Design strategies of covalent organic framework-based electrodes for supercapacitor application

Abstract

Supercapacitors (SCs) have been recognized as a promising electrochemical energy storage (EES) device, thanks to their high-power density, long lifespan, fast charge–discharge capability, and eco-friendliness. The breakthrough of electrode materials that determine the electrochemical performance of SCs is urgently desired. Covalent organic frameworks (COFs), an emerging and burgeoning class of crystalline porous polymeric materials, have been found to have huge potential for application in EES devices by virtue of their unique properties including atomically adjustable structures, robust and tunable skeletons, well-defined and open channels, high surface areas, etc. In this feature article, we aim at summarizing the design strategies of COF-based electrode materials for SCs based on the representative advances. The current challenges and future perspectives of COFs for SC application are highlighted as well.

---

1. Introduction

As the environmental issues caused by the over-utilization of fossil energy have brought serious challenges to ecosystem health and social sustainability, the disparity between the looming energy crisis and growing energy requirements is among the greatest problems faced by our society today. The efficient development and utilization of renewable energy sources (e.g., wind, solar, geothermal and tidal energy) that cannot be used directly, is highly desirable. Currently, electrochemical energy storage (EES) is represented as a key technique for the development and utilization of these discontinuous clean power sources.¹ Rechargeable supercapacitors (SCs) and batteries as widely utilized EES devices have received increasing attention thanks to their excellent electrochemical performance including high energy/power density, outstanding rate performance, long lifespan, and environmental friendliness.^2,3 It should be noted that SCs integrate the merits of high energy density of batteries and superior power density of traditional capacitors.

According to previous studies, SCs are traditionally classified into two categories, namely, electrochemical double layer capacitors (EDLCs) and pseudocapacitors, which store charge through electrostatic charge adsorption and reversible faradaic redox reactions at the electrolyte/electrode interface, respectively.^4–6 Electrode materials, serving as the core component of SCs, determine the key performances including specific capacitance, rate performance, energy density, cycle stability, etc. For EDLCs, the charge–discharge process is of physical nature, and special attention has been paid to the design of electrode materials with excellent conductivity, high wettability, nanostructures, porous structures, etc.⁷ Their electrodes are mainly dominated by carbonaceous materials, including porous activated carbon, graphene, carbon nanotubes, and so forth, which generally lead to superior rate performance and high cycling stability.^8–10 For pseudocapacitors, charge storage/separation is performed via reversible chemisorption and/or redox reactions, depending heavily on the chemical properties of the electrode materials.¹¹ Transition metal compounds,¹² conductive polymers,¹³ redox-active hybrids¹⁴ and porous organic materials¹⁵ have been demonstrated to be promising candidates for pseudocapacitive electrode materials, which usually showcase large capacitances and high energy densities. Actually, new hybrid supercapacitors that combine the electrode characteristics of EDLCs and pseudocapacitors have been developed recently, featuring one constituent as an EDLC type material and the other as a pseudocapacitive one.¹⁶ Overall, the energy storage mechanism of SC devices is determined from the electrode materials, and the currently reported electrode materials can be categorized into carbons, inorganics, organics and hybrids. As shown in Fig. 1, their representatives are listed as well.

Fig. 1 Classification of electrode materials for SC devices.

To meet the higher application requirements of higher power/energy density, superior capacitance, better rate performance and longer service life, the breakthrough of novel SC electrode materials with high performance is highly desired. Nowadays, organic porous materials, especially covalent organic frameworks (COFs), have attracted great attention as electrode materials by virtue of their high specific surface area, adjustable functionalities, convenient electron/ion transport and designable redox sites.^17–21 COFs are recognized as crystalline organic porous materials constructed entirely from light elements including C, H, O, N, B, etc., first reported by Yaghi and co-workers in 2005.²² Currently, one-dimensional (1D),²³ two-dimensional (2D)²⁴ and three-dimensional (3D)²⁵ COFs have been designed and synthesized in accordance with the principle of reticular chemistry.²⁶ The key step in constructing COFs is the formation of long-range order and porous structures, which relies heavily on the reversible equilibrium of dynamic covalent chemistry, manifesting that thermodynamically stable structures are assembled through “self-correction” between building blocks.^27–29 With the increasing demand for harsh environment applications, many COFs synthesized via irreversible covalent reactions have sprung up.^30–32 Nevertheless, COFs with the abovementioned reversible/irreversible linkages have similar requirements for the spatial configuration of building blocks, where the functional groups are obliged to undergo ring closure via covalent bonds. Besides, the formed extension layer generally needs to be stacked in an orderly and repetitive manner through mechanical interlocking and weak interactions such as hydrogen bonds, π–π interactions, etc. These structural features generally individualize COFs to have atomically precise structures, tailorable skeletons, open and regular channels, light weight, high surface areas, and facile built-in functions. Owing to their unique structural merits, COFs have been found to have tremendous potential in energy storage, gas storage and separation, optoelectronics, catalysis, sensing, drug delivery, etc.^33–36 Since the initial development of COFs was dominated by the reversible boronate ester linkage that usually has poor hydrolytic and oxidative stability, it was not until 2013 that the SC application of COFs was first realized via Schiff base chemistry.³⁷ Over the following 10 years, COFs have received growing attention for SC application. Nowadays, pristine bulk and nanostructured COFs, hybrids of COFs with other electrode materials (graphene, fullerene, conductive polymers, metal ions, etc.) and COF-based carbides are being developed for SC applications. Fig. 2 shows the number of articles published so far on COF-based materials for SC application and the corresponding citations (source: web of science, time: Dec. 31st, 2022). The growing trend in the number of articles published per year along with the outstanding number of citations indicate the prominence of next-generation COF-based SC technology.

Fig. 2 Bibliometric data on the number of papers published and the number of citations per year on COF-based SCs (the data were obtained by searching the following keywords on the web of science: (“covalent organic framework” OR “COF”) AND (“supercapacitor” OR “supercapacitors”)).

So far, many good review articles on COF electrode materials are available,^{3,15,17,20,38–40} which provides us an excellent platform for tracking the relevant progress. In this feature article, we aim to give a complete picture on the nature of the recent design strategies of COF-based electrodes for SC application. The purpose is first to clarify the relationship between the structural properties and electrochemical performance of COF-based electrode materials. Then, the design strategies for COF-based SC electrode materials with high performance are systematically proposed. Besides, the current challenges and future perspectives for the design of COF-based SC electrodes are highlighted as well.

2. Electrode requirements for SCs

According to previous studies, specific capacitance, power/energy density, rate performance, and cycling stability have been considered as the main performance indicators to evaluate whether SC devices can be practically applied.^41–43 Meanwhile, these main indicators are also the key technical parameters to guide the design of electrode materials. In this section, the core parameters are first described in detail, and the advantages of COF-based electrode materials for SCs oriented by the key parameters are then presented.

2.1 Key performance indicators of SCs

2.1.1 Specific capacitance. The capacitance (F) characterizes the capacity of SC devices to store charge, which is calculated from the ratio of the charge changes to the corresponding potential variations. The specific capacitance mainly includes areal capacitance (F cm⁻²), volumetric capacitance (F cm⁻³) and gravimetric capacitance (F g⁻¹), demonstrating the difference in the capability of charge storage. With respect to electrode materials, their chemical composition and physical morphology determine the specific capacitance of SCs. Generally, SCs show better capacitive performance when electrode materials have more available redox sites and/or a higher specific surface area. The considerable differences in specific capacitance can be derived from their charge storage mechanisms. Pseudocapacitors store energy through the reversible Faraday redox reaction at the electrode–electrolyte interface, usually showcasing 1–2 orders of magnitude higher specific capacitance than that of EDLCs storing energy via physical adsorption. It is noteworthy that test conditions have a great influence on specific capacitance as well. At a low current density/scan rate, the electrochemical reaction occurring on the surface of the electrode material is usually completed, where the electrochemical reaction rate is comparable to the charge transport rate, endowing SCs with a high specific capacitance.

2.1.2 Power/energy density. The power and energy density characterize the efficiency and amount of stored and deliverable electrical energy, being calculated in terms of W kg⁻¹ or W m⁻³, respectively. Compared with rechargeable batteries, EDLCs generally have higher power density and poorer energy density, while pseudocapacitors bridge the performance gap between batteries and EDLCs. All the factors that affect the capacitance of SCs have a subsequent effect on the energy/power density. The electrical conductivity, specific surface area, and accessible redox active sites of the electrode materials greatly affect the power/energy density. Generally, the better the conductivity, the faster the ion and electron transfer rate and the charge/discharge rate. The higher the surface area and more accessible redox active sites there are, the higher the power and energy density they deliver. Porous organic materials with high surface areas and abundant redox sites would offer abundant charging–discharging sites, thereby affording an excellent power density. Besides, the voltage window of the electrode materials needs to be considered as well. The larger the voltage window of the electrode material is, the higher the energy/power density would be.

2.1.3 Rate performance. The rate performance characterizes the capacitance, retention and restorability of the electrode material at different charge/discharge currents, which is strongly related to the reversibility of the reaction occurring on the surface of the material. Electrical conductivity is a central factor that determines the rate performance. Generally, if the magnitude of current response increases with the increasing scan rate (or current density) and a weak polarization is observed at high scan rates, the electrode would show good rate performance. A smaller resistance usually leads to reduced internal friction of electrode material, resulting in a better rate performance.

