Lithium-ion batteries (LIBs) are one of the most successful commercial rechargeable energy storage devices with high energy density and long cycle life^{[1, 2]}. However, they cannot satisfy the increasing energy demands, due to the scarcity of lithium resources and the potential safety concerns particularly when lithium metal is adopted as the anodes for the chase of high energy density (high reactivity with air and H₂O)^{[3, 4]}. Zinc-ion batteries (ZIBs), a kind of emerging energy storage devices, are considered as one of the promising alternatives of LIBs. ZIBs contain a lot of advantages owing to the properties of Zn, including the higher abundance of Zn than Li (70 ppm for Zn vs. 20 ppm for Li in crust)^[5], low cost, and small ionic radius (0.74 Å), etc.^[6]. Moreover, due to the high safety of Zn metal (weak reactivity with air and H₂O), Zn metal is regarded as the most promising anode material for ZIBs, which shows high specific gravimetric capacity (820 mAh/g) and high specific volumetric capacity (5855 mAh/cm³), although the cycling stability of ZIBs with Zn metal anode is also plagued by the possible formation of Zn dendrites. Fortunately, various methods have been proved to be effective to inhibit the growth of Li dendrites in Li-metal batteries^[7-11], which could probably be also applicable to Zn metal for high cycleability^[12-14]. In addition, Zn has proper electrode potential (0.76 V versus standard hydrogen electrode), making it possible to adopt an aqueous electrolyte, which has the merits of high ionic conductivity, low cost, green and nontoxicity.

Due to the possibility of using Zn metal as anodes, the study of cathode materials to match with the Zn anode is of the greatest interest for ZIBs^[15-19]. Among the reported cathode materials for ZIBs, inorganic materials account for a large proportion, which consist of manganese-based oxides^{[20, 21]}, vanadium-based oxides^{[22, 23]} and Prussian blue analogues^{[24, 25]}, etc. They have exhibited fairish Zn-storage performance. However, the development of inorganic cathode materials for ZIBs is hampered by some issues. For example, inorganic materials normally make use of toxic or environmental-unfriendliness transition metal elements. Besides, they suffer capacity fading due to the large volume change upon the insertion of Zn²⁺ ions^{[26, 27]}. Compared to inorganic materials, organic materials are endowed with many advantages containing eco-friendliness, low-cost, lightweight, flexibility (small volume change during discharge–charge), designable molecular structure^[28-31], which stimulate the exploration of organic cathode materials for ZIBs.

On the other hand, although aqueous electrolytes have a lot of advantages and most of the reported ZIBs adopted aqueous electrolytes, it should be noted that the aqueous electrolytes face the challenge of decomposition of water under larger electrochemical window, which would lead to poor cycleability of aqueous ZIBs (AZIBs). However, the organic cathode materials have the merits of tunable electrode potential through molecular design, which therefore are particularly suitable for aqueous ZIBs. To date, the reported ZIBs with organic cathodes are mostly based on aqueous electrolytes, and the relevant research progress is focused in this review. However, to comprehensively summarize the progress of organic cathodes for ZIBs, the investigation of organic ZIBs (OZIBs) based on non-aqueous electrolytes are also discussed (the mentioned electrolytes in this review are aqueous unless the solvent is specifically noted). The challenges and strategies to improve the performance of organic cathode materials are also addressed in each part. Finally, the perspectives on OZIBs are put forward in hope of developing high-performance ZIBs and energy storage devices.

Quinones are a kind of representative organic electrode materials and have exhibited decent electrochemical performance (high capacity, high rate capability, long cycleability etc.) as cathodes in LIBs based on the stable redox reaction of carbonyls^[32]. When the quinones accept electrons, the carbonyl groups (C=O) transform into enolates (C–O–Li). However, the investigation of quinones as cathodes for ZIBs is still in the primary stage (Fig. 1), and the energy storage mechanism is still under research. It is also considered as the transformation between C=O bonds and C–O bonds; however, because water is widely adopted as the solvent in the electrolytes of ZIBs, the intercalated counter ions might also be H⁺ ions (in aqueous electrolytes) rather than Zn²⁺ ions^[33]. To date, reported quinone-based cathodes for ZIBs can be classified into two types: insertion of Zn²⁺ ions and/or other cations.

Most quinones as cathodes for ZIBs are reported to bind with Zn²⁺ ions during discharge–charge process. A series of small molecular quinone-based cathodes for ZIBs exhibited decent Zn-storage performance (Fig. 2(a))^[34]. Theoretical calculations (electrostatic potential (ESP) method) and experimental characterizations both manifested the carbonyl groups are the active sites for Zn²⁺ ion storage (Figs. 2(d) and 2(e)). Ortho-quinones (e.g. 1,2-naphthoquinone (1,2-NQ) and 9,10-phenanthrenequinone (9,10-PQ)) possess larger steric hindrance for ion insertion than that of para-quinones (e.g. 1,4-naphthoquinone (1,4-NQ), 9,10-anthraquinone (9,10-AQ) and calix[4]quinone (C4Q)), resulting in lower capacity for ortho-quinones. However, all these small molecular quinones suffered capacity fading due to dissolution of discharged products (salts) in liquid aqueous electrolytes. With the aid of the Nafion membrane, the dissolution and the shuttle of the discharged salts were suppressed and the Zn anode was protected from being poisoned by discharged products. As a result, the C4Q-based cathode exhibited good cycling stability (Fig. 2(c), the capacity retention was 87% after 1000 cycles at 0.5 A/g) and high capacity (335 mAh/g at 0.02 A/g) with a small polarization voltage (Fig. 2(b), 70 mV).

