Covalent Organic Frameworks for Ion Conduction

Abstract

Covalent organic frameworks (COFs) are an emerging class of crystalline porous materials constructed by the precise reticulation of organic building blocks through dynamic covalent bonds. Due to their facile preparation, easy modulation and functionalization, COFs have been considered as a powerful platform for engineering molecular devices in various fields, such as catalysis, energy storage and conversion, sensing, and bioengineering. Particularly, the highly ordered pores in the backbones with controlled pore size, topology, and interface property provide ideal pathways for the long-term ion conduction. Herein, we summarized the latest progress of COFs as solid ion conductors in energy devices, especially lithium-based batteries and fuel cells. The design strategies and performance in terms of transporting lithium ions, protons, and hydroxide anions are systematically illustrated. Finally, the current challenges and future research directions on COFs in energy devices are proposed, laying the groundwork for greater achievements for this emerging material.

1. Introduction

The development of society depends on the effective use of new energy, which relies on the innovation of novel energy storage technology. Environmentally friendly energy storage devices such as lithium (Li)-ion batteries have achieved great success in the fields of consumer electronics and electric vehicles due to their excellent energy density. However, the large application of Li-ion batteries is limited by the use of liquid electrolytes, which suffer from the potential risk of leakage, flammability, and narrow voltage windows [1]. In contrast, solid polymer electrolytes with greater thermal and chemical stability have been considered as the promising candidates for applications in commercial energy devices including rechargeable batteries and fuel cells [2, 3, 4]. Recent representative solid-polymer electrolytes mostly involve with high-molecular-weight fluoro-containing polymers such as Nafion and polyolefin-type membranes [5]. However, the sever capacity fade is inevitable for these membranes due to the low ionic conductivity, especially under some extreme conditions including but not limited to high temperature and low relative humidity [6, 7]. Thus, solid-state electrolytes with outstanding ionic conductivity and superior stability are in high requirement.

The porous materials with high porosity and facial functionality, such as metal-organic frameworks (MOFs), polymers of intrinsic microporosity (PIMs), porous aromatic frameworks (PAFs), and covalent organic frameworks (COFs), offer potential high performance as solid electrolytes for energy devices. However, MOFs tend to decompose during battery cycling due to the low thermal and electrochemical stabilities. However, PIMs would reshaped their ultra-micropores to meso- and macropores under alkaline conditions, which reduces their electrochemical performance. PAFs with strong carbon-carbon bonds have a stable framework in harsh acidic and alkaline environments. But the synthesis methodologies of PAFs are very limited. Compared with all these porous materials, COFs can overcome the abovementioned disadvantages and therefore act as suitable candidates for energy device applications. COFs are synthesized by the polycondensation reaction to form covalent bonds between lightweight-atom-containing monomers, such as carbon (C), oxygen (O), nitrogen (N), hydrogen (H), boron (B), etc. [8, 9]. Most COFs are synthesized under thermodynamic control to modulate the reversibility of bond formation and breakage. As a result, highly periodic networks with defined pore size, shape, topology, and crystalline lattice will be formed according to the building blocks. The porosity generated by the geometries of monomers as well as the stable covalent bonds makes COFs preferable platforms for solid-state ion conduction. To be specific, 2D or 3D nanochannels can be constructed by the pores of COFs, through which the ions can transport. For the ions such as lithium ions (Li⁺), protons (H⁺), and hydroxides (OH⁻), the conduction behavior mainly dominated by two mechanisms, which are hopping mechanism and vehicular mechanism [10, 11, 12]. For hopping mechanism, ions are preferable to hop between counter-charged adjacent sites with lower energy requirement to transport. While for vehicular mechanism, ions prefer bonding to some vehicular carriers such as H₂O (for H⁺) or anions (for Li⁺) to form large clusters, which needs higher energy to move. From this sense, the well-defined nanochannels of COFs, which could be further precisely installed ionic groups on the pore walls, provide an ideal pattern for ion conduction [13, 14]. In addition, the possibility to introduce extra charged species into the pores of COFs offers another opportunity for the high performance as ion conductors [15].