2.1.4 Cycling stability. Cycling stability of SCs is a very important indicator to estimate their service life, which is usually calculated from the retained capacitance after a certain working time or number of charge/discharge cycles. Many factors determine/influence the cycling stability of SCs. First, the charge–discharge procedures of electrode materials affect and even determine their cycling stability. Carbon materials usually possess an excellent cycle performance owing to the adsorption/desorption charge storage mechanism, which usually has little effect on their structural stability. Redox-active electrode materials such as metal compounds and organic materials that undergo charge/discharge cycles through reversible chemical reactions usually exhibit poor structural stability because of the irreversible distortion during charge–discharge cycling.⁴⁴ Second, the structural properties of electrode materials influence the cycling stability of SCs. For instance, active organic small molecules tend to dissolve in liquid electrolytes, and thus suffer from impaired cycling performance. Enhanced cycling stability can be achieved via anchoring them into a stable scaffold. Third, the influence of test conditions including current density, potential window, scan rate, test temperature, mass loading, electrolyte composition, etc., on cycling stability of SCs also needs to be considered.

2.2 Relationship between the structure of electrode materials and their SC performance

The electrochemical performance of SCs is closely related to the chemical structure of electrode materials, which might be notably different among various electrode materials. In summary, for COFs and other electrode materials, the specific surface area, type and accessibility of active sites, pore type and pore size, conjugation properties, crystalline structure and chemical linkage have significant effects on SC performance. The relationship between the structural properties of electrode materials and the electrochemical performance of SCs is summarized in Fig. 3.

Fig. 3 Relationship between the structural properties of electrode materials and the electrochemical performance of SCs.

2.3 Advantages of COF-based electrode materials for SCs

Compared to carbon materials, transition metal oxides, amorphous porous materials, and metal–organic frameworks, COFs as a type of stable covalently bonded polymer are fully predesignable and synthetically controllable in structure. Accordingly, the application-oriented controllable construction based on the structure–performance relationship makes COF materials extremely charming. The structural properties including the pore size and distribution, specific surface area, degree of conjugation and surface functionalization determine the electrochemical performance of COFs. Specifically, the structure–performance characteristics of COF-based electrode materials in SC applications can be described in the following three aspects.

(i) The covalent linkages and rigid skeletons of COFs endow them with outstanding thermal, chemical and electrochemical stability, and thus effectively suppressing the structural changes caused by ion deintercalation during the charge–discharge process and enhancing their cyclability.

(ii) The open and well-organized channels and high specific surface area facilitate the transfer of electrons/protons between COF electrodes and electrolytes, thereby improving the rate performance and power density.

(iii) The predesigned and abundant building blocks ensure that the COF platform shows great potential as electrodes for SCs: on one hand, tremendous electrochemically active organic moieties can be precisely organized into COF scaffolds, hurdling the obstacle of easy dissolution of organic molecules during the energy storage/conversion process and enabling high power density, remarkably enhanced energy density and outstanding cycle performance; on the other hand, feasible post-modification engineering allows electroactive units and other redox-active species (e.g., radicals, fullerenes, conducting polymers, metal oxides, etc.) to be decorated onto the walls or inside the channels of COFs, thereby endowing the resulting material with excellent capacitive properties.

3. Design strategies of COF-based electrodes for SCs

3.1 Introducing redox-active moieties into COFs

COFs are a versatile platform to tailor redox activities, thanks to their well-defined channels and walls, tunable skeletons, predesigned structures, and facile post-modification. In comparison to other electrode materials such as activated carbons, metal oxides, conducting polymers and amorphous porous organic materials, precise organization of electroactive units into COF skeletons is more feasible. Two approaches have been utilized to introduce redox-active moieties into COF skeletons: (i) designing redox-active building blocks for COF construction and (ii) introducing redox-active sites on the channel-walls of COF scaffolds linked by covalent bonds via post-modification. COFs with abundant redox-active sites enable not only high-speed mass transfer through their shape-preserving and orderly open channels, but also reversible high-throughput redox reactions through highly accessible redox-active centers on the channel walls, thus exhibiting large capacitance, excellent energy/power density, and high cyclability simultaneously. Herein, redox-active COFs are classified as the carbonyl/hydroxyl type, group 5A element-based type and free radical type based on the chemical structure of the redox moieties.

3.1.1 Carbonyl/hydroxyl as redox-active centers. The reversible transformation between benzoquinone and hydroquinone has been extensively employed for electrochemical energy storage processes, where an electron/proton transfer reaction is associated with the carbonyl group (Fig. 4a). Meanwhile, the electron/proton transfer reaction would be enhanced by the hydrogen-bond interaction (Fig. 4b). A considerable number of carbonyl/hydroxyl functional materials have served as advanced electrodes.^45–47 However, a significant decrease in capacitance and cycling stability of electrochemical energy storage devices caused by the dissolution of small carbonyl molecules into electrolyte solutions often occurs. Anchoring functional small molecules into polymer networks has been demonstrated to be an effective strategy. Based on the principle of network chemistry, a controllable distribution of redox-active carbonyl/hydroxyl groups into COF skeletons can be easily achieved, which is different from redox-active amorphous polymers. In view of the wide range of SC applications of carbonyl/hydroxyl type COF materials, it should be noted that discussion in this section is focused on the bulk form of carbonyl/hydroxyl type COFs, rather than thin films, nanosheets and hybrids with conductive polymers.

Fig. 4 Energy storage process of carbonyl/hydroxyl redox-active groups through reversible transformation (a) between benzoquinone and hydroquinone, and (b) between H-bond-stabilized hydroquinone and benzoquinone. The chemical structure diagram of (c) DAB-TFP COFs, (d) DAAQ-TFP COFs (reproduced with permission from ref. 37 Copyright 2013, American Chemical Society), (e) TpPa-R [TpPa-1, TpPa-(OH)₂, and TpPa-(OMe)₂], and (f) TpBD-R [TpBD, TpBD-(OH)₂, and TpBD-(OMe)₂] (reproduced with permission from ref. 48 Copyright 2017, American Chemical Society). The chemical structure diagram of (g) BD-Tp COFs, (h) 1KT-Tp COFs, (i) 2KT-Tp COFs, (j) 4KT-Tp COFs (Reproduced with permission from ref. 49 Copyright 2020, Chinese Chemical Society), and (k) TpOMe-DAQ (reproduced with permission from ref. 50 Copyright 2018, American Chemical Society).

For SC electrode application, both paraquinone- and orthoquinone-built-in COFs have been developed. Their representative chemical structures are shown in Fig. 4. Functional COFs with different carbonyl distribution densities were constructed as well. It is worth mentioning here that structural stability is always regarded as the most important parameter for the practical application of electrode materials. At the early stage of COF development, their construction was dominated by the reversible boronate ester linkage that exhibited poor hydrolytic and oxidative stability, which hindered the application of COFs to some extent. Until 2013, DAB-TFP COFs and DAAQ-TFP COFs were synthesized via combined Schiff-base condensation and enol–keto tautomerization (Fig. 4c and d) and first employed as electrode materials for SCs.³⁷ As a control, DAB-TFP COFs without redox-active groups showed a capacitance of only 15 ± 6 F g⁻¹ at 0.1 A g⁻¹ in 1 M H₂SO₄ electrolyte, while DAAQ-TFP COFs decorated with redox-active paraquinone centers exhibited a higher specific capacitance of 48 ± 10 F g⁻¹ and 83% capacitance retention over 5000 charge/discharge cycles. Under the same test conditions, the capacitance and capacitance retention of small DAAQ active units were 35 ± 7 F g⁻¹ and 60%, respectively. This pioneering work demonstrated that introduction of redox-active centers into COF skeletons is a feasible design strategy to enhance the electrochemical properties of SCs.

The electro-transformation from hydroxyl groups to benzoquinone was also employed in the EES of SCs.⁴⁸ In 2017, TpPa-(OH)₂ and TpBD-(OH)₂ (Fig. 4e and f) containing redox-active p-phenol functional moieties showcased specific capacitances of 396 and 86 F g⁻¹ at 2 mV s⁻¹ in 1 M phosphate buffer, respectively, in a three electrode configuration, while the initial specific capacitance values of contrastive TpPa-1, TpPa-(OMe)₂, TpBD-1, and TpBD-(OMe)₂ without phenolic OH groups were only 6, 13, 29, and 16 F g⁻¹, respectively, under the same test conditions, which demonstrated the key role of intralayer hydrogen-bonds in determining the SC performance. The specific capacitance of TpBD-(OH)₂ attributed to the irreversible redox transition of phenol/phenoxy radicals is merely one fifth of that of TpPa-(OH)₂ ascribed to the reversible proton-coupled electron transfer (2H⁺/2e⁻) between hydroquinone and benzoquinone (Fig. 4b) along with 43% accessibility to active sites of the COF scaffold, and this different mechanism also accounted for the significant difference in capacitance retention in a three electrode configuration: 66% after 10000 cycles for TpPa-(OH)₂ and only 2% after 1000 cycles for TpBD-(OH)₂ as the phenoxy radical would quickly decompose.

The content of redox-active moieties in the COF skeleton has a remarkable impact on the electrochemical performance of SCs, which was exemplified using orthoquinone-based COFs systematically investigated in 2020.⁴⁹ As shown in Fig. 4g–j, BD-Tp COFs, 1KT-Tp COFs, 2KT-Tp COFs and 4KT-Tp COFs containing none, one, two, and four ketone moieties in each linker units, respectively, were constructed via sequential Schiff-base condensation and enol–keto tautomerization. Compared with redox inactive groups in BD-Tp COFs or isolated carbonyl active sites in 1KT-Tp COFs, the orthoquinone moieties in 2KT-Tp COFs and 4KT-Tp COFs could not only provide denser redox-active sites but also enlarged conjugation upon reduction in the redox process. Therefore, 4KT-Tp COFs and 2KT-Tp COFs delivered higher capacitance values of 583 and 256 F g⁻¹ at 0.2 A g⁻¹ in 1 M H₂SO₄, respectively, than those of BD-Tp COF (20 F g⁻¹) and 1KT-Tp COF (61 F g⁻¹). Furthermore, all the three ketone containing COFs showed superb cycling stability with >92% capacitance retention over 20 000 cycles.

