Integrated adsorption–catalysis design enabling high-performance sodium–sulfur batteries

Abstract

Room-temperature sodium–sulfur (RT Na–S) batteries have attracted extensive attention owing to their high energy density, abundant raw materials and cost-effectiveness for large-scale energy storage applications. However, their practical application is still limited by the severe shuttle effect and sluggish sulfur redox kinetics. To tackle these issues, multifunctional materials based on adsorption–catalysis synergy have been widely reported to anchor soluble sodium polysulfides (NaPSs) and accelerate their redox reaction process. In this review, we comprehensively summarize the recent progress in electrode materials with synergistic adsorption–catalysis effects. First, we introduce the electrochemical mechanisms and critical challenges of RT Na–S batteries and elucidate the working principle of adsorption–catalysis synergy. Subsequently, the recent advances in enhancing the electrochemical behaviors of RT Na–S batteries are meticulously discussed based on an adsorption–catalysis synergistic strategy. Finally, given the gap between current lab-scale research and industrial production needs, we point out the existing challenges and future research directions toward achieving the commercial application of RT Na–S batteries.

---

Xinghao Zhang

Xinghao Zhang is a PhD student at Southeast University under the supervision of Prof. Jiarui He. His current research interests mainly include advanced lithium–sulfur and sodium–sulfur batteries.

Wanjie Gao

Wanjie Gao is a PhD student at Southeast University under the supervision of Prof. Jiarui He. His current research interests mainly include sodium–sulfur and sodium–metal batteries for electrochemical energy storage.

Yufan Chen

Yufan Chen is an undergraduate student majoring in Energy and Power Engineering at Southeast University.

Jiarui He

Jiarui He is a professor at Southeast University. He obtained his BE (2012) and his PhD (2018) in electronic information materials and devices from the University of Electronic Science and Technology of China. He went to the Texas Materials Institute at the University of Texas at Austin as a postdoctoral fellow in 2018. His research interests are in the area of electrochemical conversion and storage materials. Now, he is a professor at Southeast University. His research interests are focused on energy storage and conversion systems and their key materials.

---

1. Introduction

The global challenges of resource depletion and environmental degradation caused by excessive fossil fuel consumption have garnered significant scientific and policy attention.^1,2 Hence, changing the energy structure and developing renewable energy sources (e.g. solar photovoltaic, wind) have been universally acknowledged as a strategy imperative for sustainable development.^3,4 However, the characterization of spatial and temporal fluctuations in renewable energy generation makes it necessary to use external energy storage systems like lithium-ion batteries (LIBs).^5,6 Despite the high maturity of LIB technology, their limited energy density (300 W h kg⁻¹) and the high cost of raw materials like lithium (Li), cobalt (Co) and nickel (Ni) severely constrain their viability for grid-scale storage applications.^7,8

Benefiting from the high theoretical specific capacity of sulfur cathodes (1675 mA h g⁻¹), lithium–sulfur (Li–S) batteries (2600 W h kg⁻¹) and RT Na–S batteries (1274 W h kg⁻¹) exhibit significantly higher energy densities than commercial LIBs.^9–11 Notably, the abundance of sodium (2.3 wt%) in the earth substantially exceeds that of Li (0.0065 wt%). This natural abundance advantage significantly reduces battery production costs.^12,13 Consequently, RT Na–S batteries with high energy density and low cost are regarded as promising candidates for large-scale energy storage systems.¹⁴ Nevertheless, practical application of RT Na–S batteries is still limited by critical challenges, including low conductivity of sulfur and sodium sulfide (Na₂S), large volume expansion, serious shuttle effect, and sluggish NaPS conversion kinetics.^15,16

In view of the mentioned application prospects and technical limitations of RT Na–S batteries, researchers have pursued multiple strategies to address these challenges and enhance the electrochemical performance of RT Na–S batteries. To improve the low conductivity of sulfur, advanced conductive materials have been employed as sulfur hosts, such as carbon nanotubes, graphene, hollow carbon spheres, and conductive polymers.^17–19 Unfortunately, single physical confinement of carbon materials and even chemisorption strategies after introducing polar materials cannot completely suppress the shuttle effect.^20,21 This limitation originates from the inherent inadequacy of passive adsorption strategies, which effectively anchor NaPSs but remain incapable of modulating the sluggish sulfur redox kinetics. Especially, the sluggish sulfur redox kinetics cause NaPS accumulation and exacerbate the shuttle effect. Aiming to address this, Zhang et al. innovatively introduced the electrocatalyst into sulfur cathodes to accelerate the conversion of NaPSs.²² Adding functional electrocatalysts to the sulfur host based on adsorption–catalysis synergy greatly promotes polysulfide redox conversion and reduces the shuttle effect to achieve high-performance batteries, which is crucial for future practical applications of RT Na–S batteries.

Research on RT Na–S batteries has predominantly focused on addressing the uncontrolled soluble NaPS migration and enhancing the sulfur species conversion kinetics since the appearance of the RT Na–S battery model in 2006. Herein, this review systematically summarizes the recent progress in adsorption–catalysis synergistic strategies for RT Na–S batteries, as illustrated in Fig. 1.^23–54 First, we briefly introduce the fundamental electrochemistry, persistent challenges of RT Na–S batteries and elucidate the working mechanism of adsorption–catalysis synergy. Subsequently, we comprehensively review the latest achievements in advanced electrocatalysts, including heterostructures, single atoms, metal nanoparticles, metal carbides, metal nitrides, metal oxides, metal chalcogenides, heteroatom doping, and others, to analyze their corresponding adsorption–catalysis behaviors. Finally, we propose the gap between current lab-scale research and commercial application, providing scientific guidance toward practical large-scale application of RT Na–S batteries.

Fig. 1 Schematic diagram of the adsorption–catalysis synergy of RT Na–S batteries.

2. Electrochemical principles and the main challenges of RT Na–S batteries

As illustrated in Fig. 2a, a typical RT Na–S battery comprises four parts: (i) sulfur-based composite cathode, (ii) Na metal anode, (iii) separator, and (iv) electrolyte. The battery facilitates energy storage and release through multi-electron redox reactions between S₈ and Na metal. This section elucidates electrochemical principles and differences between two electrolyte systems to enhance the mechanistic understanding of sulfur redox behavior in Na–S batteries. Subsequently, we discuss the essential challenges restricting the commercialization of RT Na–S batteries and propose the corresponding modification strategies to achieve high-performance RT Na–S batteries.

Fig. 2 (a) Schematic illustration of a typical RT Na–S battery. (b) Representative discharge curve of an RT Na–S battery in an ether-based system. Reproduced from ref. 55 with permission from Wiley-VCH, copyright 2015. (c) Typical charge–discharge profiles of different cycles of an RT Na–S battery in a carbonate-based system. Reproduced from ref. 56 with permission from Springer Nature, copyright 2019.

2.1 Electrochemical principles of RT Na–S batteries

RT Na–S batteries exhibit a high theoretical capacity of 1675 mA h g⁻¹ and energy density of 1274 W h kg⁻¹ through the redox reaction between S₈ and sodium metal:^57,58

2Na + S ↔ Na₂S (1)

Extensive research indicates that the electrochemical conversion from S₈ to Na₂S involves a complex multi-electron transfer process accompanied by the formation and evolution of multiple NaPS intermediates.⁵⁹ Additionally, the markedly distinct solubility characteristics of NaPSs in ether-based versus carbonate-based electrolytes fundamentally dictate the divergent electrochemical performance of RT Na–S batteries across different liquid electrolyte systems.

2.1.1 Ether-based electrolytes. In ether-based systems, redox reactions are primarily dominated by reversible sulfur dissolution and precipitation processes, involving sequential solid–liquid phase conversions (i.e. S₈ → Na₂S₈), liquid–liquid phase conversions (i.e. Na₂S₈ → Na₂S_x, 4 ≤ x ≤ 6), and liquid–solid phase conversions (Na₂S₄ → Na₂S_x, 1 ≤ x ≤ 3).¹³ Consequently, the typical discharge profile (Fig. 2b) exhibits two voltage plateaus at 2.2 V and 1.65 V, which correspond to the cyclo-S₈ ring-opening and the deposition process of dissolved Na₂S₄.^55,60 Moreover, the two distinct slopes in the voltage range of approximately 2.2–1.65 V and 1.65–1.2 V can be attributed to the reduction and disproportionation reactions of the soluble high-order NaPSs and the conversion of solid-phase low-order NaPSs to Na₂S, respectively. Therefore, the discharge process can be categorized into four distinct electrochemical phases and the corresponding specific reactions are shown below.

Region I: the first voltage plateau at 2.2 V corresponds to the initial reduction of S₈ into soluble Na₂S₈via a two-electron transfer reaction as follows:

S₈(s) + 2Na⁺ + 2e⁻ → Na₂S₈(l) (2)

Region II: the sloping voltage region ranging from 2.2 V to 1.65 V represents the continued reaction of Na₂S₈ into Na₂S₄ with a liquid–liquid transition, which also leads to the formation of other intermediates (i.e. Na₂S₆ and Na₂S₅). This complex process can be simply expressed by the following equation:^61,62

Na₂S₈(1) + 2Na⁺ + 2e⁻ → 2Na₂S₄(l) (3)

Region III: the second voltage plateau at 1.65 V is attributed to the further reduction of soluble Na₂S₄ into insoluble short-chain NaPSs including Na₂S₃ and Na₂S₂, and some of the solid-state Na₂S₃ and Na₂S₂ species are further reduced to the final charge product Na₂S. Due to the higher energy barriers in insoluble NaPS species conversion and their poor electrical conductivity, the sluggish reaction kinetics become a rate-determining factor for RT Na–S batteries. The soluble NaPS precipitation process can be expressed as follows:

Na₂S₄(s) + 2/3Na⁺ + 2/3e⁻ → 4/3Na₂S₃(s) (4)

Na₂S₄(s) + 2Na⁺ + 2e⁻ → 2Na₂S₂(s) (5)

Na₂S₄(s) + 6Na⁺ + 6e⁻ → 4Na₂S(s) (6)

Region IV: the second sloping voltage region (1.65–1.2 V) relates to the conversion between Na₂S₂ and the final discharge product Na₂S. This process exhibits sluggish kinetics and significant polarization losses due to the poor electronic conductivity of both Na₂S₂ and Na₂S. The electrochemical evolution can be described as:

Na₂S₂(s) + 2Na⁺ + 2e⁻ → 2Na₂S(s) (7)

2.1.2 Carbonate-based electrolytes. The distinct solubility and chemical reactivity of NaPSs in carbonate-based electrolytes fundamentally alter their electrochemical behavior compared to ether-based systems. As illustrated in Fig. 2c, the discharge profiles lack obvious multiple-plateau characteristics, and the initial cycle exhibits a distinctive difference compared to subsequent cycles, with the discharge capacity exceeding the theoretical capacity.⁵⁶ In fact, the S cathode commonly undergoes a multi-step activation process and parasitic reactions with electrolytes in the initial cycles.⁶³

Due to the high chemical reactivity between the soluble NaPSs and carbonate solvents, some of the NaPSs undergo nucleophilic reactions with carbonate solvents and form a dense cathode electrolyte interphase (CEI) to inhibit further reaction of NaPSs with carbonate solvents. The activation process leads to additional capacity contribution from electrolyte decomposition, thus resulting in the first-cycle discharge capacity exceeding the theoretical capacity.⁶⁴

In subsequent cycles, the cyclo-S₈ undergoes a multi-step quasi-solid-phase transformation (i.e. S₈ → Na₂S₂ and Na₂S) or a solid–liquid–solid reaction reminiscent of ether-based systems (i.e. S₈ → Na₂S_x → Na₂S) with the protection of the CEI layers and the host materials. Within carbonate-based electrolytes, the large ionic radius of Na⁺ significantly impedes NaPS dissociation, effectively inhibiting their solution.⁶⁵ Therefore, RT Na–S batteries employing carbonated electrolytes can eliminate the shuttle effect and exhibit better performance after the complex activation process. However, it is crucial to mention that this suppression mechanism could compromise the electrochemical reversibility of sulfur conversion reactions, resulting in accelerated capacity decay.

2.2 Main challenges of RT Na–S batteries

The intermediate reaction mechanisms in sodium–sulfur batteries involve intricate multi-phase transitions, NaPS conversion dynamics and complex interfacial interactions between dissolved NaPSs and electrodes. These coupled processes collectively lead to the kinetic and thermodynamic complexities of the system. Consequently, the development of high-performance RT Na–S battery systems necessitates overcoming the following critical challenges:

(1) The low conductivity of sulfur (5 × 10⁻³⁰ S cm⁻¹) and its solid-state discharge product (Na₂S₂ and Na₂S) results in difficult charge transfer, severe polarization, sluggish NaPS conversion kinetics and low active material utilization. These issues collectively constrain practical energy densities.⁶⁶

(2) The substantial 170% volume expansion of sulfur during electrochemical cycling triggers severe electrode pulverization and irreversible structural degradation, significantly undermining the cycling stability of the sulfur cathode.⁶⁷

(3) Parasitic reactions between highly reactive sodium metal and electrolyte exacerbate active sodium loss and solid electrode interphase (SEI) heterogeneity. This dual effect drives electrolyte depletion and accelerates capacity fade.^68,69 Furthermore, uncontrolled sodium dendrite growth elevates risk of internal short circuits and triggers severe safety accidents.

(4) During the electrochemical conversion process of sulfur species, soluble NaPSs repeatedly migrate between electrodes driven by electric field and concentration gradient, resulting in a “shuttle effect”.⁷⁰ This phenomenon triggers irreversible active materials loss and severe sodium metal anode corrosion, resulting in rapid capacity decay, serious self-discharge behavior and low coulombic efficiency.^71,72

(5) The sluggish conversion kinetics of solid-phase NaPSs couples with the equilibrium coexistence of soluble NaPSs in the electrolyte, leading to progressive NaPS accumulation.^73,74 This factor further intensifies the shuttle effect and leads to lower utilization of sulfur.

Extensive research has conclusively established that the coupled mechanisms of shuttle effect and sluggish sulfur redox kinetics fundamentally limit the practical application of RT Na–S batteries. Therefore, engineering advanced adsorption–catalysis synergy has emerged as the critical strategy for achieving high-performance RT Na–S batteries.

