Synthesis of a Library of Transition Metal Sulfide@MoS₂ Core@Shell Nanostructures via Post‐Synthetic Cation Exchange

ABSTRACT

The rational synthesis of compositionally diverse and structurally well‐defined inorganic heterostructures remains a substantial challenge. Herein, we report a post‐synthetic cation exchange strategy for the programmable construction of a library of transition metal sulfide@MoS₂ core@shell heterostructures. Using Ag₂S@MoS₂ as a template, we demonstrate a multi‐step transformation pathway via a highly reactive Cu_2‐xS@MoS₂ intermediate. Critically, this approach successfully decouples the formation of the MoS₂ shell from the final core composition, providing a universal route to three distinct classes of architectures—metal sulfide M_xS_y (M═Cd, Cu, Co, Ni, Mn, Zn), heterostructured (Cu_2‐xS/ZnS), and homogeneous solid‐solution (Ni_xCo_yS₄) cores—while perfectly preserving the integrity of the monolayer MoS₂ shell. The potential of these materials is highlighted by their electrocatalytic performance for the oxygen evolution reaction (OER). The Co₉S₈@MoS₂ catalyst exhibits exceptional activity, achieving a low overpotential of 253 mV at 10 mA cm⁻² and remarkable long‐term stability. This work establishes cation exchange within a 2D shell as a powerful and general platform for engineering complex hybrid architectures for advanced energy applications.

A post‐synthetic cation exchange strategy enables the construction of a diverse library of transition metal sulfide@MoS₂ core@shell heterostructures. By utilizing Ag₂S@MoS₂ as a template, the method decouples shell formation from core composition, allowing programmable synthesis of single‐metal sulfide, heterostructured, and solid‐solution cores within a preserved MoS₂ shell.

1. Introduction

The rational design and synthesis of inorganic heterostructures with precise control over composition, morphology, and interfaces remains a central pursuit in materials science [1, 2, 3]. Among various nanoarchitectures, core@shell heterostructures have emerged as a class of powerful platforms for engineering functional materials with synergistic properties that transcend those of their individual components [4]. By integrating distinct chemical phases into an intimately coupled configuration, core@shell systems enable precise modulation of their physicochemical properties [5, 6, 7, 8, 9, 10, 11]. This tunability underpins their broad applications across diverse fields, such as electronics [12], biomedicine [13], pharmaceutics [14], optics [15], and catalysis [16].

Among various shell materials that have been explored, ultrathin 2D materials, especially transition metal dichalcogenides (TMDs), have recently emerged as an exceptional class of shell components due to their atomic thickness, high specific surface area, and unique layer‐dependent electronic properties [17, 18]. When coated over a functional core, the TMD shell can engage in profound electronic interactions with the core, which can modulate its surface electronic structure and optimize the adsorption energy of reaction intermediates, thereby unlocking new physicochemical properties [19, 20]. Despite this immense potential, the synthetic challenge of constructing well‐defined, uniform, and compositionally diverse core@shell structures with a continuous TMD shell remains a significant hurdle, limiting the exploration of this vast materials landscape and its associated physicochemical phenomena [21]. Traditional synthetic routes often struggle to simultaneously achieve morphological uniformity, interfacial coherence, and compositional versatility [22, 23, 24, 25]. Direct growth of a TMD shell on a preformed core is typically hindered by lattice mismatch and interfacial energy disparities, which induce heterogeneous nucleation, or incomplete coverage [26, 27, 28, 29]. Consequently, developing a general and programmable synthetic strategy that decouples the shell formation from the core composition is essential for systematic exploration of these hybrid architectures and their interface‐driven functionalities.

