A Universal Thick Anode for Aqueous and Seawater Energy Storage Devices

Abstract

Aqueous and seawater energy storage devices hold great potential for electrical grids application due to safety, affordability, and sustainability. However, their broader deployment has been constrained by the absence of a durable thick anode. Here, the first universal thick anode operating stably across 15 simple‐ion and 3 complex‐ion systems, including nonmetallic (H⁺, NH₄ ⁺), monovalent (Li⁺, Na⁺, K⁺), multivalent ions (Zn²⁺, Ca²⁺, Mg²⁺, Al³⁺), and seawater ions (>5 cations) is reported. Composed of polymer nanosheets and carbon nanotubes, this anode supports thick electrode fabrication (e.g., 100 mg cm⁻² and 1 mm) with low porosity/tortuosity, superior electrical conductivity, mechanical robustness, and chemical stability. Consequently, it achieves exceptionable cycle life (up to 380 000 cycles) in supercapacitors and ultrahigh areal capacities (6.5 mAh cm⁻²) in batteries, even under practical/extreme conditions, attributed to the formation of a water‐scarce, cation‐rich electrical double‐layer structure, as revealed by simulations. Compatible with sea salt‐based electrolytes and paired with a metal‐free cathode, the anode enables seawater batteries with thousands‐cycle life and high energy/power density. Of universal ion storage, ultrahigh‐loading capability, unlimited resources, and cost‐effectiveness, this polymer electrode is promising for practical aqueous (seawater) energy devices.

A universal thick anode composed of polymer and carbon nanotubes has been developed, demonstrating stable operation across 15 simple‐ion and 3 complex‐ion systems. It achieves exceptional cycle life in supercapacitors, ultrahigh areal capacities in batteries, and high compatibility with seawater electrolytes. With ultrahigh‐loading capability (100 mg cm⁻²) and cost‐effectiveness, this thick electrode is promising for practical aqueous (seawater) energy devices.

1. Introduction

Nonaqueous lithium‐ion batteries (LIBs) are prevalent in portable electronics and electric vehicles. However, issues related to safety, resources, and the environment hinder their deployment for grid‐scale energy storage.^{[ 1} , ^{2 ]} Among the alternatives, aqueous batteries and supercapacitors hold substantial promise due to their inherent safety and low cost.^{[ 3} , ^{4 ]} Nevertheless, the limited voltage window imposed by water decomposition (theoretical, 1.23 V) and overpotentials for hydrogen/oxygen evolution reactions (HER and OER, typically 0.3–0.8 V) restrict the range of viable electrode materials (≈2.0–4.5 V vs Li⁺/Li), in contrast to broader options of nonaqueous ones (0–5 V vs Li⁺/Li).^{[ 5} , ^{6 ]} While water‐in‐salt electrolytes can extend the voltage window and expand material choices, their cost remains a significant barrier to widespread adoption.^{[ 7 ]} Over the past decades, various materials—including metals, metal oxides, hydroxides, chalcogens, halogens, Prussian blue analogs, polyanions, MXenes, organics, and carbons—have been explored as electrode candidates for aqueous energy storage.^{[ 3} , ^{8 ]} However, few have demonstrated sufficient cycle life to meet the rigorous demands of grid applications. For instance, lead‐based acidic batteries exhibit limited cycle life due to the formation of a lead sulfate passivation layer on the anode.^{[ 9 ]} Rechargeable alkaline batteries encounter challenges including metal anode dissolution, dendritic growth, and substantial volumetric changes in metal hydride anodes during cycling.^{[ 10 ]} Near‐neutral aqueous batteries, including lithium‐ion, sodium‐ion, and zinc‐ion systems face issues such as active material dissolution, corrosion, dendrite formation, and oxygen passivation,^{[ 11} , ^{12 ]} which collectively constrain the use of metal‐containing anode materials for grid‐scale energy storage.

Seawater batteries have also faced anode challenges. Seawater‐based energy storage devices are ideal aqueous systems for grid‐scale application due to the abundant availability, negligible cost, environmental benignity, and high conductivity of seawater.^{[ 13} , ¹⁴ , ^{15 ]} However, the corrosive chloride ions and complex ionic composition (approximately five cations and five anions) pose significant challenges for developing compatible electrode materials. Efforts to develop fully aqueous and rechargeable seawater energy devices have included the use of Zn alloy anodes with zinc salts in seawater,^{[ 16 ]} activated carbon and MXenes following supercapacitive mechanisms,^{[ 17 ]} and titanium oxide anodes capable of multi‐cation storage.^{[ 18 ]} Nonetheless, these anodes suffer from low utilization, limited capacities, and short cycle life, impeding their practical application.

To address these challenges, it is of great significance to develop a universal anode capable of storing various ions, which would not only eliminate dependency on specific resources but also be potentially compatible with seawater electrolytes, paving the way for sustainable energy storage devices. However, developing such universal anodes is extremely challenging, and relevant reports, e.g., bismuth oxide,^{[ 19 ]} organic catechols,^{[ 20 ]} and quinones,^{[ 21 ]} are exceedingly scarce. Even rarer are studies evaluating full‐cell performance under practical conditions, including low negative‐to‐positive capacity (N/P) ratios, lean electrolyte, and high‐loading (thick) electrodes (to achieve 3–4 mAh cm⁻² areal capacity), let alone ultrahigh‐loading (ultra‐thick) eletrodes to minimize inactive materials and maximize cell‐level performance.^{[ 1 ]} These issues highlight the urgent need for breakthroughs in anode materials and technologies.

