Boosting the sodium storage of the 1T/2H MoS₂@SnO₂ heterostructure via a fast surface redox reaction

Abstract

The sluggish kinetics and large volume expansion arising from the large ionic radius of Na⁺ remain elusive weaknesses of sodium ion batteries (SIBs). Here, we report a transition from bulk diffusion to surface-dominant pseudocapacitive charge storage by nanoscaling and heterostructuring, which enables fast and stable charge storage kinetics for SIBs. An electronic attraction induced self-assembly strategy was developed for the synthesis of the 1T/2H MoS₂@SnO₂ heterostructure. Ultrasmall SnO₂ nanoparticles with a low crystallinity were uniformly distributed on the basal plane of MoS₂. The intercalated SnO₂ serves as an interfacial pillar to restrict the restacking of MoS₂ nanosheets, whereas dual-phase 1T/2H MoS₂ provides a continuous network for efficient charge transfer and restrains the aggregation of Na_xSn. As a result, the 1T/2H MoS₂@SnO₂ heterostructure exhibits a higher specific capacity (626 mA h g⁻¹ at 0.1 A g⁻¹), and superior cycling and rate capabilities (262 mA h g⁻¹ at 2 A g⁻¹ for 500 cycles) compared to the raw MoS₂ and 2H MoS₂@SnO₂ counterparts. Electrochemical kinetics analyses reveal that the charge transfer kinetics are boosted by the synergistic effect between the 1T/2H MoS₂ and SnO₂ nanoparticles. Quantitative examination into the origin demonstrated that the Na⁺ storage is dominated by the fast surface redox reaction, which endows the heterostructure with a durable high rate capability.

---

Introduction

Currently, lithium-ion batteries (LIBs) dominate in the field of consumer electronics owing to their inherent superiorities such as high energy density and long cycle life.^1–4 The limited reserves of lithium, however, cannot meet the increasing global demand for portable energy storage. As a homologous element to lithium, sodium has similar physical and chemical properties to those of lithium and is abundant on the earth. Therefore, sodium-ion batteries (SIBs) are considered to be the most suitable alternatives to LIBs, especially for large-scale energy storage.^5,6 Compared to Li⁺, however, Na⁺ possesses a larger ionic radius (1.02 Å for Na⁺vs. 0.76 Å for Li⁺) and higher redox potentials (0.33 V vs. Li⁺/Li), which result in a high energy barrier and sluggish reaction kinetics for Na⁺ insertion and diffusion.^7,8 For instance, the commercial LIB anode, graphite, is almost inactive for sodium storage. So far, the development of SIBs lags far behind that of the LIBs. The key issue is the lack of high-performance anode materials with a highly reversable capacity and long cycle life.

Alloying-type materials are suggested as promising anodes for SIBs because of their considerably higher theoretical capacities and low operating potentials. Among them, tin oxide (SnO₂) is environmentally friendly and rich in resources, and thus has attracted significant attention. SnO₂ can uptake 7.75 Na⁺ per formula in the full sodiation state (Na₁₅Sn₄), which gives an initial specific capacity as high as 782 mA h g⁻¹.^9–11

SnO₂ + 4Na⁺ + 4e⁻ → Sn + 2Na₂O (1)

4Sn + 15Na⁺ + 15e⁻ ↔ Na₁₅Sn₄ (2)

However, SnO₂ suffers from a large volume expansion (∼400%) during the alloying–dealloying process. This continuous volumetric change induces mechanical stress inside the SnO₂ particles, and ultimately leads to structural failure and pulverization.¹² Additionally, owing to the large ionic radius of Na⁺, the diffusion of Na⁺ within the SnO₂ lattice is restricted, resulting in a significant loss of capacity at high current densities. Downsizing the SnO₂ particle size to the nanoscale (less than 10 nm) is an effective strategy to accommodate the volume expansion and mitigate mechanical stress. Furthermore, the ultrasmall size of the nanoparticle reduces the diffusion length of Na⁺ and improves the rate capability.^13,14 So far, carbon materials have been extensively studied as a crucial component to confine the particle size of SnO₂ and accommodate the volumetric change.^15,16 However, the addition of carbon reduces the specific capacity of the anode and the particle size of SnO₂ can only be reduced to 10 nm because of their poor compatibility.

Two-dimensional (2D) nanomaterials have been reported to exhibit a surface-dominant phenomena owing to their unique structural features.^17–20 Molybdenum disulfide (MoS₂), a transition metal dichalcogenide, possesses a 2D layered structure, similar to that of graphite. The S–Mo–S trilayers are held together through weak van der Waals forces with an interlayer distance of 0.62 nm. These structural features allow storage of multiple charges through the intercalation and/or conversion reactions.^21,22 Moreover, the MoS₂ monolayer exhibits a high Young's modulus of 230 GPa.²³ This inherent strong mechanical strength makes MoS₂ nanosheets flexible and they can accommodate the large volumetric change of metal oxides such as SnO₂.²⁴ However, the basal plane of MoS₂ has a lack of active sites for the decoration of SnO₂. Furthermore, the bare MoS₂ nanosheets tend to restack during the charge/discharge process, leading to a decay in the electrochemical activity.