3. Working mechanism of adsorption–catalysis synergy

The initial conversion of S₈ to Na₂S₈ is relatively effortless. However, the subsequent transition of higher-order NaPSs to lower-order NaPSs encounters larger energy barriers and becomes increasingly difficult. Particularly, the liquid–solid conversion (Na₂S₄ → Na₂S₂/Na₂S) and the solid–solid conversion (Na₂S₂ → Na₂S) demonstrate severely sluggish kinetics due to the huge energy barrier of the nucleation process and the intrinsically low conductivity of solid-state NaPSs.⁵⁵ Therefore, the difference in reaction rates in different processes leads to NaPS accumulation, further aggravating NaPS shuttling.⁶¹

The key to achieving the practical application of RT Na–S batteries is developing high-performance sulfur cathodes, fundamentally requiring rational materials engineering to address NaPS migration and reaction kinetics enhancement. The conventional approach employs porous carbon materials as sulfur hosts.⁷¹ The superior conductivity of carbon hosts compensates for the insulation of sulfur cathodes, and their porous architectures impose physical confinement effects to restrict polysulfide diffusion. However, weak interfacial interactions between polar NaPSs and nonpolar carbon surfaces fail to provide sufficient anchoring force for NaPSs.⁷⁵ Consequently, it is imperative to enhance the polarity of the sulfur host to strengthen NaPS chemisorption and minimize their dissolution. Heteroatom doping (with N, O, S, B, P, etc.) significantly enhances the adsorption capability towards NaPSs through electronic structure modulation and surface chemistry optimization. For example, N atoms enable strong interactions with the NaPSs through electrostatic interactions, effectively inhibiting their dissolution. Therefore, N-doped porous carbon cathodes demonstrate superior electrochemical performance compared to undoped cathodes.⁷⁶ Nevertheless, the inherent chemical inertness of carbon materials fundamentally restricts their catalytic capability toward NaPS redox reaction kinetics.^77,78 Passive confinement/adsorption strategies only partially restrict NaPS migration rather than fundamentally preventing their dissolution into the electrolyte.

To mitigate soluble NaPS accumulation and migration, reducing the energy barrier of the Na₂S and enhancing the kinetics of Na₂S deposition/dissolution are required.^79–81 Catalysts operate by decreasing the activation energy of reactions and optimizing redox reaction kinetics while maintaining chemical properties throughout catalytic cycles. An effective catalyst can lower the activation energy barrier for solid-state NaPS deposition and enhance the interconversion kinetics between soluble and insoluble NaPSs.^82–84 Compared to conventional adsorption strategies, introducing a catalyst effectively minimizes NaPS intermediates' accumulation and further inhibits the shuttle effect.⁸⁵

Recent advances in RT Na–S batteries demonstrate the efficacy of synergistic strategies that couple strong NaPS adsorption with catalytic conversion capabilities. The adsorption–catalysis synergy operates through sequential processes. Following adsorption of NaPSs onto the catalytic interface, the catalysts kinetically promote their redox conversion to Na₂S.^86,87 This synergistic strategy improves active sulfur utilization and NaPS conversion kinetics, effectively alleviating the shuttle effect.^88,89 Based on extensive research, we comprehensively review materials employing adsorption–catalysis synergy, which will be discussed in the following section.

4. Adsorption–catalysis synergistic effect for high-performance Na–S batteries

In practical RT Na–S battery systems, electrodes with adsorption–catalysis synergy effectively inhibit NaPS diffusion and accelerate their reaction kinetics, thereby mitigating the shuttle effect and enhancing the electrochemical behaviors of RT Na–S batteries.^90,91 Here, we summarize recent progress in multifunctional materials based on the adsorption–catalysis synergistic effect and systematically analyze their principles and effects on the performance of RT Na–S batteries (Table 1).

Table 1 Summary of different electrodes based on adsorption–catalysis synergy and their performance characteristics

Catalyst system	Electrode	Sulfur content [%]/loading	Initial capacity [mA h g^−1]	Rate	Final capacity [mA h g^−1]/cycle numbers	Capacity retention/decay rate [%]
Heterostructures	MoC–W2C-MCNFs^23	46.2	932.7	0.2 A g^−1	640/500 cycles	68.6/—
	Mo2N–W2N@PC^24	48.8	—	0.2 A g^−1	799/100 cycles	83/—
	NOC@MoS2 ^92	63	721	0.1 A g^−1	602/100 cycles	83.5/—
	CoS2–CoSe2@CNFs^25	45	—	1 A g^−1	749/200 cycles	—/0.073
	MoS2–MoN@CC^26	1.0 mg cm^−2	692	0.2C	392/300 cycles	84/—
	Se–ZnS/HSC^27	68.2	729	5 A g^−1	335.5/1000 cycles	76/—
	TiN–TiO2@MCCFs^93	∼56.9	1308.2	0.1 A g^−1	640/100 cycles	—/—
	Ni@Ni3N/CNS^94	∼70	1205	0.5C	939/600 cycles	—/0.04
	HCS@Ni–MnO2 ^95	50.7	—	5 A g^−1	586.8/1000 cycles	—/0.02
	Co/CeO2-NPC^96	47	485	5C	377.7/1000 cycles	77.88/0.027
	Co–S–C@MC^97	6 mg cm^−2	—	0.5C	910/2500 cycles	—/0.009
	Ni–B^98	—	1045	2 A g^−1	470/1000 cycles	44.9/0.054
Single atoms	NHC–InN5 SACs^99	—	—	1 A g^−1	384.9/800 cycles	90.7/0.05
	Mn1-PNC^100	—	—	0.2 A g^−1	784.6/120 cycles	84/—
	Mn/NC^101	80	—	0.5 A g^−1	720/500 cycles	72.4/—
	Fe/NC/700 ^28	46	—	10 A g^−1	325/5000 cycles	—/—
	Con-HC^30	47	1081	0.1 A g^−1	508/600 cycles	—/—
	Co,N-MPC-10% ^31	55.02	—	0.2C	1134.63/100 cycles	93.8/—
	Co–N2O2/MOFc^102	60	—	1 A g^−1	425/1000 cycles	90/—
	3D-PNCV^103	44.39	—	5 A g^−1	561/800 cycles	79/0.026
	MoS2/MoSAC/CF@S^104	—	—	0.2 A g^−1	609.54/100 cycles	—/—
	Y SAs/NC^105	67.4	—	5 A g^−1	822/1000 cycles	97.5/0.0025
	CN/Au/S^106	56.5	—	2 A g^−1	430/1000 cycles	—/—
	Zn–N2@NG^107	66	—	2 A g^−1	293/6500 cycles	—/0.0062
Metal nanoparticles	Co@PCNFs^32	38	1115	0.5C	398/600 cycles	—/—
	Co/C/rGO^108	37.5	572.8	5C	2865/1000 cycles	—/0.01
	CNTs/Co@NC-0.25 ^33	∼54	617.3	1C	450.5/400 cycles	—/0.068
	Ni-NCFs^34	36	431	1C	233/270 cycles	—/0.17
	Ni/Co–C-12 ^35	41.4	623.6	9C	414.4/650 cycles	—/—
Metal carbides	MoC/Mo2C@PCNT^36	64	905	1.5 A g^−1	650/1000 cycles	—/0.028
	MoC@NHC-15 ^37	52	581.2	1C	414.8/1000 cycles	—/0.028
	FCNT@Co3C–Co^109	∼77	—	2C	—/500 cycles	—/—
Metal nitrides	MoN@CNFs^38	61	—	2 A g^−1	214/1500 cycles	—/—
	Mo5N6 ^110	62.9	512	1.675 A g^−1	186/10 000 cycles	—/0.0064
	YS-Fe2N@NC^111	64	651	2C	467/350 cycles	—/0.0724
Metal oxides	TiO2@SPC^39	50	—	2 A g^−1	150/400 cycles	—/—
	Nb2O5–CNR^40	42	696	0.5 A g^−1	617/600 cycles	—/0.0193
	ITO@ACC^41	6.8 mg cm^−2	684	0.5C	445/1000 cycles	—/0.035
	CoMoO4@rGO^42	52	—	4 A g^−1	212.2/1000 cycles	—/0.2
	rGO/VO2/S^112	40	526.2	2C	156/1000 cycles	—/0.7
Metal sulfides	S@BPCS^113	64.5	755	0.5C	701/350 cycles	—/0.0126
	CoS2/NC^43	50.7	—	1 A g^−1	403/1000 cycles	—/—
	MoS2/NCS^44	43.8	—	1 A g^−1	360.7/2800 cycles	—/0.0055
	FL-MoS2−x@HC^45	75	1257.3	0.1 A g^−1	—/100 cycles	85.2/0.148
	FeS2-YSB@BIT^114	70.8		2C	/1000 cycles	—/0.024
	MnS@N–C^46	68	—	0.5C	213.1/300 cycles	—/0.16
	SPAN + 400 mg ZnS^115	—	786	0.14C	519/100 cycles	0.66/0
Metal selenides	Fe3Se4@NPCN^116	55	—	1 A g^−1	340/500 cycles	72/—
	NCCS^117	∼47.6	—	1C	470.3/500 cycles	—/0.044
	2HMoSe2/N-HCS/GO^118	71.4	—	0.5C	484/500 cycles	—/0.077
Heteroatom doping	FTe0.01S0.99PAN^47	51.26	—	1C	605/2400 cycles	75.2/0.01
	S/phos-C^48	70	600	1C	480/2000 cycles	—/0.013
	N-MXene@MWCNT-MP^49	73.82	—	2C	450.1/1000 cycles	—/—
	BN-C^50	51.5	—	1 A g^−1	528/1300 cycles	—/—
Others	FeNi3@HC^51	50.1	—	2 A g^−1	591/500 cycles	—/0.07
	MoTe2 ^52	∼87	1081	0.1C	901/350 cycles	—/−0.05
	Ti3C2Tx/Ni(OH)2/C^53	62.8	—	2.2C	673.5/800 cycles	—/0.032
	MMPC^54	52	1404.39	4 A g^−1	526.1/1000 cycles	58.8/—

---

4.1 Heterostructures

As discussed above, the operation mechanism of RT Na–S batteries involves intricate multi-step electrochemical reactions. Current materials systems predominantly exhibit single catalytic property, limiting their capability to modulate the multi-step conversion pathways of NaPSs. Heterostructures are composite materials formed by two or more different materials through chemical bonding.¹¹⁹ Rationally designed heterostructures can strengthen NaPS anchoring through the synergistic interaction between different materials and accelerate NaPS redox kinetics by reconfiguring the electronic configuration.¹²⁰ Recent studies have achieved significant advancements in heterostructure design for RT Na–S batteries, for example, MoC–W₂C, Mo₂N–W₂N, NOC@MoS₂, CoS₂–CoSe₂, Se–ZnS, MoS₂–MoN, TiN–TiO₂, Ni–Ni₃N, and Ni–MnO₂.^{23–27,92–95} These heterostructures establish innovative pathways to address the shuttle effect and sluggish redox kinetics.

Zhang et al. prepared a MoC–W₂C heterostructure embedded in multichannel carbon nanofibers (MoC–W₂C-MCNFS) by electrostatic spinning and high-temperature calcination (Fig. 3a and b).²³ The formation of Mo–S bonds facilitates the breaking of Na–S bonds and enhances NaPS redox kinetics, while the W₂C effectively anchors NaPSs through polar W–S bonds, suppressing shuttle effects and improving cycling stability. The adsorption energies on MoC–W₂C (Fig. 3c) are calculated to be −11 eV and −6.4 eV for Na₂S₈ and Na₂S₆, which are significantly stronger than those on MoC and W₂C single components. The lower decomposition energies of Na₂S on MoC–W₂C (0.15 eV) (Fig. 3d) further confirm the accelerated NaPS conversion kinetics. Furthermore, the Gibbs free energy profiles (Fig. 3e) demonstrate that the MoC–W₂C heterostructure shows lower energy barriers (2.3 eV for the Na₂S₈ → Na₂S₆ transition) compared to MoC (2.7 eV, Na₂S₆ → Na₂S₄) and W₂C (4.7 eV, Na₂S₄ → Na₂S₂), demonstrating its superior catalytic activity in facilitating multi-step sulfur redox reactions. Benefiting from excellent chemisorption and catalytic dual functionality of the MoC–W₂C heterostructure, the cathode retains a capacity of 200 mA h g⁻¹ after ultralong-term 3500 cycles.

Fig. 3 (a) SEM image of an S/MoC–W₂C-MCNFs composite. (b) HRTEM characterization of the S/MoC–W₂C-MCNFs. (c) Comparison of adsorption energies of Na₂S_n across MoC, W₂C, and MoC–W₂C. (d) Decomposition energy barriers of Na₂S on MoC, W₂C, and MoC–W₂C. Reproduced from ref. 23 with permission from Wiley-VCH, copyright 2024. (e) Free energy evolution during NaPS reduction reactions on MoC–W₂C. (f) HRTEM image of the Mo₂N–W₂N@PC. (g) XANES spectra at the Mo M-edge and N K-edge comparing Mo₂N–W₂N@PC with Mo₂N@PC. (h) UV-vis spectra and optical photographs of the Mo₂N–W₂N@PC, Mo₂N@PC, and W₂N@PC. (i–k) Potentiostatic discharge profiles (Na₂S₆-PC/FEC) of the (i) Mo₂N–W₂N@PC, (j) Mo₂N@PC, and (k) W₂N@PC. Reproduced from ref. 24 with permission from Wiley-VCH, copyright 2021. (l) SEM image of an NOC@MoS₂ composite. (m) CV curves of NOC@MoS₂ and NC@MoS₂ symmetric cells. Tafel plots of (n) NOC@MoS₂/S and (o) NOC/S cathodes with ±1% slope deviation. Reproduced from ref. 92 with permission from Elsevier, copyright 2024.

Mo₂N–W₂N heterostructures can be embedded into a spherical carbon superstructure (Mo₂N–W₂N@PC) to act as a sulfur host (Fig. 3f).²⁴ As evidenced by the data of the Mo M-edge and N K-edge X-ray absorption near edge structures (XANES) (Fig. 3g), the peak shifts to higher energy in the Mo₂N–W₂N@PC sample and this can be attributed to the chemical coupling of Mo₂N and W₂N, demonstrating the formation of a heterogeneous interface. The strong absorbability of Mo₂N–W₂N@PC also can be confirmed (Fig. 3h) in the UV-vis spectrum and insets. Besides, the Mo₂N–W₂N@PC exhibits the highest Na₂S precipitation capacity and fastest conversion response (Fig. 3i–k) among all tested materials, confirming its superior catalytic performance in promoting efficient sulfur redox reactions. Therefore, the S/Mo₂N–W₂N@PC cathode showed an excellent capacity retention of 83% after 100 cycles.

Wu et al. obtained an N,O-codoped flower-like carbon (NOC) through controlled carbonization and acid treatment of a Ni-metal–organic framework (Ni-MOF) precursor.⁹² Subsequently, MoS₂ nanosheets were grown on the surface of NOC via a solvothermal reaction and the NOC@MoS₂ heterostructure was constructed (Fig. 3l). The higher response current density (Fig. 3m) and smaller Tafel slopes (Fig. 3n and o) demonstrate that the NOC@MoS₂ heterostructure effectively lowers the energy barriers and promotes the charge transformation. Thus, the RT Na–S batteries based on NOC@MoS₂ maintained a capacity of 390 mA h g⁻¹ after 1000 cycles at high current density.

