S-La₂Mo₂O₉ solid solution: a sulfur cathode with a non-shaped matrix enables a better lithium–sulfur battery

Abstract

A prefabricated matrix is normally used as the cathode host for lithium–sulfur batteries to address the shuttle effect problem. Unconventionally, herein we present a non-shaped matrix for a sulfur cathode that enables a better lithium–sulfur battery. The fast oxide-ion conductor La₂Mo₂O₉ is introduced into the sulfur cathodes for the first time. Specifically, La₂Mo₂O₉ is highly dispersed in sulfur to form a solid solution (LMO-in-S), in which the two components are homogenously mixed to a molecular level, which is completely different from the conventional model. The non-shaped matrix provides enormous surface contact with sulfur and high catalytic ability for the conversion of polysulfides to deliver a high discharge capacity and satisfactory cycle stability. LMO-in-S, which exhibits a high tap density, delivers a high gravimetric capacity of 1374.1 mAh g⁻¹, corresponding to a volumetric capacity of 2294.8 mAh cm⁻³ at a 0.1C rate. Notably, LMO-in-S exhibits satisfactory cycle stability with a low fade rate of 0.07% per cycle over 400 cycles at 1C rate. Furthermore, it allows an ultra-high sulfur content (92.6 wt%) to deliver a high capacity of 1076.5 mAh g⁻¹ at a 0.1C rate. Objectively, this work breaks through the original concept of sulfur cathode structures and provides a novel possibility for developing high-performance lithium–sulfur batteries.

---

New concepts

In this manuscript, the concept of an S-La₂Mo₂O₉ solid solution as a non-shaped matrix sulfur cathode has been demonstrated to improve the electrochemical performance of lithium–sulfur batteries. S-La₂Mo₂O₉ solid solutions are prepared by the co-precipitation of sulfur and La₂Mo₂O₉, in which the host material La₂Mo₂O₉ is highly dispersed into sulfur to form a solid solution and provides a non-shaped matrix for sulfur, which is completely different from the conventional prefabricated matrices for sulfur. The ultra-small La₂Mo₂O₉ can provide a very high surface area for the adsorption and catalytic conversion of sulfur and polysulfides to achieve a high-performance lithium–sulfur battery. This study highlights the effectiveness of non-shaped matrix sulfur cathodes, which overturns the traditional understanding of sulfur electrodes. Furthermore, it is novel and interesting that S and La₂Mo₂O₉ can form a solid solution with a flower-like morphology. This work therefore demonstrates a new concept of sulfur cathode structure, and reveals new possibilities for developing high-performance lithium–sulfur batteries.

Introduction

The current age of information, mobility and connectivity is in great need of high-energy portable rechargeable battery systems that can support longer and more intensive usage of portable electronic devices, electric vehicles, drones, and artificial intelligence equipment.^1,2 In this regard, lithium–sulfur batteries based on sulfur as the cathode and metallic lithium as the anode, which offer a high theoretical energy density of 2600 Wh kg⁻¹, have been intensely considered as next-generation rechargeable batteries. Moreover, sulfur is abundant, cost-effective, and environmentally friendly compared to conventional electrode materials such as LiCO₂ and LiNi_1−x−yCo_xMn_yO₂.^3–6 However, the actual performance of lithium–sulfur batteries is far from the theoretical limit because of incomplete electrochemical operation, the electrically insulating nature of sulfur, soluble polysulfide intermediates in organic electrolytes, and the polysulfide shuttle effect, which lead to low utilization of sulfur and in turn, low discharge capacity and rapid capacity decay.^7–9 Therefore, increasing the utilization of sulfur and enhancing the cycle performance are significant factors for attaining high energy.^8,9 Over the past decade, a popular strategy for relieving these problems has been to employ a carbon or conductive polymer matrix for supporting sulfur, which improves the utilization of sulfur and immobilizes polysulfides.^10–12 However, the weak interaction between nonpolar carbon materials and polar polysulfides not only inhibits long-term cycle stability but also slows the release of active material.^13–15 Moreover, the large pore volume reduces the tap density of sulfur-based composites, thereby limiting the volumetric energy density of lithium–sulfur batteries.^16,17 Furthermore, to enhance the cycle stability and increase the absorbability of polysulfides, many highly polar compounds containing nitrides, carbides, sulfides, phosphides, MXene and metal oxides have been introduced into sulfur-based hosts.^18–22 Some metal-based compounds show good electro-catalytic activity and mediators during the electrochemical redox process, which is effective for anchoring the soluble polysulfide species, resulting in increased utilization of sulfur.^23–27 Irrespective of the matrix material, there is a consensus that the structure of the matrix is very important to its effectiveness, as it provides the charge transport networks for the electrode reactions, which directly influence the electrochemical performance. Therefore, almost all the matrix materials for sulfur cathodes in the reported literature have a nanotube, nanofiber, two-dimensional or sphere-like morphology to form matrix networks. Sulfur is loaded on the prefabricated matrices via gas/vapour phase deposition or a wet chemical method.^28–30 The matrix materials are usually electrochemically inactive and should be used in the smallest quantities possible.⁹ Additionally, to enhance the adsorption and catalysis of polysulfides, the material size should be minimized to provide a large surface area for contact with sulfur and polysulfides. Moreover, unlike the intercalation–deintercalation mechanism of lithium-ion batteries, the dissolution–deposition mechanism in lithium–sulfur batteries requires that electrode reactions occur only at sites where a sulfur or polysulfide molecule can simultaneously interact with lithium ions in the electrolyte and electrons from the external circuit. More specifically, the electrode reaction takes place at the interface where electrons are transferred to the sulfur or polysulfides via the electron transport networks of the electrode.¹⁴ When nonconductive metal compounds are used as the matrix, electron transport occurs mainly via the conductive agent mixed into the cathode materials due to the semiconductor properties of the metal compounds and the insulating nature of sulfur. This means that, as long as there is a sufficiently strong interaction between sulfur or polysulfides and the matrix material, and electrons can reach the adsorption sites, the electrode reaction can proceed smoothly. Therefore, a shaped matrix is not necessary for the sulfur cathode, and there is space to find a better solution for fabricating sulfur cathodes.

