1 Introduction

Sodium ion batteries (SIBs) have aroused great interests as a promising substitute to conventional lithium ion batteries (LIBs) owing to the overwhelming superiority in low cost and abundant natural resources [1,2,3,4]. However, the sluggish reaction kinetics induced by the larger ionic radius of Na ions deteriorates their electrochemical properties seriously [5, 6]. The key point to achieve high-performance SIBs relies on the rational design of electrode materials with satisfactory Na ions insertion/extraction performance during repeated cycles, especially for the active anode host materials. Among the various anode materials, transition metal sulfides (MxSy) have drawn extensive attentions due to their high sodium storage capacity [7,8,9,10]. Moreover, the weaker M–S bond than M–O bond is more kinetically favorable for electrochemical conversion reaction, leading to a better redox kinetics and reversibility [11]. As a typical example, iron sulfide (Fe1−xS) has risen to prominence because of its high theoretical capacity (~ 610 mAh g−1), natural abundance and environmental benignity [12, 13]. Nonetheless, there are still obstacles to its commercial applications mainly due to the poor intrinsic conductivity and severe volume changes during the sodiation/desodiation processes, thus resulting in inferior cycling stability and rate capability [14]. Besides Fe1−xS, 2D layered molybdenum disulfide (MoS2) is another promising anode candidate for SIBs due to its analogous structures to graphite [15,16,17]. In particular, the large interlayer spacing (0.62 nm) and high mechanical strength of MoS2 are able to facilitate Na+ migration and alleviate structural deformation during the discharge/charge processes [18]. Considering the advantages of these two metal sulfides, a hybrid composite by combining Fe1−xS with MoS2 can be rationally designed and fabricated to expectedly achieve the desirable performance.

Constructing unique heterostructures to create diverse interface effects offers unprecedented opportunities in various fields, such as solar cells, photocatalysis and electrocatalysis [19,20,21,22,23,24,25,26,27]. The heterointerfaces deriving from coupling nanostructures with different properties can greatly accelerate charge transport and improve reaction kinetics [28]. Nishitani et al. [29] have studied the charge-transfer effects in CdO/SnTe heterointerfaces, which revealed a large fourfold enhancement of electron mobility. Yin et al. [30] have investigated the charge carrier transfer in the bulk heterostructures and indicated the balanced hole/electron mobilities are key factors to attribute high solar cell performance. Hints from the previous reports indicate that rational interface designing could be of great importance for the improvement of conversion reaction kinetics in rechargeable SIBs. Some emerging heterostructures such as Ni3S2/MoS2, Sb2S3/SnS2 and Sb2S3/MoS2 have demonstrated outstanding electrochemical performance, which is difficult to realize in a signal material system [31,32,33,34,35,36]. However, the detailed synergistic effects and heterointerface properties are rarely revealed.

Herein, we design a Fe1−xS/MoS2 heterostructure with abundant “ion reservoir” to provide fast Na+ diffusion channels and interpret specifically the correlation between heterointerface and sodium ion diffusion. The DFT calculations further confirm that the heterointerface significantly reduces the diffusion barrier and facilitates the charge-transfer kinetics, hence enabling excellent cycling stability and rate capability. Our findings not only provide in-depth understanding of the dynamic relationship between heterointerface and electrochemical performance, but also carve a new path for engineering rationally heterostructures toward high-performance energy storage devices.

2 Experimental Section

2.1 Material Preparation

Synthesis of the Prussian blue (PB) nanocubes. PB nanocubes were synthesized by a hydrothermal method reported previously with modifications [37]. In a typical process, 12 g polyvinylpyrrolidone (PVP, MW ≈ 40,000) and 0.6 g potassium hexacyanoferrate (II) (K4Fe(CN6)·3H2O) were added into 0.1 M hydrochloric acid (HCl, 37%) under magnetic stirring, to form a clear solution. Subsequently, the solution was transferred in a programmable oven and then heated to 80 °C for 24 h. Finally, PB nanocubes were obtained by centrifuging and washing with water and ethanol and then dried in a vacuum oven overnight.

Synthesis of nitrogen-doped porous nanocubic structure (FeCN). FeCN nanocubes were obtained by calcining the as-synthesized PB nanocubes at 500 °C for 4 h under argon flow with a heating rate of 1 °C min−1.