Here, for the first time, we report a universal, stable, and thick organic anode compatible with aqueous energy storage devices across 15 simple‐ion chemistries and 3 seawater batteries containing complex cations and anions. Composed of polymer nanosheets and carbon nanotubes (CNTs), this anode can be made into ultra‐thick electrodes exceeding 100 mg in mass and 1 mm in thickness. Thick electrodes exhibit favorable structural characteristics, such as low tortuosity and uniform component distribution, along with outstanding physical and (electro)chemical properties, including excellent electrical conductivity, robust mechanical compressibility, low electrolyte solubility, HER passivation capability, and record‐high areal capacity (11.1 mAh cm⁻²). Consequently, this anode‐based supercapacitors achieve unprecedented cycle life of up to 0.38, 0.2, 0.15, and 0.2 million cycles in NH₄ ⁺, K⁺, Mg²⁺, and Ce³⁺ ion systems, respectively. Even under high loadings (30 mg cm⁻²) and low N/P ratios (1.2), the anode enables practical Zn²⁺ ion capacitors with a 20 000‐cycle lifespan. In battery applications, the combination of thick anode with triflate electrolyte demonstrates exceptional performance in full cells under practical conditions, achieving ultrahigh areal capacities and ultralong cycle life in lithium (6.5 mAh cm⁻2, 2400 cycles), ammonium (2.73 mAh cm⁻², 7720 cycles), zinc (3.6 mAh cm⁻², 3300 cycles), calcium (1.4 mAh cm⁻², 5000 cycles), and aluminum (1.4 mAh cm⁻², 1000 cycles) ion systems, surpassing nearly all previous records. This unparalleled performance is attributed to the formation of a favorable electrical double‐layer structure with a water‐scarce, cation‐rich inner Helmholtz plane, as revealed by molecular dynamics simulations. The universal anode was further paired with a rationally selected metal‐free cathode for seawater batteries, delivering long life (up to 7000 cycles) and high energy/power densities. Featuring universal ion storage, ultrahigh‐loading capability, and ultra‐stable cycling characteristics, this polymer anode is promising for aqueous and emerging seawater‐based energy storage devices.

2. Results and Discussion

2.1. Making of Polymer Material, Electrode, and Structural Characterization

The polyimide (PI) derived from 3,4,9,10‐perylenetetracarboxylic dianhydride (PTCDA) emerges as a highly competitive candidate for universal anodes in aqueous batteries and supercapacitors, attributed to its suitable redox potential (2.4 V vs Li⁺/Li, Figure S1, Supporting Information) and the presence of multiple carbonyl active sites that follow an enolization‐coordination mechanism (Figure  1A–C).^{[ 22} , ²³ , ²⁴ , ²⁵ , ²⁶ , ^{27 ]} PI and PI/CNT nanocomposites were synthesized via a one‐pot polycondensation reaction using PTCDA and ethylenediamine as monomers, with or without CNT as a conductive substrate. The combination of large π‐conjugated PTCDA and flexible‐chain ethylenediamine was strategically selected to provide both high conductivity and mechanical flexibility, facilitating efficient electron, ion and matter transport/storage. The PI structure was validated by FTIR spectrum and XRD pattern (Figure S2A,B, Supporting Information). UV–vis spectrum and the corresponding Tauc plot revealed a narrow bandgap of 1.66 eV (Figure S2C,D, Supporting Information), suggesting high electrical conductivity. Scanning electron microscopy (SEM) images showed that the composite consists of polymer nanosheets and CNTs, with the latter coated by the former (Figure 1D).

Figure 1.

Material selection, electrode preparation, and structural characterization. A) Potentials of different electrode materials and theoretical water electrolysis, B) Universal ion storage mechanism of PI electrode and C) representative charge‐discharge profiles of PI in different ion systems (NH₄ ⁺, Na⁺, Ca²⁺, and Ce³⁺). D) SEM images of PI/CNT composite powder. E–I) Microstructure characterization: Digital photos (E), cross‐sectional (F,G) and top‐view (H) SEM images and (I) elemental mapping images of a thick PI/CNT electrode, J–M) Synchrotron‐based X‐ray image of PI/CNT electrode (J) and its rendered image of PI (blue), CNT (green), and pores (red) (K), interconnected pores (L), and isolated pores (M).

This composite can be easily dry‐pressed into a free‐standing tablet electrode with high mass loading (20–100 mg cm⁻²) and thickness (0.3–1 mm). For example, Figure 1E shows digital photos of a thick PI/CNT electrode with a dark green color, 100 mg loading, 0.73 mm thickness, and 1.27 cm² area. Thick PI and PI/CNT electrodes with varying PI content were prepared, including PI100, PI95/CNT, PI90/CNT, PI80/CNT, and PI70/CNT, corresponding to 100%, 95%, 90%, 80%, and 70% PI, respectively (Figure S3 and Table S1, Supporting Information). In subsequent discussions, the default PI/CNT electrode is PI80/CNT. Cross‐sectional SEM images reveal a compact stacking of PI and CNT throughout the thick electrode (Figure 1F). High‐resolution SEM further reveals the presence of porous channels and homogeneous distribution of PI and CNT, in which PI nanosheets are supported by a CNT segregated network, forming a smooth surface (Figure 1G,H, Supporting Information). SEM elemental mapping confirms the uniform distribution of carbon, oxygen, and nitrogen elements throughout the thick electrode (Figure 1I).

