Like other metal-air batteries, ORR and OER are the two key processes of sodium-based seawater^[25,26]. Where ORR is a multi-electron and proton coupled electrochemical reaction. In this process, oxygen molecules adsorbed on the electrode surface are reduced to H₂O (acidic) or OH^- (alkaline) via a four-electron reaction path, or to H₂O₂ (acidic) or {{\mathrm{HO}}_2}^{-} (alkaline) via a two-electron reaction path^[27,28]. In addition, Cl^- is easily adsorbed on the active sites of the catalyst, which hinders the cleavage of O — O bonds, thus reducing the activity of the catalyst^[29]. For OER, which is considered to be the rate-limiting step of sodium-based seawater batteries, the reaction process is extremely complex and closely related to the structure of the catalyst surface^[30⇓~32]. Nowadays, according to the adsorbate evolution mechanism (AEM), OER involves a variety of oxygen reaction intermediates (^*O, ^*OOH, and ^*OH), and the binding energies of the intermediates follow a proportional relationship of ΔG_OOH=ΔG_OH+3.2±0.2 eV and are linearly correlated, which results in a minimum theoretical overpotential of 0.37 V^[33,34]. Therefore, there is an urgent need for catalysts with high catalytic activity to accelerate the OER process. The OER process of sodium-based seawater battery is a four-electron reaction involving three intermediates, while the oxidation of Cl^- is a two-electron reaction involving a single intermediate, so there is a competitive relationship between the oxidation of Cl^- in seawater and OER, and the Cl^- oxidation reaction is easy to occur when the catalyst with OER overpotential greater than 490 mV is used. Generally, Pt/C and RuO₂ are catalysts with high catalytic activity for ORR and OER, respectively, but the high cost and scarcity of noble metals limit their wide application in sodium-based seawater batteries^[35,36]. In addition, Pt-based materials are highly sensitive to Cl^- and easily react with them to form soluble chloro complexes (such as {\mathrm{PtCl}}_4^{2-}), thus reducing the structural stability of noble metal catalysts^[37,38]. Therefore, non-noble metal catalysts should be designed that are efficient and selective for oxygen/chlorine catalysis.

In addition, the actual discharge voltage plateau of the battery is lower than its standard voltage plateau, while the charge voltage plateau is higher than the standard voltage plateau due to the internal voltage losses of the battery, such as activation polarization, ohmic polarization, and concentration polarization^[39]. To sum up, the sodium-based seawater battery catalyst should have the following functions: (1) shorten the reaction path, reduce the reaction activation energy, and accelerate the reaction rate; (2) large specific surface area, hierarchical porous structure and high porosity; (3) excellent mechanical and chemical stability; (4) a highly conductive network (good conductivity) promotes electron transfer; (5) Strong oxygen/chlorine selectivity^[40⇓~42].

In addition to the selection of catalysts, battery evaluation is also one of the important links in the research and development of high-performance batteries. Battery evaluation metrics include electrochemical performance, cost, and safety. Electrochemical properties include overpotential, capacity, power/energy density, and cycle performance. The overpotential is related to the catalyst. Specifically, the ORR/OER corresponds to the discharge/charge curve of the battery. The weaker the ability of the catalyst to reduce the activation energy of the reaction, the greater the overpotential of the battery. For example, the hybrid sodium-air batteries assembled by Liang et al. Using different catalysts (MOF-NCNTs, Pt/C, and RuO₂) showed different overpotentials due to the different ability of the catalysts to reduce the reaction activation energy^[11]. The capacity of the battery is related to the solid electrolyte and organic electrolyte, in which the solid electrolyte and electrolyte with high ionic conductivity and stability can reduce the interface impedance, accelerate the Na⁺ transfer, improve the utilization efficiency of the anode, and make the battery show higher discharge capacity^[43][44]. The energy density and power density of the battery are related to the catalyst and solid electrolyte. The efficient catalyst can reduce the activation energy, accelerate the reaction rate, and make the battery have a higher discharge voltage, thereby improving the power density of the battery^[45]. The highly conductive solid electrolyte can accelerate the discharge process and improve the energy/power density of the battery. Cycle performance is the most important evaluation standard of rechargeable batteries. At present, the optimal cycle performance of sodium-based seawater batteries can reach 600 times, but it still can not meet the practical application^[46]. As for the battery cost, the energy cost of lithium-ion and sodium-ion batteries is estimated to be $1600~1800 kW·h^-1和$ 1400~1600 kW·h^-1 at the current material price, while the energy cost of sodium-air batteries is only $100~200 kW·h^-1. In addition, the sodium-based seawater battery uses cheap seawater as catholyte, so its cost is lower^[47,48]. Cathode open structure and circulating seawater can effectively accelerate the heat transfer of the battery, thus reducing the burden of heat management and significantly improving the safety of the battery in application.