Synthesis of FeCN/MoS2 nanocomposite. Twenty-five milligrams of FeCN nanocubes was dispersed in a mixed solvent containing 15 mL dimethylformamide and 15 mL ethanol under ultrasonication for 30 min. Then, 50 mg of ammonium tetrathiomolybdate ((NH4)2MoS4) was added to the above suspension under magnetic stirring. After stirring for 12 h, the mixture was transferred into a Teflon-lined stainless-steel autoclave and heated at 210 °C for 10 h. Then, the autoclave was cooled to room temperature naturally. The obtained FeCN/MoS2 nanocomposite was centrifuged and washed with water and ethanol for several times, and dried in a vacuum oven at 80 °C overnight.

Synthesis of Fe1−xS/MoS2 nanocomposite. The as-prepared FeCN/MoS2 composite and sodium hydrosulfide (NaHS·H2O) as sulfurization precursor were placed in an alumina boat that was inserted into a tubular furnace. Subsequently, the tubular furnace was heated to 400 °C for 4 h and then heated to 600 °C for 2 h under argon flow with a heating rate of 2 °C min−1. Finally, Fe1−xS/MoS2 nanocomposite was obtained.

For comparison, bare Fe1−xS nanocubes were synthesized through the same method without (NH4)2MoS4 in similar conditions.

2.2 Material Characterization

Crystal structure of all samples was performed by X-ray diffraction techniques (XRD, Bruker D8 Advance) with Cu Kα radiation operated at 40 kV and 25 mA. The morphology and microstructure were observed using field emission scanning electron microscope (FESEM, JEOL JSM-7600F), transmission electron microscopy and high-resolution transmission electron microscopy (TEM and HRTEM, JEOL JEM-2010). The surface electronic states were analyzed by X-ray photoelectron spectroscopy (XPS, PHI Quantera II) with Al Kα source operated at 1486.6 eV. Nitrogen adsorption–desorption isotherms were carried out on a Quantachrome Autosorb-IQ analyzer. For in situ XRD measurement, each scan was collected between 10° and 50° using an electrochemical cell module equipped with metal beryllium (Be) and carbon paper as the window and current collector, respectively.

2.3 Electrochemical Measurements

To fabricate working electrodes, active materials (Fe1−xS/MoS2 or Fe1−xS), acetylene black and carboxymethyl cellulose (CMC) at a weight ratio of 7:2:1 were dispersed in deionized water to form a homogeneous slurry. The resultant slurry was coated on the copper foil and then dried at 120° for 12 h in a vacuum oven. In a typical assembly process of CR2032 coin cells, sodium foil was used as the reference and counter electrodes, and glass fibers (GF, Whatman) were used as separators. 1 M NaPF6 was dissolved into a mixed solution of ethylene carbonate (EC) and diethyl carbonate (DEC) (EC/DMC, 1:1/v: v) with 2.5 wt% fluoroethylene carbonate (FEC) as electrolyte. The galvanostatic charge/discharge tests were conducted on a battery testing system (Neware) at different densities in a voltage range of 0.01–3.0 V. Cyclic voltammetry (CV) was carried out using an electrochemical workstation (VMP3, Bio-Logic) at different scan rates. Electrochemical impedance spectroscopy (EIS) was performed on the same workstation in a frequency range from 100 kHz to 10 mHz.

2.4 DFT Calculations

All the calculations are based on density functional theory (DFT) using the plane-wave pseudopotentials [38, 39] with exchange–correlation of Perdew–Burke–Ernzerhof (PBE) [40, 41] formation as implemented in the Vienna Ab initio Simulation Package (VASP) [42]. A cutoff energy of 450 eV is employed for the plane-wave expansion of the wave functions. The Brillouin zone is sampled with 3 × 3 × 1 Monkhorst–Pack k-point mesh [43] for the structural optimization. The convergence criteria for the total energy and ionic forces were set to 10−4 eV and 0.1 eV Å−1, respectively. The construction with a 20 Å vacuum zone in the z direction to minimize the interactions between adjacent images. The climbing-image nudged elastic band (NEB) method [44] is used to seek the minimum energy pathways and determine the diffusion energy barriers.