The superiority of COFs as high-performance ion conductors can also be described by the point of diffusion energy barrier of ionic migration [16]. For the liquid electrolytes, the charge carriers are surrounded by uniform and homogeneous solvents and thus can be quickly conducted by the exchange with solvating molecules, which generally produce high ionic conductivity. Thus, the diffusion energy barrier for ion conducting in liquid electrolytes can be considered flat (Figure 1a). However, the ion transporting in the solid-state conductors needs to overcome high energy barrier, which is related to the migration of charge carriers through the segmental motion of polymers or periodic crystalline space of inorganic solid (Figure 1b) [17, 18]. While for COFs, the charge species tend to migrate along the nanochannels due to the large free volume combined with the inside ionic sites, thus resulting in a lower diffusion energy barrier than that in typical solid-state conductor (Figure 1c). Thus, COFs are considered to be an excellent solid ion conductor.

Figure 1.

Schematics of the ion-conducting behavior and diffusion energy barrier in (a) liquid conductor, (b) typical solid conductor, and (c) COF conductor, respectively. Reproduced with permission from Ref. [16]. Copyright 2022 Wiley-VCH GmbH.

In this context, we focus on the application of COFs as solid-state polyelectrolytes in energy devices, especially lithium-based batteries and fuel cells. The design strategies, nanostructures, and performance in terms of transporting Li⁺, H⁺, and OH⁻ are systematically illustrated. Finally, the current challenges and future research directions for the utilization of COFs in energy devices are proposed.

2. COFs for lithium-ion conduction

Li-ion batteries have been considered as the mainstream devices in commercial portable electronics. However, traditional Li-ion batteries suffer from serious safety risk due to the utilization of liquid electrolytes consisting of Li salts and flammable organic solvents. The long-term performance of Li-ion batteries is also limited due to the narrow voltage windows of liquid electrolytes. It means that the charging process would induce the decomposition of organic solvents and result in obvious capacity fade [19]. Additionally, the separators that are essential in liquid-electrolyte batteries commonly exhibit low conductivity, also inducing the decrease in the performance of Li-ion batteries [20]. Thus, the development of polymer electrolytes for all-solid-state batteries can address the above safety problems. The third key parameter is the Li⁺ transference number (t_Li+), which represents the number of Li⁺ transferred per Faraday of charge during charge and discharge process. The values of t_Li+ for most polymer electrolytes are always lower than 0.5. These low t_Li+ values would generate a ion concentration gradient between cathode and anode, leading to the polarization and limiting the charging/discharging rate and lifetime of batteries [3, 21]. Based on this background, the state-of-the-art polymer electrolytes should possess high ionic conductivity, high t_Li+ values, and better electrochemical stability.

Due to the particular nanostructures and properties, COFs can act as an excellent Li⁺ conductor due to some intrinsic merits. Firstly, the well-defined open channels of COFs provide fast pathways for the conduction of ions by reduced diffusion energy barrier. Second, the organic nature endows COFs’ flexibility to introduce functional group to facilitate the dissociation of lithium salt and enhance the trapping of anions, which is beneficial to improve t_Li+ value. Thirdly, the superior electrochemical and thermal stabilities make COFs available in the practical application.