Herein, we report a novel and versatile cation‐exchange‐assisted strategy for the programmable synthesis of a broad class of well‐defined transition metal sulfide@MoS₂ core@shell heterostructures (Scheme 1). Our approach decouples the synthesis of the shell from the formation of the final core, thereby overcoming the interfacial compatibility constraints that limit traditional methods. Owing to the intrinsic favorable mechanical flexibility of the MoS₂ shell [30, 31], the strain accumulated during the crystal phase transformation of the core can be effectively accommodated via defect formation, stretching, and bending, thereby preserving the structural integrity of the core@shell architecture. This approach effectively generates a library of metal sulfide@MoS₂ core@shell nanostructures from a single Ag₂S@MoS₂ parent template. As a proof‐of‐concept application, these M_xS_y@MoS₂ are used as electrocatalysts for the oxygen evolution reaction (OER). Benefitting from the synergistic effects between the transition metal sulfides core and the MoS₂ shell, the obtained catalysts exhibit significantly enhanced OER activity compared with their single‐component counterparts. Specifically, Co₉S₈@MoS₂ demonstrates a particularly exceptional alkaline OER performance, achieving a low overpotential of 253 mV at 10 mA cm⁻² and long‐term durability. The strategy developed in this work not only provides a versatile route for engineering novel core@shell sulfide‐based heterostructures but also offers new insights into the rational design of highly efficient and durable electrocatalysts for water splitting and other applications.

SCHEME 1.

Sequential cation exchange process from Ag₂S@MoS₂ to M_xS_y@MoS₂ (M═Ni, Mn, Zn, Co, and Ni‐Co).

2. Results and Discussion

Our programmable synthesis strategy begins with the fabrication of a uniform Ag₂S@MoS₂ core@shell heterostructure, which was synthesized via a one‐pot wet chemical method (see Experimental Section for details). Transmission electron microscopy (TEM) images and the size distribution analysis demonstrate well‐defined nanoparticles with an average size of 12.6 ± 0.9 nm, in which a monolayer MoS₂ shell was uniformly coated on the surface of Ag₂S core (Figure S1a,b). High‐resolution TEM (HRTEM) image clearly resolves lattice spacings of 0.26 and 0.27 nm, corresponding to the (022) plane of Ag₂S and the (100) plane of MoS₂, respectively (Figure S1c). The phase purity is confirmed by X‐ray diffraction (XRD) pattern, which identifies the monoclinic phase of Ag₂S, while the absence of distinct MoS₂ diffraction peaks is attributed to its low content and ultrathin nature (Figure S2a). However, the presence of the MoS₂ shell is revealed by Raman spectroscopy, which displays the characteristic E¹ _2g and A_1g vibrational modes of the 2H‐phase at 377 and 403 cm⁻¹ [32], respectively (Figure S2b).

The creation of a diverse library of MoS₂‐based core@shell structures hinges on preserving the integrity of the monolayer MoS₂ shell during cation exchange. To test and validate this capability, we first exchanged Ag⁺ in the Ag₂S@MoS₂ template with Cd²⁺ to form CdS@MoS₂. This initial exchange was strategically designed to be mild and efficient: the high mobilities and comparable ionic radii of Ag⁺ and Cd²⁺ enable a low‐strain transformation that minimizes disruption to the delicate MoS₂ shell [33]. Crucially, the core@shell architecture was fully retained after the cation exchange, as evidenced by TEM imaging and particle size analysis (Figure 1a and Figure S3). XRD confirms the complete phase transformation from Ag₂S to wurtzite CdS (Figure 1b). HRTEM image (Figure 1c) resolves interplanar spacings of 0.36 and 0.31 nm, corresponding to the (100) and (101) planes of CdS, respectively, and matching the crystallographic model in Figure 1d. The interfacial chemistry of CdS@MoS₂ was further characterized by X‐ray photoelectron spectroscopy (XPS) (Figure S4). Collectively, these findings demonstrate the complete transformation from Ag₂S@MoS₂ to CdS@MoS₂, which retains the original core@shell architecture and particle size.

FIGURE 1.

Cation exchange reaction of Ag₂S@MoS₂ with Cd²⁺ and CdS@MoS₂ with Cu⁺. TEM images, XRD patterns, HRTEM images, and crystallographic structures of the S²⁻ sublattice of (a–d) CdS@MoS₂ and (e–h) Cu_2‐xS@MoS₂.

