thermal stability. Compositing these carbon materials with MOFs, the conductivity and mechanical properties of pristine MOFs are remarkably improved. For example, Dong et al. 27  respectively. Both Co-MOF@rGo and Cd-MOF@rGO composites performed better electrochemical performance than pristine MOFs owing to the existence of the conductive rGO, in which the rGO greatly improved the electrical conductivity of the pristine MOFs. Gao et al. 69 uniformly covered Al(OH)(O 2 C-C 6 H 4 -CO 2 ) (Al-MOF) by rGO to form Al-MOF@rGO (AMG) composites ( Figure 6A). They found a pronounced structural change occurred in the AMG particles, which delivered an order-disorder phase transition during the lithiation/ delithiation processes. This transition led to the formation of more open channels, benefiting the diffusion and storage of Li ions. At 100 mA g −1 , the AMG electrode delivered a higher capacity of 400 mAh g −1 , while the pristine Al-MOF retained only 60 mAh g −1 after 100 cycles ( Figure 6B). Zhang et al. 70 synthesized zeolitic imidazolate framework-8@CNT (ZIF@CNT) composite by in situ growth method ( Figure 6C). The ZIF@CNT with hierarchical porous structure exhibited high areal, volumetric capacities, superior rate capability, and excellent cycling stability. When ZIF@CNT was used as anode material for LSBs, 85% capacity retention can be obtained after 300 cycles at 1 C ( Figure 6D). Inspired by the time evolution growth process of MOFs, Guo et al. 72 prepared [Ni 3 (HCOO) 6 ]/CNTs composite, which also exhibited enhanced performance than pristine MOF.
Integrating the MOFs host with conductive polymers can also improve electrical conductivity. Geng et al. 73 combined the advantages of ZIF-67 and polypyrrole (PPy) to construct a special hollow-structured ZIF-67-S-PPy composite for LSBs. The hollow structure can effectively alleviate the volumetric expansion during the redox reaction. The ZIF-67-S-PPy material exhibited excellent electrochemical performance with a high initial capacity of 1092.5 mAh g −1 at 0.1 C.
In addition to the conductive materials mentioned above, other functional materials also could be deployed to composite with MOFs. Wang et al. 74 hybridized the polyoxometalate (POM) with MOFs, which integrated the advantages of both POMs and MOFs and possessed excellent chemical and thermal stability. When applied as the anode material in LIBs, it maintained a high reversible capacity of 640 mAh g −1 after 100 cycles at 100 mA g −1 . Jin et al. 71 synthesized a 2D few-layer phosphorus/NiCo-MOF (BP/NiCo-MOF) composite ( Figure 6E). The BP/NiCo-MOF provided abundant redox active sites and achieved a high Li + storage capacity. The 2D porous nanostructure can provide fluent charge transport pathways and buffer the volume change during cycling, demonstrating excellent cycling stability and rate capability. This composite also exhibited excellent Na + storage properties with a high specific capacity of 853 mAh g −1 at 5 A g −1 and long cycling stability with 398 mAh g −1 capacity retained after 1000 cycles ( Figure 6F).
Owing to the favorable surface structure features of pristine MOFs, they can be selected to serve as the host to composite with other electrode materials, such as silicon (Si) and metal oxides/sulfides, by which the volume expansion of these electrode materials is effectively relieved during the repeated cycles. Si-based materials have high theoretical capacities and low working potentials, which are one of the most ideal anode materials for the nextgeneration LIBs. However, their large volume change and low conductivity obstruct their commercial applications. 75 Constructing Si with MOFs to form compounds is a common strategy to relieve the volume expansion and improve the electrochemical properties of the Si anodes. Han et al. 76 synthesized Si@MOF through a mechanochemical synthesis method enabling MOF materials to clad on the surface of the Si anodes ( Figure 7A). After simple pyrolysis, the obtained Si@MOF electrode exhibited excellent electrochemical properties with lithium storage capacity up to 1050 mAh g −1 and excellent cycling stability with about 99% capacity retention after 500 cycles at 200 mA g −1 ( Figure 7B). Similarly, Nazir et al. 77 designed Si nanoparticles and Ni 3 (2,3,6,7,10,11-hexaiminotriphenylene) 2  showed good cycling stability and rate performance ( Figure 7C).