In 2016, Zhang and coworkers synthesized a novel type of ionic COFs (ICOFs) containing sp³ hybridized boron anions by the formation of spiroborate linkages [22]. After immobilization of Li⁺ into the nanochannels, the obtained ICOF-2 exhibited an ionic conductivity of 3.05 × 10⁻⁵ S/cm and an average t_Li+ value of 0.80 at room temperature, which was ascribed to the predesigned pathway for ion conduction. However, insertion of Li⁺ into the bare nanochannels of COF can only produce a limited ionic conductivity. Inspired by the poly(ethylene oxide) (PEO)-based electrolytes, which can solvate Li⁺ for fast ion conduction by the segmental motion, Jiang and coworkers firstly integrated the flexible oligo(ethylene oxide) chains on the pore walls of COFs [23]. Compared with the bare COF (Li⁺@TPB-DMTP-COF), the EO-modified COF (Li⁺@TPB-BMTP-COF) can exhibit an enhanced ionic conductivity and a lower energy barrier (Figure 2a). Furthermore, Horike et al. developed a bottom-up method to accumulate different concentration of glassy PEO moieties into the nanochannels of COF (Figure 2b) [24]. As a result, the material containing highest density accumulation of PEO (COF-PEO-9-Li) exhibited the superior ionic conductivity over 10⁻³ S/cm at 200°C. The similar trend was also demonstrated by branched PEO-functionalized COFs as solid conductors [26]. Recently, Horike used the similar side-chain engineering strategy to fabricate a gel-state COF (COF-Gel) with facile processability by implanting soft branched alkyl chains as internal plasticizers into the nanopores of COFs (Figure 2c) [25]. Benefiting from the gel-state morphology, the COF-Gel can be easily manufactured into gel electrolyte with controlled shape and thickness with enhanced ion conducting kinetics and reduced interfacial resistance, which showed a better performance in LiFePO₄-Li cell than the typical liquid electrolyte (Figure 2d).

Figure 2.

(a) Structural representation of Li⁺@TPB-DMTP-COF with bare pore walls and Li⁺@TPB-DMTP-COF with EO-modified pore walls. Reproduced with permission from Ref. [23]. Copyright 2018 American Chemical Society. (b) Structural representation of COFs containing different concentration of PEO moieties. Reproduced with permission from Ref. [24]. Copyright 2018 American Chemical Society. (c) Structural representation and the corresponding state of COF with alkyl chains. (d) Rate performance of COF-Gel in a LiFePO₄-Li cell. Reproduced with permission from Ref. [25]. Copyright 2021 Wiley-VCH GmbH.

The simple permeation of lithium salts into the nanopores of COFs usually results in a relatively low ionic conductivity and poor ion diffusion kinetics due to the closely associated ion pairs. While enhancing the binding interaction between the anions of lithium salt and COF backbones, the transfer of Li⁺ would be promoted. In 2018, Chen et al. synthesized a cationic COF incorporated with LiTFSI to improve the ionic conductivity [27]. Compared with the neutral framework, the cationic skeleton can generate stronger dielectric screening to split Li⁺ and the related anions, increasing the amount of free Li⁺ and resulting in an improved solid-state ionic conductivity up to 2.09 × 10⁻⁴ S/cm at 70°C (Figure 3a). Similarly, Feng and coworkers employed imidazolium monomer as the building block to construct cationic COF to enhance the trapping of counter ions of lithium salts and improve the lithium ionic conductivity (Figure 3b) [28]. The obtained Im-COF-TFSI possessed the ionic conductivity as high as 4.64 × 10⁻⁴ S/cm at 80°C and 4.04 × 10⁻³ S/cm at 150°C, respectively. Recently, Han’s group utilized the defects of COFs to introduce imidazolium groups onto the pore walls via the Schiff-base reaction [29]. After ion exchange with TFSI⁻, the resultant dCOF-ImTFSI-Xs not only had the 2D pathways for Li⁺ transport, but also contained the cationic moieties to promote the dissociation of lithium salts.

Figure 3.

(a) Structural representation of the cationic COF for binding with anions as Li⁺ conductors. Reproduced with permission from Ref. [27]. Copyright 2018 American Chemical Society. (b) Synthesis of cationic Im-COF-TFSI. Reproduced with permission from Ref. [28]. Copyright 2020 Royal Society of Chemistry.