To further elucidate the material distribution and porous structure within the thick electrode, synchrotron‐based X‐ray transmission imaging was conducted. Based on density differences, three components were distinguished and rendered in different colors (Figure 1J), including PI (high density, blue), CNT (medium density, green), and pores (low density, red). The excellent mix of three colors indicates a homogeneous dispersion of components throughout the thick electrode (Figure 1K; Video S1, Supporting Information). Pore network modeling revealed that porous channels comprise a high proportion of connected pores (95%, purple) and a minor fraction of isolated pores (5%, yellow), with pore sizes centered at 33 µm in the PI/CNT electrode (Figure 1L,M; Figure S4, Supporting Information). Based on tomography images and calculations, the electrode porosity, density, and tortuosity were determined to be 25%, 1.08 g cm⁻³, and 2.5, respectively. The low tortuosity and homogeneous distribution of the active PI, conductive CNT, and interconnected porous channels are advantageous for rapidly transporting electrons, ions, and matter during electrochemical reactions.

2.2. Physical and (Electro)chemical Properties of Polymer Material/Electrode

To gain deeper insight into the compression behavior of PI, FTIR spectra and color changes under varying pressures were recorded using a diamond anvil cell with KBr as the transmitting medium and ruby powder as the pressure indicator (Figure S5, Supporting Information). As pressure increased from 0 to 15 GPa, characteristic peaks, including carbonyl groups (1650 and 1700 cm⁻¹), exhibited a blue shift, indicating weakened chemical bonding due to enhanced electron delocalization and conjugation following high‐pressure treatment (Figure  2A; Figure S6, Supporting Information).^{[ 28 ]} The alteration in electronic structure was further evidenced by a visible color change (Figure 2B), transitioning from red at 0 GPa to violet at 3 GPa and green at 10 GPa. Upon releasing pressure back to 2 and 0 GPa in the KBr medium without air exposure, the polymer demonstrated reversible piezochromic responses and corresponding peak shifts in the FTIR spectra (Figures S6 and S7, Supporting Information),^{[ 29 ]} highlighting the polymer's high compression resistance and structural integrity after pressure treatment.

Figure 2.

Physical and chemical property of PI and PI/CNT. A) FTIR patterns and B) Color changes of PI under different pressure. C) Compression strength‐strain curves of PI/CNTs and D) PI80/CNT under cyclic load. E) Electrical conductivity of PI/CNT electrode with different PI content. F) Soaking of PI pellet in different electrolyte before (above) and after 50 days (below). G,H) Digital photos of PI pellet (50 days) and Zn foil (7 days) after soaking in 2 M ZnSO₄ and I–L) corresponding SEM images: (I–J) PI and (K–L) Zn. Scale bars: (I,K) 10 µm and (J,L) 1 µm. M) HER performance of PI, Zn, and GC in 5 m NaNO₃ and 2 m ZnSO₄. N) Plot of areal capacity (capacity) against mass loadings at 0.04 Ag⁻¹. Inset show photos of PI/CNT electrodes with different mass loadings (24.7, 40.2, 73, and 101.8 mg cm⁻²), scale bar: 2 mm.

The mechanical property of PI and PI/CNT electrodes was further assessed in the compression mode. Stress‐strain curves revealed the brittle nature of the PI pellet, which exhibited a high compression stress of 99 MPa but a low strain of 16%. The incorporation of CNTs enhanced the compression strain in PI/CNT composites, with strains increasing from 17.3% in PI95/CNT to 52.8% in PI70/CNT as the CNT content increased (Figure 2C). To evaluate the reversibility of the compression strain, PI/CNT pellets were subjected to cyclic loading with different strain limits. Results indicated high reversible strain (15%, 26.7%, and 60.6% for PI90/CNT, PI80/CNT, and PI70/CNT, respectively, Figure 2D; Figure S8, Supporting Information), which is advantageous for buffering volumetric changes during battery operation. A free‐fall experiment from a height of 2 m confirmed the mechanical integrity of the PI/CNT electrode, revealing its mechanical robustness (Video S2, Supporting Information).

The electrical conductivity of PI/CNT electrodes was measured by linear scanning voltammetry (LSV), with the electrode sandwiched between two steel plates. The intrinsic conductivity of PI is 4.2 × 10⁻⁴ mS cm⁻¹ (Figure S9A, Supporting Information), which surpasses most reported organic electrode materials and is comparable to commercially available LIB electrodes (Figure S9B and Table S2, Supporting Information),^{[ 30} , ^{31 ]} mainly attributed to the large π‐conjugated structure and narrow bandgap. Conductivity increased with CNT content, ranging from 0.0137 mS cm⁻¹ in PI95/CNT, to 0.4 in PI90/CNT, 0.72 in PI80/CNT, and 1.11 mS cm⁻¹ in PI70/CNT (Figure 2E). The conductivity was further fitted according to the power‐law model of percolation theory.^{[ 32 ]} The percolation threshold was found to be as low as 4.95 wt. % (Figure S10, Supporting Information), indicating a good dispersion of CNT and the formation of an effective conductive network in PI/CNT electrodes.