Metal oxides and metal sulfides have higher theoretical capacities due to the conversion reaction mechanisms. 81,82 However, the large volume expansion during cycling leads to fast capacity fading. Integrating the above materials with MOFs can effectively alleviate the volume expansion and enhance their electrochemical performance. MOFs usually grow on the surface of these metal oxides/sulfides to form hollow structures. Another situation is that metal oxides/sulfides are attached to the surface of the MOFs. Sun et al. 78 selected MOFs as buffer layers to coat them on the surface of the Fe 2 O 3 and synthesized Fe 2 O 3 @MOF core-shell microspheres material ( Figure 7D). The Fe 2 O 3 @MOF anode material maintained a reversible capacity of 1002 mAh g −1 after 100 cycles at 100 mA g −1 in LIBs ( Figure 7E), which was higher than that of bare Fe 2 O 3 counterpart (696 mAh g −1 ). Moreover, it still delivered a high reversible capacity of 429 mAh g −1 after 70 cycles as the current density increased from 0.1 to 2 A g −1 . Wang et al. 79 synthesized CuS@Cu-1,3,5-benzenetricarboxylate (CuS@Cu-BTC) composites, which had large specific surface areas and high electrical conductivity. The synergistic effect between Cu-BTC and CuS can accommodate the volume change, achieve stress relaxation, and facilitate Li + transport, thus greatly suppressing the structural damage during the discharging and charging processes. The CuS@Cu-BTC composites presented an ultrahigh theoretical capacity of 1609 mAh g −1 and an excellent rate capacity of about 490 mAh g −1 at the current density of 1000 mA g −1 ( Figure 7F). Based on the findings from previous studies, Gao et al. 83 encased SnO 2 nanoparticles into Al-MOF, which was further wrapped by rGO, and ultimately obtained SnO 2 @MOF/rGO anode materials for LIBs. This modification strategy effectively improved the cycling stability of SnO 2 . The SnO 2 @MOF/rGO delivered a high reversible capacity of 800 mAh g −1 with a better capacity retention of 81% after 100 cycles at 100 mA g −1 , which was superior to the pure SnO 2 . By melt diffusion and chemical modification method, Ye et al. 80 synthesized a sandwiched ZIF-67@Se@MnO 2 composite as the cathode of lithium-selenium (Li-Se) batteries, in which selenium nanoparticles were attached on the surface of the ZIF-67 and MnO 2 was uniformly loaded on the surface of Se ( Figure 7G). This special structure can inhibit the shuttle effect and the dissolution of polyselenides. ZIF-67@Se@MnO 2 cathode retained a high capacity of 329 mAh g −1 at 1 C after 100 cycles and delivered stable reversible capacities of 273 and 232 mAh g −1 at 2 and 5 C, respectively ( Figure 7H).
Recent advances in various pristine MOFs and MOFsbased materials as electrode materials are systematically elucidated. For pristine MOFs, effective strategies like morphology and nanostructure modification, elemental doping, design of conductive MOFs, and creation of MOFs-based composites can improve the electrochemical properties of these materials. However, there are still many challenges waiting to be overcome. The electrochemical mechanisms of most MOF electrode materials still remain unclear and need further investigation. In future studies, applying more advanced characterization techniques is of great necessity to explore their structure and property changes during redox processes.

| SEPARATORS
As an indispensable part of the batteries, separators can avoid the internal short circuit and provide channels for ion transfer. Traditional polyolefin microporous separators are extensively used in commercial applications due to their low cost. However, these separators demonstrate low porosity, poor liquid electrolyte wettability, and weak heat resistibility, showing poor electrochemical performance and high safety risks. [84][85][86] The modified strategies of the separators mainly focus on porosity improvement and surface treatments. 87,88 With favorable porous structures, large specific areas, and abundant functional groups, MOF-based materials have been extensively investigated as separators for batteries. 89 Bai et al. 89 presented a MOF-based separator in LSBs. The MOF-based separator, as an ionic sieve, selectively sieved Li ions while suppressed the polysulfides efficiently. The LSBs with MOF-based separators achieved excellent cycling performance. The engineering strategies of MOF-based materials as separators include free-standing pristine MOFs as separators and coating pristine MOFs on the conventional separators (polypropylene [PP], glass fibers, etc.) as well as coating MOFs-based composites on the separators.