Co nanoparticles encapsulated in carbon nanofibers (Co NPs@CNFs) were fabricated via carbonization of a cobalt-containing precursor, followed by sequential in situ sulfidation and selenization processes to convert Co NPs into CoS₂–CoSe₂@CNFs heterostructures (Fig. 4a).²⁵ UV-vis adsorption spectra of Na₂S₄ solution (Fig. 4b) showed that the Na₂S₄ signal disappeared completely upon CoS₂–CoSe₂@CNFs addition. And the solution containing CoS₂–CoSe₂@CNFs host materials became clarified after standing still for 24 h, demonstrating the excellent adsorption ability of CoS₂–CoSe₂@CNFs. Besides, density functional theory (DFT) calculations and Na₂S nucleation experiments revealed that the CoS₂–CoSe₂@CNFs significantly enhance NaPS conversion kinetics. Due to the adsorption–catalysis synergy, the CoS₂–CoSe₂@CNFs cathode exhibited an outstanding rate capability and reversible capacity.

Fig. 4 (a) SEM image of S/CoS₂–CoSe₂@CNFs. (b) UV-vis spectra and optical photographs of different materials. Reproduced from ref. 25 with permission from Wiley-VCH, copyright 2022. (c) SEM image of MoS₂–MoN@CC. (d) HRTEM image of MoS₂–MoN@CC. (e) Tafel analysis for MoS₂@CC, MoS₂–MoN@CC, and MoN@CC. (f) Comparison of peak current in the Na–S cells with MoS₂@CC, MoS₂–MoN@CC and MoN@CC. Reproduced from ref. 26 with permission from Elsevier, copyright 2021. (g) TEM image of S@Se–ZnS/HSC. (h) HRTEM image of S@Se–ZnS/HSC. (i and j) Electronic structure differences for Se–ZnS and ZnS. (k) Visual NaPS adsorption test. (l) Potentiostatic polarization curves and Tafel plots. Reproduced from ref. 27 with permission from Wiley-VCH, copyright 2024.

The MoS₂–MoN@CC heterostructures can be constructed by annealing the hydrothermally synthesized MoS₂@CC at 650 °C for 3 hours in an NH₃ atmosphere (Fig. 4c and d).²⁶ During battery operation, the MoS₂ suppresses the shuttle effect by chemically anchoring soluble NaPSs, and MoN enhances NaPS redox kinetics through accelerating charge transformation. The catalytic effect in the MoS₂–MoN@CC can be confirmed by the lower Tafel slopes (Fig. 4e) and higher diffusion coefficient of Na ions (Fig. 4f). The obtained cathode with MoS₂–MoN@CC manifested a high initial capacity of 786 mA h g ⁻¹.

The modified S and Se–ZnS nanocrystals encapsulated by hollow and hierarchical carbon spheres (S@Se–ZnS/HSC) (Fig. 4g and h) can be prepared by selenide annealing, hydrofluoric acid etching and sulfur vapor permeation, and the Se–ZnS nanocrystals are formed by partial substitution of S in ZnS with Se.²⁷ The introduction of Se alters the electron density distribution of ZnS (Fig. 4i and j) and makes the Zn atom the electron-rich center to attract NaPSs. The electronic structure restructuring improves the electrical conductivity of ZnS, enhancing its adsorption capability and catalytic activity toward NaPSs. The higher adsorption energies and lower diffusion barriers calculated by DFT indicate that the Se–ZnS/ZnS heterostructures can strengthen the binding with NaPSs and enhance the interfacial kinetics of Na–S batteries. Besides, the electrochemical tests (Fig. 4k and l) further confirm the effective adsorption capability and catalysis of Se–ZnS/HSC. Therefore, the battery using the S@Se–ZnS/HSC cathode exhibits an outstanding capacity of 378.5 mA h g⁻¹ after 50 cycles at −10 °C. And it can deliver superior discharge capacity even at −20 °C, −30 °C, and −40 °C.

The TiN–TiO₂ heterostructures can grow in multichannel carbon fibers (TiN–TiO₂@MCCFs) (Fig. 5a) through convenient electrospinning and nitriding annealing.⁹³ DFT calculations reveal that the conductive TiN enhances the Na⁺ transport path (Fig. 5b) and the polar TiO₂ exhibits strong interaction with NaPSs (Fig. 5c). Benefiting from the synergistic combination of strong NaPS anchoring, superior catalytic activity and structural integrity, TiN–TiO₂@MCCFs effectively suppress NaPS shuttling while accelerating solid-phase NaPS nucleation kinetics as shown in Fig. 5d. Thus, the obtained S@TiN–TiO₂@MCCFs exhibits exceptional capacity retention after long-term cycling.

Fig. 5 (a) SEM morphology of S/TiN–TiO₂@MCCFs (inset: electrodes obtained by an electrospinning method). (b) Na⁺ diffusion pathways on the TiN surface. (c) DFT-optimized binding configurations of Na₂S_x species on TiO₂. (d) Proposed adsorption–conversion mechanism for Na₂S_x intermediates. Reproduced from ref. 93 with permission from American Chemical Society, copyright 2021. (e) HRTEM image of a Ni@Ni₃N/CNS sample. (f) CV of symmetric cells with varied electrodes. Reproduced from ref. 94 with permission from American Chemical Society, copyright 2021. (g) HRTEM image of a S/HCS@Ni–MnO₂ composite. (h) The adsorption energies of Na₂S_n (n = 2, 4, 6, 8) and S₈ molecules on the surfaces of Ni, MnO₂ and Ni–MnO₂ heterostructures. (i and j) Diffusion coefficients of sodium ions and GITT curves. Reproduced from ref. 95 with permission from Elsevier, copyright 2023.

The Ni@Ni₃N heterostructure (Fig. 5e) can be synthesized via chemical blowing and subsequent nitridation in ammonia.⁹⁴ The Ni@Ni₃N heterogeneous interface optimizes the electronic structure and promotes the liquid–solid transition of NaPSs. It can be seen from CV profiles (Fig. 5f) that the Ni@Ni₃N embedded into the 3D carbon nanosheet (Ni@Ni₃N/CNS) exhibits sharper redox peaks and smaller peak separation, further confirming enhanced catalytic efficiency of Ni@Ni₃N. Therefore, the cell with Ni@Ni₃N/CNS achieved a high initial capacity of 1345 mA h g⁻¹.

Moreover, the Ni–MnO₂ heterostructure can be constructed on the hollow carbon sphere (HCS@Ni–MnO₂) (Fig. 5g) through high-temperature annealing and alkaline etching of the SiO₂@PDA precursor.⁹⁵ Such a binary system synergistically integrates the features of Ni metal clusters' high conductivity and abundant chemisorption sites on the polar MnO₂ surface. DFT simulations (Fig. 5h) reveal that the Ni–MnO₂ heterostructures exhibit intermediate adsorption energies between Ni and MnO₂, enabling synergistic optimization of NaPS adsorption and desorption behaviors. Furthermore, the higher Na⁺ diffusion coefficients of S/HCS@Ni–MnO₂ (Fig. 5i and j) confirm improved electron transfer behavior and reaction kinetics. Due to the dual functionality of Ni–MnO₂ heterostructures, this cathode demonstrated exceptional cycling stability with 586.8 mA h g⁻¹ retained after 1000 cycles at 5 A g⁻¹.

Heterostructure catalysts play a pivotal role in regulating sulfur nucleation behavior through synergistic interfacial effects. By integrating the inherent properties of distinct components, these catalysts effectively anchor NaPSs and accelerate their redox kinetics, offering a promising strategy to enhance the performance of RT Na–S batteries. However, the complex and costly synthesis processes of existing heterostructures significantly hinder their large-scale industrial application. Therefore, it is imperative to develop novel heterostructures with simplified synthesis processes and compatibility with current industrial manufacturing systems.

4.2 Single atoms

Generally, the catalytic activity is predominantly governed by the number and intrinsic activity of exposed active sites on the surface. Compared to conventional catalysts, single atom catalysts (SACs) with large specific surface area have more active sites, along with specificities such as tunable coordination environments, quantum size effects and high atom utilization (nearly 100%).¹²¹ Due to their unique atomic-scale advantages, SACs have been extensively employed in various fields including photocatalysis, CO₂ reduction, hydrogen evolution via water electrolysis, oxygen reduction reaction and NaPS conversion. In the field of RT Na–S batteries, SACs combine the advantages of strong NaPS adsorption and fast charge transfer, thereby promoting the conversion between NaPSs and Na₂S.¹²² Furthermore, SACs can function as the sulfophilic site to regulate the nucleation and dissociation of Na₂S during the charge/discharge process.^123,124 These synergistic effects effectively mitigate the shuttle effect originating from the NaPS dissolution and sluggish kinetics.

To achieve reduced diffusion and lower reaction energy barriers of NaPSs, a nitrogen-doped hollow carbon-supported In–N₅ single-atom (NHC–InN₅ SAC) was synthesized under the guidance of DFT.⁹⁹ Wu et al. enhanced the orbital overlap degree (OOD) between the s-orbitals of Na in NaPSs and the p-orbitals of In through modulating the coordination structure of In (InN₄ → InN₅), resulting in strengthened orbital hybridization.⁹⁹ DFT calculations revealed that enhanced OOD significantly improved the NaPS reaction kinetics, which was confirmed by electrochemical tests (Fig. 6a and b). Peak 2 (the production of Na₂S) of the distribution of relaxation time (DRT) profiles of S/NHC–InN₅ (Fig. 6c) showed lower intensity and narrower width, exhibiting enhanced Na₂S nucleation kinetics. The combination of InN₅ with a hollow carbon framework achieves atomic-level dispersion, enabling efficient adsorption and accelerated conversion kinetics of NaPSs. Therefore, the Na–S pouch cells maintained 490.7 mA h g⁻¹ at 2 A g⁻¹ with a low E/S ratio of 4.5 μL mg⁻¹ due to the superior electrocatalytic activity of NHC–InN₅.

Fig. 6 (a) Contour maps of the CV curves of S/NHC–InN₅ SAC batteries. (b) CV curves of symmetric batteries of NHC–InN₅ SACs, NPC-InN₄ SACs, NHC, and a blank sample. (c) Decoupling of the DRT peak. DRT of S/NHC–InN₅ SACs, S/NPC-InN₄ SACs, and S/NHC. Reproduced from ref. 99 with permission from Wiley-VCH, copyright 2024. (d) Comparison of E_ds values from the DFT calculations and the full-fit results using the ML algorithm for gradient boosted regression (GBR). Reproduced from ref. 100 with permission from Springer Nature, copyright 2024. (e) Potentiostatic discharge profiles of the Na₂S₆ solution across different electrodes. (f) Stepwise sulfur evolution thermodynamics: ΔG for Mn/NC vs. NC. (g and h) Schematic diagram of the electrochemical redox process of sulfur species confined in different matrices. Reproduced from ref. 101 with permission from Chinese Chemical Society, copyright 2024.

Traditional SAC screening predominantly relies on extensive trial-and-error experiments, which are inefficient and costly. Lei et al. combined machine learning (ML) algorithms (Fig. 6d) with DFT calculations to predict that the Mn₁ single-atom site exhibited optimal adsorption strength, thus promoting S–S bond cleavage while circumventing desorption difficulties caused by excessive adsorption.¹⁰⁰ Guided by preliminary theoretical calculations, the Mn₁ was synthesized through in situ deposition on porous N-doped carbon nanospheres (Mn₁-PNC). Comprehensive structural characterization and electrochemical evaluation systematically confirmed the high catalytic activity and adsorption capacity of Mn₁-PNC. Thus, the Mn₁ accelerated the sulfur redox conversion (S₈ → Na₂S₄ → Na₂S₂/Na₂S) and inhibited long-chain NaPS diffusion to suppress the shuttle effect.

Moreover, other studies also corroborated the exceptional catalytic performance of Mn SACs. For example, the hierarchical sandwich-structured carbon matrix with atomically dispersed Mn–N₄ facilitated efficient chemical adsorption via forming Mn–S bonds with NaPSs through strong Lewis acid–base interactions.¹⁰¹ The faster current response and higher capacity of Mn/NC in Na₂S deposition profiles (Fig. 6e) exhibited superior Na₂S nucleation kinetics. And the reduced activation barriers for both Na₂S₄ → Na₂S reduction and Na₂S decomposition processes confirmed the excellent catalytic activity of Mn/NC (Fig. 6f). Benefiting from the effective suppression of the shuttle effect and enhanced sulfur redox kinetics (Fig. 6g and h), the S@Mn/NC cathode enabled pouch cells to achieve a high energy density of 840 W h kg⁻¹.

To design a high-performance sulfur cathode, Ruan et al. constructed linearly interlinked iron single-atom catalysts (IFeSACs) in an interconnected columnated carbon channel (Fe/NC) (Fig. 7a).²⁸ IFeSACs can promote electron transfer and provide highly active sites to anchor NaPSs. X-ray absorption near-edge structure (XANES) analysis at the Fe K-edge (Fig. 7b) revealed a shift toward the higher energy region in the S@Fe/NC/700 during discharge. This shift can be attributed to the elevation of the electron binding energy caused by the enhanced nuclear charge of Fe. Specifically, Fe can provide electrons to NaPSs during the discharge process, promoting the conversion from long-chain intermediates to short-chain products (Fig. 7c). Consequently, the S@Fe/NC/700 cell could deliver an excellent reversible capacity of 325 mA h g⁻¹ even at 10 A g⁻¹ after 5000 cycles.

Fig. 7 (a) HAADF-STEM image and inverse FFT pattern of Fe/NC/700, and the 3D electron intensity profile of Fe/NC/700, Fe/NC/600 and Fe/NC/800. (b) Ex situ Fe K-edge XANES spectra of the S@Fe/NC/700 sample during the first discharge and charge processes. (c) Mechanism diagram for S@Fe/NC/700. Reproduced from ref. 28 with permission from Wiley-VCH, copyright 2024. (d) HAADF-STEM image of atomic cobalt distribution in S@Con-HC. Reproduced from ref. 30 with permission from Springer Nature, copyright 2018. (e) SEM image of the 3D-PNCV. (f) UV-vis absorption spectra of the Na₂S₆ solution after the adsorption experiment in the presence of 3D-PNC and 3D-PNCV (inset: digital images). (g) Adsorption energy of various S species adsorbed on 3D-PNC and 3D-PNCV. Reproduced from ref. 103 with permission from Wiley-VCH, copyright 2022. (h) Charge/discharge profile and in situ Raman contour plots and in situ XRD patterns of the RT Na–S battery with MoS₂/CF@S during the initial cycle. (i) In situ Raman spectra and in situ XRD spectra of MoS₂/CF@S measured by galvanostatic charge–discharge measurements. Reproduced from ref. 104 with permission from American Chemical Society, copyright 2024.

DFT calculations reveal that the Co atom demonstrates strong adsorption energy toward Na₂S and effectively lowers the energy barriers for Na₂S₄ → Na₂S conversion.^29,125 Zhang et al. successfully reconstructed Co nanoparticles into atomically dispersed structures through a sulfur-diffusion-induced atomic migration strategy.³⁰ The atomic Co supported in micropores of hollow carbon nanospheres served as an advanced sulfur host (S@Co_n-HC) (Fig. 7d). The strong chemical adsorption via Co–S bonds and physical confinement by HC effectively suppressed the dissolution of NaPSs. And the high electrochemical catalytic activity of atomically dispersed Co accelerated the sulfur species redox reaction kinetics. Benefiting from the adsorption–catalysis synergistic effect, S@Co_n-HC cathodes showed a high initial capacity.