3 Results and Discussion

The synthetic procedures for the Fe1−xS/MoS2 nanocomposite are illustrated in Fig. 1a. PB nanocubes were first synthesized using a simple hydrothermal method. The as-prepared PB shows the typical diffraction peaks ascribed to Fe4[Fe(CN6)]3 with a face-centered-cubic structure (JCPDS NO. 73-0687) and a well-defined nanocubic morphology with an average size of about 700 nm (Fig. S1a, b). Next, PB nanocubes were pyrolyzed to nitrogen-doped porous nanocubic structure (FeCN) nanocubes in an inert atmosphere. As shown in Fig. S1c, the cubic framework is well maintained. MoS2 nanosheets were then uniformly grown on surface of FeCN nanocubes, resulting in an enlarged nanocube size of about 1 μm (Fig. S1d). Finally, Fe1−xS/MoS2 composite was obtained without obvious structural change after a sulfurization process (Fig. 1d). Figure 1b shows XRD pattern of the as-synthesized Fe1−xS/MoS2 composite. The diffraction peaks at 30.3°, 33.7°, 43.5° and 53.0° are assigned to the (200), (2011), (2022) and (220) planes of hexagonal pyrrhotite Fe1−xS (JCPDS NO. 29-0726), respectively. Besides, a series of strong peaks at 14.3°, 29.8°, 32.7°, 39.5°, 44.2°, 49.7°, 58.2° and 60.3° can be readily indexed to the (002), (004), (100), (103), (006), (105), (110) and (008) planes of MoS2 (JCPDS No. 37-1492). XPS analysis was performed to investigate the chemical compositions and surface electronic states of Fe1−xS/MoS2 composite. The survey spectra confirm the presence of Fe, Mo, S, C, and N elements (Fig. S2). In the Fe 2p XPS spectrum, the peaks located at 712.7 and 726.2 eV are corresponded to the Fe-S bonds along with a corresponding satellite at 720.3 eV, whereas the observed peaks at 715.5 and 728.2 eV indicate the presence of Fe–O bonds (Fig. 1c) [45]. In the S 2p XPS spectrum, peaks at 162.5 and 163.7 eV are correlated to the S 2p1/2 and S 2p3/2, respectively, while the peak centered at 169.4 eV is attributed to S–O bond from superficial oxidized sulfur species [46]. The Mo 3d XPS spectrum shows two obvious peaks at 229.6 and 232.8 eV corresponding to Mo 3d3/2 and Mo 3d5/2, respectively, which are the characteristic of Mo4+, whereas the peak at 236.2 eV is attributed to Mo–O bond [47]. TEM image confirms that MoS2 nanosheets are well anchored on the Fe1−xS nanocubes, and there is an obvious interface between the Fe1−xS and MoS2 (Fig. 1e). The average thickness of MoS2 is revealed to be around 150 nm, well consistent with SEM observation. High-resolution TEM image shows well-resolved lattice fringes with d-spacing of 0.62 and 0.23 nm, corresponding to the (002) and (103) planes of MoS2, respectively (Fig. 1f). No obvious lattice fringes of Fe1−xS can be observed due to the coverage of thick MoS2 nanosheets. EDS elemental mapping images manifest the existence of Fe, Mo, S, C, and N elements, further confirming the successful synthesis of Fe1−xS/MoS2 heterostructure (Fig. 1g). The existence of C and N elements could be attributed to the formation of N-doped carbon, which is expected to enhance the electronic conductivity of electrode materials [48, 49]. For comparison, bare Fe1−xS nanocubes were also synthesized by the same route without MoS2. As shown in Fig. S3a, all diffraction peaks of Fe1−xS could be ascribed to the hexagonal pyrrhotite Fe1−xS (JCPDS NO. 29-0726) without any other impurity phases. SEM and TEM images show that the nanocubic-shaped morphology is well retained with a much rougher surface and porous structure (Fig. S3b-e). HRTEM image clearly displays the lattice spacing of 0.29 nm, corresponding to (200) plane of Fe1−xS (Fig. S3f). In addition, the nitrogen adsorption–desorption isotherms in Fig. S4 indicate that Fe1−xS/MoS2 composite possesses higher BET surface area of 45.6 m2 g−1 than that of pure Fe1−xS (12.5 m2 g−1) (Fig. S4). The composite structures with high porosity can provide abundant contact area and diffusion channels for electrolyte.