| Creation of free-standing MOFs separators
The pristine MOFs with intrinsic micro-porosity and chemical tunability can provide the contact areas to store liquid electrolytes and improve the ions migration. Additionally, free-standing pristine MOF-based separators are designed to inhibit the dendrite growth of the anodes, extending the overall operational life of the batteries. For instance, Wang et al. 90 synthesized a mixed-matrix membrane separator based on an anionic UiO-66-SO 3 Li MOF and poly(vinylidene fluoride) (PVDF) and applied it in LSBs ( Figure 8A). The tunnels in UiO-66-SO 3 Li facilitate the Li + transportation and Li deposition, achieving over 1000 h long cycling at 0.5, 2, and 5 mA cm −2 ( Figure 8B). The anions in MOFs demonstrate strong suppression of polysulfide shuttle, promote redox activity, and enhance the utilization of the sulfur cathode. Barbosa et al. 91 reported three pristine MOF-based separators with similar surface areas and topologies by solvent casting technique of thermal-induced phase separation. Compared to the conventional polymer separators, these pristine MOFsbased separators exhibited lower resistance, a higher discharge current density as well as a better rate capability ( Figure 8C-E).

| Coating pristine MOFs on a conventional separator
There are two synthetic strategies for coating pristine MOFs on conventional separators' surfaces, including mechanically coating the complete MOFs directly on separators and in situ chemically growing MOFs on separators. Chen et al. 92 designed a pore-network rearranged MOF (FJU-90) and coated it directly on Celgard 2400 by vacuum filtration. When applied it in LSBs, the large specific surface area and abundant catalytic sites of FJU-90 facilitated the Li ion conduction, inhibited polysulfide shuttle, and boosted the polysulfide catalytic conversion, thus enhancing the rate capability and cycling stability in high sulfur-loading cathodes ( Figure 9A). The cells with an FJU-90 separator can achieve a high areal capacity of 3.96 mAh cm −2 at a current density of 3.1 and 6.2 mA cm −2 with 78 wt.% high sulfur content ( Figure 9B). While, Liu et al. 93 designed a thermotolerant poly-m-phenyleneisophthalamide (PMIA)-based gel polymer electrolyte separator (P-PMIA@ZIF-8) by in situ growing ZIF-8 secondary nanostructures on the P-PMIA separator ( Figure 9C). With abundant multilevel pore structures, the modified P-PMIA@ZIF-8 separator exhibited high ionic conductivity. Moreover, the tensile strength and puncture force was up to 15 MPa and 0.95 N, respectively, and the structure could maintain without distortion at 200°C, showing high mechanical properties and excellent thermal stability. The LSBs with P-PMIA@ZIF-8 separator exhibited a high initial capacity of 1097.9 mAh g −1 at 80°C and retained 821.8 mAh g −1 after 100 cycles at 0.2 C ( Figure 9D). Moreover, under lean-electrolyte, with a high sulfur loading of 6.93 mg cm −2 and 200 mm thick lithium, the batteries still exhibited high capacity and long cycling stability.