With the aim of overcoming the weaknesses of any single component, in this work, we report a charge-induced self-assembly strategy to fabricate a 1T/2H MoS₂@SnO₂ heterostructure to synergistically enhance the charge transfer kinetics for sodium storage. In order to effectively restrict the particle size of SnO₂, the MoS₂ nanosheets were first activated by chemical exfoliation. The Sn²⁺ was then absorbed on the negatively charged MoS₂ nanosheets and converted in situ to SnO₂. Owing to the abundant absorption sites, SnO₂ particles with a low crystallinity and ultrasmall particle size were obtained, which served as interfacial linkers to connect MoS₂ nanosheets and prevent their restacking. The dual-phase 1T/2H MoS₂ nanosheets, on the other hand, served as the substrate and buffered the volume change of SnO₂. The MoS₂ nanosheets combined with ultrasmall SnO₂ particles provide fast surface redox sites, which boosts the charge transfer kinetics and leads to the transition from bulk diffusion to pseudocapacitive charge storage, as revealed by quantitative analysis. As a result, the 1T/2H MoS₂@SnO₂ exhibits a high reversible capacity (626 mA h g⁻¹ at 0.1 A g⁻¹) and excellent high-rate durability (262.2 mA h g⁻¹ after 500 cycles at 2 A g⁻¹) when applied as the anode for SIBs.

Experimental section

Preparation of the exfoliated MoS₂

The MoS₂ was exfoliated through a lithium intercalation–exfoliation method that has been reported previously.²⁵ Typically, 0.5 g MoS₂ powder was immerged in 5 ml n-butyllithium solution (1.6 M in n-hexane) and stirred for 3 d in a glovebox. The Li-intercalated MoS₂ (Li_xMoS₂) was then centrifuged and washed with hexane several times to remove the redundant butyllithium and organic residues. The obtained Li_xMoS₂ was exfoliated in 200 ml deionized water with the assistance of ultrasonication. The exfoliated MoS₂ nanosheets were isolated by centrifugation and re-dispersed in 100 ml ethyl alcohol and kept in reserve.

Preparation of 1T/2H MoS₂@SnO₂ heterostructure

Typically, 0.5 g SnCl₂·2H₂O was dissolved in 20 ml ethyl alcohol. The solution was added dropwise into the MoS₂ suspension with gentle stirring overnight. The precipitate (Sn²⁺ intercalated MoS₂) was centrifuged and washed with ethyl alcohol several times to remove the residual Sn²⁺. After drying at 100 °C in an oven, the product obtained was mixed with graphene with a mass ratio of 10 : 1, followed by grinding in an agate mortar for 20 min. The final product was denoted as 1T/2H MoS₂@SnO₂. For comparison, the 2H MoS₂@SnO₂ was also prepared using the same procedure except that the precipitate obtained was further annealed at 300 °C for 3 h under an argon atmosphere.

Material characterization

The crystal structures of the samples were investigated by X-ray diffraction (XRD, PANalytical, Empyrean) with Cu-Kα radiation at a scanning rate of 0.02° s⁻¹. Raman spectra were obtained on a Raman spectrometer (Renishaw, inVia) with a 532 nm diode-pumped solid-state laser. The morphology of the samples was investigated using field-emission scanning electron microscopy (FE-SEM, JSM-7800F) and high resolution transmission electron microscopy (HRTEM, FEI Tecnai G2 F30). The chemical compositions and surface element states were studied using X-ray photoelectron spectroscopy (XPS, Thermo Fisher Scientific) with an Al Kα X-ray source (1486.6 eV). Thermogravimetric analysis (TGA) was conducted on a thermogravimetric analyser (NETZSCH, STA409PC) between room temperature and 1000 °C in air with a heating rate of 10 °C min⁻¹.

Electrochemical measurements

First, 1T/2H MoS₂@SnO₂, acetylene black and polyvinylidene fluoride (PVDF) were stirred to give a slurry with a mass ratio of 8 : 1 : 1 in N-methyl-2-pyrrolidone (NMP). The resultant slurry was cast onto copper foil, and dried in a vacuum oven at 100 °C overnight. Subsequently, the electrodes were punched into 12 mm diameter disks, and the mass loading of the active material was about 1.2 mg. The CR2032 coin-type cells were assembled in an argon-filled glovebox (MBraun, Inc.) with a H₂O and O₂ content of less than 0.1 ppm. 1 M NaPF₆ in ethylene carbonate/dimethyl carbonate (EC/DEC = 1 : 1 vol%) with 5 vol% fluoroethylene carbonate (FEC) as an additive was used as the electrolyte. Home-made Na foils were used as the counter electrode and the commercial Celgard 2500 film was employed as the separator. The fresh cells were tested between 0.01 and 3.0 V (vs. Na⁺/Na) via a multichannel Land battery testing system. Cyclic voltammetry (CV) tests were conducted on an IVIUMNSTAT electrochemical workstation between 0.01 and 3.0 V versus Na⁺/Na with the stated scan rates. The electrochemical impedance spectrum (EIS) was measured at the open circuit potential with an amplitude of 10 mV over the frequency range of 100 kHz to 10 mHz. For the full-cell test, commercial Na₃V₂(PO₄)₃ (NVP) was used as the cathode. The cell was cycled between 0.5 and 3.5 V at 0.5 A g⁻¹. The capacity was calculated based on the weight of the anode.