Herein, we present a new concept for sulfur cathodes in rechargeable lithium–sulfur batteries. La₂Mo₂O₉, a fast oxide-ion conductor,^31,32 is used as an immobilizer and catalyst for sulfur and polysulfides for the first time; more importantly, La₂Mo₂O₉ is not a prefabricated matrix for loading sulfur, but instead a companion mixed with sulfur at the molecular level to form a solid solution. This situation allows the La₂Mo₂O₉ particles to be of sub-nanometer size, which provides a surface area tens-to-hundreds of times larger than those of conventional matrix materials with sizes of tens or hundreds of nanometers, as shown in Fig. S1. The dispersive morphology allows it to adsorb soluble polysulfide dynamically. Moreover, even in the absence of bulk polarization, the surface phases of La₂Mo₂O₉ are obviously polar, which is helpful to stimulate the catalytic conversion of polar intermediate polysulfides.³³ Additionally, La₂Mo₂O₉ is denser (5.24 g cm⁻³) than carbon materials, enabling a sulfur cathode with high volumetric capacity. Compared to the conventional La₂Mo₂O₉ matrix, this new model clearly shows superior electrochemical performance. Furthermore, the non-shaped matrix allows an ultra-high sulfur content (92.6 wt%) to realize a true high-energy lithium–sulfur battery.

Results and discussion

Preparation and properties of LMO-in-S

To achieve the homogenous mixing of sulfur and La₂Mo₂O₉ at the molecular level, the co-precipitation method was employed (Fig. 1). First, La₂Mo₂O₉ nanofibers, which are shown in Fig. 2a, were reacted with ethylene diamine (EDA) to form a La(OH)₃ suspension (eqn (1)). The other reactant, sulfur, was dissolved in EDA to form an ionic solution (eqn (2)). The products of both eqn (1) and (2) remain stable after mixing in the alkaline solvent EDA. Subsequently, diluted HNO₃ was added to the mixture until the pH reached 7. As a result, La₂Mo₂O₉ and sulfur were generated at the same time to form the composite LMO-in-S (eqn (3) and (4)). These reactions were verified more solidly using the series of control experiments listed in the SI and Fig. S2.

La₂Mo₂O₉ + 3H₂NCH₂CH₂NH₂ + 6H₂O → 2La(OH)_3(s) + 3(H₃NCH₂CH₂NH₃)²⁺ + Mo₂O₉⁶⁻ (1)

2S₈ + 2H₂NCH₂CH₂NH₂ → (H₃NCH₂CH₂NH₃)²⁺ + (S₈NHCH₂CH₂NHS₈)²⁻ (2)

2LA(OH)_3(s) + Mo₂O₉⁶⁻ + H⁺ → La₂Mo₂O_9(s) (3)

(S₈NHCH₂CH₂NHS₈)²⁻ + 2H⁺ → 2S_8(s) + H₂NCH₂CH₂NH₂ (4)

Fig. 1 Schematic of the preparation process of the S-La₂Mo₂O₉ solid solution (LMO-in-S).

Fig. 2 SEM images of (a) La₂Mo₂O₉ nanofibers and (b) LMO-in-S. (c) and (d) High-resolution TEM image and SAED pattern of LMO-in-S.

The preparation of the LMO-in-S solid solution involves a series of key reactions that are critical for the nucleation of the composite. As HNO₃ is added dropwise, the two primary reactions described in eqn (3) and (4) collectively facilitate the nucleation process, with a preference for nucleation at the insoluble La(OH)₃. The morphology of the La₂Mo₂O₉ precursor plays a significant role in determining the final structure of the LMO-in-S. Specifically, the nanofiber morphology of the precursor is essential for forming the desired flower-like LMO-in-S structure, as depicted in Fig. 2a and b. In contrast, an irregular precursor leads to the formation of irregular LMO-in-S particles, as shown in Fig. S3. These particles interact through van der Waals forces and hydrogen bonding, forming a network that results in hexagonal nanoplates.³⁴ As the reaction progresses, free molecules in the mixture begin to redeposit on slightly larger particles, driving the system towards a thermodynamically stable state.³⁵ This redeposition depletes smaller particles, ultimately leading to the development of a large, flower-like morphology. The SEM image in Fig. 2b illustrates the surface morphology of the LMO-in-S, revealing a structure composed of numerous long hexagonal nanoplates. These nanoplates measure 2–3 μm in length, approximately 0.5 μm in width, and about 50 nm in thickness. The image clearly shows that the flower-shaped branches of the product grow in various directions and are produced in large quantities with nearly uniform sizes.

