Engineering edge-rich, highly dispersed, and stable MoS₂ sites for the catalytic splitting of H₂S for H₂ production

Engineering edge-rich, highly dispersed, and stable MoS₂ sites for the catalytic splitting of H₂S for H₂ production | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Engineering edge-rich, highly dispersed, and stable MoS₂ sites for the catalytic splitting of H₂S for H₂ production Pedro Castaño, Hend Mohamed, Vijay Velisoju, Mohammed Obaid, Rafia Ahmed, and 14 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8597696/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Hydrogen sulfide (H 2 S) decomposition offers a carbon-neutral route for hydrogen production, but it remains limited by sluggish kinetics and catalyst deactivation. Here, we report an electrospinning–sulfidation strategy to engineer a confined molybdenum disulfide (MoS 2 ) catalyst with stable edge-rich active sites. During electrospinning, confined Mo–S 4 ²⁻ complexes are dispersed within the nanofibers, which then nucleate into MoS 2 during calcination. The sulfidation stage induces exfoliation, forming highly dispersed MoS₂ domains that interact strongly with the carbon nanofibers. The prepared catalyst has a high surface area (~ 180 m 2 g⁻ 1 ), abundant sulfur-vacancy edge sites, and strong support interactions, stabilized by coordinated Mo atoms. In-situ spectroscopy and ab-initio calculations reveal that these interfaces facilitate H 2 S dissociation, leading to a 3-fold higher intrinsic rate and improved long-term stability (> 50 h at 973 K) compared to bulk analogues. This work establishes design principles for fabricating grafted, stable, and highly dispersed sulfide catalysts. Physical sciences/Chemistry/Catalysis/Heterogeneous catalysis Physical sciences/Engineering/Chemical engineering MoS2 H2S decomposition electrospinning hydrogen nanofiber Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 INTRODUCTION Catalyst design for industrial hydrogen production requires precise control over the structure of the active sites, mechanistic insights derived under realistic conditions, and scalability without performance loss. 1 , 2 These requirements are particularly important for the catalytic splitting of hydrogen sulfide (H 2 S), 3,4 which is a promising alternative to the energy-intensive Claus process. 4 , 5 Despite its simplicity, this endothermic reaction remains kinetically sluggish and prone to deactivation due to the high sulfur partial pressure and temperature required. Molybdenum disulfide (MoS 2 ) is the most widely studied catalyst for the splitting of H 2 S due to its intrinsic tolerance to sulfur and its versatile chemistry. 6 – 9 However, the basal planes of layered MoS 2 prepared using the standard hydrothermal method are inaccessible, 5,10 and the edge sites responsible for S–H bond activation are difficult to engineer, prone to sintering, and can be deactivated through vacancy “healing” under standard reaction conditions. 4 Overcoming these limitations requires the use of other supports and synthesis routes. 11 – 14 For example, carbon supports can enhance dispersion and stabilize sulfides, 15–18 while electrospinning can be used to more efficiently graft sulfide sites. 19 In particular, the electrospinning–reduction–sulfidation route can be employed to transform MoS 2 into a hierarchical dual-structure catalyst consisting of (i) extended monolayer MoS 2 nanosheets conformally anchored to carbon nanofibers (CNFs) and (ii) atomically deposited MoS₂ stabilized by the CNFs. Molecularly dispersing the Mo–S 4 2- precursor in polyacrylonitrile (PAN) during electrospinning leads to sulfide grafting and limits nanoparticle growth. Subsequent sulfidation induces exfoliation, preventing restacking and seeding atomic-limit MoS 2 on the CNFs. In this work, we compare an electrospun catalyst (MoS 2 |CNF-ES) with a hydrothermal bulk analogue (MoS 2 –HT), a hybrid electrospinning–hydrothermal material (MoS 2 –HT/CNF), and commercial bulk MoS 2 (MoS 2 –Comm). Combined spectroscopic, microscopic, and kinetic analyses demonstrate that electrospinning enabled more precise MoS 2 site engineering, enabling stable operation in highly concentrated H 2 S environments. This work establishes a method for the accurate engineering of transition-metal sulfides with a high dispersion and defect density for high-temperature H 2 S waste valorization and the production