In 2018, reversible charge transfer between p-benzoquinone and hydroquinone for energy storage was reported.⁵⁰ As shown in Fig. 4k, the chemical structure of TpOMe-DAQ is quite similar to that of DAAQ-TFP COFs,³⁷ employing 2,4,6-trimethoxy-1,3,5-benzenetricarbaldehyde instead of 1,3,5-triformylpholoroglucinol as the trialdehyde building block. For TpOMe-DAQ, the interlayer hydrogen bond interaction between the N element of imine units (C=N) and the neighboring methoxyl group could effectively protect imine covalent linkages from hydrolysis, thus showing ultra-high cyclic performance with no significant capacitance degradation and Coulombic efficiency upon 10 000 charge–discharge runs. The gravimetric and areal capacitance of TpOMe-DAQ was 169 F g⁻¹ and 1600 mF cm⁻² at 3.3 mA cm⁻² in 3 M H₂SO₄, respectively. The superior areal capacitance was attributed to the thin sheet (200 μm) morphology. Besides, in the three electrode configuration, TpOMe-DAQ showcased excellent cyclic stability (>100 000) with a capacitance increase during cycling originating from the improved accessibility to redox-active centers. Even in a solid-state SC device, 65% capacitance retention over 50 000 cycles was achieved.

Although benzoquinone functionalized COFs have been widely utilized in SC energy storage devices, it should be noted that not all COFs containing benzoquinone exhibit pseudocapacitive behaviour. For instance, electrochemical measurements of DTFP-1 COFs did not exhibit characteristic redox curves, but rather charges were stored primarily through an EDLC mechanism.⁵¹ Generally, with extended π conjugation, inherent microporosity, and uniform channels, the vast majority of redox-active COFs store energy through both the reversible faradaic process and the EDLC mechanism, but the former dominates. Extending conjugation to improve conductivity and increasing the content and accessibility of redox moieties to amplify the faradaic redox effect are powerful approaches to improve the electrochemical performance of COF-based SCs.

3.1.2 Group 5A elements as redox-active centers. Introducing group 5A elements such as nitrogen, phosphorus, and antimony with lone-pair electrons into the carbon skeleton can help adjust the electronic structure, thus affecting the supercapacitive properties. So far, various heteroatom-modified COFs with redox activity have been constructed. In 2019, outstanding supercapacitor performance of TPT-DAHQ COFs (constructed via the condensation of 2,5-diaminohydroquinone dihydrochloride and 2,4,6-tris(4-formylphenyl)triazine) arising from heteroatoms (i.e., oxygen and nitrogen atoms) was observed, but the detailed electro-process during the charge–discharge cycles was not discussed.⁵² Recently, some novel COFs containing N-, P-, and Sb-based redox-active sites have been created. Among them, COFs with N-based groups as electrochemical active sites are the most widely reported, including pyridinic N,^53,54 triazinyl N^55,56 and triphenylamine N;⁵⁷ P and Sb were only found in triphenylphosphine⁵⁸ and triphenylantimony,⁵⁹ respectively. Their energy storage mechanisms are presented in Fig. 5.

Fig. 5 Energy storage mechanisms of (a) pyridinic N-, (b) triazinyl N-, and (c) triphenyl N/P/Sb-based redox-active moieties.

A large number of COFs containing N-based moieties have been constructed, but electrochemically active N centers are extremely rare. In 2016, a pyridine-based COF (TaPa-Py COF, Fig. 6a) was reported to exhibit electrochemical activities, which was the pioneering work on heteroatom-based COFs exploited in SCs.⁵³ CV measurement illustrated that the TaPa-Py COF showed a reversible faradaic process ascribed to the proton-coupled electron transfer of pyridine moieties originating from the intercalation of H⁺ into the COF (Fig. 5a). To demonstrate the pseudocapacitive behavior of pyridine moieties, the electrochemical behavior of the contrastive DAB-TFP COF based of the p-phenylenediamine linker instead of diaminopyridine (DAP) was also tested, and only an EDLC behavior was observed. Benefiting from a combination of pseudocapacitance provided by pyridine moieties and EDLC capacitance originating from the ordered porous structure, TaPa-Py COFs delivered a capacitance of 209 F g⁻¹, which was much higher than that of redox-inactive DAB-TFP COFs (98 F g⁻¹) and DAP-monomers (47 F g⁻¹) in a three electrode configuration. A similar advantage of TaPa-Py COFs in comparison with DAB-TFP COFs was also observed in a two electrode configuration. Due to a stable chemically linked crystalline structure, TaPa-Py COF based devices exhibited a high cyclability with 92% capacitance retention at 2 A g⁻¹ over 6000 charge–discharge cycles.

Fig. 6 Chemical structure representation of redox-active group 5A element N/P/Sb-based COFs.

In 2019, a series of pyridine-rich COFs (termed IISERP-COF10-12) were developed as SC electrode materials based on the pyridinal reversible faradaic process.⁵⁴ As shown in Fig. 7a–c, the three IISERP-COFs share high structural similarities, where the keto–enol tautomerism along with hydrogen bonding interactions led to high stability. The abundant pyridyl and triazine groups ensure reversible interactions with H⁺ from the acidic electrolyte, enabling rapid and immense charge storage. Generally, the high surface area contributes primarily toward the EDLC behavior, while redox-active functional sites store energy in a pseudocapacitive way. In a three electrode configuration, IISERP-COF 10 delivered the highest specific capacitance of 546 F g⁻¹, while IISERP-COF11 and IISERP-COF12 showed capacitance values of 310 and 400 F g⁻¹, respectively, which correspond to their high specific surface area (IISERP-COF10, 1233 m² g⁻¹; IISERP-COF11, 921 m² g⁻¹; and IISERP-COF12, 1067 m² g⁻¹). Meanwhile, IISERP-COF12 with a greater number of OH groups predominantly adopts the keto form that shows higher affinity toward a protic electrolyte than the enolic form, thus exhibiting the highest pseudocapacitive contribution ratio among the three IISERP-COFs (Fig. 7d–f). Their capacitance retention follows the order of IISERP-COF10 (70%) < IISERP-COF11 (75%) < IISERP-COF12 (82%), which was due to the increased number of keto–enol stabilized hydroxyl groups as well. In solid-state devices, IISERP-COF10 yielded a record high areal capacitance of 92 mF cm⁻² at 0.5 mA cm⁻² thus far among all the COF-based solid-state SCs.

Fig. 7 Chemical structure diagram of (a) IISERP-COF 10, (b) IISERP-COF 11, and (c) IISERP-COF 12, and (d–f) graphical representation of their capacitance contribution coming from the EDLC and pseudocapacitance in 1 M H₂SO₄. Reproduced with permission from ref. 54 Copyright 2019, American Chemical Society.

Triazine was found to be a novel type of redox active center as well. In 2019, pseudocapacitive characteristics of PDC-MA-COFs with a high nitrogen content of 47.87% (Fig. 6b) synthesized from piperazinedicarboxaldehyde (PDC) and melamine (MA) were reported.⁵⁵ The high nitrogen content and specific surface area, and abundant pores endow PDC-MA-COFs with outstanding electrochemical characteristics, while the interlayer hydrogen bonding provides it excellent stability. As illustrated in Fig. 5b, the triazine active moiety could experience a redox transition between aromatic and quinone structures. PDC-MA-COFs thus delivered a high specific capacitance of 335 and 94 F g⁻¹ at 1.0 A g⁻¹ in 6 M NaOH in three- and two-electrode configurations, respectively, and 88% capacitance retention after 20 000 charge–discharge cycles was achieved in the two electrode configuration.

In 2020, our group reported the pioneering work on phosphine-based COFs (Phos-COF-1)^60,61 for the controlled synthesis of metal nanoparticles. Meanwhile, its supercapacitive performance as an electrode was investigated where group 5A triphenyl moieties (i.e., triphenylphosphine) acted as redox centers. Phos-COF-1 delivered a decent capacitance of 100 F g⁻¹ at 1 A g⁻¹ in 3 M Na₂SO₄ with 90% capacitance retention over 5000 charge–discharge cycles in a three electrode configuration.⁵⁸ Later, we developed Sb-COFs with triphenylantimony active sites in 2022.⁵⁹ Sb-COFs demonstrate a similar PXRD pattern with Phos-COF-1, and the atomically structural difference is only that the Sb atoms in Sb-COFs take the place of P atoms in Phos-COF-1 (Fig. 6d). In electrochemical experiments, Sb-COFs showed a higher specific capacitance of 260 F g⁻¹ at 2 A g⁻¹, a high Coulombic efficiency of 99%, and 72% capacitance retention over 10 000 cycles in a three electrode configuration, manifesting its high energy transfer efficiency and good electrochemical stability. CV measurements demonstrated that Phos-COF-1 and Sb-COFs exhibited similar redox behavior, but the latter had a superior electrochemical performance, which was due to the higher triphenylphosphine site utilization and better conductivity associated with its micromorphology (see Section 3.2.1). The electrochemical properties of Phos-COF-1 were further improved by compositing with graphene aerogels, which will be summarized in Section 3.2.2.

The redox activity of triphenylamine-based COFs (TPA-COF) was also exploited for electrochemical energy storage of SCs in 2021.⁵⁷ The chemical structure of TPA-COFs is shown in Fig. 6c. TPA-COFs exhibited a uniform nanofiber-like structure along with a high specific surface area of 398.59 m² g⁻¹. Due to the presence of redox active triphenylamine units, the TPA-COF electrode exhibited pseudo-capacitance characteristics and showed a high specific capacitance of 263.1 F g⁻¹ at 0.1 A g⁻¹ in 1 M H₂SO₄ in a three electrode configuration. Its capacitance retention was 111% over 5000 charge–discharge cycles, which was attributed to the pore expansion originating from the shuttling of ions during the continuous charge–discharge process.