In terms of crystallographic structure, the La₂Mo₂O₉ nanofibers were assigned to a cubic structure (JCPDS 28-0509), and the diffraction peaks of the La₂Mo₂O₉ nanofibers were clearly identified in the flower-like LMO-in-S composite after the introduction of sulfur (Fig. S4). According to the XRD pattern of LMO-in-S, orthorhombic sulfur is dominant, and relatively weak diffraction peaks of monoclinic sulfur can be seen at 2θ = 20.3° and 13.4°. Some literature reports indicate that monoclinic sulfur delivers higher initial capacity for lithium–sulfur batteries because it inhibits the formation of high-order polysulfides.^36,37 In addition, the diffraction peaks of La₂Mo₂O₉ in the LMO-in-S are relatively weak, revealing its high dispersion in sulfur. This is obviously different from the XRD results for S@LMO via the typical method with La₂Mo₂O₉ nanofibers as the matrix (Fig. S5), in which the diffraction peaks of La₂Mo₂O₉ are much stronger than those of LMO-in-S (Fig. S4). The sulfur content measured via thermogravimetric analysis (TGA) is 77.1 wt% and 92.6 wt% for the two LMO-in-S samples, and 73.79 wt% for S@LMO (Fig. S6a and b). The other control sample, carbon-nanofiber-supported sulfur (S/CNFs), has a sulfur content of 73.02 wt% (Fig. S6c and S7). EDS mapping shows that the elements La, Mo, O and S are homogeneously dispersed within the LMO-in-S (Fig. S8). Very importantly, HRTEM observation reveals that the sulfur grains in the long hexagonal nanoplates have a size of less than 5 nm, and no lattice plane corresponding to the La₂Mo₂O₉ grains can be observed (Fig. 2c). It is further confirmed by the SAED results that only the electron diffraction pattern of sulfur, but not any diffraction spots of La₂Mo₂O₉, can be observed (Fig. 2d), revealing the very low crystallinity and high dispersion of the component. These results indicate that the host material La₂Mo₂O₉ is highly dispersed in the sulfur to form a solid solution, which provides a non-shaped matrix for sulfur that is completely different from the conventional prefabricated matrices for sulfur. The ultra-small La₂Mo₂O₉ can provide a very high surface area for the adsorption and catalytic conversion of sulfur and polysulfides (on the level of ∼1000 m² g⁻¹ according to the model in Fig. S1), although it is impossible to directly measure the surface area of such a non-shaped model. N₂ adsorption–desorption analysis shows that LMO-in-S has a larger specific surface area of 308.2 m² g⁻¹ compared with the conventional prefabricated matrix S@LMO (254.4 m² g⁻¹) and S/CNFs (199.4 m² g⁻¹) (Fig. S9 and Table S1).

Electrochemical performance

The initial charge–discharge curves of the LMO-in-S, S@LMO and S/CNFs composites are shown in Fig. 3a. The LMO-in-S composite delivers a high discharge capacity of 1374.1 mAh g⁻¹ based on the mass of the composite, corresponding to 1784.5 mAh g⁻¹ based on the mass of sulfur at a 0.1C rate with a coulombic efficiency of 101.4%. For lithium–sulfur batteries, an initial coulombic efficiency exceeding 100% is common because discharging occurs before charging, during which a small fraction of sulfides cannot be reversibly oxidized back to sulfur. In this case, is noted that the initial discharge capacity is even higher than the theoretical capacity of sulfur (1675 mAh g⁻¹). This can be attributed to the irreversible reaction of LMO and lithium ions at 1.7–1.8 V in the initial cycle, as confirmed by the CV result for LMO scanned in the range 1.0–3.0 V (vs. Li/Li⁺). In the second cycle, the discharge capacity is 1246.8 mAh g⁻¹, which is equal to 1617.1 mAh g⁻¹ sulfur, indicating that the non-shaped matrix promotes sulfur utilization to the largest extent. The S@LMO composite with a conventional matrix structure delivers a discharge capacity of 1142.7 mAh g⁻¹ and a coulombic efficiency of 94%, while the S/CNFs composite delivers a discharge capacity of 798 mAh g⁻¹. The gravimetric capacity delivered by the LMO-in-S composite is 1.2 and 1.72 times higher than that of the S@LMO and S/CNF composites, respectively. More importantly, the LMO-in-S composite provides an increased discharge potential and decreased overpotential, indicating higher sulfur utilization and faster electrode process kinetics. Due to the substantial difference between the tap density of the fast oxide-ion conductor La₂Mo₂O₉ and lightweight porous carbon substrates, a more dramatic change in capacity can be identified in the volumetric capacity than in the gravimetric capacity (Fig. 3c). The estimated tap density of the La₂Mo₂O₉ (LMO-in-S ∼1.67 g cm⁻³ and S@LMO ∼1.65 g cm⁻³) is larger than that of S/CNF (∼0.77 g cm⁻³). The calculated volumetric capacities of LMO-in-S and S@LMO are 2294.8 mAh cm⁻³ and 1885.5 mAh cm⁻³, respectively, while that of the S/CNFs is 614.5 mAh cm⁻³ (1.22 and 3.73 times higher than that of the S@LMO and S/CNF composites) (Fig. 3d). In terms of rate performance (Fig. 3b), the LMO-in-S composite delivered high gravimetric discharge capacity as compared with the S@LMO and S/CNF composites. The S@LMO delivers discharge capacities of 590.5, 483.3, 425.7, 135.4 and 112.6 mAh g⁻¹, respectively, at different discharge rates (0.1, 0.2, 0.5, 1, and 2C), while the S/CNF composites show gravimetric capacities of 590.5, 483.3, 425.7, 135.4 and 112.6 mAh g⁻¹. Under the same conditions, the LMO-in-S composite exhibits high gravimetric capacities of 1127.5, 893.7, 793.5, 709.7 and 605.6 mAh g⁻¹ at different discharge rates (0.1, 0.2, 0.5, 1, and 2C), confirming the improved reaction kinetics and excellent electronic/ionic transport properties. Specifically, a gravimetric capacity of 839.1 mAh g⁻¹ is retained when the discharge rate is switched back to 0.1C, which demonstrates the excellent stability of the LMO-in-S composite. Thus, the LMO-in-S composite can maintain good stability upon cycling at different discharge rates. The prolonged cycle performance of the LMO-in-S, S@LMO and S/CNFs composites at 1C rate was compared (Fig. 3e). The S@LMO and S/CNFs composites deliver a gravimetric capacity of 616.8 mAh g⁻¹ and 354 mAh g⁻¹, respectively, corresponding to volumetric capacities of 1017.7 and 272.6 mAh cm⁻³. Meanwhile, the LMO-in-S composite delivers a high gravimetric capacity of 696.7 mAh g⁻¹, corresponding to volumetric capacities of 1163.5 mAh cm⁻³, which are 1.14 and 4.3 times higher than the volumetric capacities of the S@LMO and S/CNFs composites (Fig. 3c and d). The reversible gravimetric capacity of the LMO-in-S composites is stabilized at 494.4 mAh g⁻¹ after 400 cycles with a low fading rate of 0.07% per cycle. The obviously superior cycling stability of LMO-in-S suggests that the LMO-in-S composites could improve the utilization of the active material in lithium–sulfur batteries. Even at a high rate of 5C, a high initial gravimetric capacity of 365.3 mAh g⁻¹ is delivered by the LMO-in-S composite, while the S@LMO and S/CNF composites delivered gravimetric capacities of only 211.7 mAh g⁻¹ and 154.6 mAh g⁻¹, respectively (1.73 and 2.36 times higher than those of the S@LMO and S/CNF composites) and maintained a reversible gravimetric capacity of 258.3 mAh g⁻¹ after 300 cycles (Fig. 3f). In addition, the performances of the sulfur cathode with the unshaped La₂Mo₂O₉ matrix are also better than those of the cathodes with a shaped La₂MoO₆ matrix in our previous work.³⁸ From Fig. 3c–f, it can be observed that LMO-in-S has a relatively stable coulombic efficiency with an average value higher than 97%, whether at low or high rates. However, considering the two reference samples, S@LMO shows a relatively high and stable coulombic efficiency at high rate compared with that at a low rate, while S/CNFs shows the opposite tendency. This may be because there are more opportunities for side reactions to occur at the surface of LMO at low rate, while the non-catalytic host CNFs have unstable coulombic efficiency due to limited kinetics at high rates.