Although COFs have been demonstrated as a new concept of solid-state Li⁺ conductors, most of the reported works are involved with the incorporation of lithium salts into the nanopores of COFs to facilitate ion transfer, thus failing to realize a single Li⁺ conduction, which performs the t_Li+ value close to unity (Figure 4a). To address this issue, Lee and coworkers presented a sulfonated COF (TpPa-SO₃Li) with anionic framework and well-defined directional channels (Figure 4b) [30]. As the sulfonates were covalently anchored to the pore walls, a real single Li⁺ conductor can be achieved. The obtained TpPa-SO₃Li showed the t_Li+ value as high as 0.9 at room temperature, thereby possessing a more stable Li plating/stripping on lithium anode than the previously reported COF-based ion conductors (Figure 4c). Recently, Loh et al. used a post-synthetic method to graft sulfonate groups on the pore wall of COFs and exfoliated COF powders to nanosheets to create better ion conduction channels because the repulsive forces between Li⁺ were stronger than the van der Waals interactions between frameworks (Figure 4d) [31]. As a result, the sulfonated COF nanosheets achieved the ionic conductivity of 0.9 × 10⁻⁵ S/cm at −40°C and 1.17 × 10⁻⁴ S/cm at 100°C with t_Li+ value of 0.92 when equipped in an all-solid-state battery, indicating the possibility to be practically applied under harsh conditions.

Figure 4.

(a) Conceptual illustrations of ion transport phenomena in the porous crystalline ion conductors: previous approaches (top) and this study (bottom). (b) Chemical structure of lithium sulfonated COF (TpPa-SO₃Li). (c) Galvanostatic Li plating/stripping profile of the Li/Li symmetric cell containing TpPa-SO₃Li. Reproduced with permission from Ref. [30]. Copyright 2019 American Chemical Society. (d) Schematic illustration describing the fabrication of all-solid-state organic Li-ion battery using lithiated COF nanosheets. Reproduced with permission from Ref. [31]. Copyright 2020 American Chemical Society.

3. COFs for proton conduction

Due to the high energy conversion efficiency, low emission, fuel flexibility, and mild operation accessibility, fuel cells are considered as electrochemical power plants, which convert chemical energy to electrical energy by the cost of specific fuels. Among the typical fuel cells, which differentiated by the type of electrolytes (phosphoric acid, proton exchange membrane, oxide, and alkaline), the proton exchange membrane fuel cell (PEMFC) has attracted the most attentions due to the relatively lower operation temperature [32]. Except for PEMFC, proton-conducting materials are also the key component for other electrochemical devices such as supercapacitors, proton sieving, proton transistors and hydrogen sensors [33]. Up to now, the universally utilized proton-conducting membranes are sulfonated polymers such as Nafion. Nafion can display proton conductivity as high as 10⁻¹ S/cm in fully hydrated state at a moderate temperature [5]. However, when the temperature exceeds 100°C, the proton conductivity would dramatically decrease due to the loss of water. Other polymer-based proton conducting materials such as polybenzimidazole (PBI)-based membranes can perform excellent properties in 150–200°C. While the heterogeneous random structures and the amorphous nature hinder the efficient analysis of conducting mechanism and determination of structure-property relationship at the molecular level [34]. In this regard, COFs with precise structural designability, synthetic controllability, and available functionality hold the promise to work as superior proton conductors.

Inspired by the structure of Nafion, sulfonated COFs would be the ideal candidates for proton conduction and fuel cell practical application. In 2016, Banerjee’s group firstly synthesized sulfonic-acid-based COFs (TpPa-SO₃H) by a Schiff-base reaction of 1,3,5-triformylphloroglucinol and 2,5-diaminobenzenesulfonic acid to form periodic intervals with free -SO₃H groups within COF backbones to promote the proton hopping along the hexagonal 1D channels [35]. The intrinsic proton conductivity of TpPa-SO₃H can reach 1.7 × 10⁻⁵ S/cm at 120°C under anhydrous condition. After shortly, Zhao et al. reported two sulfonated COFs, NUS-9(G) and NUS-10 (G), by liquid-assisted grinding strategy at room temperature (Figure 5a) [36]. The obtained NUS-9(G) and NUS-10 (G) exhibited hexagonal architecture and displayed eclipsed AA stacking layers (Figure 5b). Due to the pre-implanted free -SO₃H groups, NUS-9(G) showed a proton conductivity of 1.5 × 10⁻⁴ S/cm at room temperature and 33% relative humidity (RH), which was increased to 3.96 × 10⁻² S/cm at 97% RH. While for NUS-10 (G), which possessed twice as many free -SO₃H groups as NUS-9(G), the intrinsic proton conductivity increased to 2.8 × 10⁻⁴ S/cm at 33% RH and 3.96 × 10⁻² S/cm at 97% RH with long-term stability. Very recently, Zhu’s group used surface-initiated condensation polymerization to synthesize the same sulfonic COF TpPa-SO₃H (Figure 5c) [37]. Through the precise control of polymerization time, the thickness of SCOF layer can be tuned from 10 to 100 nm, overcoming the processable challenge of COFs. The obtained free-standing COF membrane exhibited a proton conductivity of 0.54 S/cm at 80°C under fully hydrated state.