The chemical properties, including solubility and corrosion, were examined by soaking PI pellets in 23 different electrolytes with varying ions and pH values (Figure 2F; Figure S11, Supporting Information). For comparison, zinc foil was also soaked in the same electrolyte. After 50 days, PI pellets remained intact, and solutions stayed colorless in most electrolytes except 1 m KOH and saturated KNO₃, suggesting the low solubility and high chemical stability of PI in acidic/neutral electrolytes. After extraction from the electrolyte (2 m ZnSO₄), the PI pellet retained its smooth surface (Figure 2G), in stark contrast to the zinc foil, which showed corroded dark pits after just 7 days of immersion (Figure 2H). SEM and energy‐dispersive X‐ray spectroscopy (EDX) analyses of the surfaces provided further insights. While the PI surface exhibited uniformly grown particles, probably heterogeneously nucleated zinc sulfate crystals (Figure 2I,J; Figure S11, Supporting Information), the Zn foil surface was covered with numerous flakes of zinc sulfate hydroxide resulting from corrosion (Figure 2K,L; Figure S12, Supporting Information). In alkaline conditions, hydroxide anion attacks the imide group of PI, forming soluble carboxylic salts with a greenish color, as shown in 1 m KOH (Figure 2G). However, the PI pellet maintained its tablet shape in this solution, indicating its low solubility, probably due to strong π–π and hydrogen bonding interactions within the compressed pellet.

The electrochemical behavior, including HER activity and high‐loading capability of PI/CNT electrodes, was examined. HER activity was assessed using LSV in a three‐electrode configuration (Figure 2M). For comparison, glassy carbon (GC) and Zn foil were tested under the same conditions. In a near‐neutral electrolyte (5 m NaNO₃), PI/CNT, GC, and Zn required overpotentials of 1.375, 1.225, and 1.08 V, respectively, to generate hydrogen at a current density of 2 mA cm⁻², indicating the lowest HER activity of PI/CNT. This behavior was further confirmed in strong/mildly acidic electrolyte (1.8 m H₂SO₄ and 2 m ZnSO₄, Figure S13, Supporting Information), revealing the HER‐passivation property of PI,^{[ 33 ]} which could reduce side reactions from water decomposition. On the other hand, thick PI/CNT electrodes were evaluated in half‐cells, with four mass loadings investigated: 24.7, 40.2, 73, and 101.8 mg cm⁻² (Figure 2N; Figures S14 and S15, Supporting Information). As mass loading increased, a linear rise in areal capacity was observed, with maximum output capacity reaching 11.1 mAh cm⁻² (14.09 mAh). Such high mass loadings and areal capacities achieved are unprecedented for organic electrodes (Figure S16 and Table S3, Supporting Information) and far surpass commercial LIBs (3–4 mAh cm⁻²), highlighting the superior ultrahigh‐loading capability of PI/CNT electrodes. Overall, these physical, chemical, and electrochemical properties of PI/CNT electrodes are beneficial for their applications in aqueous energy storage.

2.3. Universal Polymer Anode for Aqueous Supercapacitors

The ion storage capability of the PI electrode was initially investigated in 23 types of electrolytes containing various cations (H⁺, NH₄⁺, Li⁺, Na⁺, K⁺, Zn²⁺, Ca²⁺, Mg²⁺, Sr²⁺, Ba²⁺, Ce³⁺, and Al³⁺) and anions (NO₃⁻, SO₄²⁻, Cl⁻, and OH⁻) within a three‐electrode configuration. Results demonstrated consistent charge/discharge profiles and capacities close to the theoretical values across different ion systems, suggesting the PI electrode's universal cation storage capability and diverse anion compatibility (Figures S17–S20, Supporting Information). Even in alkaline electrolyte, the polymer electrode exhibited reversible charge/discharge behavior over hundreds of cycles, despite its solubility in such media (Figure S21, Supporting Information).

Following electrolyte optimization and anode/cathode mass balancing (Figure S22, Supporting Information), the PI/CNT anode was paired with an activated carbon (AC) cathode in full‐cell setups for various ion supercapacitors, including nonmetallic (NH₄⁺), monovalent (Na⁺, K⁺), divalent (Zn²⁺, Mg²⁺, Ca²⁺, Sr²⁺), and trivalent ions (Ce³⁺, Al³⁺). The pH values and ionic conductivities of the electrolytes used are summarized in Table S4 (Supporting Information). Due to HER‐suppression capability of PI and the use of highly concentrated electrolytes, these supercapacitors demonstrated a relatively high operational voltage window (1.4–1.7 V), surpassing that of typical aqueous supercapacitors (0.5–1 V). Taking NH₄⁺, K⁺, Mg²⁺, and Ce³⁺ ion supercapacitors as examples, specific capacitances of 150, 151, 138, and 125 F g⁻¹, respectively, were achieved at a low rate of 0.1 A g⁻¹. Even at a higher rate of 5 A g⁻¹, their respective capacitance remained as high as 102, 92, 100, and 97 F g⁻¹, indicating excellent fast‐rate capability (Figure  3A–D; Figure S23, Supporting Information). Post‐rate testing, these supercapacitors exhibited long‐term cycling stability. The NH₄⁺ ion supercapacitor demonstrated the longest life, achieving up to 0.38 million cycles (294 days) with a high capacitance retention of 87.5%. Similarly, the K⁺, Mg²⁺, and Ce³⁺ ion capacitors also exhibited long lifespans of 0.2 (130 days), 0.15 (112 days), and 0.2 (109 days) million cycles, with retention values of 95.5%, 75.2%, and 67.7%, respectively (Figure 3E). Figures S24–S27 (Supporting Information) show that other ion supercapacitors (Na⁺, Ca²⁺, Sr²⁺, and Al³⁺) also demonstrated high capacitance, fast‐charging capability, and long cycling stability. The long life cycles achieved are one of the longest reported for these ion systems.^{[ 34} , ³⁵ , ^{36 ]}

Figure 3.