In order to promote the practical application of rechargeable sodium-based seawater batteries, researchers have designed and optimized the battery structure. First, Kim's group designed a pouch cell (Figure 3A), in which the anode chamber consists of a nickel current collector and an organic electrolyte, the exposed conductive film is connected to the carbon felt cathode current collector, and the cathode is immersed in a beaker filled with seawater^[13]. The battery has the advantages of simple assembly and low cost, but its low safety and poor cycle performance make it difficult to be applied on a large scale. Fig. 3B shows a coin cell in which a solid electrolyte (NASICON) is fixed to an open circular cover made of polypropylene material, sodium metal is attached to a circular stainless steel gasket filled with an organic electrolyte and sealed with epoxy glue, and finally assembled with a cathode current collector^[49]. The battery can be uniformly evaluated electrochemically as a coin cell, but it is difficult to prepare and transport on a large scale due to the rigidity (fragility) of the NASICON electrolyte. In order to solve the above problems, the researchers designed a rectangular cell (Figure 3C), in which 12 rectangular NASICONs were fixed on a stainless steel mesh, and then sodium biphenyl was injected to form an anode chamber, which was sealed with carbon cloth^[50].

The battery can maximize the space utilization of the battery, and at the same time, the low cost and processability provide possibilities for its application, but the energy and power density of the battery still can not meet the needs of practical applications. In order to improve the power density, Kim's team developed a prismatic battery, as shown in Figure 3D. Its structure is similar to that of a rectangular battery, but the difference is that the anode chamber uses a larger area of NASICON and the cathode current collector is designed to resist seawater corrosion^[51]. The cell has high energy density (242 Wh·L^-1) and power density (5.8 mW·cm^-2), but its cycling stability and energy efficiency are poor.

To sum up, although the battery performance can be improved through the battery structure design, there is still a gap from the practical application. It is necessary to further explore the impact of various component materials on the battery, optimize the internal structure and space of the battery, and improve the utilization efficiency, so as to design a low-cost, high-performance and long-life battery. Based on this, this paper will summarize the latest progress of anode part, organic electrolyte, solid electrolyte and cathode catalyst to help the commercialization process of sodium-based seawater battery.

The uneven deposition of sodium metal on the copper foil current collector is easy to cause the growth of sodium dendrite, which leads to the reduction of battery cycle performance. In order to inhibit the growth of dendrites, sodiophilic materials can be introduced on the surface of the anode current collector, thereby reducing the nucleation barrier of Na⁺ and helping the uniform deposition of Na⁺ on the anode. For example, Choi et al. Prepared graphene-coated copper composite (Gr/Cu) by chemical vapor deposition and used it in sodium-free anode sodium-based seawater battery^[57]. In order to explore the sodium dendrite growth on Cu and Gr/Cu current collectors, the cell (Cu or Gr/Cu | 1 M NaOTF/DME | Na) was tested at constant current density (3.25 mA·cm^-2), and the sodium dendrite growth process was recorded by time-lapse camera. As shown in Figure 4A, no dendrite is formed on the Cu current collector in the first 50 min, while sharp needle-like sodium dendrites appear and grow rapidly to the opposite sodium electrode in the last 50 min. The irregular growth of sodium dendrites in the full battery will pierce the separator, resulting in short circuit of the battery and potential safety hazards. However, the Gr/Cu current collector can significantly delay the appearance of sodium dendrites (90 min) and slow down the growth rate of sodium dendrites (Fig. 4F). This is because the corrugated surface of the heterogeneous interface of the Cu current collector can reduce the nucleation energy barrier of Na⁺, which is conducive to the formation of smaller sodium dendrites (Fig. 4C). The monolayer graphene is covered on the Cu foil (Figure 4D), which can not only eliminate the chemical instability of the Cu current collector, but also maintain the morphology of the Cu current collector. It can be seen from fig. 4E that the corrugated Cu foil is covered by graphene with carbon hexagons, so that the Gr/Cu current collector interface becomes uniform, thereby delaying the appearance time of sodium dendrites and reducing the number of dendrite nuclei. In the electrochemical test, the sodium nucleation overpotential is represented by the difference between the potential drop value (<0 V vs Na⁺/Na) and the voltage plateau of the sodium deposition process. In order to study the effect of different current collectors on the sodium nucleation overpotential, galvanostatic plating and stripping tests were conducted (Figure 4G). Compared with the Cu current collector, the symmetrical cell with Gr/Cu as the current collector showed a lower nucleation overpotential (40 mV) and a more stable sodium plating/stripping process, and its coulombic efficiency remained unchanged after 200 cycles (150 H). This is due to the fact that the uniform interface of Gr/Cu current collector in the electrochemical reaction can provide abundant nucleation sites as well as weaken the volume change of the sodium plating/stripping process. In addition, the full cell with Gr/Cu as the current collector (Na | NASICON | seawater) has a longer cycle life. Fig. 4h is a schematic diagram of the charge-discharge structure of the battery. Compared with the Cu current collector, the metal sodium on Gr/Cu has a larger grain size, fewer crystal nuclei, and a smooth surface. This indicates that the uniform surface of Gr/Cu current collector is beneficial to control the nucleation rate of sodium metal, thus forming a uniform and smooth sodium metal deposition layer, which is conducive to the reversible anodic sodium deposition/stripping process.