Fig. 1
figure 1

a Schematic illustration of the synthesis process for Fe1−xS/MoS2 nanocomposite. b XRD pattern and c high-resolution XPS spectra (Fe 2p, Mo 3d, and S 2p) of Fe1−xS/MoS2 composite. d SEM, e TEM and f HRTEM images of Fe1−xS/MoS2 composite. g Element mappings of the Fe1−xS/MoS2 composite

Full size image

Figure 2a shows the CV curves of Fe1−xS/MoS2 heterostructure for the first five cycles. During the first discharge process, three obvious cathodic peaks at around 1.10, 0.60, and 0.15 V can be assigned to the multistep sodiation processes involving the intercalation of Na+ into Fe1−xS/MoS2 and the formation of Mo0/Fe0 [13, 35, 50]. Meanwhile, the strong peak at about 0.60 V is also associated with the formation of SEI films, and the peak intensity decreases obviously in the subsequent cycles [38, 51]. For the anodic scan, two obvious peaks at about 1.60 and 1.90 V correspond to the stepwise desodiation process [52]. From the second cycle onward, the reversible reactions of Na2-yFe1−xS2/Na2Fe1−xS2 and MoS2/Mo0 enable the reversible Na storage in composite electrode [12, 53]. These peaks related to the phase transformation during sodiation/desodiation process will be further discussed on the basis of in situ XRD analysis. The CV curves of Fe1−xS/MoS2 composite electrode basically overlap from the second cycle onward, indicating excellent electrochemical reversibility and structural stability of the heterostructure. In contrast, bare Fe1−xS nanocubes show inferior electrochemical reversibility (Fig. S5).

Fig. 2
figure 2

a CV curves of Fe1−xS/MoS2 composite electrode for the first five cycles. b Galvanostatic charge–discharge profiles of Fe1−xS/MoS2 composite electrode at 100 mA g−1. c Cycling performance at 100 mA g−1. d Rate capability at various current densities. e Galvanostatic charge–discharge profiles of Fe1−xS/MoS2 composite electrode at various current densities. f The comparison of the rate capability with other iron sulfide-based anode materials previously reported [9, 12,13,14, 38, 55, 56]. g Long-term cycling performance at 1.0 A g−1

Full size image

Figure 2b presents the galvanostatic charge–discharge curves of Fe1−xS/MoS2 composite electrode under different cycles at 100 mA g−1. The Fe1−xS/MoS2 composite exhibits an initial discharge and charge capacities of 1072.1 and 645.0 mAh g−1, respectively, showing an initial Coulombic efficiency of 60%. The irreversible capacity loss arises from the formation of SEI layer on the electrode surface [54]. However, the bare Fe1−xS electrode delivers a lower initial Coulombic efficiency of 58% at the same current density (Fig. S6). After 100 cycles, the Fe1−xS/MoS2 heterostructure demonstrates a high reversible capacity of 584.7 mAh g−1 with capacity retention of approximately 91%, and high Coulombic efficiency from the second cycle onward (Fig. 2c). In contrast, the bare Fe1−xS shows rapid capacity decay and lower Coulombic efficiency during the whole process.