To further suppress NaPS shuttling, researchers have achieved enhanced catalytic performance in Co SACs via coordination environment optimization. For example, Jin et al. designed a Co, nitrogen co-doped microporous carbon matrix (Co,N-MPC) as a sulfur host.³¹ Due to the superior NaPS anchoring effect and lower activation energy for conversion reactions by constructing a Co–N_x coordination environment, this novel catalyst accelerated the nucleation process of Na₂S. Moreover, Hu et al. adopted a N/O dual-coordination strategy to adjust the local coordination environment of Co atoms, replacing the Co–N₄ structure with Co–N₂O₂ coordination.¹⁰² This modification induced a reduction in d-orbital electron density of the Co center and facilitated d–p hybridization between S atoms and Co atoms. The reconstructed electronic structure enhances both NaPS adsorption and their redox kinetics, enabling the assembled battery with a Co–N₂O₂ modified cathode to deliver a specific capacity of 350 mA h g⁻¹ even at 10 A g⁻¹.

DFT calculations demonstrate that the V SACs exhibit a lower energy gap (E_d) between the metal element's d-band center and the sulfur p-band center compared to other transition metals (Fe, Co, Mn).¹²⁶ The lower E_d optimizes d–p orbital hybridization and significantly reduces Na₂S decomposition energy barriers. Furthermore, V atoms can effectively anchor NaPSs through the formation of strong V–S bonds. For example, Jiang et al. successfully incorporated atomic-dispersed V into a porous N-doped carbon (3D-PNCV) after the heat treatment (Fig. 7e).¹⁰³ V atoms coordinate with N in the carbon matrix to form V–N bonds and establish V–S bonds with NaPSs. This dual chemical bonding configuration effectively enhanced adsorption capability. In NaPS adsorption experiments (Fig. 7f), 3D-PNCV decolorized Na₂S₆ solution after adsorption for 3 h and significantly reduced the UV-vis adsorption spectra peak intensity, validating the predictions of DFT (Fig. 7g). Consequently, the batteries with the 3D-PNCV achieved excellent cycling stability and rate performance.

The preparation of SACs often involves complex pre-synthesis and functionalization steps, significantly hindering their practical production. To address this issue, Zhong et al. developed an electrochemical self-reconstruction strategy for Mo single atoms.¹⁰⁴ The key mechanism involves voltage-controlled phase transitions of MoS₂ during the initial discharge/charge cycles, enabling in situ generation of Mo single atoms and their integration with unconverted MoS₂ to form a composite catalytic phase (MoS₂/Mo). Highly active sites of Mo SACs and stable anchoring sites of MoS₂ synergistically enhanced NaPS adsorption and conversion kinetics. The change of MoS₂ characteristic peaks observed in in situ Raman spectra and in situ XRD patterns of MoS₂/CF@S (Fig. 7h) demonstrated the evolution of MoS₂ conversion behavior during discharge/charge cycles. In situ Raman spectra revealed the dynamic evolution of S–S and S–Mo bonds, evidencing strong chemical interaction between Mo SACs and NaPSs. And in situ XRD patterns (Fig. 7i) further confirmed their catalytic activity through electrochemical reconstruction. This work provided new insights for high-performance RT Na–S batteries' practical application.

Benefiting from the abundant active sites and high catalytic activity of SACs, these catalysts effectively anchor NaPSs and accelerate their conversion, thereby significantly suppressing the shuttle effect. However, the high surface energy of metal single atoms renders them prone to agglomeration in the synthesis and working process, causing structural degradation and catalyst failure. This characterization necessitates low metal loading in current SAC systems, severely limiting their catalytic performance. Meanwhile, the complex preparation process and high cost of raw materials also restrict their industrialization application. To overcome these issues, developing SACs with high metal loading, robust stability and facile fabrication is critically important.

4.3 Metal nanoparticles

Metal nanoparticles (NPs) exhibit outstanding advantages in anchoring NaPSs and catalyzing their conversion due to their high specific surface area and exceptional conductivity. Despite exhibiting lower atomic utilization efficiency and lower catalytic activity compared to single-atom catalysts, the mature synthesis protocols for nanoparticles substantially reduce their fabrication complexity and production cost. Thus, the metal nanoparticles have been extensively employed in constructing high-performance RT Na–S batteries.

Chain-mail catalysts with porous nitrogen doped carbon nanofibers (PCNFs) encapsulating Co NPs (Co@PCNFs) (Fig. 8a) can be designed as a sulfur host.³² Co NPs encapsulated with N-doped carbon nanofibers function as electron donors to accelerate the electron transfer from Co to NaPSs. This configuration significantly reduces the activation energy barriers for both NaPS reduction and Na₂S oxidation. A visual Na₂S₆ adsorption experiment and calculated binding energies (Fig. 8b) clearly showed the strong NaPS anchoring capability of Co@PCNFs. In situ XRD analysis (Fig. 8c) revealed rapid disappearance of the signal of Na₂S_x intermediates during the discharge process, with only Na₂S₄ detectable when discharged to 1.5 V. This phenomenon indicated that Co@PCNFs catalytically converted long-chain NaPSs to short-chain NaPSs. Notably, the absence of Na₂S₂ during the charge/discharge process confirmed the accelerated conversion kinetics between Na₂S₄ and Na₂S. Due to the adsorption–catalysis synergistic effect, Co@PCNFs/cathode exhibited a high reversible capacity under a high sulfur loading.

Fig. 8 (a) SEM morphology of a Co@PCNFs composite. (b) Comparative binding energies of NaPS intermediates on N-G vs. Co@N-G hybrids. (c) In situ XRD patterns of Co@PCNFs/S during cycling with corresponding voltage profiles at 0.1C. Reproduced from ref. 32 with permission from Wiley-VCH, copyright 2021. (d) Schematic illustration of the advantages of S@Co/C/rGO. Reproduced from ref. 108 with permission from Elsevier, copyright 2020. (e) SEM image of S@CNTs/Co@NC-0.25. Reproduced from ref. 33 with permission from Wiley-VCH, copyright 2021. (f) FESEM of the Ni-NCFs composite. (g) XPS spectra of Ni 2p from Ni-NCFs and S@Ni-NCFs. Reproduced from ref. 34 with permission from Wiley-VCH, copyright 2019. (h) FESEM image of Ni/Co–C-1. (i) In situ Raman tracking of sulfur speciation in S@Ni/Co–C during Na–S cell operation. Reproduced from ref. 35 with permission from American Chemical Society, copyright 2021.

Similar Co NPs encapsulated with a carbon matrix were also reported by Ma et al.¹⁰⁸ They prepared the carbon-coated Co nanocomposites supported by graphene aerogel (Co/C/rGO) as a multifunctional sulfur host. The adsorption–catalysis synergy of Co NPs enhanced NaPS chemisorption and conversion kinetics (Fig. 8d), enabling the RT Na–S batteries to deliver a high initial specific capacity even at 5C.

Self-grown CNTs and Co nanoparticles embedded into N-doped carbonaceous materials were also designed as a 3D sulfur host (S@CNTs/Co@NC) (Fig. 8e).³³ DFT calculations and experimental analyses exhibited that the synergistic effects arising from Co NPs and NC established multiple active sites, effectively suppressing the shuttle effect while catalyzing NaPS conversion through electronic structure optimization. Benefiting from the enhanced adsorption capacity and catalytic activity, the S@CNTs/Co@NC electrode achieved an initial capacity of 1200.3 mA h g⁻¹.

Ni-NCFs composed of Ni hollow spheres and N-doped carbon fibers (Fig. 8f) can act as an efficient electrocatalyst for fast NaPS conversion.³⁴Ex situ XPS spectra (Fig. 8g) revealed the coexistence of Ni⁰ and Ni²⁺ phases. And Ni 2p peaks underwent a shift toward higher binding energy after sulfur loading. These indicated that the Ni nanoparticles can accept electrons from NaPSs to form Ni–S bonds, thus strengthening adsorption capability. The polarization curve and EIS of Na₂S₆ symmetric cells confirmed the improved NaPS reaction kinetics. As a result, the Ni-NCFs cathodes allowed the RT Na–S batteries to exhibit a stable cycle life.

Compared to single-metal materials, alloy nanoparticles enable enhanced catalytic performance in multi-step NaPS redox reactions through synergistic interactions between diverse metal components. For example, Ma et al. reported Ni/Co bimetallic nanoparticles embedded in carbon spheres with porous channels (Ni/Co–C) (Fig. 8h).³⁵ They tuned the molar ratio of Ni/Co to achieve sequential kinetics modulation in NaPS conversion. In situ Raman spectra (Fig. 8i) showed distinct Na₂S characteristic peaks in the S@Ni/Co–C-12 cathodes after being discharged to 0.5 V, validating the accelerated Na₂S nucleation kinetics through Ni–Co synergistic interactions. The significant catalytic activity of alloy nanoparticles can reduce the aggregation of NaPSs, thus suppressing the shuttle effect. The coin cells assembled with S@Ni/Co–C-12 exhibited superior rate performance even at 9C current rates.

4.4 Metal carbides/nitrides

Metal carbides dynamically reconfigure the electronic configuration through the synergistic effect of transition metal and carbon, enhancing NaPS adsorption and promoting phase evolution from NaPSs to Na₂S.

For example, Hao et al. reported a novel sulfur host based on MoC/Mo₂C nanoparticles in situ grown in porous carbon nanotubes (MoC/Mo₂C@PCNT) (Fig. 9a).³⁶ The calculated formation energy of Na₂S_x on MoC/Mo₂C (Fig. 9b) is significantly lower than that on graphene, confirming the accelerated sulfur redox kinetics of MoC/Mo₂C@PCNT-S. And the UV-vis spectra (Fig. 9c) showed that the adsorption peaks of Na₂S₄ obviously declined after exposure to MoC/Mo₂C@PCNT, which validated the outstanding NaPS anchoring ability of MoC/Mo₂C@PCNT. The polar MoC/Mo₂C with PCNT synergistically achieves efficient capture of NaPSs and catalytic acceleration of NaPS redox kinetics, thus the RT Na–S batteries with MoC/Mo₂C@PCNT cathodes can exhibit a high reversible capacity of 576 mA h g⁻¹ after 10 cycles under a high sulfur loading (5.7 mg S cm⁻²).

Fig. 9 (a) HRTEM imaging of MoC/Mo₂C@PCNT. (b) DFT calculated formation energy of Na₂S_x phases per mol of S. (c) UV-vis spectra of the Na₂S₄ blank solution, and the solutions after 1 h exposure to PCNT and MoC/Mo₂C@PCNT. Reproduced from ref. 36 with permission from Wiley-VCH, copyright 2022. (d) SEM image of MoC@NHC-15. (e) Post-adsorption Mo 3d chemical states by XPS. (f) Comparison of adsorption energies of Na₂S_x (x = 6 or 1) species on MoC@NHC and NHC. (g and h) Energy profiles of Na₂S decomposition on NHC and MoC@NHC. Reproduced from ref. 37 with permission from Elsevier, copyright 2022. (i) HRTEM image of FCNT@Co₃C–Co/S. Reproduced from ref. 109 with permission from Elsevier, copyright 2020.

A similar synergistic effect between Mo and the carbon matrix can be found in α-MoC_1−x nanoparticles distributed on N-doped hollow porous carbon spheres (MoC@NHC) (Fig. 9d).³⁷ Characteristic Mo–C and Mo–N bonds in the Mo 3d spectra (Fig. 9e) confirmed the strong interfacial coupling between α-MoC_1−x and NHC. And the formation of Mo–S bonds confirmed the strong chemical interaction between polar Mo sites and NaPSs, thus effectively restraining NaPS diffusion (Fig. 9f). The CV curves of symmetric cells, Tafel slopes and lower Na₂S decomposition energy barriers on MoC@NHC (Fig. 9g and h) highlighted the exceptional catalytic activity of MoC@NHC in enhancing conversion kinetics. The polar α-MoC_1−x embedded within NHC frameworks regulated sulfur redox pathways through suppressing NaPS migration and catalyzing the Na₂S oxidation process. The enhanced anchoring capability and accelerated kinetics of the sulfur host effectively inhibit the shuttle effect, thus achieving outstanding cyclability at 1C and 5C.

3D fluorinated-doped carbon nanotube arrays encapsulated in a porous Co₃C–Co framework (FCNT@Co₃C–Co) (Fig. 9i) can be fabricated as a novel sulfur host.¹⁰⁹ The FCNT@Co₃C–Co is favorable for NaPS immobilization due to the strong chemical interaction of F and Co with Na⁺. The superior conductivity of Co₃C ensures rapid electron transfer from sulfur to the electrode and an accelerated long-chain NaPS reduction process. And Co NPs with FCNT synergistically promote the Na₂S nucleation process. The synergistic effect between Co₃C–Co and FCNT achieves the multi-step reaction pathway optimization, enabling efficient sulfur species conversion and effective inhibition of the shuttle effect. Consequently, the RT Na–S batteries deliver an excellent initial capacity of 1363 mA h g⁻¹ under a high S loading with a lean E/S ratio.

Metal nitrides are covalent compounds consisting of highly electronegative N and metal elements. The metal atoms and N atoms on the surface strengthen the polarity of the sulfur host to effectively anchor the NaPSs. Combined with superior electrical conductivity, metal nitrides exhibit enhanced adsorption and conversion kinetics of NaPSs.

MoN can be introduced into the carbon nanofibers (MoN@CNFs) (Fig. 10a) as both sulfur (S/MoN@CNFs) and Na (Na/MoF@CNFs) dual-functional hosts.³⁸ DFT calculations revealed a 0.16 eV energy gap between the d-orbital center of MoN and the p-orbital center. This minimized d–p band center energy gap of the Mo–Na₂S₄ system reflects its strong electronic coupling and high Fermi level electron density. XANES spectra (Fig. 10b) exhibited a distinct shift of the Mo M-edge toward the higher energy region after the S loading, indicating the strong chemical interaction and localized electronic coupling between S and MoN. The calculated reaction barrier for Na₂S₄ → Na₂S₂ conversion (Fig. 10c) decreased from 0.61 eV to 0.27 eV, and higher precipitation capacity and faster current response (Fig. 10d) demonstrated accelerated NaPS redox kinetics. And in situ Raman spectra (Fig. 10e) and in situ XRD spectra (Fig. 10f) further validated the exceptional catalytic activity of MoN in NaPS conversion. The contour plot (Fig. 10g) showed that the MoN catalyst maintained a stable crystalline phase during the discharge and charge processes, confirming its structural stability. As a Na anode host, the MoN@CNFs with superior sodiophilic characteristics enabled homogeneous sodium nucleation and effective dendrite suppression. Therefore, the designed S/MoN@CNFs|Na/MoN@CNFs batteries exhibited a high specific capacity of 990 mA h g⁻¹ after 100 cycles.