of C-free hydrogen. RESULTS Synthesis and structural evolution of MoS 2 catalysts : To elucidate how synthesis dynamics regulate active-site accessibility and catalytic stability, we prepared a series of MoS 2 catalysts, including the proposed electrospun catalyst (MoS 2 |CNF-ES), a hydrothermal analogue (MoS 2 –HT), a hybrid electrospinning–hydrothermal material (MoS 2 –HT/CNF), and a commercial reference (MoS 2 –Comm) (Methods and Fig. 1 ). In the electrospinning route, ammonium tetrathiomolybdate (ATTM) and PAN were dissolved in N,N-dimethylformamide (DMF) and subjected to electrohydrodynamic stretching under a high electric field, which immobilized the Mo–S 4 2 ⁻ complexes along the fiber axis. This restricted precursor mobility and suppressed plane growth and stacking during the subsequent transformation (Fig. 1 a). Under reductive carbonization at 723 K, ATTM decomposed within the CNF scaffold to form few-layered MoS 2 clusters (Fig. 1 a). Subsequent sulfidation in H 2 S/H 2 /N 2 at 973 K induced exfoliation, yielding two types of MoS 2 sites: (i) monolayer MoS 2 slabs conformally anchored to the CNFs and (ii) atomically dispersed MoS 2 stabilized within the support (Fig. 1 a). This synthesis route exploited growth control and support interaction to produce a monolayer, edge-enriched MoS 2 catalyst with unique features (see Supplementary Fig. S1 a for more details). The electrospinning parameters, including the polymer concentration, were systematically optimized to balance the CNF integrity, precursor dispersion, and final MoS 2 loading. PAN concentrations of 5–15 wt/vol% were evaluated, with 10 wt/vol% PAN yielding the most uniform fibers (Figs. S1–S3). In contrast, hydrothermal nucleation (MoS 2 –HT; Fig. 1 c) produced densely stacked 2H-MoS 2 lamellae (Fig. S4), while the physically mixed MoS 2 –HT/CNF (Fig. 1 b) exhibited intermediate behavior, with few-layered MoS 2 structures on the CNF surface. Together with MoS 2 –Comm, these materials constituted a structural continuum ranging from a bulk multilayer material to atomically thin, edge-rich MoS 2 (Fig. 1 and Table S1 ). Powder X-ray diffraction (XRD) analysis of MoS 2 |CNF–ES revealed only in-plane (100) and (110) reflections of 2H-MoS₂ at 2θ = 32.3° and 57.0°, respectively, whereas the characteristic (002) basal-plane reflection at 14.5° was almost completely suppressed (Fig. 1 ), confirming the absence of long-range stacking and the dominance of monolayer or highly exfoliated MoS 2 . 20 This structural signature likely originated from the electrospinning environment, in which electrohydrodynamic stretching and rapid solvent evaporation immobilized Mo–S₄²⁻ complexes within the polymer jet. The resulting limited stacking dispersed the precursor at the nanoscale level, avoiding large flakes. During sequential reduction and sulfidation, ATTM decomposed, combined with gas-evolution-driven expansion and sulfur-mediated layer delamination, further suppressing plane growth and stacking, yielding exfoliated MoS 2 sheets firmly anchored to the CNFs. In contrast, both MoS 2 –HT and MoS 2 –Comm. exhibited intense (002) reflections, while MoS₂–HT/CNF contained a (002) peak of moderate intensity, which was consistent with heterogeneous nucleation, controlled growth, and limited stacking (Fig. 1 d and Figs S1 –S4). These findings were in accordance with the Raman spectroscopy analysis (Fig. 1 e and Figs. S3 and S4). In the commercial and bulk MoS 2 samples, pronounced E 2 g¹ (~ 380 cm⁻¹) and A 1 g (~ 405 cm⁻¹) modes were observed. However, for the CNF-supported catalyst (MoS 2 |CNF–ES), these MoS 2 phonons were not detectable because the strong D- and G-bands of the PAN-derived CNFs dominated the spectra. The carbon features confirmed the partial graphitization and the formation of CNFs. In contrast, the sharper, more symmetric Raman modes in MoS 2 –HT reflected its highly ordered, defect-poor multilayer stacking, consistent with the prominent (002) reflection observed in the XRD results. 