Following this validation, we executed a pivotal step in our strategy: the transformation of the CdS core into a Cu_2‐xS intermediate. This step is critical because the high Cu⁺ ion mobility and abundant cation vacancies in Cu_2‐xS create a kinetically favorable intermediate, significantly lowering the activation barrier for subsequent exchanges and enabling library synthesis [34, 35]. The successful formation of phase‐pure roxbyite Cu₇S₄@MoS₂ is confirmed by comprehensive characterization. The material retained the core@shell morphology and a uniform particle size of 12.5 ± 0.7 nm (Figure 1e and Figure S5). XRD analysis clearly identifies the roxbyite Cu₇S₄ phase (Figure 1f), which is further corroborated by HRTEM and the crystallographic model (Figure 1g,h). XPS analysis (Figure S6) further verifies the complete chemical conversion and the preservation of the MoS₂ shell. Notably, HRTEM images of both CdS@MoS₂ and Cu_2‐xS@MoS₂ reveal discernible gaps within the MoS₂ shell (as indicated by red arrows), suggesting a general structural response. These voids originate from strain induced by the crystal phase transformation of the core, which is effectively accommodated through defect formation in the flexible MoS₂ shell, thereby preserving the overall integrity of the core@shell heterostructure.

With this highly active Cu_2‐xS@MoS₂ platform in hand, we demonstrated the full power and versatility of our strategy by performing a divergent set of cation exchange reactions to synthesize a broad library of transition metal sulfide heterostructures. Based on Hard‐Soft Acid‐Base (HSAB) principle [36], a standard experimental procedure was employed using tri‐n‐octylphosphine (TOP) as a soft base to selectively extract the soft Cu⁺ acid, allowing the incoming M²⁺ cations (M═Ni, Mn, Zn, and Co) to fill the vacated sites.

TEM analysis confirms the exceptional robustness of the MoS₂ shell, with the core@shell architecture universally preserved across the entire library of derivatives (Ni₃S₂@MoS₂, MnS@MoS₂, ZnS@MoS₂, Co₉S₈@MoS₂, Figure 2a–d). While the shell remains intact, a discernible reduction in the average particle size is observed for all derivatives (Figure S7). Interestingly, the Ni₃S₂@MoS₂ exhibits a pronounced change in shape, becoming flattened and elongated relative to Cu_2‐xS@MoS₂, whereas the other derivatives largely retain the original shape. While Mn²⁺, Zn²⁺, and Co²⁺ exchanges yielded their thermodynamically stable phases (MnS, ZnS, and Co₉S₈), the Ni²⁺ exchange stabilized the metastable heazlewoodite Ni₃S₂ phase (Figure 2e–h). HRTEM (Figure 2i–l) further verifies the structural integrity following these transformations, displaying continuous monolayer MoS₂ shells encapsulating highly crystalline cores with characteristic lattice spacings of 0.29 nm for Ni₃S₂ (110), 0.35 nm for MnS (100), 0.33 nm for ZnS (100), and 0.29 nm for Co₉S₈ (222), respectively. These crystallographic assignments are in excellent agreement with both XRD data and simulated atomic models (Figure 2m–p).

FIGURE 2.

Cation exchange of Cu_2‐xS@MoS₂ with Ni²⁺, Mn²⁺, Zn²⁺, and Co²⁺. TEM images (a–d), XRD patterns (e–h), HRTEM images (i–l), and crystallographic structures of the S²⁻ sublattice (m–p) of Ni₃S₂@MoS₂, MnS@MoS₂, ZnS@MoS₂, and Co₉S₈@MoS₂, respectively; EDS elemental mapping of Co₉S₈@MoS₂ (q).