Universal polymer anode for aqueous supercapacitors. A–D) Charge–discharge profiles of PI/AC supercapacitors in (A) NH₄ ⁺, (B) K⁺, (C) Sr²⁺, and (D) Ce³⁺ ion electrolyte. (E) Long cycling performance based in four electrolyte. F,G) Thick PI/CNT‐based zinc ion supercapacitors, (F) areal capacitance versus rate, (G) rate and cycling performance.

To further illustrate the potential of PI/CNT electrode for practical supercapacitors, thick electrodes (PI anode: 29.2 mg cm⁻², AC cathode: 15.6 mg cm⁻²) with a low N/P ratio of 1.2 were integrated into the device. The resulting zinc ion supercapacitor achieved a high capacitance of 150.3 F g⁻¹ at 0.1 A g⁻¹, consistent with the values obtained in low‐loading supercapacitors. This practical supercapacitor also maintained capacitances of 136.0, 117.1, 94.2, and 65.9 F g⁻¹ at 0.2, 0.5, 1.0, and 2.0 A g⁻¹, respectively. The corresponding areal capacitances ranged from 1.03 to 2.35 F cm⁻² at current densities from 1.56 to 31.3 mA cm⁻² (Figure 3F). At a high rate of 7.8 mA cm⁻², the supercapacitor exhibited a long cycle life of 20 000 cycles (140 days) with a capacitance retention of 62.5%, and an accumulated capacitance of 34 957 F (Figure 3G; Figure S28, Supporting Information). The capacity decay may probably be caused by proton storage causing local pH increase and gradual hydrolysis at the surface of polymer structure (Figure S29, Supporting Information). The overall structural stability of the PI electrode after long‐term cycling was investigated by disassembling the cell. The self‐supporting thick PI electrode remained intact and could be easily separated from other components (Figure S30a, Supporting Information). XRD patterns and FTIR spectra (Figure S30b,c, Supporting Information) confirmed that the polymer structure was well‐maintained even after long cycling, demonstrating the robust stability of the PI electrode. This exceptional performance demonstrates the practicality of this supercapacitor and the potential applicability of thick electrodes across other ion systems.

2.4. Universal Polymer Anode for Aqueous Batteries

In addition to their ultra‐stable cycling performance in supercapacitors, thick PI anode was further evaluated in various ion batteries (Li⁺, NH₄⁺, K⁺, Zn²⁺, Ca²⁺, Al³⁺) under practical conditions, viz. high loadings (8.3–61 mg cm⁻²), low N/P ratios (1–3), and lean electrolyte (2–9 µL mg⁻¹). Electrolyte engineering plays a crucial role in optimizing battery performance. Taking the battery based on a thick PI anode and lithium manganese oxide (LMO) cathode as an example, three electrolytes—4 m LiOTf, 4 m LiNO₃, and 2 m Li₂SO₄—were investigated. The cell with 4 m LiOTf delivered the highest capacities of 112.6, 108.7, 101.4, 95.3, 91.4, and 87.5 mAh g⁻¹ at respective rates of 0.05, 0.1, 0.2, 0.3, 0.4, and 0.5 A g⁻¹, outperforming both 4 m LiNO₃ and 2 m Li₂SO₄ cells (Figure  4A). Regarding reversibility, the 4 m LiOTf cell exhibited a high initial Coulombic efficiency (CE) of 94.43%, superior to 92.45% and 89.44% in the 4 m LiNO₃ and 2 m Li₂SO₄ cells. It maintained the highest CE values ranging from 98.53% to 99.96% across rates of 0.05–0.5 A g⁻¹ among three cells (Figure S29, Supporting Information). Furthermore, the 4 m LiOTf cell achieved the longest cycle life, with 100% capacity retention after 200 cycles, significantly outperforming the 82% and 75% retention in the 4 m LiNO₃ and 2 m Li₂SO₄ cells, respectively (Figure 4B).

Figure 4.

Universal polymer anode for aqueous batteries. A) 3D profiles of specific capacity‐rate‐electrolyte, B) Cycling performance of lithium batteries in three electrolytes. C) Electrical double‐layer structure of the anode under 2 V in three electrolytes simulated by MD. D,E) number density of (D) water and (E) lithium ions under different potential (0, 0.5, 1, 1.5, 2 V) simulated by MD. F,G) Charge–discharge profiles of thick PI anode enabled batteries in (F) NH₄ ⁺, Li⁺, and (G) Zn²⁺ and Al³⁺ ion systems. H) Long cycling performance of batteries based on thick PI anode in four ion systems. I) Comparison of mass loading, areal capacity, and cycle life in thick electrodes.