4 , 20 Corresponding nitrogen adsorption–desorption isotherms (Fig. S5a) showed that MoS 2 |CNF–ES had a specific surface area of ≈ 180 m 2 g⁻ 1 , nearly an order of magnitude higher than that of MoS 2 –HT and MoS 2 –HT/CNF (≈ 40 m 2 g⁻ 1 ), highlighting the effectiveness of the reduction–sulfidation sequence in generating hierarchical porosity and dispersing the active phases. The full results of the thermal analysis, including the BET surface area and TGA profiles, are provided in Figs. S5a,b and Table S1 . In addition, the morphology of the MoS 2 |CNF-ES, MoS 2 –HT, and MoS 2 –HT/CNF catalysts was investigated using scanning electron microscopy (SEM; Figs. S6 and S7) and high-resolution transmission electron microscopy (HR-TEM; Figs. S8–S13). The SEM imaging revealed the continuous coverage of monolayer MoS 2 on the CNFs following sulfidation, whereas the hydrothermal samples retained a stacked morphology, and the hybrid material had an incomplete conformal coating. In the synthesis of the MoS 2 |CNF–ES sample, the as-spun ATTM–PAN fibers formed uniform precursor filaments, which evolved into rigid CNFs decorated with MoS 2 when heated at 723 K under a H 2 atmosphere (Figs. S6a and S11a). Subsequent sulfidation in H 2 S/H 2 /N 2 at 973 K induced pore formation and exfoliation (Figs. S6b and S11b), producing a highly porous CNF network uniformly coated with a monolayer of MoS 2 . In contrast, MoS 2 –HT had a multilayer, densely stacked morphology (Fig. S7a), while the hybrid MoS 2 –HT/CNF sample (Fig. S7b) contained heterogeneously distributed few-layered MoS 2 platelets that did not fully conform to the carbon fibers. These differences indicated that the electrospinning strategy enabled optimized control, resulting in highly dispersed MoS₂ within the catalyst. HR-TEM, high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM), and energy-dispersive X-ray (EDX) mapping (Fig. 2 ; Figs. S8–S13) were used to investigate the nanoscale architecture of MoS 2 |CNF–ES. HR-TEM and HAADF-STEM analysis confirmed the presence of atomically thin MoS 2 nanosheets uniformly distributed across the CNFs, with no evidence of restacking or aggregation. In line with this, EDX mapping showed homogeneous dispersion of Mo and S, with no aggregated nanoparticles. In contrast, MoS₂–HT contained well-stacked multilayer lamellae with 0.62 nm (002) fringes, which was reflective of equilibrium-controlled growth and minimal defect density. MoS 2 –HT/CNF exhibited few-layered MoS 2 clusters that assembled vertically and partially aggregated. The electron energy loss spectroscopy (EELS) spectra for MoS 2 |CNF–ES (Fig. S12) revealed broader S L₂,₃ and Mo M₄,₅ edges, indicating the presence of defects in the sulfide structure, with under-coordinated Mo/S atoms. These features were significantly weaker in the hydrothermal and commercial catalysts. X-ray photoelectron spectroscopy (XPS) analysis of Mo 3d spectra for all samples revealed dominant Mo 4+ species (229.1 and 231.9 eV) characteristic of MoS 2 , 21 with minor Mo 6+ contributions (233.6 and 235.3 eV) from surface oxidation (MoO 3 ) and/or edge sites (Figs. 3 a and S14). 4 , 22 The S 2p spectrum confirmed the presence of S 2– species, while MoS 2 |CNF–ES exhibited an additional peak attributable to bridging disulfide (S 2 2– ) ligands at the edges, consistent with an edge-rich structure that was also observed using in-situ diffuse-reflectance Fourier-transform infrared spectroscopy (DRIFTS) analysis (Fig. 3 b). 23 , 24 The shift in the binding energy for Mo 3d and S 2p observed for MoS 2 |CNF–ES and MoS 2 –HT/CNF was indicative of strong sulfide–support interactions. Quantitative nitric oxide (NO) chemisorption analysis confirmed that electrospinning enhanced both metal dispersion and defect accessibility (Table S1 ). MoS 2 |CNF–ES had an adsorption rate of 132 µmol NO g cat −1 , roughly six times that of MoS 2 –HT, while its O 2 uptake was also considerably higher, indicating a higher concentration of active sites associated with vacancies. 11 , 12 In-situ O 2 DRIFTS spectra (Fig. 3 b) contained intense 960–900 cm − 1 bands associated with terminal Mo = O groups at the edges for MoS 2 |CNF–ES, whereas MoS 2 –HT and MoS 2 –HT/CNF contained additional 860–700 cm − 1 features corresponding to in-plane Mo–O–Mo linkages on the basal plane. 