If hydrogen bonds are formed between heteroatoms and hydrogen atoms in the COF skeleton, electron transfer is expected to accelerate, thereby enabling materials with enhanced or even nascent electrochemical properties, which was recently realized in COF scaffolds containing heteroatoms N, S, etc.^48,62 As presented in section 3.1.1 (Fig. 4b), intramolecular hydrogen bonding promoted redox processes.⁴⁸ In 2020, the effect of intramolecular hydrogen bonds on electrochemical performance of PG-BBT COFs was demonstrated (Fig. 8a).⁶² PG-BBT COFs were constructed via sequential reversible Schiff base condensation and irreversible oxidation. When PG-BBT COFs underwent a redox process, each proton of the phenol group was transferred to the adjacent N atom to form a hydrogen bond, and therefore 6e⁻/6H⁺ transfer for each unit (Fig. 5b). The capacitance of PG-BBT was 724 F g⁻¹ at 1 A g⁻¹ along with 96% capacitance retention over 10 000 charge–discharge cycles, indicating outstanding electrochemical performance.

Fig. 8 (a) Chemical structure diagram, and (b) energy storage mechanism via the intramolecular hydrogen bond process of PG-BBT. Reproduced with permission from ref. 62 Copyright 2020, The Royal Society of Chemistry.

3.1.3 Free radicals as redox-active centers. Organic free radicals have the signature of unpaired electrons that can offer unique redox abilities. Decorating COF skeletons with organic radicals is an efficient strategy to construct redox active COF electrode materials. In 2015, an organic free radical (4-azido-2,2,6,6-tetramethyl-1-piperidinyloxy, TEMPO) was introduced into [HC = C]x-NiP-COFs (x = 50%, 100%) (Fig. 9).⁶³ TEMPO is a type of classical free radical that stores charges via reversible structural transformation between the neutral radical form and the oxidation state of the ammonium cation (Fig. 9a). [TEMPO]_50%-NiP-COFs and [TEMPO]_100%-NiP-COFs showed capacitance values of 124 and 167 F g⁻¹, respectively, at 0.1 A g⁻¹. The CV curves showed that the peak spacing between the reduction and oxidation waves of [TEMPO]_50%-NiP-COFs and [TEMPO]_100%-NiP-COFs is 155 and 186 mV, respectively, suggesting that the former has a faster charge–discharge cycle, which resulted from the denser free radicals. Moreover, the free radical type COFs do not produce individual anionic and cationic free radicals, and their molecular structures do not change during cycling, so they therefore usually have excellent chemical stability, reversibility and fast charge and discharge performance. In this case, both [TEMPO]-NiP-COFs exhibited good stability over 100 charge–discharge cycles owing to the covalent linkage between TEMPO and the COF skeleton. On the other hand, the high chemical activity of radicals makes the construction of free radical type COFs extremely challenging. Although many radical-bearing polymers have been developed as electrode materials for SCs,⁶⁴ radical-bearing COF⁶⁵ electrode materials have been rarely reported. Nevertheless, the synthetic strategy of [TEMPO]-NiP-COFs opens up new possibilities to develop novel electrochemically active COF electrode materials from traditionally inert COF scaffolds.

Fig. 9 (a) Energy storage mechanism of TEMPO based on a one-electron redox reaction, and the chemical structure diagram of (b) [TEMPO]_50%-NiP-COFs and (c) [TEMPO]_100%-NiP-COFs. Reproduced with permission from ref. 63 Copyright 2015, Wiley-VCH.

3.2 Improving the accessibility of redox-active moieties

Thanks to the outstanding performance-oriented designability, ordered and well-established channels and easily functionalized walls, COFs have exhibited tremendous application potential in many fields. However, the poor processability, low electrochemical accessibility to the active sites and inferior electrical conductivity severely limit their practical applications. For SC electrodes, the long-range columnar channels of bulk COFs formed by stacked atomic layers and edge defects usually have a negative influence on efficient ion transport and high accessibility to the redox-active sites. In this section, the strategy via morphology control including nanosheets standing on a substrate, free standing thin films, and nanostructures for enhancing the accessibility of redox-active moieties will be summarized and analyzed.

Nanosheet-like COFs with a thickness of single- or few-atom layers and nanostructured COFs with various shapes can shorten the length of ion transport channels and increase the exposure rate of active sites, thus enhancing capacitive properties. Pristine powder COFs usually need to be mixed with carbon black and a binder before studying their electrochemical performance. As shown in Fig. 10a, the grain boundaries and random orientation remain large obstacles to achieve high conductivity and fast mass transfer. The first COF (DAAQ-TFP COF) developed as a SC electrode delivered a low capacitance of 48 ± 10 F g⁻¹, with only ∼3.0% utilization rate of redox-active DAAQ sites, which was due to random orientation of microcrystalline DAAQ-TFP COF powder.³⁷ In 2015, Dichtel and co-workers demonstrated that the oriented COF thin films could improve the accessibility to redox-active sites and shorten the ion transport path, thus enhancing the electrochemical performance.⁶⁶ DAAQ-TFP COF thin films were prepared on a Au substrate electrode (Fig. 10b). The thickness could be modulated by varying the initial monomer concentration. A much higher utilization efficiency (80–99%) of DAAQ moieties in thin films was obtained in comparison to the randomly oriented polycrystalline powder (∼3.0%). Consequently, the oriented COF thin films exhibited a 4-fold increase in areal capacitance in comparison to the powder COF electrode. 3D graphene has also been employed as a support for the growth of electrochemically active COF films.⁶⁷ In 2015, a π-conjugated surface COF (COF_DAAQ-BTA) with anthraquinone moieties was incorporated into 3D graphene through a noncovalent way. The layered COF_DAAQ-BTA-3DG hybrid delivered an areal capacitance of 31.7 mF cm⁻² with an enhanced accessibility of 13.45% (vs. 2.5% for 2D COF powder) of the DAAQ moieties in 1 M KOH.

Fig. 10 Schematic diagram of the existing morphology of (a) microcrystalline COF powder and (b) oriented COF thin film on the electrode surface. Reproduced with permission from ref. 66 Copyright 2015, American Chemical Society.

Free-standing thin films were also prepared for SC electrodes. So far, substantial efforts have been devoted to DAAQ-functioned COFs as redox-active materials. Very representatively, the TpOMe-DAQ COF (chemical structure shown in Fig. 4k) was constructed via a p-toluenesulfonic acid (PTSA) catalyzed mechanochemical grinding approach (Fig. 11a).⁵⁰ It presented an independent electrochemical behavior without any substrate, support, or additives (Fig. 11b), and delivered a gravimetric capacitance of 169 F g⁻¹ and an areal capacitance of 1600 mF cm⁻², thanks to the thin flake morphology (∼200 μm). Slightly earlier, using a similar film-forming strategy, another flexible DAAQ-based COF electrode was prepared, where two different linkers including π-electron-rich anthracene (Da) and redox active anthraquinone (Dq) were simultaneously constructed into a β-ketoenamine linked COF.⁶⁸ The effect of the ratio of the two employed linkers on the electrochemical and mechanical properties of COFs was further investigated. The larger the molar ratio of Dq: Da in diamine building blocks, the better the capacitive performance and the lower the mechanical strength of the COF thin film.

Fig. 11 (a) Fabrication procedure diagram of the TpOMe-QAQ COF thin film and its flexible nature, and (b) free-standing TpOMe-QAQ COFs as the working electrode without any support or additives. Reproduced with permission from ref. 50 Copyright 2018, American Chemical Society.

Constructing nanostructured COFs is a useful way to enhance the accessibility of redox-active groups, although its development is still in infancy. Currently, the growth and morphology of COFs are mainly controlled by the thermodynamics of the reaction system or the added template agent. In terms of SC electrode materials, the study on COF nanostructures is still relatively rare. As summarized in Section 3.1.2, the chemical structures of Sb-COFs and Phos-COF-1 are slightly different. Their pore size distribution, specific surface area, and electrochemical behavior were all quite similar as well. However, the accessibility to redox-active sites (42% vs. 12%) and specific capacitance (260 F g⁻¹ at 2 A g⁻¹vs. 100 F g⁻¹ at 1 A g⁻¹) differed significantly, which was largely due to the COF particle size.^58,59 Sb-COFs had a nanostructured morphology with an average diameter of 100 ± 20 nm and a length of ∼200 nm statistically derived from SEM and TEM images (Fig. 12a and b), while Phos-COF-1 existed in an aggregated bulk formed by a great deal of particles with an average size of ∼10 μm (Fig. 12c and d). Besides, nanostructured Sb-COFs with intrinsic opening channels could enable effective electrolyte infiltration, and the small particle size can shorten the transport and diffusion path lengths for electrolyte ions, thus providing fast kinetics and high charge–discharge rates. Moreover, a stable crystalline structure could endure strain/stress in the electrode matrix, ensuring enhanced cyclability.⁶⁹ The superior capacitive performance of Sb-COFs indicated that the nanostructured COFs possess great potential in SC applications.

Fig. 12 (a) SEM and (b) TEM images of Sb-COFs. Reproduced with permission from ref. 59 Copyright 2022, The Royal Society of Chemistry. (c) SEM and (d) TEM images of Phos-COF-1. Reproduced with permission from ref. 58 Copyright 2021, Springer.

3.3 Enhancing the electrical conductivity

Combining high surface areas and notably abundant redox-active sites, COFs can synergize with energy storage principles of EDLCs and pseudocapacitors in SC application. However, their electrochemical performance is severely limited by the low charge/discharge rates, which are fundamentally caused by the inferior electrical conductivity of COF architectures. To overcome the obstacle, compositing conductive additives with COFs was implemented. It is noteworthy that the enhanced electrical conductivity would be beneficial for a better accessibility to redox-active sites as well. The most classical electrochemically active DAAQ-COF has been intensively studied as a model pristine COF material.