Results and discussion

The 1T/2H MoS₂@SnO₂ was prepared through a charge-induced self-assembly strategy. The synthesis procedure is illustrated in Fig. 1a. First, bulk MoS₂ was chemically exfoliated to activate the surface sites on the basal plane. The Li ions were intercalated into the van der Waals gap of MoS₂ owing to its weak interlayer interaction, causing an expansion of the interlayer spacing. Meanwhile, the thermodynamically stable 2H-MoS₂ underwent a first-stage phase transition to the metastable 1T phase accompanied by the sliding and ordering of the S–Mo–S layers (Fig. 1b).²⁵ Once in contact with water, the intercalated Li was oxidized and the H₂ produced pushed the MoS₂ layers apart, forming a tawny suspension of the MoS₂ monolayers (Fig. 2a). The exfoliated MoS₂ nanosheets were negatively charged owing to the electron transfer between butyllithium and MoS₂.^26,27 As a result, the suspension was stable over a week without restacking. The ethyl alcohol solution of SnCl₂ was then added dropwise. The positively charged Sn²⁺ was absorbed on the basal plane of MoS₂ owing to the electrostatic attraction. This neutralization of the surface charges caused a flocculation of the suspension (Fig. 2a). Post-treatment, the SnO₂ nanoparticles that were formed in situ were anchored on the surface of the MoS₂ nanosheets, forming the 1T/2H MoS₂@SnO₂ heterostructure.

Fig. 1 (a) Schematic illustration of the charge-induced self-assembly process for 1T/2H MoS₂@SnO₂. (b) Crystal structures and phase transition of 2H to 1T MoS₂ from the top and side views.

Fig. 2 (a) Optical photographs of the exfoliated MoS₂ suspension after storing for a week (left) and the precipitation obtained after the addition of SnCl₂ (right). (b) XRD patterns of 1T/2H MoS₂@SnO₂, exfoliated MoS₂ nanosheets and raw MoS₂. Note that the intensity of raw MoS₂ is greatly reduced to fit the figure frame. (c) Raman spectra of 1T/2H MoS₂@SnO₂ and raw MoS₂. SEM images of (d) raw MoS₂ and (e) 1T/2H MoS₂@SnO₂. (f) TEM and (g) HRTEM images of 1T/2H MoS₂@SnO₂. (h) and (i) STEM image of the 1T/2H MoS₂@SnO₂ and the corresponding element mappings.

The self-assembly process and formation of the 1T/2H MoS₂@SnO₂ heterostructure was monitored by XRD. As shown in Fig. 2b, the XRD pattern of raw MoS₂ exhibits distinct peaks at 14.4°, 32.7°, 39.6°, 49.8° and 58.4°, corresponding to the (002), (100), (103), (105) and (110) planes of hexagonal 2H MoS₂, respectively. All of the peaks disappeared after the Li⁺ intercalation and exfoliation, indicating the formation of single-layered MoS₂. The randomly arranged MoS₂ single-layers converted to an order pillared structure when the Sn²⁺ was added, as reflected by the two peaks that emerged at 6.5° and 12.5° which correspond to the (001) and (002) reflections of Sn_xMoS₂, respectively (Fig. S1†). The XRD pattern of Sn_xMoS₂ is similar to that of Li_xMoS₂, indicating the existence of a water bilayer.^27,28 During the drying process, Sn²⁺ was in situ hydrolysed and oxidized to SnO₂, as confirmed by the following TEM and XPS results. The MoS₂ monolayers were rearranged to nanosheets as a result of the removal of the absorbed Sn²⁺. The (002) peak shifts to a lower angle compared to that of raw 2H MoS₂, indicating an expansion of the interlayer distance. This expansion increases the surface defective sites and is beneficial for the insertion and diffusion of Na⁺.²⁵ Moreover, the peak intensities of MoS₂ are also sharply decreased compared with that of raw MoS₂. This could be ascribed to the decoration of SnO₂ on the surface of MoS₂, which hinders its restacking during the self-assembly process. The existence of SnO₂ is also confirmed by XRD. The diffraction peak at 26.6° corresponds to the (110) crystal planes of SnO₂ (PDF#41-1445), while the other characteristic peaks of SnO₂ are not well distinguished. The crystallinity and size of SnO₂ are extremely restricted as a result of the abundant active sites on MoS₂. It is reported that a decrease in the crystallinity of SnO₂ reduces the energy barrier for Na⁺ diffusion and is beneficial for sodium storage of SnO₂ owing to its intrinsically isotropic nature.²⁹

The Raman spectra of raw MoS₂ and 1T/2H MoS₂@SnO₂ are shown in Fig. 2c. Typical in-plane (E¹_2g) and out-of-plane (A_1g) vibration peaks of MoS₂ are observed for both samples. For the 1T/2H MoS₂@SnO₂, two additional peaks appear at 150 and 219 cm⁻¹, which correspond to the J₁ and J₂ mode of 1T MoS₂, respectively.^22,30 The peak position of E¹_2g shifts from 375 to 381 cm⁻¹, indicating the few-layered structure of the MoS₂ substrate.³¹ Therefore, the Raman results confirm the existence of the 1T/2H MoS₂ nanosheets. The J₁ and J₂ modes disappear after calcination, showing a phase transformation from 1T to 2H (Fig. S2†). The morphology of 1T/2H MoS₂@SnO₂ was studied using FESEM. As shown in Fig. 2d and S3,† the raw MoS₂ shows an inherent flat surface and orderly stacked layer structure, while the MoS₂ nanosheets become wrinkled and their size is significantly reduced after the exfoliation and self-assembly process. However, the typical layered-structure is still maintained as can be seen from the cross-section view (Fig. 2e). The microstructure of the 1T/2H MoS₂@SnO₂ was further examined using TEM. Fig. 2f shows a typical image of 1T/2H MoS₂@SnO₂ with approximately two layers. Owing to the abundant absorption sites on the exfoliated MoS₂, numerous SnO₂ nanoparticles (dark spots) are homogeneously dispersed on the MoS₂ nanosheets without agglomeration. The size of the SnO₂ was successfully confined to 5–10 nm. The ultrasmall size of SnO₂ ensures sufficient contact with the electrolyte and shortens the diffusion length of Na⁺, which increases the activity of the surface redox storage of Na⁺.¹² The HRTEM image (Fig. 2g) reveals the tight contact between SnO₂ nanoparticles and the MoS₂ substrate. The abundant hetero-interface created may act as active sites for fast Na⁺ storage.³² Two different atomic arrangements, the honeycomb and parallelogram, can be observed, corresponding to the 2H and 1T polymorphs of MoS₂, respectively. The lattice fringe with d = 0.26 nm corresponds to the (101) plane of rutile SnO₂. For comparison, the XRD pattern and structure of the 2H.