Previous studies on cation exchange of Cu_2‐xS with Ni²⁺ typically yielded Ni₉S₈ [37] and Ni₃S₄ [38] phases. Here, we report the transformation from roxbyite Cu_2‐xS to heazlewoodite Ni₃S₂ via cation exchange. We propose that this process is structurally guided by a shared, hexagonally‐derived close‐packed S²⁻ sublattice, which manifests as an ABAB… stacking sequence along different crystallographic directions ([100] in Cu_2‐xS, [001] in Ni₃S₂) [39]. The exchange induced a slight in‐plane expansion but a pronounced contraction perpendicular to the stacking layers (Figure S8). This anisotropic lattice evolution directly accounts for the observed flattening and elongation of the Ni₃S₂@MoS₂. Importantly, the flexible yet strong MoS₂ shell accommodates this significant volume change without rupture. This size‐reduction mechanism, driven by crystal structure transformation, is also observed in other derivatives [40]. The transformations to wurtzite MnS and ZnS are structurally facile, as their hcp S²⁻ sublattices are closely related to the parent distorted hcp framework [41]. Even the conversion to Co₉S₈, which involves a transition to a cubic close‐packed (ccp) sublattice, is accessible by sliding A and B layers in the [010] direction of Cu₇S₄ [38]. Consequently, the coherent particle size reduction observed across these systems results from varying degrees of interlayer contraction and intralayer atomic rearrangement during cation exchange (Figure S9). These structural transformations are expected to generate varying degrees of lattice strain, which is reflected in the morphological evolution observed in TEM. Specifically, Ni₃S₂@MoS₂ and Co₉S₈@MoS₂ exhibit more flattened morphologies. Meanwhile, HRTEM images reveal local stretching of the MoS₂ shell in ZnS@MoS₂ and bending in Co₉S₈@MoS₂. These structural distortions are attributed to strain induced by more extensive anion sublattice reconstruction in the core, which is accommodated by stretching and bending of the MoS₂ shell, thereby preserving the core@shell architecture.

Complementary energy‐dispersive X‐ray spectroscopy (EDS) mapping (Figure 2q and Figure S10) demonstrates elemental homogeneity, with spatially correlated signals of M (Co and Mn), Mo, and S, confirming the MoS₂ encapsulation. Furthermore, XPS analysis (Figures S11–S14) delineates the chemical evolution throughout the multi‐step cation exchange. All systems exhibit minor oxidation of the MoS₂ shell, which is likely attributable to surface oxidation upon air exposure during post‐processing. These systematic XPS findings collectively validate the successful construction of the core@shell architectures and confirm the targeted chemical compositions.

Furthermore, Raman spectroscopy was employed to characterize the Cu_2‐xS@MoS₂ intermediate and the Co₉S₈@MoS₂ product after multi‐step cation exchange, further confirming the structural stability of the MoS₂ shell (Figure S15). Similar to the Ag₂S@MoS₂ template, both Cu_2‐xS@MoS₂ and Co₉S₈@MoS₂ exhibit characteristic Raman peaks located at comparable positions, corresponding to the E¹ _2g and A_1g vibrational modes of 2H‐MoS₂. In addition, slight shifts in both vibrational modes are observed in the two samples compared with the starting template. These minor deviations are likely associated with lattice strain induced by structural reconstruction of the core during cation exchange. Such strain can lead to subtle variations in Mo‐S bond lengths and interlayer spacing, thereby causing small shifts in both the in‐plane (E¹ _2g) and out‐of‐plane (A_1g) modes [42, 43]. Notably, owing to the intrinsic flexibility of MoS₂, the induced strain can be partially relaxed, resulting in only minor peak shifts. Moreover, a more pronounced decrease in the intensity of the A_1g mode is observed, which may be related to a slightly increased degree of surface oxidation, leading to the formation of MoO_x species that partially suppress the intrinsic A_1g vibrational response of MoS₂ [44, 45]. Collectively, the combined results from TEM, XPS, and Raman spectroscopy consistently demonstrate the structural robustness of the MoS₂ shell throughout the multi‐step cation exchange process.

To probe the dynamics of cation exchange and its ability to craft complex architectures, we conducted time‐dependent partial exchange using Zn²⁺. Varying the reaction time yielded a series of Cu_2‐xS/ZnS@MoS₂ intermediates with preserved core@shell structure (Figure 3a). HRTEM analysis reveals the progressive formation of a Janus‐type core, with ZnS incorporation increasing from ∼1/3 (3 min) to ∼1/2 (5 min) (Figure 3b and Figure S16). This asymmetric, unidirectional progression creates an internal ZnS/Cu_2‐xS heterojunction within the core. Elemental mapping shows uniform Mo and S distribution, while spatially resolved EDS identifies distinct Zn‐rich and Cu‐rich domains that correlate with HRTEM phase boundaries (Figure 3c). This result underscores the power of our strategy in synthesizing well‐defined, complex heterostructures.