| Coating MOF-based composites on the separator
Pristine MOF materials are usually nonconductive. They are often coupled with porous carbon materials, which then were coated on the surface of the separators. The porous carbon materials can act as secondary charge collectors to promote electrical conductivity inside the batteries. Jin et al. 94 combined Cerium-MOF (CSUST-1) with CNTs to form a CSUST-1/CNTs composite and used it as the separator coating material. Ce IV in MOFs has unique oxidative ability and the Ce IV /Ce III redox couple could perform as an effective mediator to further improve the redox catalytic activity. The additional CNTs can enhance the electrical conductivity of the separators and show a smaller interfacial resistance ( Figure 10A). The LSBs with this MOF composite-based separators exhibited a high initial specific capacity of 1468 mAh g −1 at 0.1 C and 541 mAh g −1 at 5 C, respectively ( Figure 10B). It also demonstrated excellent cycling stability with an ultralow capacity decay rate of 0.037% per cycle at 2 C. Except for the carbon materials, SiO 2 is also combined with MOFs to be applied as the coating material on separators. SiO 2 as the common separator coating material has some problems such as low thermal stability and wettability. For this, Suriyakumar et al. 95 successfully synthesized UiO-66-NH 2 @SiO 2 composites and coated them on commercial Celgard 2320 membranes ( Figure 10C). The modified membrane not only enhances the thermal stability and wettability of the electrolyte but also improves the ionic conductivity of the batteries. When applied in LSBs, this optimized separator can assist the battery to achieve long cycling stability with 98.5% capacity retention after 40 h ( Figure 10D). In brief, a deep understanding of the enhanced mechanisms is conducive to designing separators with excellent properties.

| ANODE INTERFACE MODIFIERS
The dendrite growth in the anodes of the batteries (LIBs, SIBs, ZIBs, etc.) causes safety hazards and hinders their practical applications. 96,97 In response, many strategies are applied to inhibit the growth of the dendrites. Among them, valid anode interface modification is one of the most common approaches. 98,99 Inspired by the excellent separation effect of the permselective MOFs at the angstrom scale, MOFs interlayers can be constructed directly on the anode surfaces, which then serves as effective artificial solid electrolyte interphase (SEI) layer to protect metal anodes and enhance the cycling stability of the batteries. For example, Han et al. 100 coated a MOF layer on the Si anode surface. This MOF layer acted as an efficient protective layer, reducing the volume expansion of Si anodes, decreasing the interface impedance but also improving the Li + diffusion, and thus greatly suppressing the dendrite growth of the anodes. Currently, two approaches based on MOFs can be considered to construct the anode interface modifiers, that is, pristine MOFs and MOFs-based composites.

| Pristine MOFs as interlayers
Pristine MOFs with a large number of organic functional groups are supposed to be effective anode interface modifiers. Qian et al. 101 coated a porous and robust MOF (MOF-199) on the Li metal anodes. The MOF-199 could act as the shield to effectively suppress the growth of the Li dendrite, rendering a homogeneous ion concentration and releasing excess SEI formation ( Figure 11A). Owing to the coating of the MOF-199 layer on the Li anodes, the batteries can achieve 350 cycles with a Coulombic efficiency above 97% at 10 mA cm −2 in LIBs. Thinner Li metal layers could be observed by SEM in the MOF-199-coated Li anode surface after Li deposition at 1 mAh cm −2 ( Figure 11B). Subsequently, Yang et al. 102 proposed a strategy of coating MOFs on the surface of separators to construct the supersaturated electrolyte front surface. The MOF layers rejected the large fraction of water by partial desolvation in advance. Then the Zn (H 2 O) 6 2+ SO 4 2− salt in the electrolyte was desolvated with water and bare Zn anodes were passivated, achieving homogeneous Zn deposition ( Figure 11C). The modified Zn anodes with MOF interlayers can maintain an ultralong lifespan of 3000 h at 0.5 mA cm −2 ( Figure 11D).