MoS₂@SnO₂ are shown in Fig. S4.† As indicated from the XRD pattern, the MoS₂ nanosheets restack together and the interlayer distance shrinks to 0.62 nm. Compared to 1T/2H MoS₂@SnO₂, SnO₂ with a high crystallinity is found on the surface of MoS₂. The SnO₂ nanoparticles move and cluster together to minimize the surface energy under heat, causing a phase separation from the MoS₂ substrate. The lattice fringes were measured and found to be 0.26 and 0.34 nm, corresponding to the (101) and (110) planes of the SnO₂, respectively. The energy dispersive X-ray (EDX) and elemental mapping analyses of the 1H/MoS₂@SnO₂ are shown in Fig. 2h and i and S5.† Elements of S, Mo, Sn and O were detected, which were uniformly distributed throughout the sample. The weight percentages of SnO₂ in 1T/2H MoS₂@SnO₂ were determined by TGA (Fig. S6†). The mass ratio of SnO₂, MoS₂ and graphene is around 6 : 4 : 1.

XPS analysis was then conducted to further investigate the compositions and binding states of each element. Fig. 3a shows the survey spectrum of the 1T/2H MoS₂@SnO₂. Five elements of C, O, Mo, S and Sn were detected without impurities, which is consistent with the EDS results. Fig. 3b–d shows the high resolution XPS spectra of Sn 3d, Mo 3d and S 2p, respectively. Two peaks corresponding to the 3d_5/2 and 3d_3/2 of Sn⁴⁺ were observed at around 487.1 and 495.2 eV, showing the Sn atoms are in the form of SnO₂.^12,29 In the Mo 3d region, both the 2H and 1T phases of MoS₂ can be identified. The binding energies of Mo 3d_5/2 and 3d_3/2 of the 1T phase are 231.1 and 228.1 eV respectively, which are approximately 1.2 eV lower than their 2H counterparts. The XPS S 2p spectrum of 1T/2H MoS₂@SnO₂ is shown in Fig. 3d. Similarly, the peaks located at 161.2 and 162.2 eV can be assigned to the S 2p_3/2 and S 2p_1/2 of 1T MoS₂ respectively, while the binding energies of S²⁻ in the 2H phase are 0.8 eV higher.^27,33,34 According to the XPS peak deconvolution, the relative fraction of 1T phase in 1T/2H MoS₂@SnO₂ is 53.4%. As the 1T phase MoS₂ is 10⁷ times more conductive than the 2H phase, the retained metallic 1T phase greatly enhances the conductivity of the composites.^17,30,35 In contrast, the XPS spectrum of 2H MoS₂@SnO₂ is shown in Fig. S7.† Only two peaks located at 229.3 and 232.3 eV are observed in the Mo 3d region, corresponding to the Mo 3d_5/2 and 3d_3/2 of the 2H phase, respectively.

Fig. 3 (a) XPS survey spectrum, high resolution XPS spectra of (b) Sn 3d, (c) Mo 3d, and (d) S 2p of 1T/2H MoS₂@SnO₂.

The electrochemical performances of the 1T/2H MoS₂@SnO₂ as an anode material for SIBs were then investigated. The CV test was first conducted to gain a deep insight into the sodiation/desodiation mechanism of the 1T/2H MoS₂@SnO₂ (Fig. 4a). In the cathodic process, the reduction peak at about 1.02 V is attributed to the reduction of SnO₂ to Sn and the formation of a solid electrolyte interphase (SEI).^36,37 The peak at 0.55 V corresponds to the conversion reaction of MoS₂ to Na₂S/Mo, while the peak below 0.5 V represents the multistep alloying of Sn with Na⁺ that forms Na_xSn. During the anodic scan, a broad oxidation peak ranging from 0.25 to 0.65 V can be assigned to the stepwise de-alloying reactions of Na_xSn,^36,38 while the peak that emerges at 1.6 V is attributed to the desodiation process of Na₂S. The oxidized peak at 1.6 V gradually shifts to a lower potential in the subsequent cycles, indicating the reduced polarization owing to the enhanced interfacial activity and electron transport kinetics. The CV curves are almost coincident after the first cycle, showing an excellent cycling stability. In contrast, two sharp cathodic peaks appear at around 0.79 and 0.59 V for 2H MoS₂@SnO₂ and bulk MoS₂, which correspond to the intercalation of Na⁺ into the band gap of MoS₂, accompanied by a phase change from 2H to 1T (Fig. 4b and S8†).^39,40 The absence of a peak at 0.79 V for 1T/2H MoS₂@SnO₂ indicates the reduced diffusion kinetics of Na⁺ owing to the expanded interlayer spacing and the existence of the 1T phase. In the anodic process, a higher oxidation peak of 1.76 V is observed for 2H MoS₂@SnO₂, indicating high polarization. The charge–discharge reactions of bulk MoS₂ are not stabilized, even after five cycles, further elucidating the synergistic effect between the MoS₂ nanosheets and SnO₂ nanoparticles on the electrochemical properties. Furthermore, compared with raw MoS₂ and 2H MoS₂@SnO₂, the CV curves of 1T/2H MoS₂@SnO₂ show an inconspicuous conversion and broadening of the peaks. This can be attributed to its surface dominant nature that results from the small particle size and tight contact in the hetero-interface.²⁹ The charge/discharge profiles of 1T/2H MoS₂@SnO₂ for the first four cycles are shown in Fig. 4c. No obvious potential plateau can be observed, which is consistent with the CV results.

