Qiannian
Li
a,
Yanting
Xia
a,
Jinmao
Fang
a,
Junwei
Chen
a,
Yan
Zhang
*a,
Wenpei
Kang
*b and
Jun
Xu
*a
aSchool of Microelectronics, Hefei University of Technology, Hefei 230009, P. R. China. E-mail: zhangyan@hfut.edu.cn; apjunxu@hfut.edu.cn
bSchool of Materials Science and Engineering, China University of Petroleum (East China), Qingdao, 266580, PR China. E-mail: wpkang@upc.edu.cn
First published on 9th October 2025
As a layered two-dimensional material, MoSe2 exhibits interlayer-tunable properties and exceptional theoretical capacities, making it a promising candidate for sodium/potassium ion storage systems. Nevertheless, its inadequate conductivity and irreversible reactions during charge and discharge seriously affect its electrochemical performance. Herein, a hierarchical interlayer-expanded MoSe2/C (IE-MoSe2/C) hybrid architecture with interlinked hollow nanospheres is engineered via a two-stage fabrication process combining hydrothermal self-assembly and controlled pyrolysis. The interlayer spacing increases to 1.02 nm, which accelerates the transport of sodium and potassium ions. Additionally, the strong interface between carbon and MoSe2 improves the conductivity, thereby enhancing the electrochemical kinetics of sodium-ion batteries (SIBs) and potassium-ion batteries (PIBs). On the other hand, the unique hierarchical IE-MoSe2/C structure with hollow nanospheres can effectively mitigate variations in volume during cycling. Thus, IE-MoSe2/C exhibits outstanding electrochemical characteristics as an anode material for SIBs and PIBs. Specifically, IE-MoSe2/C exhibits better rate capability (98 mAh g−1 at 20 A g−1 in SIBs) and cycling performance (269/174 mAh g−1 at 2.0/5.0 A g−1 over 1100 cycles in SIBs and 133/96 mAh g−1 at 1.0/2.0 A g−1 over 1000 cycles for PIBs). Additionally, at 0.5C, the full cell of IE-MoSe2/C||Na3V2(PO4)3 can display a consistent capacity of 93 mAh g−1, demonstrating the potential for future practical applications.
It is found that two-dimensional (2D) layered transition metal dichalcogenides (LTMDs) show promise as anode materials for SIBs and PIBs, due to their wide interlayer distances that enable efficient alkali metal ion (Na+/K+) diffusion and high theoretical capacities.16,17 Among these, as a typical sandwich-layer structure, MoSe2 has received extensive attention as a sodium/potassium ion host, owing to the wide interlayer spacing of 6.5 Å originating from the weak stacking of Se–Mo–Se monolayers, and good electrical conductivity based on the small band gap.15,18 The structural merits are beneficial for the storage of larger-sized sodium/potassium ions. However, during repeated charge–discharge cycles, large sodium/potassium ions still induce interlayer expansion of MoSe2, ultimately causing structural collapse of the 2D framework. This will induce decreased reaction kinetics and fast capacity decay. To tackle the above issues, conductive carbon compositing is considered an effective approach for enhancing the electrochemical attributes of MoSe2/C nanocomposites in SIBs/PIBs.19,20 Generally, the conductive carbon network in MoSe2/C nanocomposites facilitates electron transfer, maintains structural stability and optimizes interlayer spacing, all of which can effectively boost the rate performance and cycle life. For instance, hierarchically structured MoSe2/N–C nanorods with an expanded interlayer were synthesized via a solvothermal approach employing an ethylenediamine-modified MoO3 precursor.21 Nitrogen-doped carbon (N–C) was intercalated into MoSe2 interlayers to expand the d-spacing and served as a conductive scaffold to suppress nanosheet restacking. Thus, MoSe2/N–C possessed superior electrochemical properties in SIBs and PIBs. On the other hand, carbon coating serves as an effective approach for modulating the crystal growth of MoSe2. This regulation thereby enhances electron transfer kinetics and alleviates volumetric expansion during Na+/K+ insertion. For instance, ultrathin 2D MoSe2 nanosheets featuring expanded interlayer spacing were uniformly integrated within a three-dimensional (3D) macroporous carbon framework (labeled as MoSe2@C).22 This architecture demonstrates enhanced sodium-ion storage capacity. The MoSe2@C nanohybrid manifests exceptional high-rate performance coupled with prolonged cyclability, attributable to multiscale electron/ion transport channels enabled by 3D-interconnected conductive networks and interfacial charge redistribution effects. This can promote efficient Na+ intercalation/deintercalation, accelerate electron/ion transport kinetics, and mitigate structural expansion during cycling. Consequently, carbon plays multiple roles in the construction of the MoSe2-based hierarchical structure, thereby improving the reaction kinetics of Na+/K+ transfer and enhancing structural stability to buffer volume changes.