12 The absence of these basal-plane modes in MoS 2 |CNF–ES indicated that oxidation and vacancy formation were confined to the edges rather than distributed over the lattice. According to time-resolved DRIFTS analysis, these Mo = O bands remained stable with long-term O 2 exposure (Fig. 3 c). X-ray absorption near-edge structure (XANES; Fig. 3 d) analysis of the MoS 2 -based catalysts revealed similar edge positions for MoO 3 and Mo 2 C, consistent with mixed + 4 and minor + 6 oxidation states and in accordance with the XPS results. Fourier transform–extended X-ray absorption fine structure (FT–EXAFS) (Fig. 3 e) analysis revealed Mo–S (1.8–2.5 Å) and Mo–Mo (2.6–3.2 Å) coordination analogous to metallic Mo foil. In addition, quantitative EXAFS fitting (Table S2) suggested that MoS 2 |CNF–ES had the lowest CN Mo–Mo /CN Mo–S ratio (0.70), indicating that it contained the highest fraction of coordinatively unsaturated Mo atoms. This reduced coordination likely originated from its monolayer structure and sulfide–support interactions, which stabilized the edge sites against sulfur replenishment. Temperature-programmed in-situ XANES (Fig. 3 f) analysis under a H 2 atmosphere revealed a progressive shift at the edges to a lower energy, confirming the H 2 -induced reduction of Mo species and concomitant sulfur removal. These edge defects served as dual-function centers, providing adsorption sites for H 2 S and facilitating S–H bond activation, leading to the superior intrinsic activity and long-term stability observed during catalytic H 2 S decomposition. 11 , 25 Catalytic evaluation and structure–performance relationships Evaluation of the catalysts under H 2 S decomposition conditions (973 K, 1 bar) revealed that the synthesis pathway and interfacial structure had a strong influence on intrinsic activity (Fig. 4 ). Space–time adjustments were employed to account for variation in the bulk density, and equilibrium conversions were validated using Aspen Plus simulations (Figs. 4 a, Figs. S3 and S4). MoS 2 |CNF–ES achieved near-equilibrium H 2 S conversion at a lower contact time, significantly outperforming both its hydrothermal (MoS 2 –HT and MoS 2 –HT/CNF) and commercial (MoS 2 –Comm) counterparts (Figs. 4 a,b). Intrinsic kinetic parameters were obtained from a kinetic model that accounted for equilibrium limits (Figs. 4 c–e and Tables S3–S5), yielding a rate constant of 1.9 × 10 − 3 mol H 2 S atm⁻¹ g cat⁻¹ s⁻¹ for MoS 2 |CNF–ES, which was 9-fold higher than MoS₂–Comm and 3-fold greater than MoS 2 –HT and MoS 2 –HT/CNF. The corresponding apparent activation energy (Eₐ ≈ 63 kJ mol⁻¹) was also markedly lower than that of the layered references, indicating facilitated H–S bond activation at the edge sites (Fig. 4 d). Overall, MoS 2 |CNF–ES had the highest activity, followed by MoS 2 –HT, MoS 2 –HT/CNF, and MoS 2 –Comm, and it also compared favorably to previously reported catalysts (Table S6). The faster kinetics of MoS 2 |CNF–ES likely reflected the highly dispersed nature of the active sites, with a high edge density, and their interactions with the support. The intrinsic rate of H 2 S decomposition also scaled linearly with NO chemisorption (Fig. 4 f), establishing a quantitative link between catalytic performance and the density of coordinatively unsaturated Mo atoms. Because NO selectively binds to exposed Mo edge sites and sulfur vacancies, this correlation confirmed that the activity of MoS 2 |CNF–ES was governed by accessible edge defects rather than the total Mo content. The similar proportionality between the rate constants and defect density derived from EXAFS coordination analysis and O 2 chemisorption (Table S1 ) confirmed that the concentration of sulfur vacancies was the dominant kinetic descriptor. The lower activation barrier of MoS 2 |CNF–ES can be explained by its polarization at the sulfide–support interface, as evidenced by the XPS and XANES results (Figs. 4 a,d–f). This interaction with the support weakens the Mo–S bonds at the edges, enhancing S–H bond dissociation and facilitating H 2 evolution. This interaction enables MoS 2 to maintain a reaction turnover without deactivation. MoS 2 |CNF–ES exhibited exceptional long-term stability, maintaining the steady-state conversion of H 2 S feeds of varying concentrations (2–15 vol%) over 8 h and 50 h of continuous operation (Figs. 4 g,h). The reaction rate fully recovered when reverting to the initial feed conditions, demonstrating the reversibility of sulfur accumulation. Post-reaction analysis confirmed the preservation of the structure and oxidation state of the proposed catalyst. XRD and XPS analysis (Figs. S14 and 4i) revealed unchanged 2H reflections and Mo 4 ⁺/Mo 6 ⁺ ratios, while corresponding HR-TEM images contained intact monolayer sheets without stacking or sintering. The carbon framework remained structurally robust, underscoring that the sulfide–support interaction