Fig. 10 (a) TEM image and HRTEM image of the MoN@CNFs. (b) Mo M-edge XANES spectral shifts for MoN@CNFs vs. S/MoN@CNFs. (c) Gibbs free energy profiles of S and NaPS on the MoN (200) surface. (d) Potentiostatic Na₂S₆ reduction profiles across catalysts. (e) In situ Raman and (f) in situ XRD analysis of the S/MoN@CNFs in RT Na–S batteries at different discharge and charge stages for the first cycle. (g) The contour plot of MoN during the discharge and charge processes. Reproduced from ref. 38 with permission from Wiley-VCH, copyright 2022. (h) HAADF-STEM image of Mo₅N₆. (i) In situ XRD patterns paired with the discharge profile for S/Mo₅N₆. Reproduced from ref. 110 with permission from Springer Nature, copyright 2021. (j) Visual polysulfide adsorption test. (k) UV-vis absorbance spectra of electrolytes after adsorption of polysulfides. Reproduced from ref. 111 with permission from Elsevier, copyright 2021.

Similarly, Ye et al. prepared Mo₅N₆ (Fig. 10h) materials as the sulfur host for high performance Na–S batteries.¹¹⁰ The high-coordination Mo atom configuration in Mo₅N₆ facilitates robust chemical bonding with N atoms, creating electron-rich active centers. This electronic environment significantly enhances sulfur adsorption capability, and it can be confirmed by the Na₂S₅ adsorption energy of −9.23 eV derived from DFT calculations. In situ XRD spectra (Fig. 10i) revealed a phase evolution pathway: Na₂S₅ → Na₂S₄ → Na₂S₂ → Na₂S. More importantly, the peak of Na₂S emerged prior to Na₂S₂ phase formation, exhibiting the superior catalytic capability of Mo₅N₆ in accelerating sulfide conversion kinetics. Thus, the RT Na–S batteries using Mo₅N₆ cathodes demonstrated an exceptional cycling stability over 10 000 cycles.

The Fe₂N with N-doped carbon yolk–shell (YS-Fe₂N@NC) as a polar yolk–shell nanostructure catalyst could strengthen physical confinement and chemical adsorption toward NaPSs.¹¹¹ This can be confirmed by the visualized adsorption test and UV-vis spectra (Fig. 10j and k). And the polar Fe₂N and NC synergistically accelerate NaPS conversion kinetics. Benefiting from the combined effects of physical confinement, chemical anchoring and catalytic conversion, the YS-Fe₂N@NC composite achieves a superior initial capacity, stable cycling performance and good rate performance.

Metal carbides/nitrides have attracted significant attention due to their intrinsic high electrical conductivity. However, complex synthesis processes severely hinder their industrial application. Metal carbides typically experience a high-temperature carbonization process, leading to high energy consumption. And the production of metal nitrides demands precursor nitridation under ammonia atmospheres, accelerating equipment degradation and elevating production costs. Thus, developing simplified and energy-efficient synthesis strategies is a critical pathway to promote the practical application of metal carbides and metal nitrides.

4.5 Metal oxides

Leveraging the strong polarity between oxygen anions and metal atoms, metal oxides form strong chemical bonds with NaPSs to effectively anchor the soluble NaPSs. Compared to other transition metal compounds, metal oxides can provide more stable adsorption surfaces at a lower cost. However, the low specific surface area and poor conductivity of metal oxides necessitate integration with carbon-based hosts to fully exploit their adsorption and catalytic functions.

Among metal oxides, TiO₂ is a classic sulfur host due to its strong adsorption capability and tunable structural properties. For example, TiO₂ nanoparticle modified porous spherical carbon (TiO₂@SPC) (Fig. 11a) can be prepared as a sulfur host after the spray drying and solution soaking treatments.³⁹ In the XANES analysis (Fig. 11b), only the adsorption edge of Ti shows a shift toward lower energy levels after sulfur loading, indicating Ti valence state reduction and oxygen vacancy formation. The O²⁻ anions strengthen NaPS entrapment, and the ex situ UV spectra (Fig. 11c) further confirm its improved anchoring capability. Besides, the electrochemical deposition profiles show that the TiO₂@SPC has a larger integral area. This suggests the enhanced Na₂S deposition kinetics. The adsorption–catalysis synergy enables the TiO₂@SPC-S cathode to deliver a high reversible capacity under high sulfur loading.

Fig. 11 (a) SEM image of TiO₂@SPC. (b) Ti L-edge XANES electronic structure analysis. (c) Ex situ UV-vis adsorption comparison. Reproduced from ref. 39 with permission from Elsevier, copyright 2025. (d) TEM characterization of a S/Nb₂O₅–CNR composite. (e) Proposed mechanism for enhanced sulfur redox kinetics. Reproduced from ref. 40 with permission from Elsevier, copyright 2022. (f) EPR monitoring during Na₂S₆ catholyte addition to ITO. (g) NaPS binding geometries of ITO. Reproduced from ref. 41 with permission from Elsevier, copyright 2021. (h) Potentiostatic analysis of Na₂S₂/Na₂S deposition on rGO vs. CoMoO₄@rGO. (i) Cottrell equation linearization of deposition kinetics. (j and k) XPS spectra of Co 2p and Mo 3d in CoMoO₄@rGO and S/CoMoO₄@rGO. Reproduced from ref. 42 with permission from Elsevier, copyright 2023.

Nb₂O₅ nanoparticles can be implanted in a rod-like carbon nanoreactor (Nb₂O₅-CNR) (Fig. 11d) as the sulfur host.⁴⁰ Nb₂O₅ effectively anchors soluble NaPSs through synergistic Nb–S bonding and Na–O interactions. And the continuous conductive pathways between Nb₂O₅ and NaPSs significantly reduce conversion activation energy barriers and accelerate redox kinetics. The synergistic effect of adsorption and catalysis (Fig. 11e) effectively inhibits the shuttle effect and enhances Na₂S decomposition efficiency enabling Na–S batteries based on Nb₂O₅-CNR cathodes to achieve a high initial specific capacity of 1377 mA h g⁻¹.

Beyond binary metal oxides, ternary metal oxide catalysts have attracted widespread attention. These materials synergistically integrate the ideal properties of various metal oxides to enhance NaPS chemical anchoring and catalytic efficiency. Furthermore, the electronic structure of ternary metal oxides optimizes the redox reaction pathways of sulfur species. In a report by Kumar et al., indium tin oxide nanoparticles can be loaded onto activated carbon cloth (ITO@ACC) by solvothermal reaction and high-temperature treatment.⁴¹ During the thermal process, ITO nanoparticles formed oxygen vacancies (OVs). These OVs capture single electrons to form the single electron trapped oxygen vacancies (SETOVs). As evidenced by EPR profiles (Fig. 11f), the signal intensity progressively decreases with increasing Na₂S₆ electrolyte concentration. This trend demonstrates that these SETOVs undergo a free-radical coupling process with S₃⁻ from NaPSs, accelerating long-chain NaPSs to short-chain NaPS conversion. XPS analyses and DFT calculations (Fig. 11g) collectively confirm strong binding between ITO and NaPSs. The synergistic adsorption–catalysis effects of ITO@ACC effectively accelerate the redox kinetics of soluble NaPSs, enabling assembled batteries with ITO@ACC to achieve a reversible capacity of 310 mA h g⁻¹ after 100 cycles even at −10 °C.

Orbital hybridization-induced charge redistribution at metal active sites on the CoMoO₄ surface reduces the NaPS conversion energy barrier and accelerates the Na₂S nucleation process.⁴² The larger integral area (Fig. 11h) and the curve with a higher slope of 3.82 (Fig. 11i) according to the Cottrel equation collectively validate its superior catalytic performance. The shifts of Co 2p and Mo 3d binding energy to lower values (Fig. 11j and k) indicate the formation of Co–S and Mo–S bonds between CoMoO₄ and NaPSs, enabling effective NaPS anchoring. Moreover, the CoMoO₄–Na₂S₆ system exhibits a high adsorption energy of −3.07 eV, confirming the significant adsorption capability. Thus, CoMoO₄ modified RT Na–S batteries deliver an enhanced long-term performance.

4.6 Metal chalcogenides

Metal chalcogenides typically consist of metal elements and S/Se. The high electronegativity of S/Se lowers the d-band center of transition metals, thereby effectively suppressing excessive NaPS adsorption. Thus, the metal chalcogenides simultaneously achieve strong chemisorption and rapid electrochemical reaction kinetics. The abundant S and metal atoms on the polar surface form S–S bonds and M–S bonds with NaPSs to strongly anchor soluble NaPSs. Furthermore, the high catalytic activity of metal sulfides lowers the Na₂S deposition energy barrier, and it reduces the aggregation of soluble NaPSs. Therefore, metal sulfides serve as functional sulfur hosts, separator modifiers or interlayers for high performance RT Na–S batteries.

Co-based sulfides are widely used as dual-functional additives in Na–S batteries due to their excellent NaPS adsorption and catalysis. A unique hollow polar bipyramid prism catalytic CoS₂/C can be prepared as a sulfur host (S@BPCS) (Fig. 12a).¹¹³ The polar surface of CoS₂ effectively anchors NaPSs through Co–S bonding. XPS demonstrated a Co 2p binding energy reduction after discharging, confirming the strong Co–S chemical interaction between CoS₂ and NaPSs. The Na₂S₆ solution clarified after BPCS incorporation, further evidencing the exceptional adsorption capability of CoS₂. The d–p orbital hybridization between Co²⁺ and S enhances charge transfer and accelerates NaPS conversion. In situ XRD analysis (Fig. 12b) revealed no intermediate Na₂S₅ phase formation during charging, confirming the efficient catalytic activity of CoS₂ in promoting the Na₂S decomposition process. Consequently, the S@BPCS composite delivered a high reversible capacity of 545 mA h g⁻¹ under 9.1 mg cm⁻² sulfur loading.

Fig. 12 (a) FESEM image of BPCS. (b) In situ XRD pattern with initial discharge/charge profiles at 0.5C. Reproduced from ref. 113 with permission from Springer Nature, copyright 2020. (c) SEM image of the CoS₂/NC composite. Reproduced from ref. 43 with permission from American Chemical Society, copyright 2021. (d) Raman spectra of Na₂S, CoS, and Na₂S + CoS. Reproduced from ref. 127 with permission from American Chemical Society, copyright 2020. (e₁) Voltage profiles of the S/MoS₂/NCS electrode. (e₂ and e₃) Ex situ Raman spectra and (e₄ and e₅) ex situ XPS spectra of S/MoS₂/NCS electrodes at different states. (f) Na₂S_x adsorption energies on MoS₂ (002) vs. Na_xMoS₂ surfaces. Reproduced from ref. 44 with permission from Wiley-VCH, copyright 2021. (g) SEM image of a broken fiber and the internal particles. (h) The core–shell hybrid fiber architecture schematic. Reproduced from ref. 114 with permission from Elsevier, copyright 2021. (i) HRTEM image of SPAN + 400 mg ZnS. (j) High-resolution N 1s XPS spectra of SPAN + ZnS 400 mg. Reproduced from ref. 115 with permission from Royal Society of Chemistry, copyright 2025.

A similar report can be found regarding the mesoporous CoS₂/N-doped carbon sulfur host (CoS₂/NC) (Fig. 12c).⁴³ The NaPS adsorption capability is evidenced through visual adsorption experiments and DFT calculations. In adsorption experiments, UV-vis spectra revealed that the characteristic peaks of NaS₄ almost completely disappeared after CoS₂/NC adsorption, thereby validating its strong NaPS anchoring capability. Besides, the calculated adsorption energy of Na₂S₆ on the (100) surface of CoS₂ (−1.74 eV) was higher than that on NC, further confirming the superior chemisorption of CoS₂. As a result, the assembled batteries demonstrate superior cycling stability with high reversible capacity during long-term cycling. Moreover, Zhang et al. proposed an innovative strategy via the conversion of Co → CoS to facilitate the Na₂S nucleation process.¹²⁷ Raman spectra (Fig. 12d) revealed a new broad peak after the mixing of CoS and Na₂S, and this can be attributed to Na₂S (457 cm⁻¹) and Na₂S₄ (437 cm⁻¹). The result confirmed that CoS catalyzes the oxidation process from Na₂S to NaPS intermediates. Thus, the metal–organic framework-derived Co containing N-doped porous carbon (CoNC) cathodes exhibited enhanced rate performance.

To construct high-performance cathodes, MoS₂ can be used as an electrocatalyst to enhance NaPS adsorption and accelerate its redox kinetics. Wang et al. prepared MoS₂-embedded N-doped carbon spheres as a sulfur host (S/MoS₂/NCS) after a one-pot reaction, synchronized sulfurization and sulfur encapsulation process.⁴⁴ Exposed Mo and S atoms at (002) plane edges act as catalytic sites for NaPS chemisorption and conversion. Notably, MoS₂ underwent sodiation to form Na_xMoS₂ upon low-voltage discharge, and Raman and XPS analyses collectively confirmed the formation of Na_xMoS₂. Ex situ Raman spectra (Fig. 12e₂ and e₃) showed a shift of the MoS₂ A_1g peak to higher frequency after discharging, indicating strengthened interlayer interactions of MoS₂. And ex situ XPS spectra (Fig. 12e₄ and e₅) confirmed Mo³⁺ formation when discharged to 0.8 V. DFT calculations (Fig. 12f) demonstrated the enhanced adsorption capacity and catalytic activity of Na_xMoS₂. Benefiting from the synergistic effect of MoS₂ and Na_xMoS₂, S/MoS₂/NCS cathodes effectively suppressed the shuttle effect and achieved an excellent capacity retention after 2800 cycles.

Moreover, MoS_2−x nanosheets with surface vacancies can also be prepared as an effective catalyst. Luo et al. introduced sulfur vacancies in MoS₂@HC via H₂O₂ treatment, generating electron-rich regions and unsaturated coordination bonds. The optimized electron structure significantly enhanced NaPS adsorption and catalytic efficiency.⁴⁵ Due to these advantages, the assembled batteries with the MoS_2−x/C composite exhibited outstanding rate capability and cycling performance.

Beyond Co- and Mo-based sulfides, numerous metal sulfides with polar surfaces and abundant catalytic active sites serve as adsorbents and catalysts to suppress soluble NaPS diffusion. For example, built-in yolk–shell FeS₂@C nanoboxes in the center of the 1D carbon shell (FeS₂-YSB@BIT) (Fig. 12g) were prepared as a flexible sulfur host.¹¹⁴ Fe atoms on the FeS₂ surface form Fe–S bonds with NaPSs to realize chemisorption toward NaPSs. And heterogeneous fibers composed of yolk–shell FeS₂@C and a 1D carbon shell limit the migration of NaPSs through physical entrapment. Thus, the FeS₂-YSB@BIT effectively anchors NaPSs and catalytically accelerates their redox conversion (Fig. 12h). Furthermore, FeS₂@C/CNT can serve as the modifying layer on the separator to build the second defense and suppress the shuttle effect through adsorption–catalysis synergy. Consequently, the batteries with FeS₂-YSB@BIT cathodes and FeS₂@C/CNT separators exhibit an area capacity of 8.84 mA h cm⁻² under a sulfur loading of 9.0 mg cm⁻². A similar yolk–shell architecture material was also reported in MnS@N–C.⁴⁶

ZnS can be uniformly dispersed within the SPAN matrix through simple physical mixing and high-temperature annealing (Fig. 12i).¹¹⁵ The integration of ZnS into SPAN markedly reduces the HOMO/LUMO gap from 0.45 eV to 0.04 eV, enabling improved electronic conductivity and accelerated charge transfer kinetics. Notably, ZnS optimizes the nitrogen coordination environment, as evidenced by XPS spectra (Fig. 12j) showing increased pyridinic C content (from 35% to 65%) in the ZnS/SPAN system. The pyridinic N with higher electron density facilitates NaPS adsorption and redox kinetics. Benefiting from the suppression of the shuttle effect, the assembled pouch cell with SPAN + 400 mg ZnS demonstrates stable cycling performance, showing its practical potential for high performance RT Na–S batteries.