Herein, a hierarchical interlayer-expanded MoSe2/C (IE-MoSe2/C) composite with enlarged interlayer spacing and assembled hollow nanospheres was constructed using self-curled MoSe2 layers. This hierarchical IE-MoSe2/C structure has several advantages for boosting Na+/K+ storage. Firstly, the enlarged interlayer spacing (1.02 nm) in MoSe2 could reduce Na+/K+ diffusion barrier energy, facilitating their insertion/extraction. Secondly, the hierarchical architecture constructed from hollow nanospheres effectively alleviates volumetric expansion, enhancing cycling stability. Thirdly, the carbon matrix can prevent the collapse of MoSe2. Consequently, IE-MoSe2/C displays excellent energy storage properties. In SIBs, the IE-MoSe2/C anode demonstrates exceptional cycling stability, maintaining stable capacities of 269/174 mAh g−1 at 2.0/5.0 A g−1 for 1100 cycles. This robust cycling performance originates from both the expanded interlayer spacing (1.02 nm), which facilitates rapid Na+ diffusion, and the carbon matrix, which enhances electron transport. Remarkably, even at 20.0 A g−1, a significant specific capacity of 98 mAh g−1 is retained. For PIBs, the anode exhibits outstanding rate performance, delivering 133/96 mAh g−1 at 1.0/2.0 A g−1 over 1000 cycles, attributed to increased pseudocapacitive K+ storage behavior confirmed by kinetic analysis. The full cell coupled with Na3V2(PO4)3 (NVP) (IE-MoSe2/C||NVP) further validates practical feasibility, achieving 93 mAh g−1 after 300 cycles at 0.5C.
![]() | ||
| Fig. 1 Schematic illustration of the formation mechanism and structure of IE-MoSe2/C hollow nanospheres. | ||
The X-ray diffraction (XRD) patterns (Fig. 3a) show that pristine MoSe2 presents strong diffraction peaks at 13.6°, 31.6°, 37.8°, and 55.8°, respectively, which can be indexed to the (002), (100), (103) and (110) diffractions of 2H-MoSe2 (PDF#29-0914).32 As for IE-MoSe2/C, the (002) peak becomes weak and shifts to a lower angle of 8.0°.33 The IE-MoSe2/C interlayer spacing is determined to be 1.02 nm using the Bragg equation, which is in agreement with HRTEM observations. The component was further analyzed using Raman spectra. In Fig. 3b, the two distinct Raman peaks at 237 and 284 cm−1 correspond to the out-of-plane A1g and in-plane E2g1 of MoSe2.34,35 Notably, the A1g peak of IE-MoSe2/C has a slight red shift to 239 cm−1, indicating that the interlayer interaction is weakened by the carbon layer insertion. Additionally, Raman analysis of the IE-MoSe2/C composite reveals distinct vibrational signatures at 1356 and 1583 cm−1, which are attributable to the disorder and graphitic lattice modes of carbonaceous materials.36 This confirms the coexistence of defective carbon structures and ordered sp2-hybridized domains within the composite architecture. The intensity ratio (ID/IG) is about 0.98, illustrating the amorphous nature of the carbon with plenty of defects, which can supply sufficient active sites for the Na+/K+ reaction.37–39 Thermogravimetric analysis was employed to quantify the carbon mass content of the IE-MoSe2/C composite under an oxygen atmosphere (Fig. S3). A carbon mass ratio of 34% in the IE-MoSe2/C composites is determined via weight changes during the thermal oxidation process converting MoSe2 into MoO3.40
![]() | ||