Figure 5.

(a) Synthetic scheme of NUS-9(G) and NUS-10(G) via liquid-assisted grinding at room temperature. (b) Crystal structure and representation of the eclipsed AA stacking for NUS-9(G) and NUS-10(G), respectively. Reproduced with permission from Ref. [36]. Copyright 2016 American Chemical Society. (c) Preparation of the SCOF membrane grafted on silicon wafers and the molecular structures TpPa-SO₃H. Reproduced with permission from Ref. [37]. Copyright 2021 Wiley-VCH GmbH.

For these sulfonated COFs, the high water-assisted proton conductivity could be attributed to the presence of aligned -SO₃H groups on the pore walls of COFs, which not only enhance the adsorption of water but also facilitate the formation of hydrophilic domains to generate proton conducting pathways. At a high RH percentage, the proton transportation is mainly dominated by the hopping mechanism benefiting from the continuous hydrogen bonds between H₂O and -SO₃H groups. However, under low humidity condition, the significant decrease of proton conductivity would be still inevitable, which is similar to Nafion. Otherwise, the preinstallation of proton conducting groups onto the pore wall of COFs is sometimes difficult. Thus, incorporation of guest protonic species into the nanochannels of COFs would provide a more effective approach to improve the proton conductivity.

In one aspect, some proton donors can incorporate with COFs to trigger proton conductivity by donating more protons or facilitating the formation of hydrogen bonds. In 2016, Jiang and coworkers developed a highly robust COF, TPB-DMTP-COF, with hexagonally aligned, dense, mesoporous channels (Figure 6a) [38]. By loading the N-heterocyclic proton carriers, 1,2,4-triazole (trz) and imidazole (im), the anhydrous proton conductivity of trz@TPB-DMTP-COF and im@TPB-DMTP-COF can reach the maximum of 1.1 × 10⁻³ S/cm and 4.37 × 10⁻³ S/cm at 130°C, respectively. The activation energy calculation also demonstrated that protons were transported by hopping along the interconnected hydrogen bonding networks of proton carriers.

Figure 6.

(a) Chemical structure and hexagonal structure of TPB-DMTP-COF and graphic representation of 1,2,4-triazole and imidazole in the channels. Reproduced with permission from Ref. [38]. Copyright 2016 Nature Publishing Group. (b) Synthesis route of COF-Fx-H. (c) Illustration of proposed proton conducting mechanism. Reproduced with permission from Ref. [39]. Copyright 2020 American Chemical Society.

Besides, phosphoric acid (H₃PO₄) is also a good proton donors for the extrinsic incorporation with COFs [40]. In 2020, Horike’s group reported perfluoroalkyl-functionalized COFs (COF-Fx-H) with super hydrophobic well-defined 1D channels to accommodate a large amount of H₃PO₄ (Figure 6b) [39]. Due to the interactions between H₃PO₄ and the NH groups of framework as well as the fluorinated side chains, the guest H₃PO₄ could be anchored onto the pore walls through P∙O…H∙N, OH…N∙C, and O∙H…F∙C hydrogen bonding networks, which further generated an efficient proton conducting pathway (Figure 6c). After 62 wt% loading of H₃PO₄, the maximum anhydrous proton conductivity reached 4.2 × 10⁻² S/cm. Recently, Jiang’s group also designed polybenzimidazole COFs in conjunction with H₃PO₄ to achieve stable and ultrafast proton conduction over a wide range of temperature [41]. Due to the presence of imine linkage and the benzimidazole chains, H₃PO₄ could be tightly locked by the electrostatic and hydrogen binding interactions. More importantly, the N atom of benzimidazole moieties could be protonated by H₃PO₄ and release open H₂PO4⁻ anion. Thus the proton conduction would be facilitated by the activated proton networks. As a result, the H₃PO₄@TPB-DABI-COF realized a hydrous proton conductivity of 8.35 × 10⁻³ S/cm at 160°C.