suppressed both sulfur condensation and Mo migration, the principal deactivation pathways in conventional MoS 2 catalysts. Density Functional Theory calculations : To evaluate how the graphene support influences H 2 S activation, three basal-plane models were constructed: (i) a freestanding MoS 2 monolayer (ML–MoS 2 ), (ii) a freestanding bilayer (BL–MoS 2 ) representing a bulk-like configuration, and (iii) a graphene-supported MoS₂ monolayer (ML–MoS 2 /graphene) simulating the experimentally observed CNF-supported catalytic domains (Fig. 5 a). For each of these models, single sulfur vacancy (1-Vs) was generated by removing a top-layer S atom, while a second adjacent vacancy (2-Vs) was introduced to model multi-vacancy environments known previously as more catalytically active sites for S–H scission. 20 , 26 The corresponding optimized structures for the pristine, 1-Vs, and 2-Vs versions of the models are presented in Figs. S17–22. The H 2 S adsorption and dissociation energetics for all three systems are summarized in Figs. 5 a,b. In the 1-Vs models, H 2 S bound weakly to BL–MoS 2 (− 0.02 eV) and slightly stronger to ML–MoS 2 (− 0.20 eV) and ML–MoS 2 /graphene (− 0.24 eV), indicating that both reduced dimensionality and the graphene interface enhance initial H 2 S binding. The first S–H bond cleavage to form HS* + H* was mildly endothermic for all three 1-Vs systems (0.19–0.38 eV) together with the second cleavage leading to H* + H *+ S* formation (0.31–0.43 eV, Fig. 5 b), though this was lowest for ML–MoS 2 /graphene. In contrast, all of the dissociation steps became exergonic at the 2-Vs sites. Across all models, the adsorption (− 0.32 to − 0.47 eV) and stepwise dissociation (− 0.34 to − 0.15 eV) of H 2 S were thermodynamically favorable, demonstrating that multi-vacancy ensembles stabilize sulfur-containing intermediates and enable spontaneous S–H cleavage. Graphene also promoted hydrogen-assisted vacancy formation (Fig. 5 c). For 1-Vs, \(\:{\Delta\:}{\text{E}}_{\text{v}\text{a}\text{c}}\) decreased from + 0.01 eV (ML–MoS 2 ) and + 0.17 eV (BL–MoS 2 ) to − 0.03 eV for ML–MoS 2 /graphene, confirming the stabilizing influence of the conductive support. Similarly, 2-Vs formation was most favorable for ML–MoS 2 /graphene (− 0.62 eV), indicating that graphene promotes sulfur extraction under reductive conditions. Including finite-temperature corrections (973 K) from NIST thermochemical data reduced \(\:{\Delta\:}{\text{G}}_{\text{v}\text{a}\text{c}}\) by ~ 0.8 eV per sulfur atom, meaning that vacancy formation was thermodynamically feasible under the experimental activation conditions (Table S7). Overall, these results demonstrate that graphene not only enhances H 2 S binding but also lowers the energetic penalty for vacancy formation and sulfur dissociation. The ML–MoS 2 /graphene system had the lowest energy barriers among the three models, providing a basis for accelerated H 2 S decomposition at supported basal planes, consistent with our experimental findings. DISCUSSION Collectively, the operando, kinetic, and theoretical analyses produced a coherent mechanistic overview of H 2 S decomposition on MoS 2 -based catalysts. Within the electrospun MoS 2 |CNF–ES catalyst, MoS₂ is grafted onto CNFs and interacts strongly with the support, resulting in active, stable S–H scission sites. As indicated by pulse chemisorption analysis, the electrospun catalyst has a far higher density of accessible Mo edge sites than either the hydrothermal or the bulk analogues. This enrichment likely arises from the electrospinning-induced grafting of MoS 2 precursors within polymer fibers, yielding uniformly dispersed monolayer MoS 2 domains following sulfidation. The resulting CNF architecture prevents sintering while introducing abundant defect sites. The sulfide-support interactions polarize the Mo–S bonds and weaken the S–H activation barrier, a phenomenon directly observed using in-situ spectroscopy. O 2 -DRIFTS and XANES–EXAFS analyses suggest that sulfur vacancies form and heal dynamically under alternating H₂S/H₂ environments. The downshift of the Mo K -edge and the reduction in Mo–S and Mo–Mo coordination are indicative of the formation of undercoordinated MoS₂ centers. XPS corroborates this by identifying mixed Mo 4+ /Mo 6+ species stabilized by the support interface. These support interactions allow the edge vacancies to persist and remain catalytically