Fig. 3 Electrochemical performance of LMO-in-S, S@LMO and S/CNFs. (a) Initial discharge curve. (b) Rate capacity. (c) Gravimetric capacity. (d) Volumetric capacity. (e) and (f) Cycle stability at 1C and 5C rate.

Moreover, high areal sulfur loading and low electrolyte usage are also the key factors for the practical application of lithium–sulfur batteries with high energy density.³⁹ LMO-in-S composites with high sulfur loadings of 4 and 5.73 mg cm⁻² (Fig. S11a) exhibit high gravimetric discharge capacities of 969.2 and 838.2 mAh g⁻¹ at a 0.1C rate, respectively, with good cycle performance over 100 cycles, ensuring the good electrochemical performance of LMO-in-S. Theoretically, decreasing the amount of electrolyte could increase the viscosity due to the dissolution of polar intermediate polysulfide species in the electrolyte, leading to poor utilization of sulfur.⁴⁰ However, in our case, at a low electrolyte/sulfur (E/S) ratio of 5 μL mg⁻¹, the LMO-in-S composite exhibits high gravimetric discharge capacities with good cycle stability at a 0.1C rate (Fig. S11b). Furthermore, the LMO-in-S composite also shows excellent electrochemical performance with an ultra-high sulfur content of 92.6 wt%. A high initial gravimetric capacity of 1076.5 mAh g⁻¹ at 0.1C rate and 691.7 mAh g⁻¹ at a high rate of 1C is delivered by LMO-in-S (Fig. S11c and d). In addition, to evaluate the practical potential of S/LMO-in-S, a pouch cell with an electrode size of 4.5 cm × 5.7 cm was assembled, as shown in Fig. S12. The areal loading of sulfur was 9.82 mg cm⁻², and the electrolyte/sulfur (E/S) ratio was 7 μL mg⁻¹. The initial gravimetric capacity delivered by the pouch cell is 863.1 mAh g⁻¹ at a 0.03C rate with a retained gravimetric capacity of 821.5 mAh g⁻¹ after 21 cycles, demonstrating stable cycling. The instability in the 4th and 5th cycles was due to unbalanced pressure during testing. This is very attractive for the practical application of lithium–sulfur batteries using an S/LMO-in-S electrode.

Electrochemical mechanism. In lithium–sulfur batteries, the catalytic conversion of polar soluble intermediate polysulfides is an important phase, because the polar soluble intermediate polysulfides species diffuse to the electrode–electrolyte interface and are adsorbed on the surface of the matrix material. In situ electrochemical impedance spectra (recorded before discharge and in the fully charged state after various numbers of cycles) are presented to examine the kinetic behavior of LMO-in-S with its new structure (Fig. 4a and d). The Nyquist plots consist of two semicircles (in the medium and high frequency regions) and a sloped line (in the low frequency region), which are associated with charge transfer, adsorption, and semi-infinite Warburg diffusion processes, respectively.⁴¹ The LMO-in-S composite exhibits smaller semicircles, which indicate that the charge transfer resistance (R_ct) and adsorption impedance (W_s) of the LMO-in-S composite decrease abruptly and are sustained at lower values compared to those of the conventional-structure S@LMO and S/CNF composites, confirming its excellent electro-catalytic activity, better redox kinetics and strong adsorption of soluble intermediates. Significantly, the stable and small Warburg impedance (W_o) values indicate good diffusion of the soluble intermediate polysulfide species in the LMO-in-S composite, which is very important for the adsorption and electrocatalysis processes. In contrast, the R_ct, W_s and W_o values of S@LMO are lower than those of the S/CNF composites (Fig. 4b and e). Hence, it is confirmed from the better redox kinetics, adsorption and diffusion that the LMO-in-S composites are favorable for solving the problem of capacity decay in sulfur-based cathodes for lithium–sulfur batteries.

Fig. 4 Nyquist plots and the simulated values for each element in the equivalent circuit with increasing cycle number. (a) and (d) LMO-in-S. (b) and (e) S@LMO. (c) and (f) S/CNFs.