| MOF composites as interlayers
Some MOFs with poor stiffness as anode modifiers could not endure continuous volume expansion/shrinkage during cycling. In view of this point, MOFs are usually hybridized with polymer materials to enhance the stiffness of the anode interface modifiers. Fan et al. 103 prepared layers composed of combining Zn-MOF with polyvinyl alcohol (PVA) as the artificial SEI films which in situ grew on the surface of the Li metal anodes. The addition of PVA can modify the flexibility of the MOF. When applied in LIBs, these Zn-MOF/PVA artificial interlayers with high ion conductivity can allow Li + to pass through quickly and deposit uniformly on the Li anodes and inhibit the dendrite growth ( Figure 12A). Meanwhile, owing to the flexibility and rigidity of Zn-MOF/PVA anode interlayers, the Li anodes showed much smaller volume changes during the redox process. Therefore, the Li anodes with Zn-MOF/PVA films exhibited excellent electrochemical performance. At 3 mA cm −2 , the electrode can remain a capacity retention of 97.7% after 250 cycles ( Figure 12B). The presence of Zn dendrites is a major concern for the safety of ZIBs, and there are many works focusing on the modification of the Zn anode interfaces. Liu et al. 104 prepared the MOF-PVDF composite as the artificial protective layers on the Zn anodes to reconstruct the Zn/electrolyte interfaces. The hydrophilic MOF nanoparticles realized the nanolevel wetting effect and regulated the electrolyte flux on the Zn anodes ( Figure 12C). When this MOF-PVDF composite served as the anode interface modifier coated onto the surface of the Zn anodes, the assembled ZIBs delivered long cycling stability without Zn dendrite generation beyond 500 cycles at both 1 and 3 mA cm −1 (Figure 12D,E). As the anode interface modifiers, MOFs play a key role, in directing the shuttle and deposition of the metal ions. In addition, they also act as facilitators, promoting the penetration of the solvent molecules, and reducing the interfacial impedance between the electrolytes and anodes. Nowadays, electrolyte modification has attracted various attention in solving the safety problem of batteries. Solid state electrolytes (SSEs), 105,106 as physical barriers, can prevent metal anode dendrites from penetrating, avoid short-circuits and reduce capacity loss of the batteries, which have become a frontier research subject.
MOFs, owing to their rich porosity, periodic crystal structures, tunable surface polarities, and poor electrical conductivities, demonstrate great possibilities for being applied in advanced SSEs. 107 The periodic structures with rich porosity of MOFs could provide suitable pathways for ion diffusion, facilitating ion transport. Meanwhile, the tunable surface polarities of MOFs can be utilized to immobilize the anions and guide the uniform distribution of metal ions to construct dendrite-free solid-state batteries. 108 In an early report, Gerbaldi et al. 109 designed an aluminum (III)−1.3.5-benzenetricarboxylate (Al-BTC) and applied it as one MOF-laden nanocomposite polymer electrolyte. The existence of MOF enhanced the migration rate of Li + and improved the rate performance of the battery. Currently, the roles of MOFs in SSEs are discussed in three different systems: (1) pristine MOFs with internal mobile ions as the host of SSEs, (2) IL-laden MOF hybrids as SSEs, (3) MOFs as electrolyte additives.

| Pristine MOFs as the host
When pristine MOFs are applied as the host of SSEs, the corresponding target metal salts should be introduced within MOF hosts to enhance ion transportation. Bai et al. 108  natural angstrom-level pores in this MOF host could control the migration of anions, facilitate the homogeneous Li + flux, and enable stable Li anodes by inhibiting the growth of Li dendrite ( Figure 13A). The overgrown Li dendrites in pristine electrolytes were clearly observed after 120 h cycling at a specific current density of 10 mA cm −1 . However, there were no Li dendrites in MOF-modified electrolytes that can be observed after 1000 h cycling at 2.5 mA cm −1 , as well as 800 h cycling at 5 and 10 mA cm −1 , respectively ( Figure 13B). The formed SEI layers on the Li anodes were more complete in MOFs-based SSEs than in pristine electrolytes. The excellent electrolyte stability (over 2000 cycles at 5 C) contributed to the superior cycling stability of the batteries ( Figure 13C). In other representative works, Fischer et al. 110 loaded free Li + into the anionic Al-MOFs and synthesized a novel Li[Al(C 6 H 4 O 2 ) 2 (Al-Td-MOF-1) as the electrolyte for LIBs ( Figure 13D). The Al-Td-MOF-1 electrolyte featured a high Li + loading with high ionic conductivity of 5.7 × 10 −5 S cm −1 ( Figure 13E).