active under conditions where unsupported MoS₂ rapidly deactivates. DFT calculations provided atomistic support for this mechanism. The graphene-supported MoS 2 monolayer, which is an analogue of the CNF-grafted catalyst, has the lowest energetic cost for hydrogen-assisted sulfur removal (ΔE Svac = − 0.03 eV for 1-V_S and − 0.62 eV for 2-V_S) and exergonic H 2 S dissociation for all intermediates. The carbon substrate stabilizes these vacancies through support interactions, reducing the S-vacancy formation barrier and promoting continuous reactive turnover. According to finite-temperature corrections, vacancies form at 973 K, which was in line with the experimental observations of sustained activity and vacancy regeneration. The CNFs interact with the MoS 2 and stabilize the active sites. The faster kinetics of the MoS 2 |CNF–ES catalyst arise from a high density of edge and vacancy sites and support interactions that stabilize these highly dispersed active sites. CONCLUSIONS The edge-enriched MoS 2 |CNF catalyst developed in the present study delivered a 3-fold increase in H 2 S decomposition activity and remarkable stability over 50 h of operation compared to bulk and hydrothermally synthesized MoS 2 , while maintaining long-term stability under harsh conditions. This exceptional performance was due to the unique combination of (i) highly dispersed monolayer MoS 2 domains anchored on CNFs, (ii) strong sulfide–support interactions that suppress sintering and stabilize undercoordinated edge atoms, and (3) a dense population of sulfur vacancies. In situ spectroscopy and DFT calculations jointly confirmed that the stabilized edge vacancies serve as active sites for H 2 S dissociation and hydrogen evolution. The proposed electrospinning–sulfidation route establishes a material-design paradigm for engineering metastable, edge-rich sulfides through confinement and stabilization. The grafting of sulfides onto CNFs and the resulting defect stabilization offer a transferable blueprint for the synthesis of robust catalysts for other high-temperature, sulfur-tolerant hydrogen production processes. METHODS Catalyst synthesis Three MoS 2 -based model catalysts were prepared using hydrothermal and electrospinning methods reported in previous studies, with slight modifications and different pre-treatment protocols. 4 , 27 , 28 Synthesis of MoS 2 -HT In the typical hydrothermal synthesis route, 0.8 mmol of ammonium heptamolybdate tetrahydrate (AMT; (NH 4 ) 6 Mo 7 O 24 . 4H 2 O; Sigma-Aldrich; 99.98%) was dissolved in 30 mL of deionized (DI) water at room temperature and stirred for 30 min before the addition of citric acid (CA) or oxalic acid (OA); (0.05 g; Sigma-Aldrich; 99.5%) as a chemical agent under vigorous magnetic stirring for 20 min. The sulfide source (10 mmol of thiourea; CH 4 N 2 S; Sigma-Aldrich; 99.99%) was dissolved in another 30 mL of DI water at room temperature under vigorous magnetic stirring. The molybdenum solution was then added slowly to the sulfur solution, and the resulting mixture was transferred to a 100 mL Teflon-lined stainless-steel autoclave and heated at 473 K. After cooling to room temperature, the black product was filtered and washed several times with distilled water and absolute ethanol. The final product was dried at 333 K overnight. As a comparison, a commercial MoS 2 catalyst (98%, Alfa Aesar) was purchased from Thermo Scientific Chemicals and used as received for testing. Synthesis of MoS 2 |CNF–ES MoS 2 |CNF–ES was prepared using a modified electrospinning method. 28 , 29 PAN (1.0 g) and ATTM ((NH 4 ) 2 MoS 4 ; 0.5 g) were dissolved in dimethylformamide to obtain PAN concentrations of 5, 10, or 15 wt/vol%. The precursor solutions were stirred for 24 h at 60°C until fully homogenized, and the resulting solutions were subsequently loaded into a syringe equipped with a 23-gauge stainless-steel needle. Electrospinning was conducted at a voltage of 15 kV with a flow rate of 1 mL h - 1 and a tip-to-collector distance of 15 cm. The as-spun composite mats were collected on grounded aluminum foil and dried under a vacuum at 60°C. The obtained ATTM–PAN nanofibers were pre-treated at 723 K for 5 h in a H 2 (5 vol%)/N 2 (95 vol%) atmosphere, followed by further treatment at 973 K in 2 vol% H 2 S/N 2 for 2 h and 50 vol% H 2 /N 2 for 1 h with heating rates of 5°C min - 1 for each treatment stage to obtain a porous MoS 2 /CNF structure. As