| Fig. 3 (a) XRD patterns and (b) Raman spectra of IE-MoSe2/C and MoSe2; high-resolution XPS spectra of (c) Mo 3d, (d) Se 3d, (e) C 1s, and (f) N 1s for IE-MoSe2/C. | ||
To investigate the surface chemical states of IE-MoSe2/C, X-ray photoelectron spectroscopy (XPS) characterization was carried out. Two distinct peaks at 229.0 eV (Mo 3d5/2) and 232.15 eV (Mo 3d3/2) in the deconvoluted Mo 3d spectrum (Fig. 3c) confirm the presence of Mo4+.19,41 The broad peak at 235.2 eV is associated with the C–O–Mo bond, resulting from the dehydrogenation of the –OH group by annealing.23 Additionally, there is a minor peak at 230.4 eV, associated with the Se 3s orbital.42 The Se 3d XPS spectrum (Fig. 3d) presents characteristic binding energies of 55.3 eV (Se 3d3/2) and 54.4 eV (Se 3d5/2), attributed to the dominant Se2− oxidation state in MoSe2.18,43,44 Meanwhile, deconvolution of the C 1s spectrum (Fig. 3e) gives five distinct chemical states: C–Se at 290.2 eV, C
O at 288.7 eV, C–O–Mo at 287.2 eV, C–N at 285.4 eV, and C–C at 284.6 eV.23 Notably, the C–O–Mo bond provides evidence of interfacial charge transfer between MoSe2 and the carbon matrix.39 The N 1s spectrum (Fig. 3f) has three characteristic bonding configurations: pyridinic N (398.5 eV), pyrrolic N (400.7 eV), and graphitic N (401.9 eV).15 The XPS characterization confirmed the existence of N–C bonds, which suggests that nitrogen species are successfully incorporated into the IE-MoSe2/C composite. This N-doping can induce defective active sites and enhance charge transfer, thereby accelerating the reaction kinetics of Na+/K+ ions. Moreover, pyridinic nitrogen in the carbon matrix can produce conjugated π-electron networks, effectively boosting charge transfer kinetics.40 Therefore, IE-MoSe2/C could form covalent bonds with the intercalated carbon monolayers via C–Se/C–O–Mo bonds. This unique structure facilitates efficient electron/ion transport.23
To probe the architectural merits of the IE-MoSe2/C composite, systematic electrochemical evaluations were conducted in SIBs. Fig. 4a illustrates the first five cyclic voltammetry (CV) profiles scanned at 0.1 mV s−1 from 0.01 to 3.0 V. The weak cathodic peak at 0.85 V during the first reduction scan is predominantly governed by Na+ insertion into MoSe2 interlayers (forming NaxMoSe2), while the intense peak at 0.25 V originates from synchronous SEI film generation and subsequent phase reduction to metallic Mo/Na2Se.25,45 Accordingly, in the initial anodic process, the prominent oxidation peak observed at a potential of 1.67 V can be primarily attributed to the electrochemical conversion to MoSe2.46 The cathodic peaks shift to higher potentials and overlap well in the following four cycles. This indicates that good cycling performance can be achieved for the IE-MoSe2/C anode in SIBs. Fig. 4b displays the galvanostatic discharge/charge (GDC) profile of IE-MoSe2/C at 0.1 A g−1, where its potential plateaus align with CV peak positions. It exhibits an exceptional initial discharge specific capacity of 520 mAh g−1 and a charge specific capacity of 388 mAh g−1, yielding a notably high initial coulombic efficiency (ICE) of 74.6%. The initial capacity loss is principally due to inevitable SEI film generation in the first cycle. Subsequent GDC profiles (2nd, 5th, 10th, and 50th cycles) exhibit distinctly overlapping voltage curves. This observation confirms the outstanding cycling stability for the IE-MoSe2/C anode.