To understand the superior performance of triflate (OTf⁻) over nitrate (NO₃⁻) and sulfate (SO₄²⁻) anions, the electrolyte/electrode interface and solvation structure were simulated using molecular dynamics. An electrical double‐layer model was constructed to simulate the ion and water dynamics in the three electrolytes (4 m LiOTf, 4 m LiNO₃, and 2 m Li₂SO₄, Figure S30, Supporting Information) under different potentials.^{[ 37 ]} The simulations revealed distinct electrochemical double‐layer structures at the electrode surfaces (defined as the layer within 0–7.5 Å from the electrode surface) for each electrolyte, which determined the distribution of ions and water near the electrode surface (Figure 4C). At an applied potential of 0 V, the triflate and nitrate anions created a Li⁺‐rich and H₂O‐scarce environment at the inner Helmholtz layer (IHL, defined as the layer within 0–2.5 Å from the electrode surface, Figure 4D), in contrast to the Li⁺‐scarce and H₂O‐rich environment generated by sulfate anions. As the potential increased to 2 V, Li⁺ ions increasingly accumulated within 0.75 nm of the anode in the LiOTf and LiNO₃ electrolytes, while most Li⁺ ions remained in the bulk phase in the Li₂SO₄ electrolyte (Figure 4E). Density profiles of Li⁺ and H₂O indicated that more Li⁺ and fewer H₂O molecules accumulated near the electrode surface in the LiOTf and LiNO₃ electrolytes than in the Li₂SO₄ electrolyte (Figures S31–S33, Supporting Information), facilitating easier access to Li⁺ for redox reactions and reducing water‐related side reactions. The relatively poorer performance of LiNO₃‐based cells is likely due to the strong oxidative nature of nitrate anions, whereas triflate anions are milder and more stable in aqueous systems. The optimized PI//LMO cell based on LiOTf achieved an ultrahigh areal capacity of 6.5 mAh cm⁻², an ultra‐long cycle life of 2400 cycles (60% capacity retention), and an ultrahigh cumulative capacity of 12.77 Ah (Figure S34, Supporting Information), which is one of the best among aqueous lithium batteries (Figure 4F; Figures S34 and S35, Table S5, Supporting Information). Kinetics tests revealed a mixed diffusion‐controlled and pseudocapacitive behavior in the cell (Figure S36, Supporting Information).

Moreover, the thick PI anode demonstrates widespread applicability in ammonium, potassium, zinc, calcium, and aluminum ion batteries. Among them, NH₄⁺ ion batteries with a PI anode achieved the highest areal capacity (2.73 mAh cm⁻²), fastest rate (28.7 mA cm⁻²), longest cycle life (7700 cycles), high voltage (1.4 V), and excellent CE values (>99%) compared to previous reports (Figures S37 and S38 and Table S6, Supporting Information). Similarly, in zinc ion batteries, the PI anode enabled a high areal capacity of 3.59 mAh cm⁻² at 1.36 mA cm⁻² and 1.98 mAh cm⁻² at 38.2 mA cm⁻², with an ultra‐long cycle life of 3300 cycles (Figures S39 and S40 and Table S7, Supporting Information). The PI anode also demonstrated promising performance in potassium, calcium, and aluminum ion batteries, achieving large areal capacities of 1.4–5.4 mAh cm⁻² and respectable lifespans of up to 5000 cycles (Figures S41–S44 and Table S8, Supporting Information). These achievements underscore the universal applicability of the thick PI/CNT anode across different ion batteries, positioning it as a promising electrode for practical battery applications.

2.5. Polymer Anode for Seawater Batteries

Given its universal ion storage capability, the PI/CNT anode exhibits great potential for seawater‐based full cells. Seawater is primarily composed of Cl⁻ (≈55%), Na⁺ (≈30.6%), SO₄²⁻ (≈7.7%), Mg²⁺ (≈3.7%), Ca²⁺ (≈1.2%), K⁺ (≈1.1%), along with trace amounts of other ions (e.g., Sr²⁺, Br⁻, F⁻, HCO₃⁻) (Figure  5A).^{[ 38 ]} To construct seawater cells, the anode was paired with a rationally selected metal‐free cathode, iodine/activated carbon (I₂/AC). The selection of I₂/AC is driven by its ability to store Cl⁻ anions, forming a reversible ICl interhalogen intermediate during charge/discharge via a four‐electron transfer mechanism.^{[ 39} , ^{40 ]} The performance of I₂/AC in chloride‐containing electrolytes was first evaluated in aqueous zinc batteries, revealing two distinct redox potentials at 1.77/1.76 V and 1.26/1.21 V (vs Zn²⁺/Zn) and delivering a high capacity of 173.8 mAh g⁻¹ at 0.2 A g⁻¹ and 434.5 mAh g⁻¹ at 0.5 A g⁻¹, based on I₂/AC and I₂ only, respectively (Figure S45, Supporting Information). Such a high‐voltage, high‐capacity, chloride‐compatible cathode is thus well‐suited for seawater battery applications.

Figure 5.

Polymer anode for seawater batteries. A) Ionic composition of seawater. B) dQ/dV profiles of PI and I₂/AC electrodes relative to Zn²⁺/Zn redox potential. Solid line: high rate, shaded line: low rate. C) Proposed reaction mechanism of (−) PI//seawater//I₂/AC (+) cells for seawater batteries. D) Voltage–capacity profiles and E) dQ/dV curves of PI//I₂/AC cells in three electrolytes (VSS, PSS, HPS). F) rate and G) cycling performance of seawater batteries in HPS. H) Elemental mapping images of thick PI/CNT electrode after cycling.