Here, La₂Mo₂O₉ acts as a sulfur host as well as a good electro-catalyst in terms of the redox kinetics of the polar soluble intermediate polysulfides in the electrochemical dissolution–deposition processes, as shown by the cyclic voltammetry (CV) results (Fig. S13). The current densities of the oxidation and reduction peaks of the LMO-in-S and S@LMO composites are larger compared to those of S/CNFs, and LMO-in-S has comparatively larger current densities than S@LMO. The two sharp peaks of LMO-in-S at 2.319 V and 2.039 V, with larger peak intensity represented by C₁ and C₂ are related to the stepwise reduction of sulfur to long-chain polysulfides (Li₂S_n, 4 ≤ n ≤ 8) and then to short-chain discharge products (Li₂S₂/Li₂S), respectively. In the anodic scan, the two oxidation peaks at 2.301 V and 2.395 V represented by A₁ and A₂, associated to its reverse conversion, are also observed. The reduction of sulfur to lithium polysulfide and the oxidation of lithium polysulfide to sulfur are mutually reversible, and the peak potential difference (ΔE) is 75 mV. The reduction of high-order lithium polysulfide to low-order lithium sulfide and the oxidation of low-order lithium sulfide to high-order lithium polysulfide are mutually reversible, and the peak potential difference (ΔE) is 261 mV (Fig. S13b). Similarly, the cathodic peaks of S@LMO are observed at 2.322 V and 2.029 V and the anodic (oxidation) peaks at 2.311 V and 2.398 V, and the potential difference (ΔE) between the two reversible reactions is 76 mV and 282 mV, respectively (Fig. S13c). Compared with those of the S/CNFs, the anodic peaks of LMO-in-S and S@LMO are shifted toward lower potentials in the charge sweep, and the cathodic peaks of LMO-in-S and S@LMO are shifted toward higher potentials in the discharge sweep, indicating the faster sulfur redox reactions and significantly faster kinetics of La₂Mo₂O₉ during cycling.⁴² Conversely, the cathodic and anodic peaks of S/CNF are observed at 2.2935 V, 2.012 V and 2.3386, 2.399, respectively, with potential differences (ΔE) of 105 and 326 mV (Fig. S13d). The cathode peaks of S/CNF are not only comparatively broadened with decreased intensity but also slightly shifted toward lower potential values, which is caused by the polarization due to the slow diffusion of the electrolyte and the loss of active material during battery operation. Typically, if the reaction kinetics in the sulfur cathode are sluggish, polysulfide species accumulate in the electrolyte, increasing its viscosity, and the discharged products (Li₂S₂/Li₂S) readily precipitate on the matrix surface, forming an insulated passivation layer. The high viscosity and insulating passivation layer can further obstruct the fast diffusion of active species as well as slow the electrochemical reaction. Based on the evaluation of the diffusion kinetics, the La₂Mo₂O₉ can catalyze the conversion of soluble intermediate polysulfides effectively, resulting in increased utilization of sulfur and higher gravimetric capacity. Density functional theory (DFT) calculations were carried out to further reveal the detailed interactions and electronic structure during the adsorption of lithium polysulfides. Fig. S14a shows the stable configurations of Li₂S₆ on La₂Mo₂O₉, based on its stable (210) crystal plane. The specific interactions could also be visualized using the electron charge transfer analysis shown in Fig. S14b. The binding energy (E_b) indicates the binding strength between Li₂S₆ and La₂Mo₂O₉, which is −4.05 eV for the (210) planes in La₂Mo₂O₉. The electrochemical coupling between Li₂S₆ and the host material is remarkable. The strong interaction between lithium polysulfide and the host leads to the catalytic activity and electrochemical behaviors.

The XPS spectra detected lanthanum (La), molybdenum (Mo), oxygen (O) and sulfur (S) atoms (Fig. 5). Due to the spin–orbital j–j coupling, La 3d states split into doublets (La 3d_3/2 and La 3d_5/2). Both La 3d_3/2 and La 3d_5/2 show clear doublet peaks because of multiple splitting (Fig. 5a). The peaks on the high energy side are satellite peaks (838.38 eV and 855.38 eV), which are attributed to the transfer of an electron from O 2p to the empty 4f shell, leading to a final state of 3d⁹4f¹. This results from two screening processes, “screened” (S and S* peaks) and “unscreened” (M and M* peaks), of the empty 4f orbital that results from the creation of the 3d core hole. The empty 4f level is mainly screened by the electrons that belong to the outermost shell of the lanthanum (La) atom for the unscreened state. The unscreened peaks (M and M*) are associated with the final state 3d⁹4f⁰. For the screened state, the electrons belonging to O 2p atom are transferred to the 4f subshell to screen the 3d hole. The screened peaks (S and S*) are associated with the final state 3d⁹4f¹A, where A represents one O 2p hole. The energy difference (ΔE_j) between the M and S peaks is small (∼4 eV) because the degree of overlap between these two states is very large. The energy difference between the La 4f and O 2p levels is analogous to the Coulomb energy between the La 3d core hole and 4f electron.⁴³ The doublet peaks at binding energies of 834.75 eV and 851.73 eV are attributed to the La³⁺ state, and the energy difference between the spin–orbital splitting of La 3d_3/2 and La 3d_5/2 (ΔE_j = 16.98 eV) is almost the same as for the reference compound La₂Mo₂O₉ (ΔE_j = 17.0 eV).⁴⁴ The molybdenum (Mo 3d) spectra also show doublet peaks (Mo 3d_3/2 and Mo3d_5/2) due to the spin–orbital j–j coupling (Fig. 5b). The area ratios of the doublet peaks are given by the ratio of their respective degeneracy states (2j + 1). Thus, the intensity ratio I 3d_5/2/I 3d_3/2 of the Mo doublets should be 3/2. The energy difference between the spin–orbital splitting of Mo 3d_3/2 and Mo 3d_5/2 is almost the same (ΔE_j = 3.16 eV), and the doublet peaks at binding energies of 232.05 eV and 235.21 eV are attributed to its highest oxidation state of Mo⁶⁺.^45,46 A comprehensive study of β-La₂Mo₂O₉ showed that there exist three kinds of oxygen anions around the lanthanum (La) cations, and another three around the molybdenum (Mo) cations with three different bond lengths. The material surface is not completely free of impurities because of adsorbed carbon oxide and hydroxyl groups. In an oxygen–metal bond, the ability to eject an O 1s electron increases with ionic character, leading to lower binding energy.⁴⁷ In the O 1s spectra (Fig. 5c), the peak at ∼530 eV is due to oxides, and the peak at ∼532 eV is ascribed either to oxygen species dissolved in the metal or to adsorbed O²⁻ or OH⁻ groups.⁴⁸ In short, we can say that the surface may have an anionic nature and is mainly composed of lanthanum oxides/hydroxides and molybdenum oxides, which could be relevant to possible catalytic and electrochemical properties of interest. After the interaction of La₂Mo₂O₉ (LMO-in-S) and Li₂S₄, there were two multiple S 2p peaks, which were separated into four different sulfur environments. In the S 2p core level of Li₂S₄ (Fig. 5d), the peaks at 163.3 eV and 161.3 eV are assigned to bridging sulfur (S⁰_B) and terminal sulfur (S_T⁻¹), respectively.⁴⁹ Also, the formation of thiosulfate and polythionate complexes at binding energies of 166.1 eV and 168.33 eV can be identified, from which it is found that polythionate species can be reversibly formed via the oxidation of La₂Mo₂O₉; a similar catalytic mechanism was also confirmed previously.^50–52 The reversibly formed polythionate species act as a mediator to alleviate the shuttle effect. However, an additional peak at 162.2 eV is also observed, which corresponds to the Mo–S interaction that stems from the strong affinity between the La₂Mo₂O₉ (LMO-in-S) and polysulfides.⁵³ Generally, during the discharge–charge processes, the strong chemical bonding between La₂Mo₂O₉ and sulfur in LMO-in-S can facilitate the trapping and confinement of the polysulfides. In addition, there is a much stronger chemical interaction between Mo and S (Mo–S) in the LMO-in-S composite due to the high dispersion of La₂Mo₂O₉ and sulfur as compared with the conventional S@LMO composite (Fig. 5d and Fig. S15). In addition, the morphology of LMO-in-S can be partially retained after 100 cycles at 0.1C rate (Fig. S16). This collectively contributes to the enhanced cyclic stability of LMO-in-S.