As mentioned above, small molecular quinones usually suffered capacity decay due to the dissolution of discharged salts in aqueous electrolytes. However, it was reported that a pyrene-4,5,9,10-tetraone (PTO) cathode for AZIBs was inherently insoluble in aqueous electrolytes, as well as the discharged products, permitting high cycleability^[35]. With normal glass fiber as the separator, the PTO cathode in a coin cell showed a high capacity of 336 mAh/g at 0.04 A/g and superior fast discharge-charge ability. Besides, a high capacity retention of 70% was achieved after 1000 cycles at 3 A/g. Notably, a flexible two-dimensinal (2D) belt shape Zn//PTO battery still delivered a comparable capacity (337 mAh/g at 0.04 A/g) with that in the coin cell. After bending at different angles, the battery still maintained the capacity well (Figs. 3(a) and 3(b)) and can power LEDs and fans (Fig. 3(c)), verifying its practicability.

It is accepted that polymers usually have low solubilities in electrolytes and hence are expected to exhibit decent cycleability^[36-42]. However, it should be noted that the practical performance is highly dependent on the polymerization degree of the active materials and the low polymerization degree results in ineludible dissolution. For example, a poly(benzoquinonyl sulfide) (PBQS)-based cathode for AZIBs delivered a high initial capacity (203 mAh/g at 0.02 A/g); while the capacity loss was still observed, which may be due to its slight dissolution in aqueous electrolytes^[36]. Compared to linear polymers, porous polymers are superior as electrode materials because the inherent porous structure can facilitate the infiltration of electrolytes and fast ionic transport, which contribute to high electrochemical performance^[43-45]. For example, a HqTp covalent organic framework (COF), synthesized by poly-condensation of 2,5-diaminohydroquinone dihydrochloride (Hq) with 1,3,5-triformylphloroglucinol (Tp), possessed crystalline porous structure with pore size of 1.5 nm^[46]. The COF-based cathode exhibited a pair of redox peaks at 1.0/1.12 V vs. Zn/Zn²⁺ with a high capacity of 276 mAh/g at 0.125 A/g. Notably, with the Nafion membrane as the separator, the cycling performance was superior with a capacity retention of 95% after 1000 cycles at 0.15 A/g.

Thus, compositing polymers with conductive carbon materials is helpful for confining the polymers in the carbon materials and further inhibiting the dissolution of active materials. Moreover, the conductive carbon materials can increase the electrical conductivity of composite electrodes and thus benefits the rate capability^{[37, 38, 47, 48]}. The composite of polydopamine (PDA) and carbon nanotubes (CNTs) showed high cycleability with capacity retention of 96% after 500 cycles at 0.2 A/g as a cathode for AZIBs (Fig. 4(c))^[37]. The Zn-storage mechanism of PDA relied on the coordination of Zn²⁺ ions with the carbonyl groups (i.e. o-quinones) in PDA (Fig. 4(a)). The polycatechol (PC) electrode composited with graphene (Fig. 4(b)) (the active material content was ~23.3%) also exhibited better rate capability and cycleability (Fig. 4(d), 171 mAh/g at 10 C; retention of 74.4% after 3000 cycles at 2 C) (the active material content was ~23.3%) than the PC cathode (the active material content was 70%) (51 mAh/g at 10 C; retention of 62.2% after 2000 cycles at 2 C), which benefited from the high electrical conductivity and the confining effect of graphene, respectively^[38]. However, the low active material content in the composite electrode would cause a low energy density of the whole battery. Both of PC and PDA exhibited Zn-storage performance based on the redox of carbonyl groups.

It was widely accepted that the volume change of flexible organic electrode materials is relatively small in comparison with inorganic materials^{[26, 49-52]}. However, it was reported that the small volume change in organic electrode can still result in phase transition and lead to poor cycling performance. Experimental and density functional theory (DFT) calculations revealed the phase transition between tetrachloro-1,4-benzoquinone (p-chloranil) and Zn²⁺-p-chloranil^2–, in which the molecular columns of p-chloranil rotated for accommodation of Zn²⁺ (Fig. 5(c))^[51]. Though the volume change was small (~2.7%), the morphologies of p-chloranil cathodes changed significantly during cycling (Fig. 5(d)), leading to the detachment of active materials from the current collector and hence capacity fading (Fig. 5(a), the capacity retention was 34.15% after 30 cycles at 0.2 C). The appropriate conductive additives (CMK-3 carbon) can confine the p-chloranil from large morphological variation (Fig. 5(e)), and thus the composite cathode exhibited a higher cycling performance (Fig. 5(b), the capacity retention was 70.34% after 200 cycles at 1 C)^[51].

As mentioned above, most reported quinone-based cathodes for ZIBs are regarded to undergo the reversible coordination/de-coordination with Zn²⁺ ions during the discharge–charge process. However, some groups found that apart from Zn²⁺ ions, other ions in electrolytes can also bind with quinones and contribute capacity. Nam et al. believed that the discharge process of a triangular phenanthrenequinone-based macrocycle (PQ-Δ) was involved in the insertion of hydrated Zn²⁺ ions rather than the desolvated Zn²⁺ ions, which was revealed by the Fourier transform infrared spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS) results (Figs. 6(a)–6(c))^[53]. It was believed that the hydrated Zn²⁺ ions decreased the interfacial resistance between the electrolyte and cathode and hence was helpful for high Zn-storage performance, according to the DFT calculations that the co-insertion of H₂O molecules with Zn²⁺ ions lowered the desolvation energy. Moreover, benefiting from the robust triangular molecular structure of PQ-Δ, the inherent low solubility of the discharged/charged states in aqueous electrolytes endowed a long cycle life with high capacity (210 mAh/g and maintaining 99.9% after 500 cycles at 0.15 A/g). On the other hand, in an N, N-dimethylformamide (DMF) solution containing 0.5 M Zn(CF₃SO₃)₂, it is reported by another group that desolvated Zn²⁺ ions could coordinate with the carbonyl groups in PQ-Δ^[54].