| ILs-laden MOFs hybrids
ILs-laden MOF hybrids can be applied as SSEs. ILs have the advantages of high solubility, high ionic conductivity, and high thermal/electrochemical stability, which have been regarded as fascinating type of green solvents. The ILs can be absorbed into the special porous MOFs to form SSEs without the risk of liquid leakage. In addition, the open channels in MOFs are the benefit to the interfacial contact between the electrolytes and electrodes to facilitate the transport of ions. Recently, Qi et al. 111  polarization, demonstrating unstable Li-electrolyte interface and the growth of Li dendrites. As the research on SIBs has matured, the application of ILs-laden MOFs hybrids as SSEs in SIBs has also received much attention. Yu et al. 112 prepared a sodium-based IL (Na-IL) with sodium bis(trifluoromethylsulfonyl)imide (NaTFSI) and 1n-butyl-1-methylpyrrolidinium bis(trifluoromethylsulfonyl) imide (Bmpyr-TFSI) salts and incorporated into the Zr 6 O 4 (OH) 4 (BDC) 6 (UIO-66) MOF to synthesize Na-IL laden sulfonated UIO-66(UIOSNa) SSE ( Figure 14C), which had high Na + conductivity of 3.6 × 10 −4 S cm −1 . The asassembled Na//IL-UIOSNa//Na 3 Ni 1.5 TeO 6 battery showed remarkable cycling stability with an excellent capacity retention of 76.5% after 100 cycles and high Coulombic efficiency of 99.0%-99.9% ( Figure 14D).

| MOFs as electrolyte additives
MOFs have also been applied as electrolyte additives in composite polymer electrolytes (CPEs) to increase metal salt dissociation and polymer segmental mobility, immobilize anions and guide the uniform distribution of ions. In 2013, Yuan and coworkers 113 incorporated Zn 4 O (1,4-benzendicarboxylate) MOF (MOF-5) in poly (ethylene oxide) (PEO) based CPE to improve its electrochemical properties. As electrolyte additives, the Lewis acidic sites on MOF-5 interacted with PEO chains and a lithium salt, hindering the crystallization of PEO, accelerating the movement chains as well as increasing the salt dissociation ( Figure 15A). In addition, MOF-5 with isotropic open 3D direction can absorb the trace solvent in porous filler and provide facilitated ion transportation channels. When the molar ratio of ethylene oxide to Li was 10:1, compared with SSEs without MOF-5 additive, the ionic conductivity was increased from 7.36 × 10 −6 to 3.16 × 10 −5 S cm −1 in the presence of MOF-5 ( Figure 15B). In recent, MOFs as electrolyte additives are still hot research topics. Huo et al. 114 proposed a novel cationic metal-organic framework (CMOF) as electrolyte additives in PEO-based (P) electrolytes. By electrostatic interaction, CMOF could immobilize the anions on the surface and guide the uniform distribution of Li + on the anodes, leading to a high Li + transference number of 0.72 ( Figure 15C). When applied this electrolyte in symmetric batteries, the Li/P@CMOF/Li cells worked continuously for 400 h at 0.1 mA cm −2 without significant generation of Li dendrites ( Figure 15D). Meanwhile, the LFP/P@CMOF/Li cells also exhibited excellent cycling stability beyond 400 h and high rate capabilities among current rates of 0.1-5 C in this electrolyte ( Figure 15E).
Presently, numerous works have focused on the application of MOFs in electrolytes. The pristine MOFs with internal mobile ions as the hosts in SSEs can facilitate the homogeneous ion flux and increase ion conductivity. IL-laden MOF hybrids as SSEs improve the mechanical strength and ion transfer numbers. Additionally, MOFs as electrolyte additives in CPEs effectively increase both the metal salt dissociation and the segmental mobility of polymers. Obviously, MOFs are now expected to make a great contribution to the development of SSEs.

| CONCLUSION AND PERSPECTIVES
MOFs bear unique structural advantages of adjustable pore sizes, large specific surface areas, and accessible metal sites, which have attracted great attention in battery systems. Herein, various recent applications and progress of MOFs in battery systems are summarized, including electrode materials, separators, interface modifiers, and applications in electrolytes.