Since Se and S have comparable ionic radii, metal selenides exhibit similar crystal structures and surface polarity to metal sulfides. Since the electronegativity of Se (2.55) is slightly lower than that of S (2.58), metal selenides exhibit weaker adsorption capacity for NaPSs compared to metal sulfides. However, the Se atoms on selenide surfaces can interact with S²⁻ in NaPSs through stronger van der Waals forces, providing additional physical adsorption sites to suppress NaPS shuttling. Moreover, benefiting from the fact that the conductivity (1 × 10⁻³ S m⁻¹) of Se is much higher than that of S (5 × 10⁻²⁸ S m⁻¹), metal selenides show higher electrical conductivity and enhanced catalytic performance, effectively suppressing the shuttle effect.

A 3D nitrogen-doped porous carbon nanosheet anchoring Fe₃Se₄ nanoparticles (Fe₃Se₄@NPCN) (Fig. 13a) serves as a multifunctional interlayer.¹¹⁶ The visual adsorption experiment and UV-vis spectra (Fig. 13b) confirmed the strong NaPS anchoring effect. The polarization curves of symmetrical batteries (Fig. 13c) showed a marked increase in current and integral curve area with Fe₃Se₄ addition, validating the excellent catalytic activity of Fe₃Se₄. The synergistic adsorption–catalysis effects of Fe₃Se₄@NPCN enabled the assembled batteries to show exceptional cycling stability.

Fig. 13 (a) FESEM image of Fe₃Se₄@NPCN. (b) UV-vis absorption spectra and optical photographs of Fe₃Se₄@NPCN and NPCN for Na₂S₆ solution. (c) CV curves of symmetric cells for Fe₃Se₄@NPCN and NPCN. Reproduced from ref. 116 with permission from Elsevier, copyright 2024. (d) FESEM image of NCCS. Reproduced from ref. 117 with permission from Royal Society of Chemistry, copyright 2022. (e) SPS permeation measurement in an H-type glass cell with the 2H-MoSe₂/N-HCS/GO + GF separator. Reproduced from ref. 118 with permission from Royal Society of Chemistry, copyright 2021.

CoSe₂ has strong NaPS chemical adsorption and accelerated Na₂S nucleation kinetics. A self-supporting tessellated N-doped carbon/CoSe₂ (NCCS) can be designed as a sulfur host (Fig. 13d) to effectively suppress the shuttle effect and enhance sulfur utilization.¹¹⁷ Therefore, the obtained NCCS@S cathode delivered an outstanding cycle stability.

The few-layer 2H-MoSe₂ nanoparticles can modify the N-doped hollow carbon spheres to decorate the functional separator.¹¹⁸ The 2H-MoSe₂ layered configuration exposes abundant active sites, thereby promoting NaPS redox kinetics. The calculated energy barrier for Na₂S diffusion on MoSe₂ was only 0.21 eV, and the remarkably low energy barrier accelerated Na⁺ migration and NaPS conversion. Moreover, the glass cell with the 2H-MoSe₂/N-HCS/GO + GF separator exhibited negligible NaPS penetration even after 24 h (Fig. 13e). This phenomenon confirmed that the functional separator effectively inhibits the migration of NaPSs, enabling the batteries to achieve a high reversible capacity of 787 mA h g⁻¹ after 100 cycles.

Metal sulfides are prone to irreversible phase transitions, whereas metal selenides demonstrate superior chemical stability, enabling metal selenides to maintain stable catalytic performance over prolonged cycling. The synthesis of metal selenides involves complex preparation processes and high raw material costs, while sulfur is abundant and low-cost. Thus, metal sulfides exhibit significant economic advantages. In summary, metal sulfides and selenides each show distinct advantages and limitations, and future efforts should synergistically integrate their complementary strengths.

4.7 Heteroatom doping

Carbon materials are widely employed as sulfur hosts owing to their high conductivity, porous structure, and structural stability. However, the weak interfacial affinity between the nonpolar carbon surface and polar NaPSs fundamentally limits their ability to inhibit the shuttle effect. The doping strategy can introduce high electronegativity heteroatoms to break the electroneutrality of carbon-based materials, creating a localized charge distribution and enhancing electron transport. And polar heteroatoms establish strong chemical interactions with NaPSs, significantly enhancing intermediate products' anchoring capability and effectively suppressing the shuttle effect. Furthermore, doping-induced charge redistribution optimizes the catalytic activity of carbon-based materials and promotes efficient sulfur species conversion.

Te-doped nanoflower-like sulfurized polyacrylonitrile (FTe_0.01S_0.99PAN) (Fig. 14a) can be designed as a sulfur host for a high-performance RT Na–S battery.⁴⁷ The higher I_D/I_G value (Fig. 14b) indicated that FTe_0.01S_0.09PAN possesses more surface defects, providing additional active sites to anchor NaPSs. DFT calculations revealed that the Na⁺ diffusion energy barriers of Te-SPAN are lower than those of SPAN at various sites. And a higher Na⁺ diffusion coefficient (Fig. 14c) during reduction and oxidation processes confirmed the accelerated Na⁺ migration in the electrode materials through Te doping. Moreover, Te-SPAN exhibited a lower Gibbs free energy (ΔG = −2.27 eV) compared to SPAN (ΔG = −0.52 eV) in the Na₂S deposition step, indicating that Te-doping facilitates NaPS reduction. Thus, the assembled batteries delivered an energy density of 340.9 W h kg⁻¹ even at a high sulfur loading (24 mg cm⁻²), low E/S ratio (5 μL mg_S⁻¹) and low N/P ratio (2.1).

Fig. 14 (a) The TEM image of FTe_0.01S_0.99PAN composites. (b) Raman spectra of SPAN, FSPAN and FTe_0.01S_0.99PAN composites. (c) Na⁺ diffusivity during redox processes. Reproduced from ref. 47 with permission from Elsevier, copyright 2024. (d) Sulfur discharge thermodynamics: ΔG comparison (P-doped vs. pristine carbon). (e) XPS of P 2p. Reproduced from ref. 48 with permission from American Chemical Society, copyright 2024. (f) FESEM image of an N-MXene@MWCNT-MP sample. (g) Raman spectra of an N-MXene@MWCNT-MP/S cathode and bare Ti₃C₂T_x/S cathode with a high sulfur loading after the 1000th cycle and being discharged to 1.1 V. (h) N-MXene@MWCNT-MP coin cell self-discharge static test. Reproduced from ref. 49 with permission from American Chemical Society, copyright 2021.

A rationally designed P-doped carbon (phos-C) matrix can effectively suppress the shuttle effect and significantly enhance NaPS redox kinetics.⁴⁸ The doping of P introduces structural defects into the carbon matrix, where defect sites act as localized reactive centers to enhance reaction kinetics. Gibbs free energy analysis (Fig. 14d) showed that the energy barrier for the Na₂S₄ → Na₂S₂ process decreases from 0.789 eV (pristine carbon) to 0.141 eV (P-doped carbon), demonstrating that P-doping accelerates the nucleation process of insoluble NaPSs. XPS analysis further confirmed P–C and P–O bond formation between P and the carbon framework, significantly increasing the surface polarity of the carbon matrix (Fig. 14e). The polar surface facilitated P–S covalent bond formation, enabling strong chemisorption of soluble NaPSs. Benefiting from the adsorption–catalysis synergy, the designed RT Na–S batteries achieved a superior initial capacity of 1034 mA h g⁻¹ at 0.1C.

Porous nitrogen-doped Ti₃C₂T_x MXene@multiwalled carbon nanotube microspheres (N-MXene@CNT-MP) (Fig. 14f) can be synthesized by spray drying and annealing.⁴⁹ Abundant catalytic centers on the N-MXene@CNT-MP surface effectively enhance reaction kinetics. DFT calculations demonstrated that N-MXene@CNT-MP exhibits higher adsorption energies (−1.722, −2.757 and −3.543 eV for Na₂S₂, Na₂S₄ and Na₂S₆) compared to bare MXene (−1.297, −2.180 and −3.193 eV). These results demonstrated that N-doping significantly improves the adsorption capability of MXene toward NaPSs. Ex situ Raman spectra (Fig. 14g) of the cycled electrodes demonstrated that only the N-MXene@CNT-MP cathode detected the characteristic peaks of S₈⁻ and S₆⁻, confirming effective NaPS confinement during discharge. Moreover, self-discharge tests (Fig. 14h) revealed that N-MXene@CNT-MP maintained voltage stability for over 50 h, evidencing inhibited higher-order NaPS diffusion. The significant adsorption capability and catalytic activity enable RT Na–S batteries with N-MXene@CNT-MP to deliver high reversible capacity under a sulfur loading up to 5.5 mg cm⁻².

B,N co-doped porous carbon (BN-C) serves as an efficient sulfur host, improving the electrochemical performance of RT Na–S batteries.⁵⁰ B atoms act as electron-deficient Lewis-acid sites, forming B–S bonds to achieve NaPS selective adsorption and effectively suppress their migration. And N atoms function as electron-rich Lewis-base centers, leveraging high electronegativity to interact with Na⁺ and accelerate Na⁺ transport across the carbon framework. This co-doping strategy prepared a bifunctional host that synergistically restricts NaPS diffusion and enhances redox kinetics. Therefore, BN-C based RT Na–S batteries demonstrated an excellent capacity retention after 1300 cycles.

Compared to metal compound catalysts, heteroatom doping catalysts are more suitable for large-scale application. However, the doping process often results in inhomogeneous distribution or uncontrollable chemical states of heteroatoms, causing low utilization of active sites and poor cycling stability. To address these challenges, integrating composite architectures with dynamic protection mechanisms is essential to enhance catalytic efficiency and structural stability.

4.8 Others

Beyond adsorption–catalysis multifunctional materials mentioned above, we investigated some materials that are not easy to classify but demonstrate excellent adsorption and catalytic properties. For example, Wang et al. prepared FeNi₃ alloy nanoparticles embedded in porous hollow carbon spheres (FeNi₃@HC) (Fig. 15a) under the guidance of first-principles calculations.⁵¹ The lower electronegativity of Fe (1.83) versus Ni (1.91) enables electron transfer from Fe to Ni in FeNi₃ alloys. Bader charge analysis (Fig. 15b) indicated that each Ni atom acquires about 0.14|e| from the Fe atom, resulting in an increase of Ni-3d band electron density. This optimized electronic structure enhanced NaPS anchoring and accelerated their conversion kinetics. Besides, the charge transfer shifted the Fe-3d band into electron-deficient states, thus serving as Lewis acid sites. This facilitated Fe–S bond formation with NaPSs, strengthening chemical adsorption. DFT calculations (Fig. 15c) revealed that the Gibbs free-energy barrier from Na₂S₄ → Na₂S₂ on FeNi₃@C (0.64 eV) is lower than that on Ni@C (0.72 eV), confirming that FeNi₃ promotes nucleation of Na₂S₂ and Na₂S. The S@FiNi₃@HC cathode consequently exhibited a high initial capacity of 1344 mA h g⁻¹.

Fig. 15 (a) SEM image of FeNi₃@HC. (b) Bader charge analysis of an FeNi₃ alloy. The positive and negative values represent obtaining and losing electrons. (c) Catalytic calculation: free energy landscapes for NaPS reduction on FeNi₃@C vs. Ni@C. Reproduced from ref. 51 with permission from American Chemical Society, copyright 2021. (d and e) Charge and discharge curves of a Na–S coin cell and the corresponding diffraction patterns with C/S and MTG/S cathodes. (f) The voltage profile of the C/S cathode. (g) The voltage profile of the MTG/S cathode. Reproduced from ref. 52 with permission from Springer Nature, copyright 2023. (h) TEM image of Ti₃C₂T_x/Ni(OH)₂/C. Reproduced from ref. 53 with permission from Elsevier, copyright 2023. (i) FESEM image of MMPC. Reproduced from ref. 54 with permission from Wiley-VCH, copyright 2023.

In situ grown MoTe₂ nanosheets on reduced graphene oxide flakes (MoTe₂-rGO) can be prepared as a high-performance sulfur host to suppress NaPS migration and accelerate redox kinetics.⁵² The strong chemical interaction between the Te²⁻ and Na⁺, and the formation of covalent Mo–S bonds synergistically realized effective NaPS adsorption. And the vertically grown MoTe₂ on the rGO surface exposed abundant edge-active sites to facilitate the transformation of sulfur species. In situ XRD analysis revealed that α-S₈ peaks were detected during the whole discharge process in the C/S cathode (Fig. 15d), indicating its incomplete sulfur conversion. In contrast, the α-S₈ peaks progressively diminished during discharge with no detectable intermediate phases (Fig. 15e). The difference directly exhibited enhanced catalytic performance of MTG. Besides, the characteristic plateau in charge–discharge profiles (Fig. 15f and g) confirmed the sodiation of MoTe₂ into Na_xMoTe₂ during discharge. The fast kinetics of the intercalation reaction enhanced the Na⁺ transport rates within the electrode. Benefiting from the coupling of intercalation–catalytic synergy and the adsorption–catalysis mechanism, the hybrid cathode showed a long cycle life even under high sulfur loading and lean-electrolyte conditions.

Qian et al. synthesized a MXene derived functional sulfur host based on Ni(OH)₂ coated with a carbon nanonetwork.⁵³ The obtained MXene/Ni(OH)₂/C composite (Fig. 15h) synergistically enhances NaPS anchoring and reaction kinetics, effectively inhibiting the shuttle effect. Consequently, MXene/Ni(OH)₂/C based sulfur cathodes exhibited an exceptional reversible capacity of 1175.3 mA h g⁻¹. In addition to metal-based materials, polar functional group-modified materials also demonstrate significant catalytic activity for redox kinetics enhancement. Micro–mesopores with oxygen-containing functional groups (MMPC) (Fig. 15i) can be prepared as a sulfur host after a special multi-step annealing process.⁵⁴ XPS, FTIR and CV curves confirmed that abundant oxygen-containing functional groups (C–O, C=O, –SO²⁻) serve as active sites, significantly accelerating the conversion from long-chain NaPSs to short-chain NaPSs. And the oxygen-containing functional groups strongly anchor NaPSs through polar interactions and O–S covalent bonding. As a result, the RT Na–S batteries with S@MMPC cathodes showed an excellent capacity even at 16 A g⁻¹.