Sodium metal is easy to form dendrites in the deposition/stripping process, which increases the possibility of battery short circuit, thus causing potential safety hazards. At the same time, the shape and volume of the sodium anode are prone to change during battery cycling, resulting in the decomposition of the SEI layer and the formation of a new SEI layer on the newly exposed sodium surface. In this process, sodium and electrolyte are continuously consumed, which reduces the energy efficiency of the battery^[58]. In addition, the high reactivity of sodium metal with common electrolytes leads to the precipitation of {{\mathrm{H}}_2}^{\mathrm{}} by side reactions, which degrades the capacity and cycle performance of sodium-based seawater batteries^[59]. In order to solve the above problems, Kim et al. Assembled a sodium-based seawater battery using Na-BP-DEGDME as the anode^[59]. As shown in Figure 5A, the Na-BP-DEGDME molecule with high electronic/ionic conductivity acts as an intermediary/carrier of Na⁺ and electrons, which is beneficial to reduce the sodium deposition potential, inhibit the precipitation of H₂ and the growth of needle-like sodium dendrites, so that the battery exhibits excellent capacity retention and cycling stability (400 H) (Figure 5B). As shown in Fig. 5C – e, our research group prepared Na-BP-TEGDME (sodium-biphenyl-tetraethylene glycol dimethyl ether) and used it in a hybrid sodium-air battery, which exhibited excellent cycling stability (500 cycles), round-trip efficiency (93.26%), and power density (621 mW·g^-1)^[35].

Hard carbon (HC) can be used as a sodium-based – seawater battery anode because of its working potential (vs Na⁺/Na) close to that of metallic sodium and high reversible capacity. The feasibility of HC as the anode of sodium-based seawater battery was proved by half-cell (Na | HC) and full-cell (HC | seawater) charge-discharge experiments and simulations^[13]. Lim et al. Used HC-polystyrenesulfonic acid (HC-PSS) anode for sodium-seawater battery, and the reversible specific capacity of HC-PSS anode at 0.5 and 1 C current density was 383.2 and 295.5 mAh·g^-1, respectively, and the discharge capacity and coulombic efficiency of the whole battery were 382 mAh·g^-1 and 99.5%, respectively, after 100 stable cycles at 0.5 C, which proved that HC was an ideal anode material for sodium-seawater battery^[60].

High capacity characteristic conversion-Alloy materials are prone to volume expansion during battery cycling, resulting in the continuous formation of SEI layer, and the Na⁺ provided by the cathode is continuously consumed, resulting in capacity reduction^[61,62]. However, the sodium-based – seawater battery has an open cathode structure, and seawater can continuously supply Na⁺ to alleviate the capacity fading. Therefore, the high capacity characteristic conversion-alloy materials can be used in sodium-based seawater batteries.

The Gr/Cu current collector and Na-BP anode reduced the sodium deposition potential and inhibited the dendrite growth, thereby improving the battery life. The HC material contains a large number of defects, which is beneficial to the storage of sodium ions with a large radius and improves the reversible capacity of the battery. The open cathode structure of sodium-based seawater battery makes the battery with conversion-alloy material as the anode show excellent cycle stability. At the same time, the conversion-alloy material does not involve sodium metal in the battery manufacturing process, so it can be assembled in a low humidity environment, which significantly reduces the cost of safety measures and the purchase of sodium metal.