a comparison, after the first-stage heat treatment at 723 K for 5 h under H 2 (5 vol%)/N 2 (95 vol%), the resulting material was subjected to heat treatment at 973 K for 5 h under a N 2 atmosphere (Fig. S2). Synthesis MoS 2 –HT/CNF CNFs were synthesized via electrospinning, followed by thermal treatment and the hydrothermal deposition of MoS 2 . Initially, 1.0 g of PAN was dissolved in 7.0 g of DMF (Sigma-Aldrich, 99.8%) under continuous stirring for 24 h at room temperature. The resulting homogeneous solution was subjected to electrospinning to prepare the mats using the previously described process. The collected fibrous mats were pre-treated at 723 K for 5 h under a H 2 (5 vol%)/N 2 (95 vol%) atmosphere, followed by treatment at 973 K in 2 vol% H 2 S/N 2 for 2 h and 50 vol% H 2 /N 2 for 1 h with heating rates of 5 K min - 1 for each treatment stage to obtain CNFs without MoS 2 . A hydrothermal method was employed for MoS 2 deposition, similar to the synthesis process for MoS 2 -HT and a previously reported method. 30 Ammonium molybdate (15 wt% Mo) was dissolved in 30 mL of DI water and stirred for 30 min, followed by the addition of 0.05 g of OA under vigorous stirring for another 20 min. In parallel, 10 mmol of thiourea was dissolved in 30 mL of DI water. The molybdate solution was slowly added to the thiourea solution under stirring. The mixture was then transferred to a 100 mL Teflon-lined stainless-steel autoclave, and the required quantity of the CNF mat was added. The mixture was heated at 493 K for 18 h. After cooling, the product was filtered, washed with distilled water and ethanol, and dried overnight at 333 K. Catalyst characterization Crystallographic analysis was conducted via powder XRD in a Bruker D8 advanced diffractometer with a Bragg–Brentano geometry fitted with a Cu tube operating at 40 kV and 40 mA. Diffractograms were acquired over a 2θ range of 10–80° with a step size of 0.1° and a scan speed of 0.5 s per step. PDF-4+ (2019) was the database used for phase identification. To investigate the textural properties of the samples, N 2 adsorption–desorption isotherms were obtained at 77 K using an ASAP 2040 instrument (Micromeritics). Before the measurements were taken, the samples were degassed for 10 h at 523 K. The specific surface area (S BET ) was determined from the isotherms using the Brunauer–Emmett–Teller (BET) equation. The surface oxidation states of Mo and S were determined using XPS analysis in a K-ALPHA spectrometer (Thermo Scientific) with an Al-Kα (1486.6 eV) radiation source at room temperature under an ultra-high vacuum at 3 mA × 12 kV. The powder samples were pressed, mounted on the holder, and placed in the vacuum chamber. The C 1s peak at 284.5 eV was used to calibrate and correct the binding energy for the element, and charge compensation was achieved using a system flood gun, providing low-energy electrons. In addition, Raman spectra for the MoS 2 catalysts were obtained using a Renishaw Via Reflex confocal spectrometer with 532 nm laser excitation. The laser power was set to 5–10 mW, and the sample was irradiated for 10 s with six accumulations. The catalyst morphology was investigated using HR-TEM with a Titan Themis-Z microscope (Thermo Fisher Scientific) at an accelerating voltage of 300 kV and a beam current of 0.5 nA. Dark-field (DF) imaging employed an STEM and an HAADF detector. The STEM-HAADF data were collected at a convergence angle of 29.9 mrad and an HAADF inner angle of 30 mrad. Furthermore, an X-ray energy dispersive spectrometer (EDS) was utilized in conjunction with DF-STEM imaging to acquire STEM-EDS spectrum-imaging datasets and a post-column EELS instrument (GIF-Quantum 966; Gatan, Inc.). NO chemisorption measurements were taken using a Micromeritics AutoChem II 2920 instrument. 11 , 12 In a typical fabrication route, 0.1 g of the catalyst sample was pretreated in situ under a flow of H 2 at 573 K for 3 h, followed by purging with He at the same temperature for 60 min. Subsequently, NO chemisorption was conducted at 313 K by pulsing 2% NO in He through the catalyst every 30 min. The NO signal was monitored via mass spectrometry and observed to increase until reaching a steady state, indicating saturation of the catalyst surface. The total NO uptake was then determined from the cumulative amount adsorbed. O 2 pulse chemisorption experiments were performed with in-situ DRIFTS to quantify the active sites in the catalysts. 