The electrochemical property is strongly dependent on the electrical conductivity and structural stability of the electrode materials, while pristine MOFs are facing negative issues of poor electrical conductivity and structural instability. Thus, optimization strategies, including morphology and structure modification, elemental doping, and design of conductive MOFs and MOF composites, are proposed. The modified MOFs show high electrochemical properties. Meanwhile, due to their low electrical conductivity and high porosity, pristine MOFs can be applied as separators, and anode modification layers and are widely used in electrolytes, which are also discussed in this review. The traditional separators with poor wettability and mechanical properties exhibit slow ion migration and high safety risks. Pristine MOFs have been successfully applied in separators by designing freestanding MOFs separators, MOFs coating, and MOFs composites coating on the separators. The functional groups on MOF surfaces provide uniform nucleation sites for dendrite-free deposition of the anodes and the porous structures enhance the wettability and thermal stability of the electrolytes. Subsequently, the pristine MOFs and MOF composites can also be constructed directly on the anodes surface to apply as anode interface modifiers. As the interconnecting electrolyte reservoirs, MOF layers with high porosity enhance the wettability of the electrolyte and reduce the charge-transfer resistance to improve the reaction kinetics, but also act as protective layers successfully to avoid the dendrite deposition. Finally, the application of pristine MOFs in electrolytes is also highlighted. Pristine MOFs can be the host of SSEs, IL-laden MOF hybrids can be applied as SSEs directly and pristine MOFs can be used as electrolyte additives in CPEs. The rich porosity and rich functional groups in structural channels can assist to immobilize the anions and guide the uniform distribution of metal ions, solving the problem of poor ionic conductivity in SSEs.
Even though tremendous progress has been achieved using MOFs for applications in the battery system, there still exist many challenges that need continued effort to overcome. Herein, the following several bottlenecks and perspectives are outlined.
(1) Since the crystal size, morphology, and orientation significantly impact the electrical conductivity and stability of the pristine MOFs, it is of great importance to accurately design the crystal structures of MOFs at the atomic scale to effectively improve the intrinsic property of MOFs for batteries. In this regard, phase engineerings such as defect, strain, and amorphization can be considered as powerful strategies to modulate MOFs. (2) There are many different types of MOFs, the hybridization of two kinds of MOFs (e.g., ZIF and PBA) through ligands grafting or epitaxial growth is interesting to integrate their merits to enable their extensive applications in batteries. In addition, searching the novel MOFs with low cost, high conductivity, and favorable stability is also crucial to propelling their practical development in advanced batteries. (3) The underlying mechanisms for both the redox reactions of MOFs as the electrode materials and the reactions of MOFs in SSEs application still remain elusive. To this end, advanced characterization techniques are indispensable to investigate the fundamental mechanisms during the dynamic electrochemical processes, such as advanced in situ X-ray diffraction, X-ray absorption spectra, Raman, fourier transmission infrared spectroscopy, and in situ transmission electron microscope, and so on.   [59] Abbreviations: MOF, metal-organic framework; KIB, potassium-ion battery; LIB, lithium-ion battery; SIB, sodium-ion batter; ZIB, zinc-ion battery.
(4) Most MOFs are insulators and deliver poor conductive property. It is time-and cost-consuming to screen out the most suitable MOFs among the numerous coordination of metal ions and organic chains. To overcome this challenge, theoretical simulations including machine learning and molecular dynamics (MD) could be adopted to predict the promising modified strategies. (5) High ligands price, toxic organic reagents and tedious production processes increase the production cost of some MOFs and may be harmful to the environment, which is not conducive to the largescale application of MOFs. Further research on synthesis routes, reaction equipments, and feedstocks are highly necessary to achieve low-cost and green synthesis.
Overall, there is still a long way to go to conquer these key issues, however, the investigations in recent years in MOFs have built a strong foundation, and the pristine MOFs-based materials will have great commercialization prospects with continuous exploration (Tables 1-4).