In view of the smaller size of H⁺ ion than that of Zn²⁺ ion and the previous report that the MnO₂-based cathode for ZIBs displayed insertion of H⁺ ions^[55], it is reasonable to speculate that H⁺ ions in aqueous electrolytes can also be inserted into organic electrodes for ZIBs with the co-insertion of Zn²⁺ ions. Such a phenomenon was discovered in a quinone-based cathode (sulfur heterocyclic quinone dibenzo[b,i]thianthrene-5,7,12,14-tetraone, DTT) for ZIBs recently^[33]. The comparison of cyclic voltammetry (CV) tests of DTT in 1 M H₂SO₄ and 2 M ZnSO₄ (Fig. 6(d)) indicated that the DTT cathode in aqueous ZIBs can store H⁺ ions and Zn²⁺ ions to form DTT₂(H⁺)₄(Zn²⁺). DFT calculations verified such mechanism. It was believed that one Zn²⁺ ion was coordinated with two carbonyls and four H⁺ ions reacted with four carbonyls in two adjacent DTT molecules to form phenolic hydroxyls (the six-electron transfer for two DTT corresponded to a theoretical capacity of 213.8 mAh/g). Such stable structure of the discharged product benefited long-term cycleability. The DTT cathode exhibited high capacity (210.9 mAh/g at 0.05 A/g), fine rate capability (97 mAh/g at 2 A/g), and stable cycling performance (the capacity retention was 83.8% after 23 000 cycles at 2 A/g).

In a short summary, quinones exhibit fine Zn-storage performance with high capacity. However, it is worth noting that the research of quinone-based cathodes in ZIBs is in infancy, and they face some challenges. For instance, the dissolution behavior needs to be overcome by polymerization or compositing with carbon materials to achieve stable cycleability. Besides, the Zn-storage mechanism requires further investigation. And the redox voltages of quinones are relatively low, which restricts the energy density of ZIBs.

Conducting polymers (CPs) are promising electrode materials for secondary batteries due to their high electrical conductivity and reversible redox reactions, which generally involve the anion-insertion in the polymer backbone. They have exhibited fine anion-storage performance as electrodes for metal-ion batteries^{[56, 57]}. Compared to quinones, CPs exhibit higher redox voltage. Reported quinone-based cathodes for ZIBs all showed discharge voltages below 1 V, except p-chloranil (1.1 V)^[51] due to the electron-withdrawing effect of –Cl groups. While PANI-based cathodes for ZIBs all exhibited discharge voltages above 1 V^[58-60]. To date, reported CPs as cathodes for ZIBs include polyaniline (PANI), polypyrrole (PPy), polythiophene (PTh), poly(3,4-ethylenedioythiophene) (PEDOT), poly(p-phenylene) (PPP) and polyindole (PIn).

Among various CPs, PANI is an attractive electrode material due to its fine redox reversibility, easy synthesis from chemical or electrochemical methods, and the stability in air. PANI has exhibited fine electrochemical performance in ZIBs, however, there are still some problems that restrict its further application.

It has been accepted that among three states (leucoemeraldine, pernigraniline, emeraldine) of PANI, the emeraldine form of PANI at the half-oxidized state can be doped (protonated) after acid doping and the resulting emeraldine salt possesses high electrical conductivity; while leucoemeraldine (fully reduced state) and pernigraniline (fully oxidized state) are insulators even when doped with acid (Fig. 7)^{[61, 62]}. When charging, PANI can lose electrons, together with the acceptance of anions from the electrolytes; while reactions are reversed in the discharge process. Furthermore, it is believed that the protonation of PANI after acid doping is essential for electrochemical activity. However, the Zn anode suffers corrosion in acidic electrolytes. Therefore, it is important to select the pH value of electrolytes to balance the protection of Zn anode from corrosion (high pH value) and the high electrochemical activity of PANI (low pH value). To overcome such a problem and achieve high-performance PANI-based AZIBs, a lot of research has been done, mainly focusing on the modification of PANI.

Forming doped PANI is the most common strategy for improving the electrochemical performance, which can be realized via chemical synthesis (called self-doped PANI) or forming composites with suitable materials that can provide protons. In this case, doped PANI can keep electrochemical activity in weakly acidic electrolytes, even neutral or basic electrolytes.

3.1.1.1. Self-doped polyaniline

It is an effective method to synthesize self-doped PANI by introducing substituent groups (as shown in Fig. 8, e.g., –COOH, –SO₃H, –OH, etc) as proton reservoirs.

For example, three aniline derivatives, o-aminobenzoic acid (o-ABA), m-aminobenzoic acid (m-ABA) and m-aminobenzenesulphonic acid (m-ABS) were employed to synthesize self-doped PANI via electro-polymerization with aniline^{[63, 64]}. They all displayed redox activity in neutral electrolytes. Among them, PANI-co-m-ABS could keep electrochemical activity even in basic electrolytes with pH > 10. Shi et al. further proposed the reduction routes (Fig. 9(c)) of PANI-co-m-ABS (also called PANI-S)^[65] according to the two reduction peaks in the differential capacity curves (Fig. 9(a)). The –SO₃H group in the copolymer chain served as a proton reservoir because it could be ionized to –SO₃^– in 1 M ZnSO₄ electrolyte, and thus provided a local acidic environment around the polymer without changing the overall pH value of the electrolyte, maintaining the electrochemical activity of PANI-co-m-ABS. Accordingly, the copolymer cathode exhibited superior performance with high rate capability (184 mAh/g at 0.2 A/g, 130 mAh/g at 10 A/g) and stable cycling performance (the capacity retention was 84.6% after 2000 cycles at 10 A/g), which outperformed PANI in the same electrolyte (Fig. 9(b)). Undoubtedly, by polymerizing aniline with 5-aminosalicylic acid, which contains both –OH group and –COOH group, the co-polymer, poly(aniline-co-5-aminosalicylic acid) (PANI-co-5-ASA), also exhibited good electrochemical activity in electrolytes with higher pH values^[66].