FIGURE 3.

Synthesis of MoS₂‐based core@shell structures with heterostructured and solid‐solution cores. TEM image (a), HRTEM (b), and EDS mapping (c) of Cu_2‐xS/ZnS@MoS₂; TEM image (d), XRD pattern (e), HRTEM (f), and EDS mapping (g) of Ni_2.5Co_0.5S₄@MoS₂.

To further demonstrate the versatility of our cation exchange strategy, we targeted a homogeneous solid‐solution via dual‐cation (Ni²⁺/Co²⁺) co‐exchange. The resultant Ni_2.5Co_0.5S₄@MoS₂ retained the parental core@shell morphology (Figure 3d,e). HRTEM confirms the single‐crystalline core with lattice fringes of 0.30 nm for the (311) plane. The XRD pattern (Figure 3f) is indexed to cubic Ni_2.5Co_0.5S₄, confirming a homogeneous alloy rather than a phase‐segregated heterostructure. EDS elemental mapping (Figure 3g) reveals uniform spatial distributions of Ni, Co, Mo, and S. The overlapping Ni and Co signals demonstrate the alloyed composition of the core, which contrasts sharply with the Janus‐type structure. The chemical states of the heterostructure were further probed by XPS (Figure S17).

Recent studies have demonstrated transition metal sulfides as efficient OER electrocatalysts [46]. As a proof‐of‐concept application, the OER performance of all as‐synthesized core@shell heterostructures was evaluated in 1.0 m KOH, with bare Co₉S₈ nanoparticles (NPs) (Figure S18, detailed analysis is provided in Supporting Information) and commercial RuO₂ as benchmarks. Notably, a direct comparison between Co₉S₈@MoS₂ and bare Co₉S₈ NPs clearly reveals the significant enhancement induced by the MoS₂ shell. Specifically, linear sweep voltammetry (LSV, Figure 4a) reveals that Co₉S₈@MoS₂ requires a much lower overpotential of 253 mV to reach 10 mA cm⁻², compared to 306 mV for Co₉S₈ NPs. Meanwhile, the Tafel slope (Figure 4b) decreases from 81 mV dec⁻¹ for Co₉S₈ NPs to 66 mV dec⁻¹ for Co₉S₈@MoS₂, indicating accelerated reaction kinetics. In addition, electrochemical impedance spectroscopy (EIS, Figure 4c) shows a substantially reduced charge‐transfer resistance (R_ct) of 26 Ω for Co₉S₈@MoS₂, in sharp contrast to 102 Ω for Co₉S₈ NPs, suggesting more efficient interfacial charge‐transfer processes. Collectively, these results suggest that the MoS₂ shell is not merely a passive coating layer; instead, it likely participates in the OER process by modulating the local surface reaction environment and facilitating interfacial charge‐transfer kinetics through coupling with the Co₉S₈ core, as evidenced by the reduced overpotential, smaller Tafel slope, and significantly decreased charge‐transfer resistance. In addition, the improved structural uniformity of the Co₉S₈@MoS₂ compared to bare Co₉S₈ NPs, as revealed by TEM, may contribute to more consistent catalytic interfaces and enhanced utilization of active sites. Nevertheless, the exact electronic structure modulation at the Co₉S₈/MoS₂ interface requires further investigation by DFT calculations or operando spectroscopic techniques.

FIGURE 4.

OER performance of MoS₂‐based core@shell heterostructures. LSV curves for the OER of commercial RuO₂, Co₉S₈ NPs and synthesized core@shell heterostructures (a); Tafel slopes derived from the LSV curves (b); EIS Nyquist plots of different catalysts at 10 mA cm⁻² (c); C_dl obtained from CV graphs (d); LSV curves before (gray) and after (blue) 5000 CV cycles (e); Long‐term stability test for Co₉S₈@MoS₂ at j = 10 mA cm⁻² (f).