Fig. 4 The electrochemical performances of 1T/2H MoS₂@SnO₂. CV curves of (a) 1T/2H MoS₂@SnO₂ and (b) 2H MoS₂@SnO₂. (c) Discharge/charge profiles for the first four cycles. (d) Cycling performance of the 1T/2H MoS₂@SnO₂, 2H MoS₂@SnO₂ and raw MoS₂. (e) Rate capability and high rate cycling performance at (f) 0.5 A g⁻¹ and (g) 2 A g⁻¹. Note that the electrodes were first activated at 0.05 A g⁻¹ for a few cycles.

The electrochemical performances of the 1T/2H MoS₂@SnO₂ are shown in Fig. 4d–g. The specific capacities of 1T/2H MoS₂@SnO₂ in the first discharge/charge are 1130/516 mA h g⁻¹, respectively. The large capacity loss mainly comes from the irreversible reaction between SnO₂ to Sn, the formation of the SEI layer and the conversion between MoS₂ and Mo/Na₂S.^12,41 In subsequent cycles, the coulombic efficiency (CE) quickly stabilizes to approximately 100%. A maximum reversible capacity of 626 mA h g⁻¹ is reached at the 20th cycle, which could be attributed to the activation of the heterostructure during cycling. The reversible capacity is maintained above 537 mA h g⁻¹ at the 100th cycle, which is superior to that of raw MoS₂, 2H MoS₂@SnO₂ and exfoliated MoS₂ (Fig. 4d and S9†), demonstrating the synergistic effect between the ultrasmall SnO₂ nanoparticles and dual-phased MoS₂ nanosheets. For the raw MoS₂, a very low initial reversible capacity of 231 mA h g⁻¹ was obtained with a CE of 35.9% due to the low electrical conductivity of 2H MoS₂, while the 2H MoS₂@SnO₂ delivers an initial reversible capacity of 335 mA h g⁻¹ with a CE of 45.4%. However, the capacity of 2H MoS₂@SnO₂ quickly decays to 138 mA h g⁻¹ after 100 cycles, which could be ascribed to the aggregation of SnO₂, the phase transition of the 1T to 2H phase and the restacking of the MoS₂ layers. The relative concentrated SnO₂ causes a drastic local volume expansion and a higher barrier at the SnO₂ and 2H MoS₂ interface, causing sluggish Na⁺ transport. In addition, the 1T/2H MoS₂@SnO₂ exhibits an excellent rate capability (Fig. 4e). The reversible capacities at 0.1, 0.5, 1, 2, and 5 A g⁻¹ are 525, 444, 287, 195 and 114 mA h g⁻¹, respectively. When the current density rebounds to 0.1 A g⁻¹, a highly reversible capacity of 586 mA h g⁻¹ is obtained, indicating a robust structure for sodium storage. The high-rate cycling performance of 1T/2H MoS₂@SnO₂ is presented in Fig. 4f and g. In the initial ten cycles, a low current density of 0.05 A g⁻¹ is applied to generate a dense stable SEI film. The current density is then switched to 0.5 A g⁻¹ or 2 A g⁻¹. The high reversible capacity of 262 mA h g⁻¹ is retained after 500 cycles at 2 A g⁻¹ with a capacity retention of 81.8%, showing an excellent high-rate stability. To further verify the durability and high rate behaviour, the structural evolution of 1T/2H MoS₂@SnO₂ during cycling was characterised using FE-SEM (Fig. S10†) and TEM (Fig. 5). Fig. 5a and b shows the morphology of 1T/2H MoS₂@SnO₂ at different resolutions after the first cycle. The crystal lattice becomes indistinct owing to the formation of the SEI layer and the nano-size effect. The Sn nanoparticles, as indicated by the yellow cycles, are uniformly distributed on the substrate without agglomeration. Notably, an imperceptible lattice fringe of SnO₂ also appears, which may come from the oxidation of Sn during the repeated washing and drying processes. The elemental mapping shows good dispersion of Mo, S, Sn and O (Fig. 5c). After 10 cycles, the morphology is maintained and found to be almost the same (Fig. 5d–f), illustrating the excellent structural stability of the electrode. The electrochemical performance of T/2H MoS₂@SnO₂ was compared with those of previously reported state-of-the-art anode materials, such as dual phase MoS₂, 1T MoS₂ and some other hybrid composites.^{18,36,42–45} As shown in Table S1,† the superior performances of the 1T/2H MoS₂@SnO₂ demonstrate the importance of the synergistic effect between SnO₂ and MoS₂.

Fig. 5 TEM images and the corresponding elemental mappings of the anodes after (a)–(c) the first and (d)–(f) the 10th cycle.