Fig. 5 XPS spectra of LMO-in-S. (a) La 3d(3/2, 5/2). (b) Mo 3d(3/2, 5/2). (c) O 1s. (d) S 2p.

Experimental section

Preparation of La₂Mo₂O₉ nanofibers

The La₂Mo₂O₉ nanotubes were prepared using an electrospinning technique. First, the precursor solution was prepared by dissolving La(NO₃)₃·6H₂O (1.039 g), (NH₄)₆Mo₇O₂₄·4H₂O (0.423 g) and PVP (5.0 g) in a mixture of deionized water and ethanol (5.0 mL). After rigorous stirring of the solution overnight, the pristine nanofibers were attained via electrospinning the solution at a flow rate of 2 mL h⁻¹ with an applied voltage of 20 kV. The distance between the syringe nozzle and the collector plate was fixed at 15 cm. Finally, to attain hollow La₂Mo₂O₉ nanofibers with the PVP template completely removed, the collected pristine nanofibers were calcined at 600 °C for 3 h at a rate of 2 °C min⁻¹.

Preparation of LMO-in-S

First, the sulfur (0.4 g) was dissolved in anhydrous ethanediamine (16 mL) under continuous stirring for 30 minutes. Second, La₂Mo₂O₉ nanofibers (0.12 g) were dispersed in deionized water (200 mL), and the as-prepared S-EDA solution was added under stirring. The diluted HNO₃ was then added dropwise into the above suspension until the pH of the solution reached ∼7. Finally, the mixture solution was centrifuged, rinsed, and dried under vacuum, affording the flower-like S/La₂Mo₂O₉ molecular compound.

Preparation of the CNFs

The precursor solution for the CNF nanofibers was prepared by dissolving polystyrene (0.2 g) and polyacrylonitrile (1.0 g) in N,N-dimethylformamide (10 mL) with rigorous stirring overnight. Then, the pristine nanofibers were obtained via electrospinning the solution at a flow rate of 2 mL h⁻¹. The distance between the syringe nozzle and the collector plate was fixed at 15 cm, and the applied voltage was 15 kV. Finally, the collected pristine CNF nanofibers were calcined at 800 °C under an argon atmosphere for 3 h at a rate of 3 °C min⁻¹.

Materials characterization

The purity and structure of the as-prepared samples were examined via X-ray diffraction (XRD, Rigaku mini Flex II). To investigate and characterize the microstructure and morphology of the as-prepared samples, scanning electron microscopy (JEOL, JSM-7800F) and transmission electron microscopy (JEOL, JEM-2800) were employed. The sulfur content was measured via the TG curve (METTLER TOLEDO, TG/DSC1) under an Ar atmosphere. To calculate the pore size and specific surface area of the samples, the Brunauer–Emmett–Teller (BET) method was employed using a JW-BK112 system. The surface chemical composition was identified via X-ray photoelectron spectroscopy (XPS) using a Thermo Scientific ESCALAB 250Xi with an Al Kα source. The adsorption test was conducted via UV-vis absorption spectrophotometry (Varian Cary 100 Conc).

Electrochemical measurements

The cathodes were prepared by coating an N-methyl-2-pyrrolidone (NMP) slurry consisting of LMO-in-S, S@LMO or S/CNFs (70 wt%), super P (20 wt%) and polyvinylidene fluoride (PVDF, 10 wt%) on a carbon-coated aluminum foil and drying at 60 °C overnight. The coated foil was then punched into disks (diameter, 10 mm). The prepared electrolyte consisted of 1 M lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) and 0.2 M LiNO₃ in 1,3-dioxolane (DOL) and 1,2-dimethoxyethane (DME) (1 : 1, v/v). Coin cells (CR2032) were assembled in a glovebox using the coated cathode, a lithium anode and Celgard 2300 as the separator. The galvanostatic charge/discharge tests were carried out with a potential window of 1.7–2.8 V (vs. Li/Li⁺) using LAND-CT2001A instruments. Cyclic voltammogram (CV) tests were performed on an electrochemical workstation (CHI 600e) in the potential range of 1.7–2.8 V. Electrochemical impedance spectra (EIS) were obtained using an electrochemical workstation (Zahner IM6ex) in the frequency range of 10 MHz to 100 kHz.