The superiority of the heterostructure is also highlighted by its outstanding rate capability. The Fe1−xS/MoS2 composite electrode delivers the average reversible specific capacities of 637.2, 594.7, 552.8, 490.4, 432.3, and 372.1 mAh g−1 at the current densities of 0.1, 0.2, 0.5, 1.0, 2.0, and 3.0 A g−1, respectively. When the current density returns to 0.1 A g−1, the specific capacity can revert back to a high value of 636.7 mAh g−1 with approximately 100% capacity retention (Fig. 2d, e). Such excellent rate capability is superior to most of other iron sulfide-based anode materials reported previously, as shown in Fig. 2f. However, with the increasing current density, the bare Fe1−xS electrode shows a much lower capacity (Fig. S7). Furthermore, the long-term cycling performance of the Fe1−xS/MoS2 electrode was evaluated at 1.0 A g−1 (Fig. 2g) with an initial five cycles activation at 0.1 A g−1. After 300 cycles, a high reversible capacity of 396.8 mAh g−1 is achieved, implying the superior high-rate cycling stability. However, the bare Fe1−xS electrode shows rather poor cycle performance under the identical testing conditions. Ex situ SEM was then conducted after cycling. As shown in Fig. S8, Fe1−xS/MoS2 composite almost maintains the structural integrity without obvious damage, confirming the good structural stability for sodium storage. Such excellent cycling and rate performance might be related to the following unique hetero-nanostructure design advantages. First, the porous MoS2 nanosheets enlarge the electrode/electrolyte contact area and reduce the charge-transfer resistance along the electrode/electrolyte interface. Second, the heterostructure structure design prevents the nanocubes agglomeration and accommodates the severe structural deformation significantly [35]. More importantly, the heterointerface between the Fe1−xS and MoS2 could serve as an “ion reservoir” to significantly boost Na+ capture/storage and then accelerate the Na+ diffusion from the shell to the internal part [31].

To further interpret the outstanding sodium storage performance of Fe1−xS/MoS2 heterostructure, kinetics analysis based on CV measurements at different scan rates was carried out. As shown, the CV curves from 0.1 to 1.0 mV s−1 exhibit similar shape with broad reduction/oxidation peaks related to the sodiation/desodiation processes (Fig. S9). The relationship between the peak current (i) and scan rate (v) obeys the power law as shown in Eq. (1) [57]:

$${\text{i}} = {\text{av}}^{b}$$
(1)

where both a and b are constants. The b value reveals the charge storage mechanism (b = 0.5 represents a diffusion-controlled process; b = 1.0 refers to a surface capacitance-controlled process). As shown in Fig. 3a, the b values obtained from the log (v) versus log (i) plots for peak 1, 2, 3, and 4 are 0.84, 0.92, 0.89, and 0.92, respectively, suggesting the charge storage in Fe1−xS/MoS2 composite is dominated by surface capacitive behavior.

Fig. 3
figure 3

Kinetic analysis of sodium storage behavior for Fe1−xS/MoS2 composite. a Linear relationship between log (i) and log (v). b Capacitive and diffusion-controlled contribution at 1.0 mV s−1. c Normalized contribution ratio of capacitive capacities at different scan rates. d In situ EIS spectra evolution at different charge/discharge potentials. e First charge/discharge profile at 100 mA g−1 with labeled points for EIS. f EIS spectra after different cycles. g GITT curve. h The corresponding Na+ diffusion coefficient at different discharge/charge state of Fe1−xS/MoS2 composite and pure Fe1−xS electrodes

Full size image

To be more intuitive, the quantitative pseudocapacitance contribution at different scan rates could be calculated according to Eq. (2) [58]:

$${\text{i}}\left( V \right) = k_{1} v + k_{2} v^{1/2}$$
(2)

where k1 and k2 are constants for a fixed voltage. The k1v indicates the current response from surface capacitive contribution, while k2v1/2 represents the diffusion-controlled current. The surface capacitive contribution of the Fe1−xS/MoS2 composite electrode at 1.0 mV s−1 is about 74.4% as displayed by the shaded area (Fig. 3b). With the increase in scan rate, the pseudocapacitance contribution gradually increases from 57.9% to 85.4% (Fig. 3c). However, the bare Fe1−xS electrode shows a much lower capacitive contribution at different scan rates (Fig. S8). The higher pseudocapacitance contribution for Fe1−xS/MoS2 composite determines a more favorable electrochemical kinetics behavior at high current density, which is consistent with its excellent rate capability as shown in Fig. 2.