To validate the advantages of the heterostructure, the reaction kinetics of 1T/2H MoS₂@SnO₂ were first investigated using EIS. As shown in Fig. 6, both 1T/2H MoS₂@SnO₂ and 2H MoS₂@SnO₂ exhibit similar impedance patterns. The compressed semicircle in the high and medium frequency region corresponds to the charge transfer impedance (R_ct) at the electrode–electrolyte interface, while the straight line in the low-frequency region is associated with Warburg behaviour, which originates from the diffusion of Na⁺ in the electrode. Obviously, the 1T/2H MoS₂@SnO₂ has a much lower R_ct (87 Ω) compared to 2H MoS₂@SnO₂ (593 Ω), showing the enhanced charge storage kinetics of the heterostructure. The apparent Na⁺ transfer coefficient was then calculated based on the following equations:^29,46

D_Na⁺ = R²T²/2n⁴F⁴σ_w²A²C² (3)

Z′ = R + σ_wω^−1/2 (4)

in which R, T, n, F, A, C, and σ_w are the gas constant (8.314 J mol⁻¹ K⁻¹), temperature, the number of electrons per molecule during oxidation, the Faraday's constant (96 485 C mol⁻¹), the surface area of the electrode (1.13 cm²), the Na⁺ concentration in the electrode material (1 M) and the Warburg coefficient, respectively. As the σ_w is the slope of the real part of the impedance (Z′) as a function of the inverse square root of the angular frequency (ω^−1/2), the apparent sodium diffusion coefficient of 1T/2H MoS₂@SnO₂ was then calculated and found to be 2.4 × 10⁻¹⁶ cm² s⁻¹, which is approximately 10⁴ times higher than that of 2H MoS₂@SnO₂. These enhanced charge transfer kinetics highlight the synergistic effects between the dual phased 1T/2H MoS₂ nanosheets and well-dispersed SnO₂ nanoparticles.

Fig. 6 The EIS spectra of (a) 1T/2H MoS₂@SnO₂ and (c) 2H MoS₂@SnO₂ (insets show the equivalent circuit models). (b) and (d) The real part of the impedance (Z′) as a function of the inverse square root of the angular frequency (ω^−1/2) in the Warburg region of (a) and (c), respectively.

The galvanostatic intermittent titration technique (GITT) test was then performed to evaluate the change in the Na⁺ diffusion coefficient during sodiation/desodiation to gain a thorough insight into the dynamic Na⁺ storage behaviour of 1T/2H MoS₂@SnO₂.^4,47 The test was carried out with a pulse current of 0.1 A g⁻¹ and a time interval of 30 min after 10 cycles. Typical GITT curves of the 1T/2H MoS₂@SnO₂ and the exfoliated MoS₂ are shown in Fig. 7a and b. The V and t^1/2 show a good linear relationship (Fig. 7c). As a result, the equation can be simplified by introducing Fick's law (t = L²/2D):

in which τ denotes the constant current pulse time, and L is the distance of the diffusion path of the Na⁺ ions (cm), which is simplified as the thickness of the electrode. ΔE_s represents the change in potential (V) caused by the current pulse and ΔE_τ represents the change in the potential (V) excluding the iR drop (Fig. S11†). The calculated D_Na⁺ values at different depths of charge/discharge are shown in Fig. 7d and e. During discharge, the 1T/2H MoS₂@SnO₂ exhibits a similar D_Na⁺ to that of exfoliated MoS₂, showing that combining it with SnO₂ does not deteriorate the large charge inplane carrier mobility of MoS₂.²⁷ A sharp increase of D_Na⁺ was observed at approximately 0.45 V. This is consistent with the voltage at which the conversion reaction of MoS₂ occurs. It is therefore believed that the Mo nanoparticles formed could further synergize with Na⁺ and promote the formation of Na_xSn. The dynamic change of D_Na⁺ in the charging process is shown in Fig. 7e. The 1T/2H MoS₂@SnO₂ has a higher D_Na⁺ than that of the exfoliated MoS₂, further revealing the origin of its superior rate and cycling capabilities. Similarly, the variation in D_Na⁺ is almost consistent with the CV changes, indicating that the phase transformation has a profound impact on the charge storage properties. The above described analyses shed light on the enhanced electrochemical performances of the 1T/2H MoS₂@SnO₂ heterostructure. The boosted kinetics that originate from the nanoscaling and heterostructuring may induce a transition from the diffusion to surface-dominated redox. To quantify the contribution of the extrinsically induced surface redox charge-storage behaviour, CV curves of 1T/2H MoS₂@SnO₂ at scan rates from 0.3 to 2 mV s⁻¹ were measured. As shown in Fig. 7f, the polarization of the CV curves increases gradually with the increase in the scan rate, while their shapes remain identical, showing a good high-rate capability. The dominant charge-storage mechanism can be distinguished from the slope of the logarithm of the peak current as a function of the scan rate. Specifically, a slope of 0.5 represents an ideally diffusion-controlled mechanism, whereas 1.0 implies a surface redox capacitive behaviour. As shown in Fig. 7g, good linear relationships are obtained from the transformation peaks in CV. The slopes range between 0.9 and 1.1, indicating the reaction is controlled by the surface-dominant capacitive effect. According to Dunn's method, the total charge storage can be kinetically decoupled into the diffusion and surface controlling parts:^11,20

i(v) = k₁v + k₂v^1/2 (6)

in which k₁v represents the pseudocapacitive current and k₂v^1/2 represents the diffusive current. By plotting i(v)/v^1/2versus v^1/2, the values of k₁ and k₂ can be obtained, which are a quantitative indication of the pseudocapacitive and diffusive contributions at a specific potential and given rate. For instance, the diffusive and surface induced current response at 2 mV S⁻¹ is shown in Fig. 7h, in which the red and green areas represent the total charge stored and the fast surface pseudocapacitive contribution, respectively. It is not surprising that owing to the confinement of the particle size (large surface area) and the creation of a highly active hetero-interface, a transition from battery-type diffusion to surface redox pseudocapacitive behaviour is observed even at a low scan rate (Fig. 7i). The percentage of the pseudocapacitive contribution is 79% at 0.3 mV S⁻¹, indicating that most of the stored charges originate from the surface redox reaction. The pseudocapacitive contribution increases from 79% to 91% if the scan rate increases from 0.3 to 2 mV S⁻¹, resulting in an excellent high-rate stability.