Rate capability assessment of the IE-MoSe2/C and pristine MoSe2 anodes (Fig. 4c) reveals progressive capacity attenuation over the range of 0.1–20.0 A g−1. The IE-MoSe2/C anode delivers capacities from 368 to 116 mAh g−1 upon current increase from 0.1 to 15.0 A g−1. Remarkably, even under extreme 20.0 A g−1 conditions, the capacity is retained at 98 mAh g−1, surpassing most reported LTMD anodes.47–50 This demonstrates exceptional high-rate tolerance. When reverting to 0.1 A g−1, the IE-MoSe2/C anode achieves full capacity recovery (383 mAh g−1), which is obviously higher than that of pristine MoSe2 (211 mAh g−1). The rate capabilities of the IE-MoSe2/C composite are superior to those of pristine MoSe2. This suggests that IE-MoSe2/C has superior and stable high-rate sodium storage capacity. Furthermore, Fig. S4 shows the GDC curves of IE-MoSe2/C and pristine MoSe2 at various current densities in SIBs. The curves retain similar shapes, except for the slope shifts, which are attributed to increased polarization at higher rates. The increased interlayer spacing and the carbon matrix of IE-MoSe2/C may accelerate the diffusion of Na+ and raise the electronic conductivity. In Fig. 4d, systematic investigations were conducted on the cyclability of the IE-MoSe2/C anode at high current densities. At 2.0 A g−1, a discharge capacity of 245 mAh g−1 is initially achieved. The progressive capacity attenuation over the first 40 cycles results in 7.5% capacity decay, kinetically constrained by the high electrochemical polarization. After that, the capacity increases gradually due to the activation process. Notably, the capacity is increased to 267 mAh g−1 at the 800th cycle. Then, the capacity is stabilized and 269 mAh g−1 is retained after 1100 cycles, demonstrating outstanding stability under high-rate cycling. It should be noted that the continuous capacity increase of the IE-MoSe2/C electrode in the first 800 cycles can be attributed to the electrochemical activation process. Specifically, the nanospheres in the IE-MoSe2/C hierarchical structure are tightly stacked together. During the initial cycling, the inner nanospheres are less accessible for participating in the electrochemical process. As cycling proceeds, the nanospheres in the IE-MoSe2/C assembly become loose, as confirmed by the SEM image of the anode after 800 cycles (Fig. S5). This significantly increases the effective surface area for the electrolyte contact, which facilitates Na+ intercalation and deintercalation reactions, resulting in an increase in specific capacity. This capacity growth phenomenon has also been reported for MoSe2-based anodes.51,52 The pristine MoSe2 anode delivers a second-cycle capacity of 284 mAh g−1 (Fig. S6). However, this capacity is significantly decreased in the first 50 cycles and stabilizes at a low value of 53 mAh g−1 after 200 cycles. This demonstrates that the hierarchical structure with hollow nanospheres and carbon confinement frameworks enables significant structural integrity during sodiation/desodiation processes, thus improving the cycling stability. Additionally, with a greater current of 5.0 A g−1, a similar variation trend is observed. Following 1100 cycles, the IE-MoSe2/C anode retains 90% of its original capacity, providing a stable capacity of 174 mAh g−1. The remarkable capacity retention of the IE-MoSe2/C composite demonstrates its superior high-rate sodium storage capability. Overall, the cycling stability of IE-MoSe2/C for SIBs is superior to that of some MoSe2-based anodes (Table S1). To validate the practical feasibility, a full cell battery was constructed utilizing commercial NVP as the cathode. Comparative GDC profiles of the NVP cathodes (half-cell) and the IE-MoSe2/C||NVP full cells are depicted in Fig. 4e. The NVP cathode in the half-cell reveals a characteristic 3.4 V discharge plateau and delivers a capacity of 104 mAh g−1 at 0.2C (where 1C = 117 mA g−1), reaching its theoretical capacity.25,53,54 In the IE-MoSe2/C||NVP full cell with a voltage range of 1.0–3.8 V, it delivers a capacity of 99 mAh g−1 at 0.2C. In Fig. 4f, for the rate performance of the IE-MoSe2/C||NVP full cell, capacities of 99, 93, 90, 86, and 62 mAh g−1 can be achieved at 0.2C, 0.5C, 1C, 2C, and 5C, respectively. Remarkably, the capacity is almost recovered (96 mAh g−1) after reverting to 0.2C. For the long-term cycling at 0.5C, a capacity of 93 mAh g−1 is retained over 300 cycles (Fig. 4g), indicating no capacity decay. This illustrates that exceptional reaction kinetics and cycling stability are also achieved in the full cells, demonstrating its potential practical application.