β-alumina electrolyte was first reported in 1967^[71⇓~73]. As shown in Fig. 7 a, β-alumina has two crystal structures of hexagonal β-alumina (β-Al₂O₃) and rhombic β ″ -alumina (β″-Al₂O₃), both of which are composed of spinel layers composed of [AlO₄] tetrahedra and [AlO₆] octahedra and two sodium-oxide layers stacked alternately^[74,75]. Among them, β″-Al₂O₃ possesses higher ionic conductivity (1.4×10^-2S·cm^-1) due to its high sodium content, which exhibits excellent cycling stability when used in sodium-sulfur batteries^[76]. When the β″-Al₂O₃ solid electrolyte was used in the sodium-based – seawater battery, by comparing the X-ray diffraction (XRD) patterns of β″-Al₂O₃ before and after cycling, it was found that the AlOOH phase was formed after 10 cycles, resulting in a decrease in the cycling stability of the battery (Figure 7B)^[43]. This is due to the entry of water molecules, H⁺ and H₃O⁺ into the β″-Al₂O₃ lattice, which destroys the β″-Al₂O₃ structure and hinders the Na⁺ transfer in the electrolyte, resulting in the decrease of its conductivity^{[77⇓⇓~80]}.

NASICON-type solid state electrolyte was first proposed by Goodenough et al. In 1976, and its general formula is {{\mathrm{Na}}_{1+}}_x Zr₂Si_xP_3-xO₁₂(0≤x≤3), which is formed by partial substitution of P by Si when the parent NaZr₂P₃O₁₂ is in excess of sodium^[81]. There are rhombic and monoclinic phases of Na₃Zr₂Si₂PO₁₂ with good conductivity (Fig. 7 C), in which the corner-sharing tetrahedron composed of SiO₄ and PO₄ and the ZrO₆ octahedron structure form three-dimensional (3D) channels favorable for Na⁺ transport^[82,83]. The Na1 site located between the two ZrO₆ octahedra in the rhombic structure and the Na2 site located between the two structural bands constitute a three-dimensional channel (Na1-Na2), and the Na2 site can accommodate 3 mol Na⁺, which is beneficial to Na⁺ transport. However, in the monoclinic structure, the Na2 site is split into Na2 and Na3 sites to form Na1-Na2 and Na1-Na3 channels due to the lattice distortion caused by the replacement of PO₄ tetrahedra by larger SiO₄ tetrahedra. NASICON-type solid electrolytes with 3D Na⁺ channels and high ionic conductivity have been widely used in sodium-metal chloride batteries and CO₂ sensors^[56]. In order to evaluate the feasibility of NASICON electrolyte used in sodium-based seawater battery, Hwang et al. Immersed NASICON electrolyte in seawater at room temperature for 1440 H, and found that the phase and morphology of NASICON did not change before and after immersion in seawater through XRD and SEM (Fig. 7d), indicating that NASICAN electrolyte has excellent structural stability in seawater^[13]. In addition, the sodium-based seawater battery assembled by Kim et al. Using NASICON electrolyte can be stably cycled for 600 times, and the morphology, phase and ionic conductivity of NASICON electrolyte do not change after cycling, which further proves the feasibility of NASICON-based electrolyte for sodium-based seawater battery^{[46][46,84,85]}. As shown in Fig. 7e, f, our research group tested the stability of NASICON electrolyte in acidic solution with different pH values, and found that with the decrease of solution pH value, the proton exchange rate of H₃O⁺ and Na⁺ increased, and the decrease of Na⁺ concentration in the lattice led to the contraction of PO₄ tetrahedron and the formation of H₃O⁺-NASICON phase, which reduced its conductivity^[86].

In summary, the NASICON solid electrolyte is more stable than the β″-Al₂O₃ in the sodium-based seawater battery. The chemical stability and ionic conductivity of NASICON solid electrolyte determine the power density, cycle stability, safety performance and service life of sodium-based seawater battery, which is very important for the development of the battery system. However, NASICON solid electrolyte still faces the problems of poor interfacial wetting and low ionic conductivity. At the same time, there are few reports on the stability test of NASICON solid electrolyte with organic electrolyte and seawater at high current density, so the structural stability of NASICON solid electrolyte in practical application needs to be discussed. Based on this, the research and development of NASICON solid electrolyte with high conductivity and high stability is of great practical significance for the development of high energy density batteries.