Conclusions

In summary, the prepared LMO-in-S solid solution with a non-shaped matrix exhibits an exceptional anchoring performance towards soluble polysulfides and shuttling effect, leading to enhanced cycle stability and contributing to a high volumetric capacity. LMO-in-S with a sulfur content of 77.1 wt% delivers a high gravimetric capacity of 1374.1 mAh g⁻¹ corresponding to a volumetric capacity of 2294.8 mAh cm⁻³ at 0.1C rate, much higher than those obtained using the conventionally structured lanthanum-molybdate-nanofiber-supported sulfur (S@LMO). In particular, LMO-in-S exhibits satisfactory cycle stability with a low fading rate of 0.07% per cycle over 400 cycles at 1C rate. Furthermore, the non-shaped matrix allows an ultra-high sulfur content of 92.6 wt%, delivering a high gravimetric capacity of 1076.5 mAh g⁻¹ at 0.1C rate, which corresponds to a volumetric capacity of 1797.8 mAh cm⁻³. This is a new concept in the area of sulfur cathodes for rechargeable lithium–sulfur batteries. The La₂Mo₂O₉ in LMO-in-S is not a prefabricated matrix for loading sulfur, but instead a companion mixed with sulfur on the molecular level, like a solid solution. This enables the La₂Mo₂O₉ in LMO-in-S to provide a very high surface area for contact with sulfur and polysulfides. This work breaks through the traditional concepts of sulfur cathode structures and provides new possibilities for developing high-performance lithium–sulfur batteries.

Author contributions

HMUA: investigation, data curation, formal analysis, writing – original draft; JS: formal analysis, visualization, writing – review &editing; QZ: visualization; XG: formal analysis, resources; GL: conceptualization, funding acquisition, project administration, supervision, writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

Data are available within the article and its SI. Supplementary information: Some results of XRD, TG, SEM, N₂ adsorption, DFT, electrochemical performance of pouch cells, etc. See DOI: https://doi.org/10.1039/d5mh01572c.

Acknowledgements

Financial supports from the National Natural Science Foundation of China (22279066) are gratefully acknowledged.