In situ electrochemical impedance spectroscopy (EIS) during the first sodiation/desodiation process was conducted to investigate the kinetics differences induced by structural and phase evolution at different discharge/charge states. Figure 3d shows the EIS behavior of Fe1−xS/MoS2 composite in the first cycle with labeled points presented in Fig. 3e. All Nyquist plots show similar features with a depressed semicircle in high–medium-frequency region and an oblique line in low-frequency region. Based on the fitted equivalent circuits, at open-circuit voltage (OCV) state, the semicircle can be ascribed to the charge-transfer resistance (Rct) and the oblique line represents the Warburg impedance (W) related to the Na+ diffusion [59]. However, at other states, the semicircle corresponds to two overlapping parts of the SEI film resistance (RSEI) and Rct [60]. During the sodiation process, due to the formation of insulative Na2S matrix and SEI films accompanied by the gradual volume change, the resistance increases obviously from OCV to 0.01 V [61]. During the desodiation process, the nanoclustered NayFe1−xS2 phase and metallic state 1T-MoS2 gradually form, and meanwhile, non-conductive Na2S gradually disappears. As a consequence, the resistance decreases gradually from 0.01 to 3.0 V. In addition, EIS measurements were also performed after different cycles at current density of 100 mA g−1 to show the charge-transfer stability. With the increase in cycles, the resistance of Fe1−xS/MoS2 composite electrode decreases initially owing to the activation process and then slightly increases after several cycles (Fig. 3f). In contrast, Nyquist spectra of the bare Fe1−xS electrode show higher resistances at different cycles. Overall, the in situ EIS measurements have confirmed the enhanced charge-transfer kinetics and electronic conductivity of the Fe1−xS/MoS2 composite (Fig. S10) [62, 63].

Galvanostatic intermittent titration technique (GITT) was further performed to investigate the influence of multistep sodiation and desodiation reactions of Fe1−xS/MoS2 heterostructure on Na-ion chemical diffusion coefficient (DNa) (Figs. 3g and S12). From Fig. 3h, the Na+ diffusion coefficient fluctuates with the progress of sodiation/desodiation, and the minimum values appear at each cathodic/anodic reaction plateau, where Na ions diffuse deeply into/from the internal crystal structure. Obviously, the Fe1−xS/MoS2 composite electrode shows a higher diffusion coefficient and minor change than those of bare Fe1−xS, which can be attributed to the unique heterointerface to act as an “ion reservoir” and fast diffusion channel for Na ions. In addition, phase boundaries can suppress the growth of crystal domains, thus forming numerous defects and active sites to facilitate Na+ diffusion [64]. Furthermore, benefiting from out-of-sync electrochemical reactions of these two sulfides at different voltages, the structural stress could be effectively mitigated, which is also favorable for diffusion of sodium ions.

To further reveal the voltage-dependent phase transformation behavior of Fe1−xS/MoS2 heterostructure during the first sodiation/desodiation process, in situ XRD was performed between 0.01 and 3.0 V at the current density of 80 mA g−1 (Fig. 4). Noticed that the strong peaks located at about 26.7°, 44.0°, and 46.0° are derived from carbon paper, BeO, and Be, respectively. As shown in Fig. 4, the peaks at 29.9°, 33.7°, and 43.5° are related to the (200), (2011), and (2022) planes of Fe1−xS, and the peaks at about 14.3°, 32.7°, and 39.5° correspond to the (002), (100), and (103) diffractions of MoS2. During the sodiation process, the peaks of Fe1−xS gradually shift to a lower 2θ, suggesting the formation of intermediate Na2Fe1−xS2 through the Na+ insertion into Fe1−xS. When continuously discharging to 0.01 V, the peaks related to Na2Fe1−xS2 gradually disappear, along with the gradual increase in the diffraction peak corresponding to Na2S. Fe peaks could not be clearly observed in the process, which is mostly due to either the ultra-small crystal size or amorphous nature of resultant Fe0 [65]. On the other hand, the prominent peak at 14.3° shifts obviously during the whole sodiation/desodiation process, indicating the structural changes of MoS2. In stages I and II, there are two stages involving two-phase transitions: One is from 2H-MoS2 (2θ(002) = 14.3°) to 2H-Na0.5MoS2 (2θ(002) = 11.8°) and the other is from 2H-Na0.5MoS2 to 1T-NaMoS2 (2θ(002) = 12.6°) [66]. Subsequently, more Na ions insertion into NazMoS2 did not induce any detectable phase changes. In stage III, the intensity of diffraction peak related to Na2S gradually increases accompanied by the formation of Mo. After recharging back to 1.6 V, the (002) main peak gradually shifts toward a lower 2θ angle to form 1T-NaMoS2. In subsequent deintercalation process, the diffraction peak shifts back to a higher 2θ angle with a two-phase transition from 1T-Na0.5MoS2 to 1T-MoS2 (2θ(002) = 14.1°) rather than the initial 2H-MoS2 (2θ(002) = 14.3°) [47, 67]. Meanwhile, the diffraction peaks corresponding to Na2S and Mo gradually disappear. Interestingly, the diffraction peak related to NazMoS2 always exists in the whole conversion reaction process, implying the incomplete conversion from NazMoS2 to Mo, which is favorable for structural stability. At the desodiation process, due to the strong peak of carbon paper, the peaks at around 30° could not be clearly assigned to the Na2Fe1−xS2 and Na2-yFe1−xS2, which are similar to the previous report for iron sulfide materials [12]. Therefore, from the above in situ XRD and CV analysis, the multistep reaction mechanisms for Fe1−xS/MoS2 heterostructure at different sodiation/desodiation states could be expressed as follows:

Fig. 4
figure 4

a Contour plots of in situ XRD results and b selected diffraction patterns of Fe1−xS/MoS2 composite electrode during the first sodiation/desodiation process

Full size image

Sodiation process:

$$\begin{aligned} {\text{Stage I }}\left( {\text{intercalation}} \right) & :2{\text{Fe}}_{1 - x} {\text{S }} + 2{\text{Na}}^{ + } + 2{\text{e}}^{ - } \to {\text{ Na}}_{2} {\text{Fe}}_{1 - x} {\text{S}}_{2} + \, \left( {1 - x} \right){\text{Fe}} \\ &\quad 2{\text{H}} - {\text{MoS}}_{2} + 0.5{\text{Na}}^{ + } + \, 0.5{\text{e}}^{ - } \to \, 2{\text{H}} - {\text{Na}}_{0.5} {\text{MoS}}_{2} \\ \end{aligned}$$
$$\begin{aligned} {\text{Stage II }}\left( {\text{intercalation}} \right): & 2{\text{H}} - {\text{Na}}_{0.5} {\text{MoS}}_{2} + 0.5{\text{Na}}^{ + } + 0.5{\text{e}}^{ - } \to \, 1{\text{T}} - {\text{NaMoS}}_{2} \\ &\quad 1{\text{T}} - {\text{NaMoS}}_{2} + \left( {z - 1} \right){\text{Na}}^{ + } + \left( {z - 1} \right){\text{e}}^{ - } \to \, 1{\text{T}} - {\text{Na}}_{z} {\text{MoS}}_{2} \\ \end{aligned}$$
$$\begin{aligned} {\text{Stage III }}\left( {\text{conversion}} \right): & 1{\text{T}} - {\text{Na}}_{z} {\text{MoS}}_{2} + \, \left( {4 - z} \right){\text{Na }} + \, \left( {4 - z} \right){\text{e}}^{ - } \to {\text{Mo}} + \, 2{\text{Na}}_{2} {\text{S}} \\ &\quad{\text{Na}}_{2} {\text{Fe}}_{1 - x} {\text{S}}_{2} + \, 2{\text{Na }} + 2{\text{e}}^{ - } \to \, \left( {1 - x} \right){\text{Fe }} + \, 2{\text{Na}}_{2} {\text{S}} \\ \end{aligned}$$

Desodiation process:

$$\begin{aligned} {\text{Stage IV }}({\text{deconversion}}): & \, \left( {1 - x} \right){\text{Fe }} + \, 2{\text{Na}}_{2} {\text{S }} \to {\text{Na}}_{2} {\text{Fe}}_{1 - x} {\text{S}}_{2} + \, 2{\text{Na}} + 2{\text{e}}^{ - } \\ &\quad{\text{Mo }} + \, 2{\text{Na}}_{2} {\text{S }} \to \, 1{\text{T}} - {\text{Na}}_{0.5} {\text{MoS}}_{2} + \, 3.5{\text{Na}} + 3.5{\text{e}}^{ - } \\ \end{aligned}$$
$$\begin{aligned} {\text{Stage V }}\left( {\text{deintercalation}} \right): & {\text{Na}}_{2} {\text{Fe}}_{1 - x} {\text{S}}_{2} \to {\text{ Na}}_{2 - y} {\text{Fe}}_{1 - x} {\text{S}}_{2} + y{\text{Na}} + y{\text{e}}^{ - } \\ &\quad1{\text{T}} - {\text{Na}}_{0.5} {\text{MoS}}_{2} \to \, 1{\text{T}} - {\text{MoS}}_{2} + \, 0.5{\text{Na}} + \, 0.5{\text{e}}^{ - } \\ \end{aligned}$$