Fig. 7 (a) and (b) GITT curves of 1T/2H MoS₂@SnO₂ and the exfoliated MoS₂. (c) Plot of V versus t^1/2. (d) and (e) The calculated Na⁺ diffusion coefficients during charging and discharge. (f) CV curves at various scan rates. (g) Plots of the logarithm peak current versus the logarithm scan rate. (h) Capacitive contribution at 2 mV s⁻¹ for the 1T/2H MoS₂ @SnO₂. (i) Capacitive contribution at various scan rates.

Considering the impressive performance in half-cells, the practical application of the 1T/2H MoS₂@SnO₂ electrode in a Na⁺ full cell was further confirmed by pairing it with a commercial NVP cathode. The XRD pattern and SEM image of the NVP are shown in Fig. S12.† The anode was first pre-sodiated at 0.1 A g⁻¹ using Na foil as a counter electrode. The mass ratio of the cathode and anode was confined at 5 : 1. The full cell delivers a high reversible capacity of 313 mA h g⁻¹ in the first cycle. The reversible capacity of 248 mA h g⁻¹ can be maintained after 30 cycles at 0.5 A g⁻¹ (Fig. S13b†). The slight fading observed may be caused by the large internal resistance of the 2032-type cell. The charge–discharge curves of the 1st, 10th and 30th cycles are shown in Fig. S13c.† The 1T/2H MoS2@SnO₂//NVP Na⁺ full cell can light a small lamp, further demonstrating the practicability of the 1T/2H MoS₂@SnO₂ anode for sodium storage (Fig. S13d†).

Conclusions

In summary, a 1T/2H MoS₂@SnO₂ heterostructure was successfully prepared via a surface charge-induced self-assembly process and used as an anode in SIBs. By measuring the electrokinetic parameters, we observed a transition from sluggish diffusion to the fast surface redox storage of Na⁺, which originates from the synergistic effects between dual phased 1T/2H MoS₂ nanosheets and well-dispersed SnO₂ nanoparticles. As a result, the 1T/2H MoS₂@SnO₂ delivers an improved cycling stability and superior rate capability compared to that of raw MoS₂ and 2H MoS₂@SnO₂. The reversible capacity of 1T/2H MoS₂@SnO₂ remains at 537 mA h g⁻¹ after 100 cycles at 0.1 A g⁻¹ and 262 mA h g⁻¹ after 500 cycles at a high rate of 2 A g⁻¹. This surface-dominant phenomenon provides a promising strategy for optimizing the structure and properties of SIBs.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors are grateful for financial support from the National Natural Science Foundation of China (51904191), and the Natural Science Foundation of Guangdong province (2018A030310499).