Although great progress has been achieved for improving the proton conductivity by extrinsic incorporation with proton carriers such as acids and N-heterocycles, relatively less attention has been paid on the proton conducting property of ionic liquids (ILs) impregnated COFs. In 2021, Tang and coworkers firstly reported an IL impregnated sulfonic-acid-based COF (IL-COF-SO₃H), which further combined with silk nanofibrils (SNFs) to fabricate a composite membrane (Figure 7a) [42]. The electrostatic interactions between imidazolium anions and sulfonic acids promoted the deprotonation to release more protons and immobilized ILs. The uniform distribution of ILs in the channels of COF-SO₃H could provide a large amount of hopping sites for protons. Moreover, the hydrogen bonding networks between SNFs and IL-COF-SO₃H could provide additional proton conduction pathways. Particularly, the IL-COF-SO₃H@SNF-35, which loaded 35 wt% SNFs, acquired an ionic conductivity of 224 mS/cm at 90°C and 100% RH. Very recently, Yan’s group developed a protic ionic liquid (PIL), 1-methyl-3-(3-sulfopropyl) imidazolium hydrogensulphate ([PSMIm][HSO₄]) to incorporate with a high-density ∙SO₃H functionalized COF (TB-COF) for efficient anhydrous proton conduction (Figure 7b) [43]. As expected, the addition of PIL into the nanochannels of COFs can significant increase the ionic conductivity from 1.52 × 10⁻⁴ S/cm to 2.21 × 10⁻³ S/cm at 120°C due to the increase of hopping sites for protons.

Figure 7.

(a) Schematic of the synthesis and structure of IL-COF-SO₃H. Reproduced with permission from Ref. [42]. Copyright 2021 Elsevier B. V. (b) Structure and illustration of the proton transfer of PIL-TB-COF. Reproduced with permission from Ref. [43]. Copyright 2022 The Royal Society of Chemistry.

4. COFs for hydroxide anion conduction

Compared with PEMFC, alkaline fuel cells, which are operated on hydroxide anion transport, have attracted increasing attention due to the high energy density, rapid reaction kinetics, and low-cost catalyst [44]. As one of the critical components, the hydroxide conducting membrane affords the transfer of anions and determines the terminal electrochemical output. However, the high-performance hydroxide anion conduction is challenging because of its lower diffusion coefficient compared with protons [45]. Similar to proton conducting materials, the typical anion conducting membranes depend on polymer system, which can form percolated water channels via the microphase separation of hydrophobic/hydrophilic domains [46, 47, 48]. Multiple factors including the polarity of segments, distribution of charged moieties, and anion concentration have influence on the physical phase-separation process. Thus, precise control of the phase-segregated morphologies usually has of a large difficulty. Based on this background, COFs with structural tunability and functional pore surface build a powerful platform to achieve fast anion transport.

For example, Jiang and coworkers constructed an anion-surfaced channels for hydroxide anion conduction via supramolecular self-assembly while maintaining both the ordering topology and skeleton stability of COFs [49]. The precursor COF with ethynyl side groups was synthesized by the condensation of C₃-symmetric 1,3,5-tris(4-aminophenyl)benzene (TAPB) as knot and 2,5-bis(2-propynyloxy)terephthalaldehyde (BPTA) and 2,5-dimethoxyterephthalaldehyde at different molar ratios (DMTA) as linker (Figure 8a). Then an azide-imidazolium salt was introduced into the pores by the click reaction with the ethynyl sides. After anion exchange, the hydroxide anion conducting [OH⁻]₁₀₀-TPB-BPTA-COF was constructed (Figure 8a). The crystal structural analysis demonstrated that the imidazolium cations were extruded from pore walls and concentrated in the channel center aligning with OH⁻ at the end of cationic chains, thus creating a continuous anionic phase (Figure 8c). While for [OH⁻]₅₀-TPB-BPTA-COF in which half of the edge units were appended with ethynyl groups (Figure 8b), the hydroxide anion interface was not continuous due to the reduced anion density (Figure 8d). Consequently, the conductivity of [OH⁻]₁₀₀-TPB-BPTA-COF was 2–8 times higher than that of [OH⁻]₅₀-TPB-BPTA-COF.