5. Conclusion and perspectives

RT Na–S batteries are promising candidates for large-scale energy storage and next-generation power batteries due to their high energy density, high power density, and low cost. Nevertheless, the persistent shuttle effect seriously limits their practical applications. To address this critical challenge, extensive research efforts have focused on finding rational solutions. Among them, catalyst engineering that enables chemisorption and accelerated conversion of NaPSs has been recognized as the most viable strategy for achieving high-performance RT Na–S batteries. Here, we reviewed the electrochemical principles of RT Na–S batteries and the main challenges to their commercialization, and detailed the mechanism of the adsorption–catalysis synergistic strategy for suppressing the shuttle effect. Subsequently, various electrocatalysts with excellent adsorption capability and high catalytic activity were systematically presented, including heterostructures, single atoms, metal nanoparticles, metal carbides, metal nitrides, metal oxides, metal chalcogenides, heteroatom doping and others. Although achievements in electrocatalyst design in RT Na–S batteries have been realized, many critical challenges need to be overcome to enable the practical application of RT Na–S batteries. Therefore, engineering practical high-performance RT Na–S batteries necessitates focused attention on the following aspects:

(i) Higher practical energy density. Although the use of various catalysts achieved enhanced adsorption and accelerated reaction kinetics, most of the studies tend to be conducted with low sulfur loading, excessive electrolyte and thick sodium metal. This phenomenon reflects the gap between laboratory studies and practical energy. Thus, researchers must consider several important parameters about the practical energy density during their research process. These parameters include sulfur loading, the electrolyte-to-sulfur ratio (E/S), the negative-to-positive ratio (N/P), and so on. For example, Na–S batteries with an energy density ≥300 W h kg⁻¹ require a sulfur area loading of ≈8 mg cm⁻², along with an E/S ratio ≤5 μL mg⁻¹ and an N/P ratio ≤5.

(ii) Clear practical visibility. Despite advanced catalysts exhibiting superior catalytic performance, their practical viability is constrained by expensive raw materials, complex synthesis, and incompatibility with existing LIB manufacturing infrastructure. Research focus should shift toward earth-abundant catalysts with a facile and practical synthesis process, rather than simply pursuing excellent electrochemical properties. Besides, a dry electrode process significantly reduces manufacturing costs in large-scale production and it enables the preparation of thick electrodes with high compaction density, which can increase the sulfur loading and energy density of sulfur cathodes. Therefore, it is desirable to investigate the compatibility between advanced materials and a dry electrode process, and to evaluate the electrochemical performance of batteries fabricated via this process.

(iii) Innovative catalyst recycling technologies. In large-scale RT Na–S battery applications, advancing catalyst recycling technologies is pivotal for sustainable deployment, as these technologies enable closed-loop raw material utilization and substantial cost reduction. However, conventional catalyst designs often overlook lifecycle management, resulting in critical challenges for decommissioned catalysts, including inefficient separation, prohibitive recovery costs, and diminished catalytic activity of regenerated catalysts. Thus, future research should prioritize recyclability-driven catalyst design to realize efficient separation under mild conditions without compromising catalytic efficacy.

(iv) Pioneering sulfur host design. The difference in density between S and Na₂S can cause serious volume expansion and deformation of the cathode during cycles. It results in the loss of effectiveness of catalysts that were originally uniformly dispersed. For instance, the catalytic efficiency of single-atom catalysts embedded in the carbon matrix can be limited due to the formation of clusters. It is essential to design a sulfur host with robust structure that maintains stability during cycling to maximize adsorption–catalysis synergy.

(v) Advanced characterization technology. The complex NaPS conversion process makes it difficult to achieve the catalytic acceleration of the entire redox process by catalysts. And the dynamic evolution process of sulfur species and electrocatalysts under operating conditions poses a challenge to clearly understand the catalytic principles. Therefore, developing in situ characterization techniques is crucial to study the dynamic evolution and structure–performance relationships of NaPSs and electrocatalysts under catalytic conditions. The development of advanced characterization techniques can help reveal the catalytic mechanism and provide scientific guidance for the design of efficient catalysts.

(vi) Advanced catalyst screening technology. The development of advanced computational and theoretical chemistry provides a more efficient pathway for catalyst screening. Machine learning based on historical experiments and computational databases can train models to rapidly predict key parameters of candidate catalysts such as adsorption energy, activation energy barriers, d-band positions and so on. And the first-principles calculations can provide highly accurate theoretical support and mechanistic explanation. By integrating these techniques, the efficiency and accuracy of catalyst screening can be significantly enhanced without extensive trial and error experiments to save money and time.

In summary, advancements in electrocatalyst design have highlighted a promising direction for high-performance RT Na–S batteries. The adsorption–catalysis synergistic strategy has demonstrated exceptional effectiveness in suppressing NaPS shuttling. Consequently, developing more efficient and cost-competitive catalysts is the key to enhancing battery performance and advancing RT Na–S batteries toward commercialization.

Data availability

No primary research results, software, or code have been included in this review article, and no new data were generated or analyzed as part of this study. All data discussed within this review are sourced from previously published studies, which are appropriately cited throughout the article. Therefore, no new datasets were generated, and no additional data are available.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 52372180, 52073143, and 52131306); the Project on Carbon Emission Peak and Neutrality of Jiangsu Province (BE2022031-4); the Fundamental Research Funds for the Central Universities (2242023R10001, 2242022K40001); and the Start-up Research Fund of Southeast University (RF1028623005 and RF1028623081).