Notes and references

1. M. Armand and J. M. Tarascon, Nature, 2008, 451, 652–657.
2. M. D. Slater, D. Kim, E. Lee and C. S. Johnson, Adv. Funct. Mater., 2013, 23, 947–958.
3. B. L. Ellis and L. F. Nazar, Curr. Opin. Colloid Interface Sci., 2012, 16, 168–177.
4. Y. Wang, Y. Zhang, J. Shi, A. Pan, F. Jiang, S. Liang and G. Cao, J. Mater. Chem. A, 2018, 6, 18286–18292.
5. J. Wang, C. Luo, T. Gao, A. Langrock, A. C. Mignerey and C. Wang, Small, 2015, 11, 473–481.
6. X. Xie, T. Makaryan, M. Zhao, K. L. Van Aken, Y. Gogotsi and G. Wang, Adv. Energy Mater., 2016, 6, 1502161.
7. C. Zhao, J. Kong, X. Yao, X. Tang, Y. Dong, S. L. Phua and X. Lu, ACS Appl. Mater. Interfaces, 2014, 6, 6392–6398.
8. Z. Hu, Q. Liu, S. L. Chou and S. X. Dou, Adv. Mater., 2017, 29, 1700606.
9. L. D. Ellis, T. D. Hatchard and M. N. Obrovac, J. Electrochem. Soc., 2012, 159, A1801–A1805.
10. G. Wang, J. Feng, L. Dong, X. Li and D. Li, Appl. Surf. Sci., 2017, 396, 269–277.
11. S. M. Jung, D. W. Kim and H. Y. Jung, J. Mater. Chem. A, 2020, 8, 8244–8254.
12. H. Wang, Q. Wu, Y. Wang, X. Wang, L. Wu, S. Song and H. Zhang, Adv. Energy Mater., 2019, 9, 1802993.
13. Y. Liu, N. Zhang, L. Jiao and J. Chen, Adv. Mater., 2015, 27, 6702–6707.
14. Y. N. Sun, M. Goktas, L. Zhao, P. Adelhelm and B. H. Han, J. Colloid Interface Sci., 2020, 572, 122–132.
15. M. Wang, X. Wang, Z. Yao, W. Tang, X. Xia, C. Gu and J. Tu, ACS Appl. Mater. Interfaces, 2019, 11, 24198–24204.
16. W. Chen, K. Song, L. Mi, X. Feng, J. Zhang, S. Cui and C. Liu, J. Mater. Chem. A, 2017, 5, 10027–10038.
17. X. Geng, Y. Jiao, Y. Han, A. Mukhopadhyay, L. Yang and H. Zhu, Adv. Funct. Mater., 2017, 27, 1702998.
18. D. Sun, D. Huang, H. Wang, G. Xu, X. Zhang, R. Zhang, Y. Tang, D. Abd Ei-Hady, W. Alshitari, A. Saad Al-Bogami, K. Amine and M. Shao, Nano Energy, 2019, 61, 361–369.
19. Y. Tang, Z. Zhao, Y. Wang, Y. Dong, Y. Liu, X. Wang and J. Qiu, ACS Appl. Mater. Interfaces, 2016, 8, 32324–32332.
20. Q. Su, X. Cao, T. Yu, X. Kong, Y. Wang, J. Chen, J. Lin, X. Xie, S. Liang and A. Pan, J. Mater. Chem. A, 2019, 7, 22871–22878.
21. X. Zhao, S. Ning, W. Fu, S. J. Pennycook and K. P. Loh, Adv. Mater., 2018, 30, 1802397.
22. W. Tang, X. Wang, D. Xie, X. Xia, C. Gu and J. Tu, J. Mater. Chem. A, 2018, 6, 18318–18324.
23. B. Simone, B. Jacopo and K. Andras, ACS Nano, 2011, 5, 9703–9709.
24. Y. Chen, J. Lu, S. Wen, L. Lu and J. Xue, J. Mater. Chem. A, 2014, 2, 17857–17866.
25. C. Zhao, J. M. Ang, Z. Liu and X. Lu, Chem. Eng. J., 2017, 330, 462–469.
26. Q. Li, Z. Yao, J. Wu, S. Mitra, S. Hao, T. S. Sahu, Y. Li, C. Wolverton and V. P. Dravid, Nano Energy, 2017, 38, 342–349.
27. G. Eda, H. Yamaguchi, D. Voiry, T. Fujita, M. Chen and M. Chhowalla, Nano Lett., 2011, 11, 5111–5116.
28. M. Kan, J. Y. Wang, X. W. Li, S. H. Zhang, Y. W. Li, Y. Kawazoe, Q. Sun and P. Jena, J. Phys. Chem. C, 2014, 118, 1515–1522.
29. L. Fan, X. Li, B. Yan, J. Feng, D. Xiong, D. Li, L. Gu, Y. Wen, S. Lawes and X. Sun, Adv. Energy Mater., 2016, 6, 1502057.
30. L. Jiang, S. Zhang, S. A. Kulinich, X. Song, J. Zhu, X. Wang and H. Zeng, Mater. Res. Lett., 2015, 3, 177–183.
31. C. Lee, H. Yan, L. E. Brus, T. F. Heinz, J. Hone and S. Ryu, ACS Nano, 2010, 4, 2695–2700.
32. M. Li, Q. Deng, J. Wang, K. Jiang, Z. Hu and J. Chu, Nanoscale, 2018, 10, 741–751.
33. S. Wang, D. Zhang, B. Li, C. Zhang, Z. Du, H. Yin, X. Bi and S. Yang, Adv. Energy Mater., 2018, 8, 1801345.
34. S. Shi, Z. Sun and Y. H. Hu, J. Mater. Chem. A, 2018, 6, 23932–23977.
35. M. Acerce, D. Voiry and M. Chhowalla, Nat. Nanotechnol., 2015, 10, 313–318.
36. Z. Chen, D. Yin and M. Zhang, Small, 2018, 14, 1703818.
37. H. Chen, J. He, G. Ke, L. Sun, J. Chen, Y. Li, X. Ren, L. Deng and P. Zhang, Nanoscale, 2019, 11, 16253–16261.
38. Y. Wang, Y. Lim, M. Park, S. Chou, J. H. Kim, H. Liu, S. Dou and Y. Kim, J. Mater. Chem. A, 2014, 2, 529–534.
39. Y. Li, Y. Liang, F. C. Robles Hernandez, H. Deog Yoo, Q. An and Y. Yao, Nano Energy, 2015, 15, 453–461.
40. J. Wu, J. Liu, J. Cui, S. Yao, M. Ihsan-Ul-Haq, N. Mubarak, E. Quattrocchi, F. Ciucci and J.-K. Kim, J. Mater. Chem. A, 2020, 8, 2114–2122.
41. Y. Huang, Z. Wang, M. Guan, F. Wu and R. Chen, Adv. Mater., 2020, 32, 2003534.
42. K. Yao, Z. Xu, Z. Li, X. Liu, X. Shen, L. Cao and J. Huang, ChemSusChem, 2018, 11, 2130–2137.
43. D. Sun, D. Ye, P. Liu, Y. Tang, J. Guo, L. Wang and H. Wang, Adv. Energy Mater., 2018, 8, 1702383.
44. G. Jia, D. Chao, N. H. Tiep, Z. Zhang and H. J. Fan, Energy Storage Mater., 2018, 14, 136–142.
45. Z. Xu, K. Yao, Z. Li, L. Fu, H. Fu, J. Li, L. Cao and J. Huang, J. Mater. Chem. A, 2018, 6, 10535–10542.
46. K. Yao, Z. Xu, J. Huang, M. Ma, L. Fu, X. Shen, J. Li and M. Fu, Small, 2019, 15, 1805405.
47. H. J. Kwon, J. Y. Hwang, H. J. Shin, M. G. Jeong, K. Y. Chung, Y. K. Sun and H. G. Jung, Nano Lett., 2020, 20, 625–635.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d0ta08711d

---