The in situ Raman test was performed with different voltage states to investigate reaction mechanisms in SIBs. As shown in Fig. 5a, the characteristic A1g band of MoSe2 is clearly observed for the IE-MoSe2/C electrode before cycling. Upon discharging to 0.84 V, the A1g peak intensity undergoes a gradual attenuation, primarily attributed to Na+ intercalation within MoSe2 layers and the subsequent weakening of interlayer interaction. This demonstrates that the intercalation reaction occurs. The vanishing of the A1g peak upon discharging to 0.44 V indicates that the conversion reaction is in progress. Throughout the charging process, the A1g peak gradually emerges and becomes strong upon charging to 3.0 V, confirming the high reversibility.18,55 To clearly observe the peak changes, a contour plot is derived, as shown in Fig. 5b. Consequently, an insertion reaction combined with a conversion reaction is responsible for the Na+ storage for the IE-MoSe2/C anode in SIBs.
![]() | ||
| Fig. 5 (a) In situ Raman spectra and (b) the corresponding contour plots of the IE-MoSe2/C anode with different voltages during the first discharge and charge process. | ||
The rapid Na+ storage was probed via the CV method (Fig. 6a). Similar CV shapes are observed when the electrode is scanned at 0.1–4 mV s−1 despite peak modulation. The i = avb relationship quantifies storage behaviour.56,57 Through taking the logarithm, the b value can be inferred based on the linear plots of log(i)–log(v).58 The Na+ storage process can be predicted using the b value. A b-value of 1.0 indicates capacitance-dominated energy storage behaviour, whereas a diffusion-controlled process becomes predominant at b = 0.5. Anodic and cathodic peaks with b-values of 0.81 and 0.88 are shown by quantitative analysis in Fig. 6b, indicating mixed charge storage processes. This suggests that the pseudocapacitive mechanism is primarily responsible for promoting the rapid response kinetics. Additionally, the capacity contribution (k1v) and the diffusion-controlled contribution (k2v1/2) ratios may be obtained using the formula i(v) = k1v + k2v1/2.59,60Fig. 6c quantifies the evolving charge storage contribution ratios in the IE-MoSe2/C anodes. Pseudocapacitive contributions dominate 60.9% at 0.1 mV s−1 and increase to 97.9% at 4 mV s−1, reflecting that a capacitive process promotes the high-rate sodium storage capacity. Notably, the contribution of pseudo-capacitance for IE-MoSe2/C is higher than for pristine MoSe2 anodes at each scan rate, which induces a superior rate capability for the IE-MoSe2/C anode.28 Fig. S7 displays the Nyquist plots after 50 cycles in SIBs, which are composed of straight lines with low frequency and a semicircle with high frequency for the charge transfer resistance (Rct). The reduced Rct of the IE-MoSe2/C anode relative to pristine MoSe2 signifies increased conductivity. Typically, Na+ storage kinetics is governed by solid-state Na+ diffusion. The galvanostatic intermittent titration technique (GITT) is used to determine the sodium ion diffusion coefficient (DNa). The GITT profiles, which are similar to the GDC profiles, are shown in Fig. 6d. The following formula can be used to obtain the DNa values:
![]() | (1) |
Fig. 6e shows the DNa values of IE-MoSe2/C and MoSe2, which show similar change trends for the two electrodes. The reduction in DNa values observed alongside the phase transformation demonstrates a strong correlation with the characteristic oxidation/reduction features in the CV profiles.61,62 The DNa values for the IE-MoSe2/C electrode are from 1.2 × 10−11 to 2.1 × 10−11 cm2 s−1, which are larger than those for pristine MoSe2 (0.7 × 10−11–1.4 × 10−11 cm2 s−1). As a result, enhanced reaction kinetics can be achieved for IE-MoSe2/C owing to its expanded interlayer spacing and carbon modification, which can provide superior rate and cycling performance.