Carbon-based catalysts are considered as a potential alternative to noble metal-based catalysts because of their low cost, high specific surface area, high electrical conductivity, and excellent chemical stability^[92]. At the same time, the catalytic activity is often improved by heteroatom doping and the creation of intrinsic defects. Senthilkumar et al. Prepared a porous carbon catalyst (PC) containing edge defects and oxygen functional groups using grapefruit peel as raw material by hydrothermal method combined with chemical activation, as shown in Fig. 8 a, the half-cell with PC as catalyst exhibited the smallest overpotential (0.47 V) compared with Pt/C, IrO₂ and MnO₂ catalysts^[91]. Because the defects and oxygen functional groups in PC change the charge distribution and spin density of adjacent carbon atoms in PC, resulting in enhanced ORR/OER catalytic activity. Similarly, Kim et al. Synthesized nitrogen and phosphorus-doped pine pollen carbon powder (PPC) by heat treatment using pine pollen as raw material. Due to the high catalytic activity of PPC caused by nitrogen and phosphorus doping, the full cell with PPC as catalyst showed high capacity retention and excellent cycle stability (Fig. 8B)^[93][94].

Compared with noble metal catalysts, non-noble metal based catalysts have the advantages of low cost, good stability and abundant sources, but their catalytic activity is poor. In order to improve the intrinsic catalytic activity of non-noble metal-based catalysts, researchers have constructed nanostructured catalysts to expose more active sites. Abirami et al. Prepared porous cobalt manganese oxide nanocubes (CMO) with large specific surface area. As shown in Figure 8 C, the half cell with CMO as the catalyst showed a low overpotential (0.53 V)^[95]. By comparing the XRD and X-ray photoelectron spectroscopy (XPS) of the catalyst before and after cycling, it was found that the phase and surface electronic state of the catalyst did not change after cycling, indicating that the CMO catalyst had excellent structural stability. Shine et al. Prepared cobalt vanadate nanoparticles (Co₃V₂O₈) by solution method and used them as sodium-based – seawater battery catalyst^[96]. The porous structure of Co₃V₂O₈ can provide more active sites and enhance the ORR/OER reaction kinetics, allowing the battery to exhibit high power density (5.9 mW·cm^-2)^[97]. Due to the excellent structural stability of catalyst Co₃V₂O₈ in seawater, the full cell can be stably cycled for 20 cycles (400 H) at 0.1 mA·cm^-2 (Figure 8D).

The ORR performance of carbon material catalysts is comparable to that of platinum-based catalysts due to their excellent conductivity and chemical stability^[98,99]. However, carbon materials have low catalytic activity and are easily oxidized or corroded, resulting in the decline of battery cycle performance. On the other hand, transition metal oxides (spinel and perovskite) exhibit excellent catalytic activity for OER and ORR, but their conductivity is low. Based on the advantages and disadvantages of the above two catalysts, in order to obtain a catalyst with high activity and high conductivity, a composite catalyst is made of a transition metal material and a carbon material. Suh et al. Grew cobalt nanoparticle-doped carbon nanotubes (S-rGO-CNT-Co) on reduced graphene oxide by vapor activation combined with microwave treatment and served as a catalyst for sodium-based seawater batteries^[100]. As shown in Figure 8E, the cell exhibits a small overpotential (0.4 V), excellent cycling stability, and rate capability. This is attributed to the high catalytic activity, abundant active sites, high conductivity and chemical stability of the composite catalyst. In addition, conventional air cathodes require a polymer binder to ensure good physical contact between the catalyst and the electrode. However, the electrically insulating and hydrophobic properties of the polymer binder reduce the electrochemically active surface area of the catalyst. Binderless air electrode can maximize the activity of the catalyst and improve the energy efficiency of the battery. Therefore, Kim et al. Anchored tricobalt tetroxide (Co₃O₄) on a carbon substrate (PPy+Co₃O₄@CF) by a polypyrrole-doped carbon (PPy/C) infiltration technique and used it as a sodium-based – seawater battery catalyst^[101]. As shown in Fig. 8f, the PPy+Co₃O₄@CF exhibits excellent bifunctional catalytic activity and the cell cycling performance is comparable to that of the commercial Pt/C+IrO₂ air cathode.

Due to the low solubility and slow diffusion of oxygen in seawater, ORR and OER involve complex four-electron transfer processes, resulting in slow reaction kinetics and large cell overpotential. Therefore, in order to improve the energy efficiency of the cell and reduce the overpotential, the catalyst with large specific surface area, high efficiency and high activity should be designed to reduce the ORR and OER reaction energy barrier and improve the energy efficiency and service life of the cell. The above three types of catalysts have the advantages of low cost, high activity and wide source of raw materials (biomass materials), and have considerable development prospects in sodium-based seawater batteries.