In 2016, encapsulating poly(3,4-ethylenedioxythiophene) (PEDOT), the most widely studied conducting polymer with excellent conductivity and high stability, inside the nanochannels of thick DAAQ-COF films to accelerate the charge–discharge rate was investigated.⁷⁰ As shown in Fig. 13a, employing 3,4-ethylenedioxythiophene as a monomer, PEDOT was in situ electropolymerized inside the pores of COF films that was grown on a Au substrate. The preparation procedure of COF films was similar to the abovementioned work.⁶⁶ The obtained PEDOT-modified DAAQ-TFP film exhibited 10-fold higher current response, a 12-fold increase in volumetric power density and 30-fold higher volumetric energy density than those of the pristine DAAQ-COF film. Besides, the charge transfer rate of the PEDOT-modified DAAQ-TFP film could be up to 1600 C without deteriorating the performance. However, the PEDOT chains that penetrated deep into the channels of the DAAQ-TFP significantly weakened the intrinsic porous property, with an ultralow BET surface area of 6 cm² cm⁻² after nine electropolymerization cycles, which was detrimental to the surface capacitance. In 2019, via in situ solid-state polymerization, PEDOT linear chains were controlled to grow inside the DAAQ-TFP channels and retained a high porosity with a BET specific surface area of 131 m² g⁻¹.⁷¹ The PEDOT@AQ-COF electrode exhibited an ultrarapid charge–discharge rate (998 F g⁻¹ at 500 A g⁻¹), an superb conductivity of 1.1 S cm⁻¹, as well as a high capacitance of 1663 F g⁻¹ at 1 A g⁻¹. The above two studies demonstrated that the engineered introduction of conducting polymers into the cavities of COFs can efficiently improve the conductivity of COF-based electrodes for high performance SCs.

Fig. 13 Synthetic scheme of (a) PEDOT-modified DAAQ-TFP films, and (b) PEDOT@DAAQ-TFP powder. Reproduced with permission from ref. 70 Copyright 2016 and ref. 71 Copyright 2019, American Chemical Society, respectively; (c) Preparation procedure of DAAQ-TFP/GA. Reproduced with permission from ref. 72 Copyright 2021, the Royal Society of Chemistry.

Besides conducting polymers, 3D graphene aerogels (GAs) have also attracted a great deal of attention in energy storage application due to their excellent electrical conductivity and unique mechanics characteristics. Currently, a few COF/graphene composites have been fabricated. In 2021, DAAQ-TFP was first modified with the cationic surfactant CTAB (cetyltrimethylammonium bromide) followed by self-assembly with the negatively charged GA via electrostatic interaction (Fig. 13c).⁷² The 3D cross-linking conductive network in the obtained free-standing DAAQ-TFP/GA composite could efficiently surmount the limitations of slow electron transfer and low accessibility to the active sites in the COF skeleton. The as-obtained DAAQ-TFP/GA electrode exhibited a high specific capacitance of 378 F g⁻¹ at 1 A g⁻¹ and fast kinetics with ∼93.4% capacitive contribution at 3 mV s⁻¹. Besides, the asymmetric SC assembled with DAAQ-COFs/GA and pristine GA electrodes exhibited an energy density of 30.5 W h kg⁻¹. In early 2021, our group also made similar progress in the SC electrode application of graphene/COF composites. We chose reduced graphene oxide (rGO) aerogel as a conducting network support, and four Phos-COF-1/rGO composites (PPrGO-1, PPrGO-2, PPrGO-3, and PPrGO-4) with different mol% rGO were synthesized via a solvothermal method where the polymer was formed in the presence of rGO.⁷³ Thanks to the well-established interfacial interaction and high specific surface area (690 m² g⁻¹), PPrGO-2 exhibited the lowest charge transfer resistance (3.85 Ω cm⁻²) and series resistance (1.72 Ω cm⁻²) in 1 M KOH. Correspondingly, the as-assembled PPrGO-2//AC (active carbon) could be operated in a wide potential window of 1.6 V, displaying a high energy density of 33.3 W h kg⁻¹ with a power density of 8370 W kg⁻¹ and superb cyclability with 88% capacitance retention after 12 000 successive charge/discharge cycles. The excellent electrochemical performance of PPrGO-2 was ascribed to the conjugated structure, highly enhanced conductivity, and easily accessible redox sites. Meanwhile, Phos-COF-1 was also hydrothermally composited with rGO to give a hierarchical nanostructured PP/rGO which then served as the anode for high-performance SCs. Employing α-MnO₂ nanowires as the cathode, a high-voltage aqueous asymmetric SC was successfully assembled, which had an extended potential window of 2 V, and exhibited a significantly enhanced capacitance of 380.2 F g⁻¹ at 2 A g⁻¹ with a good rate capability of 165.0 F g⁻¹ at 4 A g⁻¹. A high specific energy of 39 W h kg⁻¹ with extraordinary cyclability with 96% capacitance retention over 10 000 cycles was achieved as well.⁷⁴

Enhancing conductivity via a metal ion assisted strategy has also been carried out to improve the capacitive performance of COF-based SCs. In 2019, Ni-COFs were solvothermally synthesized via choosing 1,2,4,5-benzenetetraamine (BTA), 2,5-dihydroxy-1,4-benzenedicarboxaldehyde (HBC) as building blocks and Ni(OAc)₂·4H₂O as an electrochemically active center.⁷⁵ The chemical structure and charge–discharge mechanism of Ni-COFs are presented in Fig. 14. The reversible transformation between hydroquinone and benzoquinone occurred during the energy storage/conversion process. Benefiting from the 2D highly conjugated structure and the incorporation of Ni(II) centers, Ni-COFs exhibited high electrical conductivity of 1.2 and 1.3 × 10⁻² S cm⁻¹ for thin film and powder, respectively. Besides, Ni-COFs showed an ultrahigh specific capacitance of 1257 F g⁻¹ at 1 A g⁻¹ with 94% capacitance retention over 10 000 cycles. The as-assembled activated carbon//Ni-COF showed a high capacitance of 417 F g⁻¹, and an excellent energy density of 130 W h kg⁻¹. This work demonstrated a feasible approach to enhance the electrochemical performance through introducing metal ion centers.

Fig. 14 Energy storage mechanism (top) and chemical structure (bottom) of Ni-COFs. Reproduced with permission from ref. 75 Copyright 2019, The Royal Society of Chemistry.

3.4 COFs as supports for other electrode materials

COFs feature with ordered channels and good structural tailorability, which are considered to be an outstanding porous support for other electrochemically active species.⁷⁶ For one thing, the easily functionalized construction of the channel-wall composition can provide a chemical platform for the controllable introduction of guest molecules; for another thing, their uniform channels can effectively confine guest species inside the channel without aggregation or loss during application. Therefore, by incorporating highly electrochemically active species such as metal oxides, hydroxides, fullerenes into the channels of COFs, high electrochemical activity and cycle performance are expected. However, the studies on COF-dispersed and stabilized electrochemically active metal oxides/hydroxides have been rarely reported. Therefore, numerous opportunities exist in the field, which will be discussed in detail in the last section.

The unique π-electron conjugated system and excellent electron affinity of fullerenes make them attractive candidates for electrode materials in EES devices. However, fullerene materials usually lack a certain degree of order on the nanoscale, resulting in low electrochemical utilization and a limited ion transport rate, thus greatly limiting their application in energy storage. In 2021, we designed an azide group-functionalized COF to hang the fullerene carbonyl/hydroxyl type into the ordered pores through “click” chemistry to create ordered nonporous fullerenes. (Fig. 15).⁷⁷ Powder X-ray diffraction and N₂ adsorption–desorption test results showed that the obtained material ([C₆₀]X-COF, X = 0.025; 0.05; 0.09; 0.016) could retain the crystallinity and porosity of the COF support, but the pore volume decreased with the increase in C₆₀ loading. Among the four [C₆₀]X-COFs with different C₆₀ contents, [C₆₀]0.05-COF electrode material showed the best electrochemical performance, which was because an appropriate C₆₀ loading can endow sufficient active sites without blocking the pores that serve as mass transfer paths. All [C₆₀]X-COFs exhibited much higher capacitance values than those (<14 F g⁻¹) of pristine COF and C₆₀ in a three electrode configuration, among which [C₆₀]0.05-COFs gave the maximum specific capacitance of 63.1 F g⁻¹ at 0.7 A g⁻¹. The assembled [C₆₀]0.05-COF//rGO delivered an energy density of up to 21.4 W h kg⁻¹ at a power density of 900 W kg⁻¹ in an operating voltage window of 1.8 V with 99% capacitance retention after 5000 cycles. This work discloses that constructing ordered nanoporous C₆₀ templated by a COF material is an efficient strategy to enhance its electrochemical accessibility and thus capacitive performance.

Fig. 15 Chemical structure of [C₆₀]X-COFs. Reproduced with permission from ref. 77 Copyright 2021, Elsevier.

3.5 COFs as precursors for porous carbon electrode materials

Porous carbon materials have been drawing intensive attention with regard to metal-free electrode application thanks to their pore structure, high specific surface areas, and high chemical and mechanical stability. Doping heteroatoms (N, B, S, P, etc.) into sp² carbon networks is a feasible approach to modify the electronic properties and provide electroactive sites.^78–80 Actually, as aforementioned, the vast majority of pristine COFs deliver weak electrical conductivity, which severely restricts the electrochemical applications of COFs. Carbonization at high temperature has been considered as an effective approach to enhance their electrical conductivities. Compared with amorphous organic porous material precursors, carbonizing crystalline COFs is more beneficial to obtain porous carbon materials with uniform pore size and controlled introduction of heteroatoms.^81,82 Although the carbonization treatment might destroy the periodical features of pristine COFs, many investigations have demonstrated that the carbides derived from COF precursors showcased enhanced electrochemical performance compared to original COFs.^83,84 Hitherto, both approaches to carbonize COFs were developed, including template-free direct carbonization and templated pyrolysis, while the former has a wider range of applications.