Figure 8.

(a) Synthesis route of [OH⁻]₁₀₀-TPB-BPTA-COF. (b) Structure of [OH⁻]₅₀-TPB-BPTA-COF. Reconstructed structures of (c) [OH⁻]₁₀₀-TPB-BPTA-COF and (d) [OH⁻]₅₀-TPB-BPTA-COF (red O, blue N, gray C, white H). Reproduced with permission from Ref. [49]. Copyright 2021 American Chemical Society.

The poor processability of COFs generally produces insoluble powders, which dramatically limits their practical application as free-standing membranes. Thus, it is highly desirable to develop efficient methods to fabricate COF-based anion conducting membranes. Recently, Jiang’s group has made remarkable achievements on engineering COF membranes via interfacial polymerization strategy [50, 51, 52]. For instance, they used hydrazide units functionalized with quaternary ammonium (QA) groups bearing different length of alkyl chains and aldehyde units to construct four quaternized COFs (COF-QAs) (Figure 9a) [53]. The well-defined ordered nanochannels with aligned QA cations provided an ultrafast pathway for anion transport (Figure 9b). In order to realize the self-standing membrane with robust crystalline framework, a phase-transfer polymerization process, which involved phase transfer of 1,3,5-triformylbenzene to polymerize with QA-functionalized hydrazide in a mesitylene-water system (Figure 9c). Due to the slight solubility in water, the aldehyde units would gradually diffuse from organic phase to aqueous solution to react with hydrazides, resulting in a stable colloidal suspension. Upon solvent removal, the COF nanoplates could further be assembled into free-standing membrane with identical crystalline structures, which exhibited the hydroxide conductivity as high as 212 mS/cm at 80°C. Very recently, they developed six QA-functionalized COFs via the assembly of hydrazides and aldehyde precursors by interfacial polymerization to systematically elucidate the impact of aldehyde size, electrophilicity, and hydrophilicity on the synthesis process as well as the anion conducting property of COFs [54]. Particularly, more hydrophilic aldehydes were preferable to react with hydrazides in the aqueous solution rather than the interface region, which led to the tight membrane. Compared with the loose membranes, the anion conductivity could improve around 4–8 times.

Figure 9.

(a) Scheme for the synthesis of COF-QAs. (b) Schematic of anion transport through the 1D channel of COF-QAs. (c) Schematic of COF-QAs membrane fabrication process. Reproduced with permission from Ref. [53]. Copyright 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

5. COFs for other ion conduction

Among various electrical devices to date, sodium-ion batteries have gained considerable attention due to its low cost and sustainability. Similar with lithium-ion batteries, sodium-ion batteries also suffer from the easy formation of dendrite in traditional liquid electrolyte and have high desire to develop solid ion conductors. To tackle these bottlenecks, Sun and coworkers studied the first example of carboxylic acid sodium functionalized COF (NaOOC-COF) as quasi-solid-state electrolyte to accelerate the transporting of Na⁺ and simultaneously restrain the dendrite growth (Figure 10a). The covalently tethered carboxylic acid sodium groups in the pore wall of COFs provided sufficient content of Na⁺ and favorable nanostructures for Na⁺ migration. Benefiting from the well-defined ion channels, NaOOC-COF displayed an excellent conductivity of 2.68 × 10⁻⁴ S/cm at room temperature and high transference number of 0.9. Finally, NaOOC-COF devoted to durable cycling performance of Na plating/stripping and outstanding performance in solid-state battery [55].

Figure 10.