It was reported that the meta-substituted ABA-based PANI derivatives showed higher electrochemical properties than the ortho-substituted counterparts^[63]. Mu et al. applied two kinds of hydroxyl-substituted anilines (aminophenol) to co-polymerize with aniline to obtain PANI-co-o-aminophenol^[67] and PANI-co-m-aminophenol^[68], respectively, and utilized these two co-polymers as cathodes for AZIBs. The hydroxyls in the copolymer chain could not only tune the pH values around the cathodes, but also can undergo redox process. On account of this, the two co-polymers both exhibited higher capacity than PANI in the same electrolytes and could keep electrochemical activity in the electrolytes with higher pH values (> 4). Moreover, the meta-OH-substituted PANI derivative^[68] showed better Zn-storage performance (137.5 mAh/g at 0.5 mA/cm²; 94.5 mAh/g at 5 mA/cm²) than its ortho-substituted analogue (103 mAh/g at 0.5 mA/cm²; 64.7 mAh/g at 5 mA/cm²)^[67].

Apart from the co-polymerization of anilines with the substituted anilines, other redox-active monomers with proton-supply ability are also taken into consideration to synthesize self-doped PANI. PANMTh, which was synthesized by electrochemical copolymerization of aniline with N-methylthionine (also called as azure C), exhibited fine electrochemical activity in aqueous electrolytes with pH value of 10^[69], which was attributed to the phenothiazine unit in N-methylthionine. The reversible redox of phenothiazine in neutral even basic aqueous electrolytes would undergo proton exchange between the electrode and the electrolytes, and thus endowed PANMTh with proton-self-supply ability, leading to its less dependence on the pH value of electrolytes. Moreover, higher capacity could also be expected due to the possible contribution from the phenothiazine. Consequently, this copolymer electrode for AZIBs showed high capacity (146.3 mAh/g at 1 mA/cm²) and fine cycleability (the capacity retention was 99.4% after 150 cycles at 2 mA/cm²)^[70]. Another phenothiazine derivative, azure B, was also reported to form a copolymer with aniline. The resulting copolymer, poly(aniline-co-azure B) (PANAB), possessed good electrochemical activity at high pH values and exhibited enhanced performance with fine rate capability (127.8 mAh/g at 1 A/g)^[71].

3.1.1.2. Mixing polyaniline with materials of proton-sup ply ability

In addition to the chemical synthesis, mixing PANI with materials that possess proton-supply ability is also effective to form the composite cathodes that are less dependent on the pH values of electrolytes. For example, poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) is an electrically conductive polymer, where the –SO₃H group in PSS can function as a proton reservoir and provide enough H⁺ ions for the protonation of PANI (Fig. 10(c))^[72]. The PANI cathodes mixed with PEDOT:PSS and carbon nanotubes (CNTs) (Fig. 10(a)) possessed enhanced electrical conductivity and proton-self-supply ability. Thus, superior rate capability (238 mAh/g at 0.2 A/g; 145 mAh/g at 10 A/g) and long cycle life (Fig. 10(b), the capacity retention was 77.9% after 1500 cycles at 10 A/g) were achieved. Besides PEDOT:PSS, graphene oxide (GO) was also favorable for the protonation of PANI, because the abundant hydroxyl groups of GO can serve as a proton reservoir^{[73, 74]}. The PANI-GO composite electrodes exhibited higher capacity and cycling performance than those of PANI^[73]. Compared to PANI-CNT, the PANI-GO composites with CNT as conductive additive also showed enhanced Zn-storage performance including capacity (233 mAh/g at 0.1 A/g; 100 mAh/g at 5 A/g) and cycling performance (the capacity retention was 78.7% after 2500 cycles at 3 A/g)^[74].

The doping of PANI can also be achieved by using other materials. For example, after the electro-deposition of PANI on carbon cloth (CC), soaking them into aqueous solution of K₃[Fe(CN)₆], and CC-PANI-FeCN composite was formed, where [Fe(CN)₆]^3– ions were reduced to [Fe(CN)₆]^4– ions by PANI during the soak process. Owing to the interactions between –NH⁺–/–NH⁺= groups in PANI and [Fe(CN)₆]^4– ions, the cycling performance of the CC-PANI-FeCN cathode (the capacity retention was 71% after 1000 cycles at 5 A/g) was superior to CC-PANI (capacity retention of 17% at same conditions) without the doping. Besides, fine rate capability (e.g. 162 and 125 mAh/g at 1 and 5 A/g, respectively) was achieved^[75].

The above examples indicated that the electrochemical performance of PANI not only relies on its electrochemical activity, but also can be influenced by its electrical conductivity. In addition to the method by synthesizing doped PANI, another effective strategy for improving the electrical conductivity of PANI is mixing PANI with special conductive materials.