3.5.1 Template-free direct carbonization. The pioneering SC electrode materials carbonized from COF precursors was reported in 2013.⁸⁵ A series of hollow spherical nitrogen-rich carbons with a microporous shell structure (Fig. 16a) were prepared through carbonizing COF-LZU1⁸⁶ at 700 °C, 800 °C, 900 °C, respectively. Porous carbon (NPC800) obtained at 800 °C had the highest specific surface area of 525 m² g⁻¹ among the three experimental conditions, delivering high specific capacitance values of 230 and 190 F g⁻¹ at 0.5 and 3 A g⁻¹ in 5 M KOH, respectively, while retaining 98% of the initial capacitance over 1500 cycles in a three electrode configuration. N-doped microporous carbon material with a nanorod-like structure (Fig. 16b) was also fabricated through carbonization of an azine-based 2D COF (ACOF1) in 2017.⁸⁷ From carbonization mechanism analysis, it was found that N₂ released from the degradation of azine bonds facilitated the formation of microporous textures during the pyrolysis of ACOF1, which was confirmed by the decreased nitrogen content (16 vs. 1.6 atom%) and enhanced specific surface area (1168 vs. 1596 m² g⁻¹) of ACOF1 before and after carbonization. In 6 M KOH, carbonized ACOF1 exhibited a much higher specific capacitance of 234 F g⁻¹ at 1 A g⁻¹ than that (191 F g⁻¹) of carbonized COF-LZU1⁸⁶ treated using a similar procedure. The enhanced capacitive performance of the carbonized ACOF1 was ascribed to its large surface area and rich exposed active sites.

Fig. 16 (a-i) Chemical structure of COF-LZU1, and SEM image of COF-LZU1 (a-ii) before and (a-iii) after carbonization. Reproduced with permission from ref. 85 Copyright 2013, American Chemical Society. (b-i) Chemical structure of ACOF1 and SEM image of ACOF1 (b-ii) before and (b-iii) after carbonization. Reproduced with permission from ref. 87 Copyright 2017, Elsevier. (c-i) Synthesis schematic illustration of MPCFs and (c-ii) its simulated fragments. Reproduced with permission from ref. 88 Copyright 2017, American Chemical Society.

The N and O atomically homogeneously doped porous carbon frameworks (MPCFs) were polymerized from 4,4′-(4-oxophthalazine-1,3(4H)-diyl)dibenzonitrile through the trimerization of cyano groups and subsequent ionothermal synthesis catalyzed by ZnCl₂ at 600 °C, 650 °C, 700 °C, respectively (Fig. 16c).⁸⁸ After pyrolysis at high temperature, the black carbonized powder (MPCFs@700) obtained at 700 °C had the highest specific surface area of 1638 m² g⁻¹ and a pore diameter of ∼2 nm, thus affording a high specific capacitance of 302 F g⁻¹ in a three electrode configuration. When 1-butyl-3-methylimidazolium tetrafluoroborate was employed as an electrolyte to broaden the special window (up to 3.5 V), an SC device based on MPCFs@700 exhibited a specific capacitance of 155 F g⁻¹ and an energy density of 65 W h kg⁻¹ at 0.1 A g⁻¹. Furthermore, the as-prepared MPCFs@700 showed excellent cyclability in aqueous, organic and ionic liquid electrolyte respectively, demonstrating its long term durability.

3.5.2 Templated carbonization. Templated carbonization for converting original COFs into porous carbons has been demonstrated to be an effective strategy to address the collapse of the porous structures of COFs at high temperature. As shown in Fig. 17a–c, polycyclotriphosphazene-co-4,4′-sulfonyldiphenol (PPZS) spheres possessing high heteroatom content is an ideal template, and TAPT-DHTA-COFs via in situ polymerization of 2,5-dihydroxyterephthalaldehyde (DHTA) and 4,4′,4′′-(1,3,5-triazine-2,4,6-triyl)trianiline (TAPT) could be coated on the surface of the PPZS spheres, resulting in TAPT-DHTA-COF_X@PPZS (X = weight ratio of COF to PPZS = 0.05, 0.1, and 0.2).⁸⁹ After pyrolysis at 900 °C, three microporous carbons, i.e., TAPT-DHTA-COF_x@PPZS₉₀₀ were prepared, exhibiting specific surface areas of 533, 456 and 421 m² g⁻¹, respectively, while TAPT-DHTA-COF₉₀₀ from template-free pyrolysis at 900 °C had a BET surface area of only 21 m² g⁻¹. Their capacitive performance was closely related to the core–shell structure. In a three electrode configuration, TAPT-DHTA-COF_0.1@PPZS₉₀₀ delivered a capacitance of 287 F g⁻¹ at 1 A g⁻¹, which was much higher than that of contrastive TAPT-DHTA-COF₉₀₀ (43.8 F g⁻¹) and PPZS₉₀₀ (132 F g⁻¹), while TAPT-DHTA-COF_0.05@PPZS₉₀₀ and TAPT-DHTA-COF_0.2@PPZS₉₀₀ exhibited capacitance values of 178 and 255 F g⁻¹, respectively.

Fig. 17 Chemical structure of (a) PPZS and (b) TAPT–DHTA-COFs, and (c) preparation schematic illustration of TAPT-DHTA-COFX@PPZS and TAPTDHTA-COF_X@PPZS₉₀₀. Reproduced with permission from ref. 89 Copyright 2017, The Royal Society of Chemistry. (d) Synthesis schematic illustration of B-N-C capsules. Reproduced with permission from ref. 90 Copyright 2019, Elsevier.

B, N-codoped carbon (B-N-C) was successfully derived via thermal conversion of COFs by using metallic Cu particles as the template core in 2019.⁹⁰ As shown in the Fig. 17d, Cu(NO₃)₂ served as both a chemical equilibrium control agent and a catalyst for Schiff-base reaction between melamine and 4-formylphenylboronic acid to synthesize Cu(NO₃)₂(OH)₃@B-N-COFs in one step. Core-capsule-like Cu@B-N-Cs with a thin carbon shell was then obtained by pyrolysis at 900 °C, 1000 °C, and 1100 °C. Hollow B-N-C capsules with ∼15 nm thickness were finally formed after removing the Cu core with HNO₃. The resulting B-N-C-1000 showcased a high specific capacitance of 230 F g⁻¹ at 5 A g⁻¹. The storage of B-N-C was mainly dominated by EDLC mechanism via adsorption of electrolyte ions onto the surface of electrode materials.

It has been confirmed that K₂CO₃ can be reduced to emit K and CO gases by carbon at a high temperature of >627 °C in an inert atmosphere.⁹¹ In 2019, O and N co-doped carbons (ONC-T1s) were obtained from K₂CO₃-assisted pyrolysis of AQ-COF.⁹² During the carbonization process, K₂CO₃ inside the channels of AQ-COFs serves not only as an activator to create pores and cause swelling of the carbon layers, but also as a template to prevent the skeleton collapse. Among the resulting ONC-T1s (carbonized at 550, 700 and 850 °C, respectively), ONC-T1-700 and ONC-T1-850 exhibited hierarchical pore size distributions from micropores to mesopores along with ultrahigh BET surface areas of 3451 and 1518 m² g⁻¹, respectively. In 1 M H₂SO₄, ONC-T1-700 and ONCT1-850 achieved high specific capacitance values of 768 and 1711 F g⁻¹ at 1 A g⁻¹, respectively, accompanied with a fast charge–discharge rate and superb cyclability.

3.6 Pros and cons of COF-based electrode materials

As summarized above, pristine COFs, hybrid COFs and carbonized COFs have been developed as electrode materials for SCs. Each COF-based electrode material has its pros and cons in terms of chemical synthesis and structural properties.

The electrochemical properties of pristine COFs are usually determined from the chemical structure and aggregation morphology. Their SC performances are regulated through the precise predesign of the chemical component, which is superior to other electrode materials. Correspondingly, due to the high requirements for the constructing conditions of COFs, the development of high-performance novel pristine COF electrode materials is also very challenging, usually requiring a reasonable combination of conjugated skeletons and highly active-redox centers.

Hybrid COFs provide a broad development prospect for SC applications of COF-based electrode materials. The SC performance of COFs could be further optimized and the cost could be easily reduced in combination with electrochemically active units such as activated carbon, metal oxides, conductive polymers, etc. However, the analysis of the structure–property relationship and the function-oriented construction of COF-based hybrids are generally difficult.

The morphology of carbonized COFs depends largely on carbonization temperature and the thermal stability of pristine COFs. The purpose of carbonization is to improve the conductivity of COFs and retain the regular pore structure in the meantime. It may also increase the accessibility to active sites and even introduce new sites through templated carbonization. However, the controllability of template-free carbonized COFs is usually poor. Although the templated carbonization can improve the controllability to some extent, the process is usually complex and the cost is usually high, which is not conducive to the commercial application.

4. Conclusions and outlook

COFs are an emerging class of organic porous materials with ordered porous properties. Due to their adjustable redox centers, high specific surface areas, ordered open channels, COFs have drawn great attention in the application of electrode materials. For SC devices, to fabricate electrodes with high specific capacitance, high energy/power density and excellent cycling stability, constructing COFs with high specific surface areas, high conductivity and abundant accessible redox active sites is highly desired. In the past few decades, a series of electrochemically active COFs exhibiting excellent SC performance have been developed, and they have demonstrated tremendous potential for SC applications. In this feature article, design strategies of COF-based electrode materials for SCs, including (i) introducing redox-active moieties into COFs, (ii) improving the accessibility of redox-active moieties, (iii) enhancing the electrical conductivity, (iv) COFs as supports for other electrode materials, and (v) COFs as precursors for porous carbon electrode materials, have been emphatically discussed via summarizing representative research studies. Furthermore, the challenges as well as perspectives in the future are highlighted as follows:

(1) Low-cost preparation of COFs on a large scale is one of the key challenges for them to achieve practical applications. Besides the traditional solvothermal synthesis, some high-throughput approaches have been developed, such as liquid crystal-directed synthesis⁹³ and an ultrasound-assisted process.⁹⁴ Batch preparation of COFs can be achieved through these high-throughput synthesis strategies. Of course, other preparation methods, such as photocatalytic synthesis, electrocatalytic synthesis, and microwave synthesis, are also expected to result in large-scale production of COFs.