The PI/CNT anode, with redox potentials of 0.66/0.58 and 0.38/0.31 V versus Zn²⁺/Zn, suggests theoretical redox potentials as 1.46/1.37 and 0.68/0.55 V for the coupled PI/CNT//I₂/AC cell (Figure 5B). The proposed reaction mechanism is illustrated in Figure 5C. As a proof of concept, PI/CNT and I₂/AC electrodes were assembled with three sea salt‐based electrolytes: Vancouver Island sea salt (VSS) from Canada, Pacific sea salt (PSS) from New Zealand, and Himalayan pink salt (HPS) from Pakistan. The corresponding batteries are denoted as the VSS, PSS, and HPS cell, respectively. Galvanostatic charge/discharge profiles and derived differential capacity (dQ/dV vs V) curves confirm the presence of two redox potentials in all cells, with high and low redox potentials recorded at ≈1.4/1.38 V and ≈0.7/0.4 V, matching well with theoretical values and supporting the occurrence of proposed redox reactions during charge/discharge. The slight variations observed between three cells may be attributed to differences in their ionic compositions and pH values (Table S9, Supporting Information). Notably, the ionic conductivity of these sea salt‐based electrolytes is exceptionally high, reaching 200 mS cm⁻¹, significantly surpassing that of aforementioned electrolytes. As a result, in an anode‐controlled configuration, the VSS, PSS, and HPS cells deliver high capacities of 112.2, 115.1, and 111 mAh g⁻¹, respectively, at a rate of 0.5 A g⁻¹. Even at high current densities, their capacities remain as high as 98.6 mAh g⁻¹ for VSS and 106.0 mAh g⁻¹ for PSS at 7 A g⁻¹, and 106.7 mAh g⁻¹ for HPS at 6 A g⁻¹, demonstrating the fast‐charging capability of these seawater batteries. Moreover, all cells exhibit long cycle life, with the HPS cell retaining 84% after 7000 cycles, the PSS cell retaining 91.4% after 1000 cycles, and the VSS cell retaining 91.1% of its capacity after 500 cycles (Figures S46 and S47, Supporting Information).

To demonstrate the potential for practical seawater batteries, thick electrodes (PI: 39.8 mg cm⁻², I₂/AC: 22.3 mg cm⁻²) were paired to construct the PSS cell. As shown in Figure S48 (Supporting Information), the PSS cell delivers a high areal capacity of 2.96, 2.41, and 2.23 mAh cm⁻² at respective rates of 4.5, 6.7, and 8.9 mA cm⁻². To our knowledge, this is the first fully aqueous seawater battery to achieve such a high areal capacity,^{[ 41} , ^{42 ]} approaching that of commercial LIBs (3–4 mAh cm⁻²). To further demonstrate real‐world applicability, thick‐electrodes‐based cell was assembled using saturated natural seawater as the electrolyte. The resulting cell exhibited stable performance over 100 cycles, achieving a high areal capacity of 2.75 mAh cm⁻² and a volumetric capacity of 38 mAh cm⁻³, highlighting its significant potential for practical application. To further verify the universal ion storage capability of the PI/CNT anode, the thick electrode was characterized post‐cycling (Figures S49–S51, Supporting Information). Both cross‐sectional and top‐view SEM images reveal a well‐maintained electrode structure, with polymer nanosheets interwoven with carbon tubes, and corresponding elemental mapping images and EDX analysis demonstrate the uniform distribution of sodium, potassium, magnesium, and calcium throughout the thick electrode (Figure 5H).

2.6. Energy, Power, Cost, and Overall Analysis of PI Anode‐Based Cells

Given the remarkable performance enabled by PI/CNT anodes, the energy and power densities, as well as the overall costs of the aforementioned cells, were analyzed to evaluate their potential for practical application. The Ragone plots (Figure  6A) demonstrate that all supercapacitors using various ions, including zinc‐ion capacitors based on thick electrodes, deliver significantly higher energy and power densities than traditional electrochemical capacitors. In comparison to lead‐acid and Ni‐MH aqueous batteries, our thick‐anode‐based batteries utilizing different ion systems and seawater batteries also offer superior power and energy densities. Among the systems, lithium‐ion batteries exhibited the highest energy density of 81.7 Wh L⁻¹ (at 36.3 W L⁻¹), specific energy of 73.7 Wh kg⁻¹ (at 32.8 W kg⁻¹), and areal energy density of 8.75 mWh cm⁻² (at 3.89 mW cm⁻²) (Figure 6B; Figures S52–S54, Supporting Information). The seawater batteries with thick electrodes achieved high specific energy and areal energy densities of 37.6 Wh kg⁻¹ and 2.33 mWh cm⁻², respectively, with comparable performance observed across other battery types.

Figure 6.

Energy, power, and cost analysis of polymer anode‐enabled energy storage devices. A,B) Ragone plots of (A) different supercapacitors, (B) aqueous and seawater batteries. C) Cost analysis of different energy devices. D,F) Radar plots of (D) supercapacitors, (E) batteries, and (F) seawater‐based energy devices in terms of cycle life, cost, sustainability, environment, fast charging, and energy density.