As one of the most efficient ways, PANI could be synthesized on the conductive materials during the electrochemical polymerization process, and thus the composite electrode with enhanced electrical conductivity can be directly used as electrodes for ZIBs. Different conductive substrates (Pt, carbon cloth (CC)^[76], carbon rod^[77], graphite^[78], Ni foam^[79]) have been adopted for the electro-polymerization of PANI. For example, the composites of PANI with graphite enhanced the electrical conductivity, facilitated the diffusion of dopant anions and increased the capacity^[78]. Notably, compared to dense materials, porous conductive materials are superior as substrates because the resulted porous composite electrodes facilitate ionic transport. Li et al. achieved a 3D hierarchical electrode architecture via the electro-polymerization of PANAC on a CC-porous carbon scaffold^[76]. The continuous porous conductive scaffold rendered fast ionic and electronic transport, and thus the CC-porous carbon@PANAC electrode showed higher capacity and rate capability than those of Pt plate@PANAC. In addition to carbon materials, other porous materials are also helpful for the fast reaction kinetics. With porous Ni foam as substrate, PANI^[79] showed high rate capability (183.28 mAh/g at 2.5 mA/cm²). Through in-situ chemical polymerization of PANI on carbon felts, the obtained PANI vertical nanostructure was favorable for the fast charge transfer^[80]. The composite cathodes exhibited a high capacity of 200 mAh/g at 0.05 A/g with an average voltage of 1.1 V. Interestingly, energy dispersive X-ray spectroscopy (EDS) and ex situ XPS tests indicated that the Zn²⁺ ions and CF₃SO₃^– ions both participated in the redox process of PANI using 1 M aqueous Zn(CF₃SO₃)₂ as electrolyte (Figs. 11(a) and 11(b)). Zn²⁺ ions can interact with –N= in half-oxidized PANI during discharge with the acceptance of electrons, while during charge process, the –NH– in half-oxidized PANI could transform to =NH⁺-, along with the loss of electrons and interaction with CF₃SO₃^– ions (Fig. 11(c)). High rate capability (95 mAh/g at 5 A/g) and superior cycling performance (the capacity retention was 92% after 3000 cycles at 5 A/g) were achieved. The high rate performance may be due to the surface-controlled electrochemical behavior.

Furthermore, it has been reported that the functionalization of carbon materials via pre-treatment or in situ process during polymerization of PANI is helpful to strengthen the interactions between carbon materials and PANI. For example, electron-cyclotron-resonance (ECR) plasma treatment is an effective method to oxidize the surface of carbon materials homogeneously. After the treatment of ECR O₂ plasma, the surface of carbon fibers (CF) had oxygen-containing functional groups (–OH, –COOH, –C=O). The functional groups enhanced the interactions with the deposited PANI, when such carbon fibers were used as substrates^[81]. Besides, the surface of CF became rough, which increased the contact area between PANI and CF. Thus, the obtained composites (PANI-OCF) possessed higher structural robustness, leading to higher cycling performance than that of PANI-CF.

The aforementioned methods, including doping of PANI and mixing with conductive materials, are mainstream ways for improving the Zn-storage performance of PANI. Beyond these, the synthesis conditions of PANI, such as the control of monomer concentration^{[58, 82]} or the ratio of different monomers for co-polymerization^[63], also affected the electrochemical performance of PANI in AZIBs.

In view of the fine Zn-storage performance of PANI and the rise of flexible electronics, flexible AZIBs based on PANI have attracted increasing attention based on gel polymer electrolytes (GPEs) recently. A sandwich-like flexible ZIB was fabricated with polyacrylamide (PAM)-ZnSO₄ GPE, nanowire array PANI cathode and nanosheet array Zn anode^[82]. Benefiting from the nano-array electrode structure, shortened ionic diffusion length promoted the fast ionic transport, and thus facilitated high capacity and rate capability. Considering the non-interfering preparation of GPE and electrodes, all-in-one ZIBs could be made by integrating the GPE and electrodes. The GPEs are beneficial for improving the interfacial contact and thus contributing to durability under bending, twisting or compressing^[82-84]. Zhang et al. fabricated an all-in-one PANI-based ZIB via a series of procedures^[84], which included the spread of three kinds of suspensions in a Teflon mold (in the sequence of PANI/reduced graphene oxide (rGO) suspension, carbon nanofiber (CNF) suspension and Zn/exfoliated graphene (EG) suspension), freeze drying and subsequent injection of polyvinyl alcohol (PVA)-based GPE. Such all-in-one ZIB achieved tight contact between the electrode and electrolyte, and the continuous porous network benefited fast ionic and electronic transport, leading to good fast charge-discharge ability (79.5 mAh/g at 2 A/g) and stable cycling (the capacity retention was 94.6% after 500 cycles at 1 A/g). Owing to the abundant hydroxyl groups and corresponding hydrogen bonds in the PVA gel, the PVA-based GPEs fabricated via a facile freeze/thaw method possessed a self-healing function^[85]. By integrating the cathode, anode and separator during the preparation of GPE, close contact and good interfacial stability were achieved between GPE and electrodes. After being broken and reconnected, the hydrogen bonds at the interface healed the function of the electrolyte and electrodes (Fig. 12(a)), and thus the ZIB could recover, which could even retain its original performance (Figs. 12(b)–12(d)). Impressively, Bi et al. found that the all-in-one PANI-based ZIBs possessed photo-detection ability when using transparent polyethylene terephthalate (PET) substrate^[86]. Photocurrent could be detected under a white light with energy that was higher than the energy band of PANI. Such a novel bi-functional device broadens the application range of flexible electronic products.

In summary, PANI is validated to be a superior Zn-storage electrode material with fine performance. However, the electrochemical activity of PANI highly depends on the pH value of electrolytes, which should be acidic and hence helpful for the easy protonation. Unfortunately, under such circumstances, the Zn anode faces corrosion. Extensive efforts have been reported to solve this challenge and effective strategies include the doping of PANI by chemical synthesis or mixing with proper materials. Besides, the capacity of PANI is relatively low and limited by the doping level. Moreover, the energy storage mechanism of PANI in AZIBs needs further investigation. Recently, Zhao et al. reported a novel poly(1,5-naphthalenediamine, NAPD) cathode for ZIBs (the repeating unit was naphthalene for poly(1,5-NAPD) and benzene for PANI; both linkages are –NH– or =N–)^[87]. The authors claimed that Zn²⁺ ions and H⁺ ions both contributed to the capacity. Simultaneously, they also believed that SO₄^2– anions could be inserted into the poly(1,5-NAPD), which can be inferred by the one more reduction peak at high voltage in the ZnSO₄-based electrolyte than in the Zn(CF₃SO₃)₂-based electrolyte. The mechanism is different from the previous reports on PANI and probably needs further investigation.