(a) The synthetic route of NaOOC-COF. Reproduced with permission from ref. 55. Copyright 2021 Elsevier Ltd. (b) Chemical structure of TpPa-SO₃Zn_0.5. Representative snapshots obtained from the MD simulations showing time-dependent ion distributions in (c) TpPa-SO₃Zn_0.5 and (d) LE. Zn²⁺: colored diversely for a clear representation of the movement, TpPa-SO₃⁻: gray, SO₄²⁻: green, H₂O: omitted for clarity. Reproduced with permission from Ref. [56]. Copyright 2020 The Royal Society of Chemistry.

Aqueous Zn-ion batteries are also a great promising energy storage system owing to the high energy density driven by multielectron redox (Zn^0/2+) and prominent safety supported by water-based electrolytes. However, the practical application has still been limited due to the lack of suitable electrolytes to ensure stable interface with electrodes. Recently, Lee and coworkers demonstrated for the first time to use COF-based single Zn²⁺ conductors, which can both secure interfacial stability with electrodes and exhibit competitive ionic conductivity [56]. A zinc sulfonated COF (TpPa-SO₃Zn_0.5, Figure 10b) with well-defined directional channels in which covalently anchored and delocalized sulfonates was designed to realize single Zn²⁺ conduction. From the molecular dynamics (MD) simulations, a significantly uniform Zn²⁺ flux was observed due to the anionic groups along the directional pores (Figure 10c). While in the control model of liquid electrolyte (LE), which is 2 M ZnSO₄ in H₂O, only randomly spread Zn²⁺ clusters can be observed due to the freely mobile SO₄²⁻ (Figure 10d). As a result, TpPa-SO₃Zn_0.5 enabled the Zn-MnO₂ cells to exhibit a long-term cycling performance.

6. Conclusion and outlook

In this chapter, we summarized the recent progress of COFs as solid-state ion conductors in energy devices, especially lithium-based batteries and fuel cells. As the emerging crystalline porous materials with controllable chemistry, tunable topology, and well-defined order channels, COFs exhibit a promising performance to conduct lithium ion, proton, and hydroxide anion. However, the development of COF-based solid ion conductors is still in its infancy, and many challenges remain to be issued.

Firstly, most ion-conducting COFs relate to ionic frameworks. Compared with neutral COFs, the examples of ionic COFs are still limited due to the more restricted synthesis conditions for crystallization and ionization. Although a series of covalent bonds have been successfully applied to construct COFs, only a few of linkages afford the formation of ionic COFs. To date, most ionic COFs are formed by the imine bonds. Thus, deep chemistry insight and novel synthetic approach to ionize COFs are in high demand. In addition, universal strategies to construct 3D ionic COFs, which are scientifically intriguing with unique properties, are also requiring since most COFs have 2D frameworks.

Secondly, most COFs are synthesized via a solvothermal method under harsh reaction conditions with powder products. Thus, the large-scale synthesis of COFs at industrial level is still challenging. To achieve more practical application, the large-scale synthesis with retaining the crystallinity and porosity of COFs is of critical significance. Moreover, the powder nature also hinders their application in electronic devices. Although some strategies such as interfacial polymerization can develop free-standing COF membranes to some extent, the limited mechanical property is always hard to meet the practical requirement. Therefore, efficient approaches to prepare COF membrane with good mechanical stability should be explored to enhance their practicality, especially in flexible electronic devices.

Thirdly, the development of COFs as ion conductors is in the initial state with most research interests focusing on the improvement of apparent performance via experimental investigations. To better clarify the structure-property relationship and guide the structural design, the theoretical simulations, which can provide more thorough insight on the ion transport mechanism in COFs, should be probed.

To sum up, COFs offer new opportunities for the solid ion conductors and exhibit tremendous advantages over other materials such as highly ordered pores, tailorable pore surface, tunable chemical composition, etc. Benefiting from the rapid development of experimental and theoretical tools, the electrochemical performance of COFs is expected to gain greater achievements.

Acknowledgments

The authors acknowledge financial support from the National Natural Science Foundation of China (21965011 and 21902092) and the Major Science and Technology Plan of Hainan Province (ZDKJ202016).

Conflict of interest

The authors declare no conflict of interest.