Additionally, Co₉S₈@MoS₂ outperforms commercial RuO₂ (270 mV), Ni_2.5Co_0.5S₄@MoS₂ (336 mV), Ni₃S₂@MoS₂ (363 mV), and MnS@MoS₂ (negligible activity) at 10 mA cm⁻². Such a low overpotential is among the best values reported for Co‐based catalysts under alkaline conditions (Table S1, Supporting Information). The superior kinetics of Co₉S₈@MoS₂ are also evidenced by its smaller Tafel slope compared to Ni_2.5Co_0.5S₄@MoS₂ (87 mV dec⁻¹), Ni₃S₂@MoS₂ (109 mV dec⁻¹), and RuO₂ (150 mV dec⁻¹). Consistently, it exhibits the lowest charge‐transfer resistance among all samples, compared with Ni_2.5Co_0.5S₄@MoS₂ (57 Ω) and Ni₃S₂@MoS₂ (282 Ω), following the same activity trend. The electrochemically active surface area (ECSA), estimated from the double‐layer capacitance (C_dl) obtained from CV measurements, is highest for Co₉S₈@MoS₂ (27.43 mF cm⁻²), compared to Ni_2.5Co_0.5S₄@MoS₂ (9.52 mF cm⁻²) and Ni₃S₂@MoS₂ (2.02 mF cm⁻²) (Figure 4d and Figure S19). Stability assessments confirm the exceptional durability of the optimal catalyst. Co₉S₈@MoS₂ exhibits a minimal overpotential increase of only 10 mV at 100 mA cm⁻² after 5000 CV cycles (Figure 4e) and maintains stable operation for over 100 h in chronoamperometric tests at both 10 and 100 mA cm⁻² (Figure 4f and Figure S20).

It has been reported that the OER activity of Co‐based catalysts originates from in situ‐formed oxides/hydroxides on their surface during catalysis [47, 48, 49]. To elucidate the catalytic mechanism of Co₉S₈@MoS₂, the catalyst after the 100 h OER stability test was investigated by TEM and XPS. HRTEM reveals lattice fringes of 0.21 and 0.22 nm, which are assignable to the (121) plane of CoOOH and the (420) plane of Co₉S₈, respectively (Figure S21). This observation suggests that CoOOH likely acted as the active phase during OER. XPS analysis further supports this transformation. The Co 2p spectrum exhibits a positive shift and line shape characteristic of CoOOH (Figure S22a) [50]. Concurrently, the O 1s spectrum is deconvoluted into components assignable to Co‐O (529.8 eV), ‐OH (531.3 eV), adsorbed water (532.6 eV), and the Nafion binder (534.2 eV) (Figure S22b) [51, 52, 53]. Moreover, the Mo and S spectra (Figure S22c,d) show exclusively oxidized states, confirming comprehensive surface oxidation. These results collectively identify Co₉S₈@MoS₂ as a highly efficient pre‐catalyst, whose performance is likely attributable to the in situ generation of CoOOH, together with the dynamic structural evolution of the Mo‐containing shell into oxidized species under OER conditions. Although no crystalline MoS₂ shell is observed in TEM after long‐term OER operation, XPS results confirm that Mo species remain on the catalyst surface in oxidized states. This suggests that the Mo component is not lost but transformed into amorphous or highly dispersed MoO_x species, which may still participate in the catalytic process by modulating the local chemical environment and electronic structure of the active CoOOH phase [49, 54].

3. Conclusion

In summary, we have developed a general cation‐exchange strategy to construct a diverse library of transition metal sulfide@MoS₂ core@shell heterostructures. The versatility of this approach enables the precise synthesis of single‐metal sulfide, Janus‐type heterojunction, and solid‐solution cores, all within a preserved monolayer MoS₂ shell. The integrity of this MoS₂ shell is retained throughout multi‐step exchanges, ensuring well‐defined interfaces. As an electrocatalyst for the oxygen evolution reaction, Co₉S₈@MoS₂ exhibits exceptional activity, favorable kinetics, and outstanding stability. This work demonstrates cation exchange as a powerful tool for engineering complex heterostructures as a versatile platform for designing metastable and multi‐component nanomaterials.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: smll73984‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.