Given the superior Na+ storage capabilities, the IE-MoSe2/C architecture was strategically adapted for K+ storage. Fig. 7 comprehensively characterizes the performance of PIBs. Fig. 7a describes the first five CV profiles recorded at 0.1 mV s−1. During the initial reduction phase, K+ intercalation into the MoSe2 interlamellar structure occurs at 1.0 V, forming KxMoSe2. The pronounced reduction signal observed at 0.07 V is linked to the phase transformation of KxMoSe2 into metallic molybdenum and K2Se compounds, accompanied by SEI layer generation. Anodic scanning shows a peak at 1.79 V (Mo → MoSe2 conversion).63 In the following cycles, the reduction peaks shift to higher voltages, which is similar to that for the SIBs. The high overlapping of the subsequent CV curves reveals the good reversibility and cycling stability of the IE-MoSe2/C anode. At 100 mA g−1 (Fig. 7b), the initial GDC cycles show 377/247 mAh g−1 discharge/charge capacities (66% ICE), attributed to SEI formation losses. Beyond the first cycle, the charge–discharge profiles exhibit minimal variation. Furthermore, the GDC curves with different currents are shown in Fig. 7c. The voltage–capacity curves show similar features, accompanied by gradual slope evolution as current densities escalate, originating from polarization effects. Rate capability assessment of IE-MoSe2/C (Fig. 7d) reveals progressive capacity attenuation upon current increase (0.1 → 5.0 A g−1), decreasing from 248 to 88 mAh g−1. Notably, upon reverting to the baseline 0.1 A g−1, the IE-MoSe2/C anode demonstrates complete capacity recovery (246 mAh g−1). In contrast to pristine MoSe2, only less than 70% of the initial capacity can be recovered (Fig. 7d). For the cycling performance at 1.0 and 2.0 A g−1 (Fig. 7e), it is found that the cycling performance is similar to that in SIBs. At 1.0 A g−1, the capacity reached 172 mAh g−1 during the 5th cycle. By the 1000th cycle, the capacity remained stable at approximately 133 mAh g−1, with a capacity decay rate of only 0.023% per cycle. Even under high-current cycling (2.0 A g−1), the capacity stabilizes at 96 mAh g−1 after 1000 cycles, retaining 72% of its initial value. Conversely, the MoSe2 electrode exhibits limited cycling stability under continuous charge–discharge cycling (Fig. S8), maintaining only a low capacity of 21 mAh g−1 after 400 cycles. Notably, the cycling stability of the IE-MoSe2/C anode is excellent, compared with the previously reported MoSe2-based anodes (Table S1).
Kinetic analysis via CV could infer high-rate K+ storage promotion for IE-MoSe2/C. In Fig. 7f, CV profiles acquired at 0.1–1 mV s−1 reveal b-values of 0.80 (anodic) and 0.81 (cathodic) (Fig. 7g). Thus, the fast reaction kinetics of the IE-MoSe2/C anode in PIBs is also dominated by the pseudo-capacitance behaviour.63 Meanwhile, the pseudo-capacitance contribution for the IE-MoSe2/C anode increases from 51.6% to 76.6% at 0.1 to 1.0 mV s−1, as shown in Fig. 7h, which is higher than that of the MoSe2 anode at each scan rate. The enhanced reaction kinetics observed in PIBs are attributable to the specific 3D hierarchical structure of the IE-MoSe2/C composite anode. This can induce the outstanding rate performance and prolonged cycling stability in PIBs for the IE-MoSe2/C anode. Moreover, Nyquist plots after 50 cycles in PIBs are displayed in Fig. S9. The reduced Rct of the IE-MoSe2/C anode indicates the increased conductivity, which can induce the enhanced reaction kinetics for the IE-MoSe2/C anode in PIBs.
| This journal is © the Partner Organisations 2026 |