References

1. Y. Chen, Y. Yao, W. Zhao, L. Wang, H. Li, J. Zhang, B. Wang, Y. Jia, R. Zhang, Y. Yu and J. Liu, Nat. Commun., 2023, 14, 7487.
2. Y. Liang, H. Dong, D. Aurbach and Y. Yao, Nat. Energy, 2020, 5, 646–656.
3. M. K. Aslam, S. S. A. Shah, S. Li and C. Chen, J. Mater. Chem. A, 2018, 6, 14083–14090.
4. H. J. Peng, J. Q. Huang, X. B. Cheng and Q. Zhang, Adv. Energy Mater., 2017, 7, 1700260.
5. A. A. Kebede, T. Kalogiannis, J. Van Mierlo and M. Berecibar, Renewable Sustainable Energy Rev., 2022, 159, 112213.
6. J. M. Tarasconand and M. Armand, Nature, 2001, 414, 359–367.
7. N. Nitta, F. Wu, J. T. Lee and G. Yushin, Mater. Today, 2015, 18, 252–264.
8. F. Wu and G. Yushin, Energy Environ. Sci., 2017, 10, 435–459.
9. Z. Pan, D. J. L. Brett, G. He and I. P. Parkin, Adv. Energy Mater., 2022, 12, 2103483.
10. A. Manthiram, Y. Fu and Y.-S. Su, Acc. Chem. Res., 2013, 46, 1125–1134.
11. D. Kumar, S. B. Kuhar and D. K. Kanchan, J. Energy Storage, 2018, 18, 133–148.
12. Y. X. Ren, H. R. Jiang, T. S. Zhao, L. Zeng and C. Xiong, J. Power Sources, 2018, 396, 304–313.
13. A. Manthiram and X. W. Yu, Small, 2015, 11, 2108–2114.
14. C. Zhao, Q. Wang, Z. Yao, J. Wang, B. Sánchez-Lengeling, F. Ding, X. Qi, Y. Lu, X. Bai and B. Li, Science, 2020, 370, 708–711.
15. D.-Y. Wang, W. Guo and Y. Fu, Acc. Chem. Res., 2019, 52, 2290–2300.
16. W.-J. Chen, B.-Q. Li, C.-X. Zhao, M. Zhao, T.-Q. Yuan, R.-C. Sun, J.-Q. Huang and Q. Zhang, Angew. Chem., Int. Ed., 2020, 59, 10732–10745.
17. I.-S. Hwang, J.-C. Kim, S.-D. Seo, S. Lee, J.-H. Lee and D.-W. Kim, Chem. Commun., 2012, 48, 7061–7063.
18. L. Li, H. Yang, D. Zhou and Y. Zhou, J. Nanomater., 2014, 2014, 187891–187898.
19. B. H. Nguyen and V. H. Nguyen, Adv. Nat. Sci.: Nanosci. Nanotechnol., 2016, 7, 023002.
20. J. Mou, T. Liu, Y. Li, W. Zhang, M. Li, Y. Xu, J. Huang and M. Liu, J. Mater. Chem. A, 2020, 8, 24590–24597.
21. T. Tang and Y. Hou, Electrochem. Energy Rev., 2018, 1, 403–432.
22. B.-W. Zhang, T. Sheng, Y.-D. Liu, Y.-X. Wang, L. Zhang, W.-H. Lai, L. Wang, J. Yang, Q.-F. Gu, S.-L. Chou, H.-K. Liu and S.-X. Dou, Nat. Commun., 2018, 9, 4082.
23. S. Zhang, M. Huang, Y. Wang, Z. Wang, H. Wang and X. Liu, Adv. Energy Mater., 2024, 14, 2303925.
24. S. Zhang, Y. Yao, X. Jiao, M. Ma, H. Huang, X. Zhou, L. Wang, J. Bai and Y. Yu, Adv. Mater., 2021, 33, 2103846.
25. J. Wu, Z. Yu, Y. Yao, L. Wang, Y. Wu, X. Cheng, Z. Ali and Y. Yu, Small Struct., 2022, 3, 2200020.
26. S.-C. Wu, Y.-H. Huang, C.-R. Liao, S.-Y. Tang, T.-Y. Yang, Y.-C. Wang, Y.-J. Yu, T.-P. Perng and Y.-L. Chueh, Nano Energy, 2021, 90, 106590.
27. S. Luo, J. Ruan, Y. Wang, M. Chen and L. Wu, Adv. Sci., 2024, 11, 2308180.
28. J. Ruan, Y. J. Lei, Y. Fan, M. C. Borras, Z. Luo, Z. Yan, B. Johannessen, Q. Gu, K. Konstantinov, W. K. Pang, W. Sun, J. Z. Wang, H. K. Liu, W. H. Lai, Y. X. Wang and S. X. Dou, Adv. Mater., 2024, 36, 2312207.
29. K. Yang, D. Liu, Y. Sun, Z. Qian, S. Zhong and R. Wang, Nanomaterials, 2021, 11, 1197.
30. B.-W. Zhang, T. Sheng, Y.-D. Liu, Y.-X. Wang, L. Zhang, W.-H. Lai, L. Wang, J. Yang, Q.-F. Gu, S.-L. Chou, H.-K. Liu and S.-X. Dou, Nat. Commun., 2018, 9, 4082.
31. F. Jin, Y. Ning, B. Wang, Z. Ren, H. Luo, Z. Zhang, N. Zhang and D. Wang, J. Power Sources, 2023, 565, 232917.
32. H. Yang, S. Zhou, B. W. Zhang, S. Q. Chu, H. Guo, Q. F. Gu, H. Liu, Y. Lei, K. Konstantinov, Y. X. Wang, S. L. Chou, H. K. Liu and S. X. Dou, Adv. Funct. Mater., 2021, 31, 2102280.
33. J. Mou, Y. Li, T. Liu, W. Zhang, M. Li, Y. Xu, L. Zhong, W. Pan, C. Yang, J. Huang and M. Liu, Small Methods, 2021, 5, 2100455.
34. B. Guo, W. Du, T. Yang, J. Deng, D. Liu, Y. Qi, J. Jiang, S. J. Bao and M. Xu, Adv. Sci., 2020, 7, 1902617.
35. Q. Ma, W. Zhong, G. Du, Y. Qi, S. J. Bao, M. Xu and C. Li, ACS Appl. Mater. Interfaces, 2021, 13, 11852–11860.
36. H. Hao, Y. Wang, N. Katyal, G. Yang, H. Dong, P. Liu, S. Hwang, J. Mantha, G. Henkelman, Y. Xu, J. A. Boscoboinik, J. Nanda and D. Mitlin, Adv. Mater., 2022, 34, 2106572.
37. J. Mou, Y. Li, L. Ou and J. Huang, Energy Storage Mater., 2022, 52, 111–119.
38. Z. Li, C. Wang, F. Ling, L. Wang, R. Bai, Y. Shao, Q. Chen, H. Yuan, Y. Yu and Y. Tan, Adv. Mater., 2022, 34, 2204214.
39. R. Ma, L. Zhang, Y. Shi, Y. Fu, D. Wang, T. Wang, G. Yang, J. Zhang, L. Wu, D. Liu, Y. Wu and D. He, J. Colloid Interface Sci., 2025, 684, 235–242.
40. X. L. Huang, X. Zhang, L. Zhou, Z. Guo, H. K. Liu, S. X. Dou and Z. Wang, Adv. Sci., 2023, 10, 2206558.
41. A. Kumar, A. Ghosh, A. Ghosh, A. Ahuja, A. Sengupta, M. Forsyth, D. R. MacFarlane and S. Mitra, Energy Storage Mater., 2021, 42, 608–617.
42. X. Ye, S. Luo, Z. Li, J. Ruan, Y. Pang, J. Yang, J. Wang and S. Zheng, J. Energy Chem., 2023, 86, 620–627.
43. F. Xiao, H. Wang, T. Yao, X. Zhao, X. Yang, D. Y. W. Yu and A. L. Rogach, ACS Appl. Mater. Interfaces, 2021, 13, 18010–18020.
44. Y. Wang, Y. Lai, J. Chu, Z. Yan, Y. X. Wang, S. L. Chou, H. K. Liu, S. X. Dou, X. Ai, H. Yang and Y. Cao, Adv. Mater., 2021, 33, 2100229.
45. S. Luo, J. Ruan, Y. Wang, J. Hu, Y. Song, M. Chen and L. Wu, Small, 2021, 17, 2101879.
46. F. Ma, P. Hu, T. Wang, J. Liang, R. Han, J. Han and Q. Li, ACS Appl. Energy Mater., 2021, 4, 3487–3494.
47. Q. Wu, W. Zhang, M. Qin, W. Zhong, H. Yan, H. Zhu, S. Cheng and J. Xie, Nano Energy, 2024, 129, 110049.
48. Y. Wang, Y. Wang, C. Xu, Y. Meng, P. Liu, C. Huang, L. Yang, R. Li, S. Tang, J. Zeng and X. Wang, ACS Nano, 2024, 18, 3839–3849.
49. W. Bao, R. Wang, C. Qian, Z. Zhang, R. Wu, Y. Zhang, F. Liu, J. Li and G. Wang, ACS Nano, 2021, 15, 16207–16217.
50. Q. Ma, H. Zou, H. He, Y. Li and Z. Fang, Chem. Eng. J., 2023, 474, 145954.
51. L. Wang, H. Wang, S. Zhang, N. Ren, Y. Wu, L. Wu, X. Zhou, Y. Yao, X. Wu and Y. Yu, ACS Nano, 2021, 15, 15218–15228.
52. J. He, A. Bhargav, L. Su, H. Charalambous and A. Manthiram, Nat. Commun., 2023, 14, 6568.
53. S. Qian, Z. Yuan, G. Li, D. Li, J. Li and W. Han, Chem. Eng. J., 2023, 471, 144528.
54. S. F. Zhao, C. Li, Z. Cui, J. Zhang, W. Hu, R. Ma and C. M. Li, Adv. Energy Mater., 2023, 13, 2302490.
55. X. Yu and A. Manthiram, ChemElectroChem, 2014, 1, 1275–1280.
56. S. Y. Wei, S. M. Xu, A. Agrawral, S. Choudhury, Y. Y. Lu, Z. Y. Tu, L. Ma and L. A. Archer, Nat. Commun., 2016, 7, 11722–11729.
57. Y. T. Wang, Y. Hao, L. C. Xu, Z. Yang, M. Y. Di, R. P. Liu and X. Y. Li, J. Phys. Chem. C, 2019, 123, 3988–3995.
58. J. W. Robertson, C. N. Struthers and D. A. Syme, PLoS One, 2018, 13, e0191828.
59. A. Y. S. Eng, D. T. Nguyen, V. Kumar, G. S. Subramanian, M. F. Ng and Z. W. Seh, J. Mater. Chem. A, 2020, 8, 22983–22997.
60. X. W. Yu and A. Manthiram, J. Phys. Chem. C, 2014, 118, 22952–22959.
61. Y. X. Wang, B. Zhang, W. Lai, Y. Xu, S. L. Chou, H. K. Liu and S. X. Dou, Adv. Energy Mater., 2017, 7, 1602829.
62. H. Liu, W. Pei, W.-H. Lai, Z. Yan, H. Yang, Y. Lei, Y.-X. Wang, Q. Gu, S. Zhou and S. Chou, ACS Nano, 2020, 14, 7259–7268.
63. X. Xu, D. Zhou, X. Qin, K. Lin, F. Kang, B. Li, D. Shanmukaraj, T. Rojo, M. Armand and G. Wang, Nat. Commun., 2018, 9, 3870–3881.
64. L. Zeng, Y. Yao, J. Shi, Y. Jiang, W. Li, L. Gu and Y. Yu, Energy Storage Mater., 2016, 5, 50–57.
65. T. Yim, M.-S. Park, J.-S. Yu, K. J. Kim, K. Y. Im, J.-H. Kim, G. Jeong, Y. N. Jo, S.-G. Woo, K. S. Kang, I. Lee and Y. J. Kim, Electrochim. Acta, 2013, 107, 454–460.
66. T. Wang, J. He, X.-B. Cheng, J. Zhu, B. Lu and Y. Wu, ACS Energy Lett., 2022, 8, 116–150.
67. Y. X. Wang, J. P. Yang, W. H. Lai, S. L. Chou, Q. F. Gu, H. K. Liu, D. Y. Zhao and S. X. Dou, J. Am. Chem. Soc., 2016, 138, 16576–16579.
68. B. Lee, E. Paek, D. Mitlin and S. W. Lee, Chem. Rev., 2019, 119, 5416–5460.
69. X. Tang, D. Zhou, P. Li, X. Guo, B. Sun, H. Liu, K. Yan, Y. Gogotsi and G. Wang, Adv. Mater., 2020, 32, 1906739.
70. K. Xie, et al., Ferroelectric-enhanced polysulfide trapping for lithium–sulfur battery improvement, Adv. Mater., 2017, 29, 1604724.
71. Q. B. Guo, S. Li, X. J. Liu, H. C. Lu, X. Q. Chang, H. S. Zhang, X. H. Zhu, Q. Y. Xia, C. L. Yan and H. Xia, Adv. Sci., 2020, 7, 1903246.
72. T. T. Yang, W. Gao, B. S. Guo, R. M. Zhan, Q. J. Xu, H. He, S. J. Bao, X. Y. Li, Y. M. Chen and M. W. Xu, J. Mater. Chem. A, 2019, 7, 150–156.
73. Y. Liu, S. Lu, S. Weng, J. Xu, H. Tu, Z. Wang, J. Xue, L. Liu, F. Zhang, G. Sun, Y. Gao, C. Qian, Z. Liu, H. Li and X. Wu, Adv. Energy Mater., 2024, 15, 2404890.
74. J. Zhang, D.-W. Wang, W. Lv, L. Qin, S. Niu, S. Zhang, T. Cao, F. Kang and Q.-H. Yang, Adv. Energy Mater., 2018, 8, 1801361.
75. Z. Yan, Y. Liang, W. Hua, X.-G. Zhang, W. Lai, Z. Hu, W. Wang, J. Peng, S. Indris and Y. Wang, ACS Nano, 2020, 14, 10284–10293.
76. J. Mou, T. Liu, Y. Li, W. Zhang, M. Li, Y. Xu, J. Huang and M. Liu, J. Mater. Chem. A, 2020, 8, 24590–24597.
77. Q. Guo, S. Sun, K. i. Kim, H. Zhang, X. Liu, C. Yan and H. Xia, Carbon Energy, 2021, 3, 440–448.
78. L. Zhou, Z. C. Zhuang, H. H. Zhao, M. T. Lin, D. Y. Zhao and L. Q. Mai, Adv. Mater., 2017, 29, 1602914.
79. W. Luo, C.-F. Lin, O. Zhao, M. Noked, Y. Zhang, G. W. Rubloff and L. L. Hu, Adv. Energy Mater., 2017, 7, 1601526.
80. Y. Yao, L. Zeng, S. Hu, Y. Jiang, B. Yuan and Y. Yu, Small, 2017, 13, 1603513.
81. G. Nikiforidis, M. C. M. van de Sanden and M. N. Tsampas, RSC Adv., 2019, 9, 5649–5673.
82. Q. Q. Lu, X. Y. Wang, J. Cao, C. Chen, K. Chen, Z. F. Zhao, Z. Q. Niu and J. Chen, Energy Storage Mater., 2017, 8, 77–84.
83. E. Yi, E. Temeche and R. M. Laine, J. Mater. Chem. A, 2018, 6, 12411–12419.
84. C. Li, D. Yang, J. Yu, J. Wang, C. Zhang, T. Yang, C. Huang, B. Nan, J. Li, J. Arbiol, Y. Zhou, Q. Zhang and A. Cabot, Adv. Energy Mater., 2024, 14, 2303551.
85. S. Xin, Y. X. Yin, Y. G. Guo and L. J. Wan, Adv. Mater., 2014, 26, 1261–1265.
86. Q. Y. Guo and Z. J. Zheng, Adv. Funct. Mater., 2020, 30, 1907931.
87. Q. Zhang, L. Bo, H. Li, L. Shen, J. Li, T. Li, Y. Xiao, Z. Tian and Z. Li, Nano Lett., 2024, 24, 10711–10717.
88. J. Zhou, Y. Yang, Y. Zhang, S. Duan, X. Zhou, W. Sun and S. Xu, Angew Chem. Int. Ed. Engl., 2021, 60, 10129–10136.
89. J. Zhu, L. Zeng, Y. Song, F. Peng, Y. Wang, T. He, L. Deng and P. Zhang, J. Colloid Interface Sci., 2023, 647, 546–553.
90. J. Zhang, G. X. Zhang, Z. S. Chen, H. L. Dai, Q. M. Hu, S. J. Liao and S. H. Sun, Energy Storage Mater., 2020, 26, 513–533.
91. C. Wu, Y. Lei, L. Simonelli, D. Tonti, A. Black, X. Lu, W. H. Lai, X. Cai, Y. X. Wang, Q. Gu, S. L. Chou, H. K. Liu, G. Wang and S. X. Dou, Adv. Mater., 2022, 34, 2108363.
92. Y. Wu, Q. Xu, L. Huang, B. Huang, P. Hu, F. Xiao and N. Li, J. Colloid Interface Sci., 2024, 654, 649–659.
93. X. Ye, J. Ruan, Y. Pang, J. Yang, Y. Liu, Y. Huang and S. Zheng, ACS Nano, 2021, 15, 5639–5648.
94. Y. Li, X. Wang, L. Wang, D. Jia, Y. Yang, X. Liu, M. Sun, Z. Zhao and J. Qiu, ACS Appl. Mater. Interfaces, 2021, 13, 48536–48545.
95. H. Zhao, Z. Li, M. Liu, S. Ge, K. Ma, Q. Jiao, H. Li, H. Xie, Y. Zhao and C. Feng, Chem. Eng. J., 2024, 494, 153006.
96. L. Ou, J. Mou, J. Peng, Y. Zhang, Y. Chen and J. Huang, ACS Appl. Mater. Interfaces, 2024, 16, 3302–3310.
97. X. Wei, Z. Zhang, D. Luo and X. Wang, Adv. Funct. Mater., 2024, 35, 2414172.
98. B. Wang, L. Wang, B. Guo, Y. Kong, F. Wang, Z. Jing, G. Qu, M. Mamoor, D. Wang, X. He, L. Kong and L. Xu, Adv. Mater., 2024, 36, 2411725.
99. G. Wu, T. Liu, Z. Lao, Y. Cheng, T. Wang, J. Mao, H. Zhang, E. Liu, C. Shi, G. Zhou, C. He, W. Hu, N. Zhao, N. Wu and B. Chen, Angew. Chem., Int. Ed., 2025, 64, e202422208.
100. Y. J. Lei, X. Lu, H. Yoshikawa, D. Matsumura, Y. Fan, L. Zhao, J. Li, S. Wang, Q. Gu, H. K. Liu, S. X. Dou, S. Devaraj, T. Rojo, W. H. Lai, M. Armand, Y. X. Wang and G. Wang, Nat. Commun., 2024, 15, 3325.
101. H. Zhang, M. Wang, X.-L. Huang, S. Lu, K. Lu and X. Wu, CCS Chem., 2024, 6, 2289–2304.
102. P. Hu, Y.-F. Wu, X.-P. Gao, L. Huang, B.-B. Cai, Y.-X. Liu, Y. Ma, S. Jiang, F. Wang and F.-P. Xiao, Rare Met., 2024, 44, 288–299.
103. Y. Jiang, Z. Yu, X. Zhou, X. Cheng, H. Huang, F. Liu, Y. Yang, S. He, H. Pan, H. Yang, Y. Yao, X. Rui and Y. Yu, Adv. Mater., 2023, 35, 2208873.
104. X. Zhong, Y. Huang, J. Cai, Y. Li, Z. He, D. Cai, Z. Geng, W. Deng, G. Zou, H. Hou and X. Ji, J. Am. Chem. Soc., 2024, 146, 32124–32134.
105. E. Zhang, X. Hu, L. Meng, M. Qiu, J. Chen, Y. Liu, G. Liu, Z. Zhuang, X. Zheng, L. Zheng, Y. Wang, W. Tang, Z. Lu, J. Zhang, Z. Wen, D. Wang and Y. Li, J. Am. Chem. Soc., 2022, 144, 18995–19007.
106. N. Wang, Y. Wang, Z. Bai, Z. Fang, X. Zhang, Z. Xu, Y. Ding, X. Xu, Y. Du, S. Dou and G. Yu, Energy Environ. Sci., 2020, 13, 562–570.
107. D. Fang, S. Huang, T. Xu, P. Sun, X. L. Li, Y. V. Lim, D. Yan, Y. Shang, B. J. Su, J. Y. Juang, Q. Ge and H. Y. Yang, ACS Appl. Mater. Interfaces, 2023, 15, 26650–26659.
108. Q. Ma, G. Du, B. Guo, W. Tang, Y. Li, M. Xu and C. Li, Chem. Eng. J., 2020, 388, 124210.
109. G. Qin, Y. Liu, P. Han, S. Cao, X. Guo and Z. Guo, Chem. Eng. J., 2020, 396, 125295.
110. C. Ye, H. Jin, J. Shan, Y. Jiao, H. Li, Q. Gu, K. Davey, H. Wang and S. Z. Qiao, Nat. Commun., 2021, 12, 7195.
111. M. K. Aslam, T. Hussain, H. Tabassum, Z. Wei, W. Tang, S. Li, S.-j. Bao, X. S. Zhao and M. Xu, Chem. Eng. J., 2022, 429, 132389.
112. W. Du, Y. Wu, T. Yang, B. Guo, D. Liu, S.-J. Bao and M. Xu, Chem. Eng. J., 2020, 379, 122359.
113. M. K. Aslam, I. D. Seymour, N. Katyal, S. Li, T. Yang, S. J. Bao, G. Henkelman and M. Xu, Nat. Commun., 2020, 11, 5242.
114. H. Wang, C. Deng, X. Li, D. Yan, M. Xie, S. Zhang and B. Huang, Chem. Eng. J., 2021, 420, 129681.
115. C. Sanjaykumar, Sungjemmenla, M. Chandra, C. B. Soni, S. K. Vineeth, S. Das, N. Cohen, H. Kumar, D. Mandler and V. Kumar, J. Mater. Chem. A, 2025, 13, 1420–1429.
116. J. Zhang, Y. Zhou, H. Shu, Z. Yan, Z. Wu, Y. Wang, Z. Zhu and X. Wang, J. Colloid Interface Sci., 2025, 678, 292–300.
117. H. Wang, Y. Qi, F. Xiao, P. Liu, Y. Li, S.-j. Bao and M. Xu, Inorg. Chem. Front., 2022, 9, 1743–1751.
118. C. Dong, H. Zhou, B. Jin, W. Gao, X. Lang, J. Li and Q. Jiang, J. Mater. Chem. A, 2021, 9, 3451–3463.
119. C. Wang, H. Wang, X. Hu, E. Matios, J. Luo, Y. Zhang, X. Lu and W. Li, Adv. Energy Mater., 2018, 9, 1803251.
120. S. Chen, L. Liang, Y. Li, D. Wang, J. Lu, X. Zhan, Y. Hou, Q. Zhang and J. Lu, Adv. Energy Mater., 2023, 13, 2204334.
121. K. Fan, Y. Ying, Z. Lin, Y. H. Tsang and H. Huang, Adv. Energy Mater., 2023, 13, 2300871.
122. Z. Li, X. Chen, G. Yao, L. Wei, Q. Chen, Q. Luo, F. Zheng and H. Wang, Adv. Funct. Mater., 2024, 34, 2400859.
123. M. Jakhar, V. Barone and Y. Ding, Nanoscale, 2024, 16, 12982–12991.
124. H. Xu, D. Cheng, D. Cao and X. C. Zeng, Nat. Catal., 2024, 7, 207–218.
125. R. Jayan and M. M. Islam, J. Phys. Chem. C, 2021, 125, 4458–4467.
126. W. Xu, T. Feng, J. Xia, R. Cao and Q. Wu, Phys. Chem. Chem. Phys., 2024, 26, 15657–15665.
127. R. Zhang, A. M. Esposito, E. S. Thornburg, X. Chen, X. Zhang, M. A. Philip, A. Magana and A. A. Gewirth, ACS Appl. Mater. Interfaces, 2020, 12, 29285–29295.

---