References

1. X. Q. Zeng, M. Li, D. A. El-Hady, W. Alshitari, A. S. Al-Bogami, J. Lu and K. Amine, Adv. Energy Mater., 2019, 9, 1900161.
2. H. J. Li, K. Xi, W. Wang, S. Liu, G. R. Li and X. P. Gao, Energy Storage Mater., 2022, 45, 1229–1237.
3. W. Ahn, K. B. Kim, K. N. Jung, K. H. Shin and C. S. Jin, J. Power Sources, 2012, 202, 394–399.
4. B. Liu, R. Y. Fang, D. Xie, W. K. Zhang, H. Huang, Y. Xia, X. L. Wang, X. H. Xia and J. P. Tu, Energy Environ. Mater., 2018, 1, 196–208.
5. Y. Son, J. S. Lee, Y. Son, J. H. Jang and J. Cho, Adv. Energy Mater., 2015, 5, 1500110.
6. R. Fang, K. Chen, L. Yin, Z. Sun, F. Li and H. M. Cheng, Adv. Mater., 2018, 31, 1800863.
7. W. Wang, K. Xi, B. W. Li, H. J. Li, S. Liu, J. A. Wang, H. Y. Zhao, H. L. Li, A. M. Abdelkader, X. P. Gao and G. R. Li, Adv. Energy Mater., 2022, 12, 2200160.
8. Y. C. Jiang, H. M. U. Arshad, H. J. Li, S. Liu, G. R. Li and X. P. Gao, Small, 2021, 17, 2005332.
9. Y. T. Liu, S. Liu, G. R. Li and X. P. Gao, Adv. Mater., 2021, 33, 2003955.
10. Z. Li, H. B. Wu and X. W. Lou, Energy Environ. Sci., 2016, 9, 3061–3070.
11. B. Zhang, X. Qin, G. R. Li and X. P. Gao, Energy Environ. Sci., 2010, 3, 1531–1537.
12. M. H. Sun, X. Z. Wang, Y. Li, Z. B. Zhao and J. S. Qiu, Energy Environ. Mater., 2023, 6, e12349.
13. L. Wang, G. R. Li, S. Liu and X. P. Gao, Adv. Funct. Mater., 2021, 31, 2010693.
14. R. Y. Deng, M. Wang, H. Y. Yu, S. R. Luo, J. H. Li, F. L. Chu, B. Liu and F. X. Wu, Energy Environ. Mater., 2022, 5, 777–799.
15. L. Wang, Z. Y. Wang, G. R. Li, S. Liu and X. P. Gao, Nano Energy, 2020, 77, 105173.
16. M. Hagen, D. Hanselmann, K. Ahlbrecht, R. Maça, D. Gerber and J. Tubke, Adv. Energy Mater., 2015, 5, 1401986.
17. S. H. Chung, C. H. Chang and A. Manthiram, ACS Nano, 2016, 10, 10462.
18. Z. Sun, J. Zhang, L. Yin, G. Hul, R. Fang, H. M. Cheng and F. Li, Nat. Commun., 2017, 8, 14627.
19. J. Choi, T. G. Jeong, B. W. Cho, Y. Jung, S. H. Oh and Y. T. Kim, J. Phys. Chem. C, 2018, 122, 7664–7669.
20. Z. Lin, Z. Liu, N. J. Dudney and C. Liang, ACS Nano, 2013, 7, 2829–2833.
21. J. Shen, X. Xu, J. Liu, Z. Liu, F. Li, R. Hu, J. Liu, X. Hou, Y. Feng, Y. Yu and M. Zhu, ACS Nano, 2019, 13, 8986–8996.
22. C. Zhang, L. Cui, S. Abdolhosseinzadeh and J. Heier, InfoMat, 2020, 2, 613–638.
23. J. Xu, W. Zhang, H. Fan, F. Cheng, D. Su and G. Wang, Nano Energy, 2018, 51, 73–82.
24. L. Y. Tian, Z. Zhang, S. Liu, G. R. Li and X. P. Gao, Energy Environ. Mater., 2021, 5, 645–654.
25. H. J. Li, Y. H. Song, K. Xi, W. Wang, S. Liu, G. R. Li and X. P. Gao, J. Mater. Chem. A, 2021, 9, 10704–10713.
26. H. J. Li, K. Xi, W. Wang, S. Liu, G. R. Li and X. P. Gao, Energy Storage Mater., 2022, 45, 1229–1237.
27. Y. C. Jiang, S. Liu, X. P. Gao and G. R. Li, Adv. Funct. Mater., 2023, 33, 2304965.
28. F. Li, M. R. Kaiser, J. Ma, Y. Hou, T. Zhou, Z. Han, W. Lai, J. Chen, Z. Guo and H. Liu, J. Wang, ACS Appl. Mater. Interfaces, 2020, 12, 5958–5967.
29. H. W. Ma, Z. S. Yu, H. C. Li, D. Y. Guo, Z. Y. Zhou, H. L. Jin, L. H. Wu, X. A. Chen and S. Wang, Adv. Funct. Mater., 2024, 34, 2310301.
30. Y. Huang, L. Lin, Y. Zhang, L. Liu, B. Sa, J. Lin, L. Wang, D. L. Peng and Q. Xie, Nano-Micro Lett., 2023, 15, 67.
31. P. Lacorre, F. Goutenoire, O. Bohnke, R. Retoux and Y. Laligant, Nature, 2000, 404, 856–858.
32. F. Qianfeng, W. Xianping, C. Zhijun and Z. Guoguang, Front. Mater. Sci. China, 2007, 1, 7–15.
33. Q. Li, T. Lu, J. Schiemer, N. Laanait, N. Balke, Z. Zhang, Y. Ren, M. A. Carpenter, H. Wen, J. Li, S. V. Kalinin and Y. Liu, Phys. Rev. Mater., 2018, 2, 041403.
34. H. Siddiqui, M. S. Qureshi and F. Z. Haque, Nano-Micro Lett., 2020, 12, 29.
35. Q. Zhang, K. Zhang, D. Xu, G. Yang, H. Huang, F. Nie, C. Liu and S. Yang, Prog. Mater. Sci., 2014, 60, 208–337.
36. B. He, W. C. Li, Y. Zhang, X. F. Yu, B. S. Zhang, F. Li and A. H. Li, J. Mater. Chem. A, 2018, 6, 24194.
37. J. H. Kim, H. J. Eun, J. Juan, S. Eom, J. H. Ahn, M. Wu, J. Suk and S. Moon, ChemElectroChem, 2023, 10, e202300019.
38. H. M. U. Arshad, S. Liu, G. R. Li and X. P. Gao, ACS Appl. Mater. Interfaces, 2022, 14, 5247–5256.
39. A. Rosenman, E. Markevich, G. Salitra, D. Aurbach, A. Garsuch and F. F. Chesneau, Adv. Energy Mater., 2015, 5, 1500212.
40. C. Shen, J. Xie, M. Zhang, P. Andrei, M. Hendrickson, E. J. Plichta and J. P. Zhang, J. Electrochem. Soc., 2019, 166, A5287.
41. Q. Q. Qiao, H. Z. Zhang, G. R. Li, S. H. Ye, C. W. Wang and X. P. Gao, J. Mater. Chem. A, 2013, 1, 5262–5268.
42. S. Wang, S. Feng, J. Liang, Q. Su, F. Zhao, H. Song, M. Zheng, Q. Sun, Z. Song, X. Jia, J. Yang, Y. Li, J. Liao, R. Li and X. Sun, Adv. Energy Mater., 2021, 11, 2003314.
43. W. Jinshu, Z. Meiling, Z. Tieyong, Z. Jiuxing and N. Zuoren, J. Alloys Comp., 2002, 339, 293–298.
44. A. Khaled, J. J. Pireaux and S. Khelili, Acta Chim. Slov., 2012, 59, 766–778.
45. J. G. Choi and L. T. Thompson, Appl. Surf. Sci., 1996, 93, 143–149.
46. K. A. Gesheva, T. Ivanova, B. Marsen, B. Cole, E. L. Miller and F. Hamelmann, Surf. Coat. Technol., 2007, 201, 9378–9384.
47. D. A. Pawlak, M. Ito, M. Oku, K. Shimamura and T. Fukuda, J. Phys. Chem. B, 2002, 106, 504–507.
48. H. L. Wu, J. H. Dong, Y. G. Zhang, L. Lin, G. Y. Gao, T. Y. Li, X. L. Yi, B. Sa, J. X. Wang, L. S. Wang, J. T. Li, K. Amine, D. L. Peng and Q. S. Xie, Adv. Funct. Mater., 2023, 33, 2303707.
49. X. Liang, C. Hart, Q. Pang, A. Garsuch, T. Weiss and L. F. Nazar, Nat. Commun., 2015, 6, 5682.
50. K. M. Abraham and S. M. Chaudhri, J. Electrochem. Soc., 1986, 133, 1307.
51. X. Liang, Y. Rangom, C. Y. Kwok, Q. Pang and L. F. Nazar, Adv. Mater., 2017, 29, 1603040.
52. X. H. Zhao, X. Y. Wu, Q. Q. Hao, Y. S. Liu, K. X. Wang and J. S. Chen, ACS Appl. Mater. Interfaces, 2024, 16, 63647–63654.
53. T. Kabe, W. Qian, K. Tanihata, A. Ishihara and M. Godo, J. Chem. Soc., Faraday Trans., 1997, 93, 3709–3713.

---