(2) The design and synthesis of COFs with high electrochemical activity is a very important research direction, which is related to whether COF-based electrodes can be applied into our daily life and industry applications. Here are three suggested strategies to obtain COFs with high capacitive performance: (i) to develop new building blocks and new linkage types with electrochemical activity; (ii) to chemically anchor the electrochemical active moieties on the backbone of COFs; and (iii) to physically composite electrochemical active centers on the COF supports. Although some COFs with good capacitive performance have been obtained using these strategies, more types of COFs with better performance need to be developed.

(3) The type, number and accessibility of redox active centers are the key to obtain COF materials with high specific capacitance, excellent energy density and cycling stability. Multi-type redox centers need to be developed and exploited. These novel COFs can be constructed via the rational design of small organic molecules with high redox activity as building blocks of COFs.

(4) More preparation strategies for free-standing thin sheets and more available COF sources are needed to meet commercial applications.

(5) Superior capacitive performances for the composites of COF scaffolds and classical electrochemically active species (i.e., conducting polymers and metal oxides/hydroxides) are reasonably expected. Although a few studies on conductive polymer functionalized COFs have been reported, bridging the vast gap in development is of great importance. Metal oxides/hydroxides featuring abundance, low cost, and satisfying electrochemical properties have also been studied as potential electrodes for SCs. However, the relatively low electrical conductivity and inferior structural stability hinder their practical application. Compositing COFs and metal oxides/hydroxides is a feasible strategy to integrate their merits: (i) the porous nature of COFs facilitate good interaction between the electrolyte and electrochemical active metal oxides/hydroxides; (ii) the ultrafine metal oxide/hydroxide nanoparticles dispersed on the surface or dwelled inside the channel of the COFs could be almost fully accessible; (iii) COF carriers with high superiority in designability can overcome the poor conductivity of metal oxides/hydroxides; and (iv) the synergistic effects of electrochemically active centers of COFs and metal oxides/hydroxides lead to a super-excellent capacitive performance.

Author contributions

Rao Tao: writing the original draft, reviewing, and editing. Tianfu Yang and Yan Wang: reviewing, and editing. Jingmin Zhang and Zhengyi Wu: making figures. Li Qiu: reviewing, editing, and supervision.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (grant 51962036), the Key Project of Natural Science Foundation of Yunnan (grant 202201AS070011), High-Level Talents Introduction in Yunnan Province (grant C619300A025), the Major Science and Technology Project of Precious Metal Materials Genetic Engineering in Yunnan Province (grants 2019ZE001-1 and 202002AB080001), and Yunnan University (grant C176220100005).