Apart from PANI, many other CPs (Fig. 13) have also demonstrated fine Zn-storage performance, such as polypyrrole (PPy)^[88-93], polythiophene (PTh)^[94], poly(3,4-ethylenedioythiophene) (PEDOT)^{[95, 96]}, polyindole (PIn)^{[97, 98]} and poly(p-phenylene) (PPP)^[99].

PPy is known as a bipolar CP, and both cations and anions could be inserted into PPy. Grgur et al. believed that the energy storage mechanism of PPy cathode for ZIBs was the insertion of anions from the electrolytes^[93]. The PPy cathode exhibited similar capacity in 0.1 M HCl and 0.1 M NH₄Cl (pH = 5), which indicated the low dependence on pH values of the electrolytes. However, another group claimed a different mechanism by utilizing the p-toluenesulfonic anions (pTS^–) doped PPy as a cathode for ZIBs^[91]. They claimed that due to the large size of anions, the energy storage mechanism of PPy/pTS^– relied on the cation exchange. During the reduction, the adsorption of cations balanced the pTS^– dopants; while during oxidization, the loss of electrons accompanied with the desorption of cations. The adsorption/desorption of cations can be monitored by the mass change of the electrodes. In this case, the 0.1 M NaCl electrolyte showed higher capacity than other electrolytes (0.1 M phosphate buffered saline (PBS), and 0.1 M simulated body fluid (SBS)), probably due to the smallest size of Cl^– anions in NaCl solution that had the weakest electro-static interaction with the cations in electrolytes. The weak interaction led to fast cation transport, which was favorable for high capacity. However, it was not clear if the mechanism was applicable to the electrolytes containing Zn²⁺ ions. Lahiri et al. investigated the Zn-storage performance of PPy in two kinds of electrolytes, of which the solutes were both ZnAc₂^[92]. The difference was that one solvent was water and the other was a mixture solvent containing choline acetate (ChAc) and water. In situ Raman spectra indicated that the coordination of Zn²⁺ ions with PPy was responsible for the capacity. PPy cathode suffered capacity decay in both electrolytes due to the phase transition that was caused by the insertion of Zn²⁺ ions in the polymer cathode. Notably, the capacity of PPy in the electrolyte with water as the solvent was lower than that in the electrolyte with a mixed solvent. The result was attributed to the ChAc that reduced the strain during the insertion of Zn²⁺ ions.

Normally, the cycleability of PPy electrode was inferior^[88] unless being composited with aerogel^[90]. Wang et al. prepared PPy on a PET substrate by using electrodeposition (Fig. 14(a)) for flexible AZIBs, in which PVA-based GPE was the electrolyte. The transparent Zn//PPy AZIBs showed different colors at different charged/discharged voltages (as shown in Fig. 14(b): black for 1.2 V, yellow for 0 V) owing to the electro-chromic function of PPy^[88]. Such unique property opened a door for the monitor of battery safety. However, PPy suffered severe capacity decay (the capacity retention was 38% after 200 cycles at 8.8 A/g). Accordingly, the electro-chromic function of the PPy cathodes became weak after tens of cycles, which can be verified by the color of the battery at 0 V after different cycles (Fig. 14(c)). On the other hand, with the PVA-based aerogel as a scaffold, the PPy composite cathode showed better cycling stability (the capacity retention was 76.7% after 1000 cycles at 8 A/g) than the previous work, which may be due to the confining effect of the PVA-based aerogel to PPy^[90]. Besides, benefiting from the porous structure of the scaffold and the nanofiber morphology of PPy, fast ionic transport was achieved in the PPy/aerogel, contributing to high rate capability (151.1 mAh/g at 0.5 A/g and 87.6 mAh/g at 16 A/g).

Similar to PANI and PPy, PIn is also a N-heterocyclic CP. It shows less dependence on the pH value of the electrolytes than PANI. In the 1990s, Pandey et al. reported PIn cathode for ZIB with 1 M ZnSO₄ electrolyte^[97]. It delivered a maximum capacity of ~90 mAh/g in the voltage range from 0.75 to 1.45 V through a one-electron transfer process, and underwent reversible charge–discharge process for 50 cycles with CE close to 100%. Cai et al. also reported a Zn//PIn secondary battery with aqueous ZnCl₂ as electrolyte^[98]. With the doping and dedoping of the Cl^– anions during the charge–discharge process, the voltage of the battery could reach 2 V and the PIn cathode delivered high rate capability (80 mAh/g at 0.02 mA/cm² and 60 mAh/g at 0.1 mA/cm²). It also showed good cycleability with slow capacity fading (2.0% after 200 cycles) and thus small energy density decay (3.5% after 200 cycles). Besides, the PIn cathode exhibited high pH durability: the area of the CV curve in solutions with pH of 10 was 80.3% of that in solutions with pH of 6. Later, poly(5-cyanoindole) fibers via the electro-spinning technique were developed as the cathode for the ZIBs^[100]. The battery achieved 2.0 V electromotive force and showed high rate performance with slow capacity decay (4% after 360 cycles at 0.2 C).