Furthermore, the batteries and supercapacitors featuring thick electrodes demonstrated exceptional cost‐effectiveness. The LIB cell exhibited the lowest material cost at $226.6 kWh⁻¹ (Figure 6C; Figure S55, Table S10, Supporting Information), while zinc‐ion supercapacitors and seawater batteries were priced at $1564 and $576.1 kWh⁻¹, respectively. These costs are highly competitive with established commercial lithium‐ion batteries ($150–500 kWh⁻¹), aqueous batteries ($100–400 kWh⁻¹), and supercapacitors ($900–6000 kWh⁻¹), underscoring the substantial commercial potential of this thick anode technology. A detailed cost breakdown revealed that PI only accounts for less than 10% of the total cost while the electrolyte is a significant cost driver, accounting for 37% and 55.2% of the total cost in the PI‐LMO and PI‐NH₄CuHCF cell, respectively (Figure S55, Supporting Information). It is anticipated that further cost reductions may be achieved through electrolyte engineering.

Overall, compared to traditional electrochemical capacitors and lithium‐ion capacitors, our supercapacitors offer the advantages of lower cost, extended cycle life, and superior environmental friendliness (Figure 6D). In contrast to lead‐acid and Ni‐MH aqueous batteries, our aqueous battery systems, such as the PI‐NH₄CuHCF cell, exhibit longer cycle life and greater sustainability, alongside lower costs and faster charging capabilities, as seen in the PI‐LMO cell (Figure 6E). Additionally, compared to previously reported seawater batteries—including primary cells, hybrid seawater batteries, and the recently developed TiO₂‐air cell reliant on metal catalysts—our seawater batteries stand out for their metal‐free electrodes and reliance on abundant resources. This makes them particularly promising for sustainable development and large‐scale applications (Figure 6F).

Compared to state‐of‐the‐art anode materials for aqueous batteries and supercapacitors, such as metals (e.g., Zn, Pb, Sn),^{[ 43} , ^{44 ]} metal oxides (e.g., Bi₂O₃),^{[ 19 ]} metal sulfides/selenides,^{[ 45 ]} sulfur,^{[ 46 ]} activated carbon, MXenes,^{[ 47 ]} organic molecules (e.g., PTCDI),^{[ 48 ]} and polymers (e.g., PAQS),^{[ 21 ]} our PI/CNT anode offers distinct advantages. These include superior electrical conductivity, mechanical robustness, corrosion resistance, low solubility, thermal stability, and dendrite‐free features, attributed to its large π‐conjugated and flexible‐chain‐bridged structure. These features enable the fabrication of ultrahigh‐loading (ultrathick) electrodes while maintaining high‐rate capability and ultralong cycling stability. However, the theoretical capacity (136.6 mAh g⁻¹) and density (1.08 g cm⁻³) of PI/CNT electrode remain lower than those of zinc (820 mAh g⁻¹, 7.14 g cm⁻³) and sulfur (1675 mAh g⁻¹, 2.07 g cm⁻³), indicating room for further optimization in energy density while retaining its structural and electrochemical advantages.

3. Conclusion

In summary, we have developed a universal PI/CNT anode capable of functioning across a wide range of simple‐ion batteries (six types) and supercapacitors (nine types) as well as complex‐ion seawater batteries (three types), involving non‐metallic (H⁺, NH₄ ⁺), monovalent (Li⁺, Na⁺, K⁺), multivalent ions (Zn²⁺, Ca²⁺, Mg²⁺, Ba²⁺, Sr²⁺, Ce³⁺, Al³⁺). This anode exhibits low tortuosity, high compression resistance, excellent corrosion resistance, superior electrical conductivity, robust mechanical strength/strain, high chemical stability, and effective HER suppression. These attributes enable the PI/CNT anode to achieve a sub‐million cycle life in NH₄ ⁺, K⁺, Mg²⁺, Zn²⁺, and Ce³⁺ supercapacitors, even under high loadings and low N/P ratios. Additionally, the thick anode demonstrates exceptional performance in lithium (6.5 mAh cm⁻², 2400 cycles), ammonium (2.73 mAh cm⁻², 7720 cycles), potassium (5.4 mAh cm⁻², 200 cycles), zinc (3.6 mAh cm⁻², 3300 cycles), calcium (1.4 mAh cm⁻², 5000 cycles), and aluminum (1.4 mAh cm⁻², 600 cycles) ion batteries, owing to the formation of a water‐scarce and cation‐rich IHP at the electrode‐electrolyte interface. The versatility of this anode was further highlighted in seawater battery applications, where it showed prolonged cycle life and high energy/power density. Given its universal ion storage and high‐loading capability along with future development of high‐performance cathodes, this PI/CNT anode paves the way for practical aqueous and seawater energy storage devices that are safe, scalable, and sustainable.

Conflict of Interest

The authors declare no conflict of interest.

Author Contributions

X.W. and G.L. supervised the research. X.W. and Z.X. conceived the idea and designed the experiments. Z.X. prepared materials, conducted characterizations, electrochemical measurements and analyzed data. P.L. conducted molecular dynamics simulations. J.Z. assisted with high‐pressure IR tests. K.H. and S.G. helped with synchrotron X‐ray imaging analysis. W.J. assisted with characterization and DFT calculation. All authors discussed the results and contributed to the manuscript writing.

Supporting information

Supporting Information

Supplemental Video 1

Supplemental Video 2

Xu Z., Li P., Zhao J., Hu K., Jia W., Gasilov S., Li G., Wang X., A Universal Thick Anode for Aqueous and Seawater Energy Storage Devices. Adv. Mater. 2025, 37, 2416427. 10.1002/adma.202416427

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.