References

1. D. Zhu, G. Xu, M. Barnes, Y. Li, C. P. Tseng, Z. Zhang, J. J. Zhang, Y. Zhu, S. Khalil, M. M. Rahman, R. Verduzco and P. M. Ajayan, Adv. Funct. Mater., 2021, 31, 2100505.
2. T. Sun, J. Xie, W. Guo, D. S. Li and Q. Zhang, Adv. Energy Mater., 2020, 10, 1904199.
3. Y. Dai, C. Liu, Y. Bai, Q. Kong and H. Pang, Nanotechnol. Rev., 2022, 11, 1005–1046.
4. G. Wang, L. Zhang and J. Zhang, Chem. Soc. Rev., 2012, 41, 797–828.
5. D. P. Dubal, N. R. Chodankar, D. H. Kim and P. Gomez-Romero, Chem. Soc. Rev., 2018, 47, 2065–2129.
6. A. González, E. Goikolea, J. A. Barrena and R. Mysyk, Renew. Sust. Energ. Rev., 2016, 58, 1189–1206.
7. C. Merlet, C. Péan, B. Rotenberg, P. A. Madden, B. Daffos, P. L. Taberna, P. Simon and M. Salanne, Nat. Commun., 2013, 4, 2701.
8. F. Béguin, V. Presser, A. Balducci and E. Frackowiak, Adv. Mater., 2014, 26, 2219–2251.
9. Y. Zhai, Y. Dou, D. Zhao, P. F. Fulvio, R. T. Mayes and S. Dai, Adv. Mater., 2011, 23, 4828–4850.
10. L. Hao, X. Li and L. Zhi, Adv. Mater., 2013, 25, 3899–3904.
11. R. Chen, M. Yu, R. P. Sahu, I. K. Puri and I. Zhitomirsky, Adv. Energy Mater., 2020, 10, 1903848.
12. G. Zhang, X. Xiao, B. Li, P. Gu, H. Xue and H. Pang, J. Mater. Chem. A, 2017, 5, 8155–8186.
13. J. Kim, J. Lee, J. You, M. S. Park, M. S. Al Hossain, Y. Yamauchi and J. H. Kim, Mater. Horiz., 2016, 3, 517–535.
14. X. Xiao, L. Zou, H. Pang and Q. Xu, Chem. Soc. Rev., 2020, 49, 301–331.
15. X. Liu, C. F. Liu, S. Xu, T. Cheng, S. Wang, W. Y. Lai and W. Huang, Chem. Soc. Rev., 2022, 51, 3181–3225.
16. D. P. Chatterjee and A. K. Nandi, J. Mater. Chem. A, 2021, 9, 15880–15918.
17. J. Lin, Y. Zhong, L. Tang, L. Wang, M. Yang and H. Xia, Nano Select, 2022, 3, 320–347.
18. S. Jin, O. Allam, S. S. Jang and S. W. Lee, InfoMat, 2022, 4, e12277.
19. L. Yang and N. Huang, J. Polym. Sci., 2022, 60, 2225–2238.
20. M. Sajjad and W. Lu, J. Energy Storage, 2021, 39, 102618.
21. K. Zhang, K. O. Kirlikovali, R. S. Varma, Z. Jin, H. W. Jang, O. K. Farha and M. Shokouhimehr, ACS Appl. Mater. Interfaces, 2020, 12, 27821–27852.
22. A. P. Cote, A. I. Benin, N. W. Ockwig, M. O'Keeffe, A. J. Matzger and O. M. Yaghi, Science, 2005, 310, 1166–1170.
23. Z. Chen, K. Wang, X. Hu, P. Shi, Z. Guo and H. Zhan, ACS Appl. Mater. Interfaces, 2020, 13, 1145–1151.
24. S. B. Alahakoon, S. D. Diwakara, C. M. Thompson and R. A. Smaldone, Chem. Soc. Rev., 2020, 49, 1344–1356.
25. X. Guan, F. Chen, S. Qiu and Q. Fang, Angew. Chem., Int. Ed., 2022, 62, e202213203.
26. O. M. Yaghi, M. J. Kalmutzki and C. S. Diercks, Introduction to reticular chemistry: metal-organic frameworks and covalent organic frameworks, John Wiley & Sons, 2019.
27. Y. Jin, C. Yu, R. J. Denman and W. Zhang, Chem. Soc. Rev., 2013, 42, 6634–6654.
28. X. Chen, K. Geng, R. Liu, K. T. Tan, Y. Gong, Z. Li, S. Tao, Q. Jiang and D. Jiang, Angew. Chem., Int. Ed., 2020, 59, 5050–5091.
29. S. Kandambeth, K. Dey and R. Banerjee, J. Am. Chem. Soc., 2019, 141, 1807–1822.
30. Y. Su, Y. Wan, H. Xu, K. Otake, X. Tang, L. Huang, S. Kitagawa and C. Gu, J. Am. Chem. Soc., 2020, 142, 13316–13321.
31. H. L. Qian, F. L. Meng, C. X. Yang and X. P. Yan, Angew. Chem., Int. Ed., 2020, 59, 17607–17613.
32. B. Zhang, M. Wei, H. Mao, X. Pei, S. A. Alshmimri, J. A. Reimer and O. M. Yaghi, J. Am. Chem. Soc., 2018, 140, 12715–12719.
33. X. Li, K. Kawai, M. Fujitsuka and Y. Osakada, Surf. Interfaces, 2021, 25, 101249.
34. N. Keller and T. Bein, Chem. Soc. Rev., 2021, 50, 1813–1845.
35. M. Wu and Y. Yang, Chin. Chem. Lett., 2017, 28, 1135–1143.
36. X. Liu, D. Huang, C. Lai, G. Zeng, L. Qin, H. Wang, H. Yi, B. Li, S. Liu and M. Zhang, Chem. Soc. Rev., 2019, 48, 5266–5302.
37. C. R. DeBlase, K. E. Silberstein, T. T. Truong, H. D. Abruña and W. R. Dichtel, J. Am. Chem. Soc., 2013, 135, 16821–16824.
38. J. Li, X. Jing, Q. Li, S. Li, X. Gao, X. Feng and B. J. C. S. R. Wang, Chem. Soc. Rev., 2020, 49, 3565–3604.
39. M. Li, J. Liu, T. Zhang, X. Song, W. Chen and L. Chen, Small, 2021, 17, e2005073.
40. S. Wang, Y. Guo, F. Wang, S. Zhou, T. Zeng and Y. Dong, New Carbon Mater., 2022, 37, 109–135.
41. C. Gao, J. Huang, Y. Xiao, G. Zhang, C. Dai, Z. Li, Y. Zhao, L. Jiang and L. Qu, Nat. Commun., 2021, 12, 2647.
42. J. C. Russell, V. A. Posey, J. Gray, R. May, D. A. Reed, H. Zhang, L. E. Marbella, M. L. Steigerwald, Y. Yang, X. Roy, C. Nuckolls and S. R. Peurifoy, Nat. Mater., 2021, 20, 1136–1141.
43. X. Chen, R. Paul and L. Dai, Natl. Sci. Rev., 2017, 4, 453–489.
44. Q. Wu, T. He, Y. Zhang, J. Zhang, Z. Wang, Y. Liu, L. Zhao, Y. Wu and F. Ran, J. Mater. Chem. A, 2021, 9, 24094–24147.
45. B. Häupler, A. Wild and U. S. Schubert, Adv. Energy Mater., 2015, 5, 1402034.
46. Y. Liang, P. Zhang and J. Chen, Chem. Sci., 2013, 4, 1330–1337.
47. P. Poizot, J. Gaubicher, S. Renault, L. Dubois, Y. Liang and Y. Yao, Chem. Rev., 2020, 120, 6490–6557.
48. S. Chandra, D. Roy Chowdhury, M. Addicoat, T. Heine, A. Paul and R. Banerjee, Chem. Mater., 2017, 29, 2074–2080.
49. M. Li, J. Liu, Y. Li, G. Xing, X. Yu, C. Peng and L. Chen, CCS Chem., 2020, 2, 696–706.
50. A. Halder, M. Ghosh, M. A. Khayum, S. Bera, M. Addicoat, H. S. Sasmal, S. Karak, S. Kurungot and R. Banerjee, J. Am. Chem. Soc., 2018, 140, 10941–10945.
51. P. Bhanja, K. Bhunia, S. K. Das, D. Pradhan, R. Kimura, Y. Hijikata, S. Irle and A. Bhaumik, ChemSusChem, 2017, 10, 921–929.
52. A. F. M. El-Mahdy, Y. H. Hung, T. H. Mansoure, H. H. Yu, T. Chen and S. W. Kuo, Chem. – Asian J., 2019, 14, 1429–1435.
53. A. M. Khattak, Z. A. Ghazi, B. Liang, N. A. Khan, A. Iqbal, L. Li and Z. Tang, J. Mater. Chem. A, 2016, 4, 16312–16317.
54. S. Haldar, R. Kushwaha, R. Maity and R. Vaidhyanathan, ACS Materials Lett., 2019, 1, 490–497.
55. L. Li, F. Lu, R. Xue, B. Ma, Q. Li, N. Wu, H. Liu, W. Yao, H. Guo and W. Yang, ACS Appl. Mater. Interfaces, 2019, 11, 26355–26363.
56. R. Xue, Y. Zheng, D. Qian, D. Xu, Y. Liu, S. Huang and G. Yang, Mater. Lett., 2022, 308, 131229.
57. S. Xiong, J. Liu, Y. Wang, X. Wang, J. Chu, R. Zhang, M. Gong and B. Wu, J. Appl. Polym. Sci., 2022, 139, 51510.
58. M. Sajjad, R. Tao and L. Qiu, J. Mater. Sci.: Mater. Electron., 2021, 32, 1602–1615.
59. K. Kang, Z. Wu, M. Zhao, Z. Li, Y. Ma, J. Zhang, Y. Wang, M. Sajjad, R. Tao and L. Qiu, Chem. Commun., 2022, 58, 3649–3652.
60. R. Tao, X. Shen, Y. Hu, K. Kang, Y. Zheng, S. Luo, S. Yang, W. Li, S. Lu, Y. Jin, L. Qiu and W. Zhang, Small, 2020, 16, e1906005.
61. Y. Liu, A. Dikhtiarenko, N. Xu, J. Sun, J. Tang, K. Wang, B. Xu, Q. Tong, H. J. Heeres, S. He, J. Gascon and Y. Fan, Chem. – Eur. J., 2020, 26, 12134–12139.
62. T. Li, X. Yan, Y. Liu, W. Zhang, Q. Fu, H. Zhu, Z. Li and Z. Gu, Polym. Chem., 2020, 11, 47–52.
63. F. Xu, H. Xu, X. Chen, D. Wu, Y. Wu, H. Liu, C. Gu, R. Fu and D. Jiang, Angew. Chem., Int. Ed., 2015, 54, 6814–6818.
64. X. Zhang, Z. Xiao, X. Liu, P. Mei and Y. Yang, Renew. Sust. Energ. Rev., 2021, 147, 111247.
65. J. Lee, S. Hong, Y. Heo, H. Kang and M. Kim, Dalton Trans., 2021, 50, 14081–14090.
66. C. R. DeBlase, K. Hernández-Burgos, K. E. Silberstein, G. G. Rodríguez-Calero, R. P. Bisbey, H. D. Abruña and W. R. J. A. N. Dichtel, ACS Nano, 2015, 9, 3178–3183.
67. Z. Zha, L. Xu, Z. Wang, X. Li, Q. Pan, P. Hu and S. Lei, ACS Appl. Mater. Interfaces, 2015, 7, 17837–17843.
68. M. A. Khayum, V. Vijayakumar, S. Karak, S. Kandambeth, M. Bhadra, K. Suresh, N. Acharambath, S. Kurungot and R. Banerjee, ACS Appl. Mater. Interfaces, 2018, 10, 28139–28146.
69. S. Chen, W. Xing, J. Duan, X. Hu and S. Z. Qiao, J. Mater. Chem. A, 2013, 1, 2941–2954.
70. C. R. Mulzer, L. Shen, R. P. Bisbey, J. R. McKone, N. Zhang, H. D. Abruna and W. R. Dichtel, ACS Cent. Sci., 2016, 2, 667–673.
71. Y. Wu, D. Yan, Z. Zhang, M. M. Matsushita and K. Awaga, ACS Appl. Mater. Interfaces, 2019, 11, 7661–7665.
72. N. An, Z. Guo, J. Xin, Y. He, K. Xie, D. Sun, X. Dong and Z. Hu, J. Mater. Chem. A, 2021, 9, 16824–16833.
73. M. Sajjad, R. Tao, K. Kang, S. Luo and L. Qiu, ACS Appl. Energy Mater., 2021, 4, 828–838.
74. M. Sajjad, K. Kang, Z. Wu, Y. Ma, R. Tao and L. Qiu, J. Energy Storage, 2021, 40, 102772.
75. T. Li, W. Zhang, Y. Liu, Y. Li, C. Cheng, H. Zhu, X. Yan, Z. Li and Z. Gu, J. Mater. Chem. A, 2019, 7, 19676–19681.
76. R. Tao, X. Ma, X. Wei, Y. Jin, L. Qiu and W. Zhang, J. Mater. Chem. A, 2020, 8, 17360–17391.
77. X. Zhao, M. Sajjad, Y. Zheng, M. Zhao, Z. Li, Z. Wu, K. Kang and L. Qiu, Carbon, 2021, 182, 144–154.
78. J. P. Paraknowitsch and A. Thomas, Energy Environ. Sci., 2013, 6, 2839–2855.
79. Y. Deng, Y. Xie, K. Zou and X. Ji, J. Mater. Chem. A, 2016, 4, 1144–1173.
80. L. Miao, H. Duan, M. Liu, W. Lu, D. Zhu, T. Chen, L. Li and L. Gan, Chem. Eng. J., 2017, 317, 651–659.
81. Q. Xu, Y. Tang, X. Zhang, Y. Oshima, Q. Chen and D. Jiang, Adv. Mater., 2018, 30, e1706330.
82. X. Hu, Y. Long, M. Fan, M. Yuan, H. Zhao, J. Ma and Z. Dong, Appl. Catal., B, 2019, 244, 25–35.
83. J. K. Sun and Q. Xu, Energy Environ. Sci., 2014, 7, 2071–2100.
84. X. Zhang, G. Zhu, M. Wang, J. Li, T. Lu and L. Pan, Carbon, 2017, 116, 686–694.
85. X. Liu, L. Zhou, Y. Zhao, L. Bian, X. Feng and Q. Pu, ACS Appl. Mater. Interfaces, 2013, 5, 10280–10287.
86. S. Ding, J. Gao, Q. Wang, Y. Zhang, W. Song, C. Su and W. Wang, J. Am. Chem. Soc., 2011, 133, 19816–19822.
87. G. Kim, J. Yang, N. Nakashima and T. Shiraki, Chem. – Eur. J., 2017, 23, 17504–17510.
88. F. Hu, J. Wang, S. Hu, L. Li, W. Shao, J. Qiu, Z. Lei, W. Deng and X. Jian, ACS Appl. Mater. Interfaces, 2017, 9, 31940–31949.
89. Q. Xu, Y. Tang, L. Zhai, Q. Chen and D. Jiang, Chem. Commun., 2017, 53, 11690–11693.
90. Z. Zhou, X. Zhang, L. Xing, J. Liu, A. Kong and Y. Shan, Electrochim. Acta, 2019, 298, 210–218.
91. D. W. McKee and D. Chatterji, Carbon, 1978, 16, 53–57.
92. D. Yan, Y. Wu, R. Kitaura and K. Awaga, J. Mater. Chem. A, 2019, 7, 26829–26837.
93. F. Tan, Y. Zheng, Z. Zhou, H. Wang, X. Dong, J. Yang, Z. Ou, H. Qi, W. Liu, Z. Zheng and X. Chen, CCS Chem., 2022, 4, 3751–3761.
94. W. Zhao, P. Yan, B. Li, M. Bahri, L. Liu, X. Zhou, R. Clowes, N. D. Browning, Y. Wu, J. W. Ward and A. I. Cooper, J. Am. Chem. Soc., 2022, 144, 9902–9909.

---