PTh, a S-heterocyclic CP, is also a potential cathode material for ZIB. The energy storage mechanism of PTh in ZIBs is considered as the doping/dedoping of anions from the electrolytes. The Zn//PTh battery showed an average discharge voltage of 1.2 V with 0.1 M Zn(ClO₄)₂ and 1 M LiClO₄ in propylene carbonate electrolyte and ClO₄^– anions could insert/extract into/from PTh reversibly during cycling^[94]. However, the coulombic efficiency (CE) decreased with the increase of discharge rate (98% at 100 µA/cm²; while 60% at 480 µA/cm²). A PTh derivative, poly(3,4-ethylenedioythiophene) (PEDOT), as a cathode for ZIBs showed fine CE in 1-ethyl-3-methylimidazolium dicyanamide [C2mim][dca] ionic liquid (IL) electrolyte^[95]. The high CE was ascribed to the inhibition of dendrite growth of metal anode in IL, although the PEDOT-based batteries still suffered capacity fading. Similar dendritic inhibition in IL was observed in Zn//PPP battery^[99]. With polystyrene (PS) spheres assembled on the Zn sheets and the usage of IL as electrolytes, the growth of Zn dendrite was inhibited during cycling, which contributed to high cycleability (the capacity retention was over 90% after 300 cycles at 1 C) of the Zn//PPP battery. The PPP cathodes exhibited high redox voltage (discharge: 1.2 V; charge: 1.7 V); however, the low capacity (48 mAh/g at 0.2 C) would lead to a low energy density. Compared to IL, poly(ionic liquid) (PIL) iongel possesses similar physicochemical properties but without the risk of leakage. A solid-state Zn//PEDOT battery was fabricated with a PIL, i.e., poly (diallyldimethylammonium bis(trifluoromethanesulfonyl)imide) [p(DADMATFSI)] and 1-ethyl-3-methylimidazolium dicyanamide [emim][dca]^[96]. It showed a wide electrochemical window (–0.9 to 3.1 V vs. Zn/Zn²⁺) and high ionic conductivity (1.1 × 10^–2 S/cm at 50 °C).

Poly(acetylene) (PAc) is the first reported CP and its application as electrode materials for secondary batteries can date back to 1980s^[101]. Häupler et al. reported a PAc-based polymer as cathodes for ZIBs, in which the side chain was 9,10-di(1,3-dithiol-2-ylidene)-9,10-dihydroanthracene (exTTF)^[102]. The polymer cathode demonstrated a rapid charge–discharge ability and an improved electron transfer owing to the incorporation of redox-active moiety into the PAc backbone. Thus, high rate capability (100 mAh/g at 10 C; 47 mAh/g at 120 C) and good cycling performance were achieved in the voltage range of 0.6–1.7 V. However, the active material content in the electrode was low (50%), leading to low energy density.

In short, CPs are endowed with advantages of relatively high redox voltage as cathodes for ZIBs. Besides, fine rate capability could be expected owing to the high electrical conductivity; which, however, is affected by the doping level and the chemical environment (e.g. the pH value of the electrolytes). Furthermore, the capacity is also limited by the doping level. Thus, the modification of CPs needs further investigation for high-performance ZIBs.

In addition to the aforementioned organic/polymeric electrode materials, there are also many other redox-active organic materials that can be used as electrodes for ZIBs.

As representative carbonyl-based materials, quinones have been demonstrated as fine Zn-storage electrode materials and were studied widely. Other materials containing carbonyl groups were also reported. For example, anhydrides and imides are also potential cathode materials for ZIBs due to the redox activity of carbonyls. Wang et al. first reported 1,4,5,8-naphthalenetetracarboxylic dianhydride (NTCDA) and 1,4,5,8-naphthalene diimide (NTCDI) as cathodes for ZIBs^[103]. NTCDA showed two pairs of redox peaks (discharge: 0.37, 0.58 V; charge: 0.54, 0.88 V) and suffered capacity fading. The peak at 0.37/0.54 V was identified to the insertion/extraction of hydronium in NTCDA. The poor cycling performance may be due to the structural damage upon the insertion of Zn²⁺ ions, which could be indicated by the new peak in the X-ray diffraction (XRD) pattern of the discharged NTCDA. NTCDI exhibited only one pair of redox peaks (discharge: 0.45 V; charge: 0.86 V), but higher capacity and rate capability (240 mAh/g at 0.1 A/g; 140 mAh/g at 2 A/g) than those of NTCDA (less than 180 mAh/g at 0.1 A/g and less than 20 mAh/g at 2 A/g). Besides, a better cycleability was achieved with a capacity retention of 73.7% after 2000 cycles at 1 A/g.

Apart from the carbonyl compounds, C=N groups containing materials are also expected to possess Zn-storage capacity due to the redox of C=N groups. For example, diquinoxalino [2,3-a:2’,3’-c] phenazine (HATN) and a HATN-based polymer have exhibited fine energy storage performance in LIBs^[104], sodium-ion batteries, and potassium-ion batteries^[105] and the active site was identified to the C=N groups, which could undergo reversible transformation to C-N group with the insertion of alkali metal-ions. Encouragingly, the Zn-storage performance of HATN is also superior with high capacity (370 mAh/g at 0.1 A/g; 123 mAh/g at 20 A/g) and good long-term cycleability (the capacity retention was 93.3% after 5000 cycles at 5 A/g)^[106]. Such outstanding performance was ascribed to the unique H⁺ coordination/de-coordination at C=N groups of HATN, which was revealed by ex situ spectra (FTIR, Raman, XPS) that the peaks of C=N disappeared while those of C–N and N–H enhanced during the discharge process. Fast kinetics of coordination/de-coordination of small H⁺ ions and the π-conjugated structure of HATN benefited good rate capability, leading to high energy density of the battery.

In view of the variety of organic materials, a series of bio-materials with C=N and C=O groups, riboflavin (RF), alloxazine (ALX) and lumazine (LMZ), were reported as electrodes for ZIBs^[107]. Among them, RF showed the lowest capacity (113.5 mAh/g at 0.03 A/g) but the highest cycling stability (the capacity retention was 92.7% after 5000 cycles at 5 A/g).