Furthermore, the structural change of Fe1−xS/MoS2 composite electrode at a high current density was also monitored by in situ XRD. As shown in Fig. S13, from the second cycle forward, the (002) diffraction peak exhibits small periodic change, confirming the stable framework and kinetic processes of this composite structure during cycling.

DFT calculations were carried out to investigate the heterointerfacial behavior to present atomic-level verification for the superior sodium storage capability of the heterostructure. The Na+ diffusion barriers in bare Fe1−xS and Fe1−xS/MoS2 heterostructure were calculated (Figs. 5b and S14). As shown, the barrier is about 0.4 eV in MoS2 side close to the interface for heterostructure, obviously lower than that in bare Fe1−xS (~ 0.5 eV). This result indicates that Na+ migration is more favorable in Fe1−xS/MoS2 heterointerface, which is beneficial for the improvement of electrochemical reaction kinetics. On the basis of the above analysis, a mechanism for enhanced electrochemical performance was proposed as illustrated in Fig. 5c. The Fe1−xS/MoS2 heterostructure is composed of hexagonal Fe1−xS nanocubes and 2D layered MoS2 nanosheets. Once discharging, the unique “stacking card” nanostructures derived from the assembled MoS2 nanosheets can accelerate the electrolyte permeation and Na+ migration. Benefiting from the low diffusion barrier, plenty of Na ions could store at the interface, then forming the so-called ion reservoir. High concentration gradient also drives Na+ transport from the interface to the internal part, hence realizing an efficient conversion reaction kinetics [34]. After full sodiation, MoS2 is transformed into metallic Mo and Na2S. During the desodiation process, metallic Mo nanoclusters can serve as “pin conductor” until 1T-MoS2 phase is formed, dramatically improving the electrical conductivity of the whole electrode and then facilitating the conversion reaction [31]. In the subsequent cycles, such enhancement effects could be retained owing to the regeneration of heterointerface. The difference is that the original 2H-MoS2 phase was completely converted into 1T-MoS2 phase along with the change of electronic states between semiconductive and metallic, which can greatly enhance electrical conductivity, as well as boost Na+ diffusion and accelerate the charge-transfer kinetics [67]. Therefore, such outstanding sodium storage performance can be mainly attributed to the unique heterostructure design: (1) Low Na+ diffusion barrier originated from Fe1−xS/MoS2 heterointerface can significantly promote reaction kinetics; (2) the phase transformation from 2H- to 1T-MoS2 after the initial cycle can greatly enhance electrical conductivity; (3) nanoarchitectures enable shorten ion diffusion pathway and mitigate the volume change; (4) high specific surface area caused by hierarchical MoS2 nanosheet decoration is able to further buffer the structural stress and facilitate electrolyte permeation.

Fig. 5
figure 5

a Schematic illustration of Fe1−xS/MoS2 interface. b The diffusion barrier profiles of Na on Fe1−xS surface and Fe1−xS/MoS2 interface. c Proposed schematic atomistic models of the illustrated reaction mechanism for Fe1−xS/MoS2 heterostructure during the sodiation and desodiation process

Full size image

4 Conclusion

In conclusion, we have demonstrated experimental and theoretical evidence that the electrochemical reaction kinetics could be significantly boosted by rationally designing heterostructure with abundant “ion reservoir” for sodium storage. The low diffusion barrier at the heterointerface greatly promotes sodium ion diffusion and charge-transfer kinetics. As a proof of concept, SIB anode based on Fe1−xS/MoS2 heterostructure exhibits superior rate capability and long cycle life. In light of the analysis about the dynamic relationship between heterointerface and electrochemical performance, our present work provides a fundamental understanding on heterostructure engineering for high-performance energy storage devices.