Organic materials containing C=O or/and C=N groups are both considered as cation-insertion electrode materials for ZIBs, and they have exhibited high capacity but the voltage is relatively lower than the anion-insertion electrode materials. In addition to conducting polymers, arylamine compounds can also be utilized as anion-insertion cathodes for ZIBs. For example, Glatz et al. reported a 1,4-bis(diphenylaminobenzene) (BDB) cathode for ZIBs in a high-concentrated electrolyte (19 M LiTFSI and 1 M Zn(CF₃SO₃)₂)^[108]. The BDB cathode exhibited two pairs of redox peaks with a capacity of 112 mAh/g at 3 C (1 C = 0.13 A/g), corresponding to the insertion/extraction of two anions (TFSI^– and CF₃SO₃^–) in each BDB molecule. The problem was that such a small molecular cathode material suffered capacity fading due to the dissolution of redox products in electrolytes. With a cellculose nanocrystal membrane as a protection layer on the cathode, fine cycling performance was achieved with a capacity retention of 82% after 500 cycles at 3 C. Besides, radical polymers are another kind of anion-insertion electrode materials and possess high redox voltage. Poly(2,2,6,6-tetramethylpiperidinyloxy-4-yl vinyl ether) (PTVE) cathode for ZIBs underwent a reversible one-electron redox process in aqueous electrolytes containing 0.1 M ZnCl₂ and 0.1 M NH₄Cl^[109]. The PTVE cathode displayed a plateau voltage at 1.73 V (vs. Zn/Zn²⁺) with discharge capacity of ~131 mA h/g at 60 C (~8 A/g) and could maintain 65% of the initial capacity after 500 cycles. In view of the anion-insertion mechanism of PTVE, it can be speculated that the anions may affect the electrochemical performance. Thus, Luo et al. further investigated the effects of three anions (SO₄^2–, CF₃SO₃^–, and ClO₄^–) on the electrochemical performance of the Zn//PTVE^[110]. The PTVE cathode delivered capacities of 58, 52, and 50 mAh/g with average voltages of 1.77, 1.58, and 1.53 V at 10 A/g in 1 M ZnSO₄, Zn(CF₃SO₃)₂, and Zn(ClO₄)₂ electrolytes, respectively. The highest voltage in ZnSO₄ electrolyte was attributed to the highest binding energy between SO₄^2– anions with PTVE according to DFT calculations. Besides, a better cycleability (with a capacity retention of 77.0% after 1000 cycles at 1 A/g) was achieved in 1 M Zn(CF₃SO₃)₂ electrolyte than those in the other two electrolytes. Molecular electrostatic potential plots indicated that CF₃SO₃^- anions possessed well-delocalized electronic structure, which was favorable for the dissociation from the PTVE host and thus contributed to the good cycling stability. However, it should be noted that the capacities in these electrolytes were much lower than the previous work in aqueous electrolyte containing 0.1 M ZnCl₂ and 0.1 M NH₄Cl, which may be due to the smaller ionic size of Cl^- anions and needs further investigation.

The investigation of organic electrode materials for ZIBs is appealing due to the flexibility, eco-friendliness, and designable molecular structure of organic compounds. Organic electrode materials have exhibited fine Zn-storage performance (as shown in Table 1). However, the research of OZIBs is still in infancy and the further development is hampered by many issues. Aimed at these challenges, various strategies have been proved to be effective. The challenges and corresponding solutions are specifically summarized as follows:

(1) The dissolution behavior: Small molecular organic materials often suffer capacity decay due to the dissolution of discharged products in electrolytes. Fortunately, such an issue can be alleviated by many methods. Polymerization is the most effective way to inhibit the dissolution of small molecules. Applying functional separators^{[111, 112]} (e.g. the Nafion membrane^[34]) or compositing with carbon materials also helps. Besides, using solid electrolytes is another potential method, though they often face the challenges of low ionic conductivity and the reported solid-state ZIBs are mostly based on GPEs.

(2) The unclear energy storage mechanism: The Zn-storage mechanism is essential for further enhancing the performance OZIBs (particularly in aqueous electrolytes due to the various ions) and needs further exploration. For example, the Zn-storage mechanism of carbonyl compounds is considered as the coordination of Zn²⁺ ions with carbonyls, but some reports showed that quinones could also bind with H⁺ ions (e.g. DTT) or hydrated Zn²⁺ ions (e.g. PQ-Δ) during discharge.

(3) Inherent restrictions resulting from the organic electrode materials: Various organic materials have been validated to be promising cathodes for ZIBs and they have exhibited fine performance. However, they often suffer their own shortcomings. For example, quinones possess high capacity due to the abundant Zn-storage active sites, but the redox voltage is not high. The potential could be elevated to a certain extent, by introducing electron-withdrawing groups. Furthermore, the low electrical conductivity of quinones restricts their rate capability. It seems that the CPs with high electrical conductivity and high voltage are more promising as cathodes for ZIBs. But the capacity of CPs is low. Particularly, the acidic electrolytes are necessary to guarantee the electrochemical activity of PANI, but such electrolytes would cause Zn corrosion.

(4) The universal problems in ZIB, such as the low electrochemical window of aqueous electrolytes, the corrosion of Zn metal and the possible growth of Zn dendrites, are out of the scope of this review and will not be further illustrated here.

In conclusion, rapid development of OZIBs is expected if the above challenges can be solved well. We hope this review can provide insight into the development of high performance OZIBs.

This work was supported by the National Natural Science Foundation of China (No. 51773071), the National 1000-Talents Program, Innovation Fund of WNLO and the Fundamental Research Funds for the Central Universities (HUST : 2017KFYXJJ023, 2017KFXKJC002, 2018KFYXKJC018, and 2019kfyRCPY099).