11 , 12 O 2 pulse titration was conducted using a Micromeritics Autochem II 2930. Typically, ~ 0.1 g of the catalyst was pretreated by H 2 at 573 K for 2 h and purged with He at 573 K for another 1 h. O 2 chemisorption was conducted at 313 K by pulsing 1 mL of 5% O 2 /He over the catalyst every 15 min until peak saturation. Effluent O 2 was detected using a mass spectrometer until its level reached a constant value, representing O 2 saturation of the catalyst, after which the total uptake of O 2 was calculated. In-situ DRIFTS measurements were taken on a Nicolet 6700 FTIR spectrometer with a liquid-nitrogen cooled MCT detector using a high-temperature reaction cell (Harrick) equipped with a temperature programmer and connected to a gas-dosing device with mass flow controllers (Bronkhorst). The catalyst was first reduced by pure H 2 at 573 K with a gas flow of 30 ml min − 1 at 1 bar for 1 h and then cooled with He to 313 K for another 1 h. Spectra obtained at 313 K under a H 2 atmosphere were used as background spectra. Subsequently, the catalyst samples were treated with 5% O 2 /He at a rate of 30 mL min − 1, followed by spectra collection throughout in-situ DRIFTS. Mo K -edge X-ray absorption spectra were recorded at room temperature using a Si (311) monochromator in the BL01B1 line of Spring-8 at the Japan Synchrotron Radiation Research Institute (JASRI) in Harima, Hyogo, Japan. The obtained spectra were analyzed and fitted using Athena and Artemis, which were equipped with ATOMS and FEFF6 in the software package Demeter 0.9.26. 31–33 After normalization at the edge height, the k 3 -weighted χ spectra were extracted. Subsequently, these spectra in the k range of 3–14 Å −1 were Fourier-transformed into the R -space. Curve-fitting analysis was conducted in an R range of 1.8–3.4 Å for Mo foil and 1.4–3.3 Å for all MoS 2 catalysts using the back-scattering amplitude and phase shift functions of Mo–Mo and Mo–S. Catalyst testing Catalytic H 2 S splitting experiments were conducted using a PID testing unit with a fixed-bed quartz reactor (8 mm ID, length 340 mm) in down-flow mode (Fig. S23). The reactor was placed in a furnace, and the flow was controlled using a mass flow controller. In the reactor, ~ 0.03 g of sieved catalyst particles (100–150 µm) were loaded onto a 0.15 g SiC (particle grit 40) bed to ensure the catalyst bed rested in the isothermal zone of the reactor. The catalyst was pre-treated in situ first with a mixture of H 2 S and H 2 (50:50) at 973 K for 1 h, followed by testing with a mixed feed containing 2 vol.% of H 2 S at 1 bar in the temperature range of 773–973 K. A flow of 1 mL min – 1 of He was added to the mixed feed as an internal standard. Reactor outputs were then analyzed using a Varian micro gas chromatograph with TCD detectors. The two-channel configuration and columns of this analytical system were used to quantify H 2 , He, H 2 S, and N 2 . The conversion (X) is expressed as follows: $$\:{X}_{i}\left(\%\right)=\frac{{F}_{i,0}-{F}_{i}}{{F}_{i,0}}\times\:100$$ 1 where \(\:{F}_{i,0}\) is the molar flow rate at the inlet of the reactor of species i (e.g., H 2 S), and \(\:{F}_{i}\:\:\) is the outlet molar flow rate of species i . Simulations, kinetic model, rate constant, and turnover frequency calculations The thermodynamic equilibrium conversion of H 2 S under each tested condition was simulated using Aspen Plus with an R-Gibbs reactor and the WILSON base method. A kinetic model was used to describe the intrinsic reaction rate by integrating the rate expression over the catalyst weight. The rate constant (k) and activation energy ( Eₐ ) were determined by minimizing the difference between the experimentally observed and theoretically predicted rates, ensuring an accurate fit to the data. The turnover frequency (TOF) was calculated based on the initial reaction rate per active site. While the standard TOF expression assumes differential operation, we employed an alternative formulation suitable for near-equilibrium conditions, based on the initial rate at low conversion. Further details of model derivation, the fitting process, and TOF equations are provided in the Supporting Information. Computational details All spin-polarized density functional theory (DFT) calculations were performed using the Vienna Ab initio Simulation Package (VASP 5.4.4). 34 The projector-augmented wave (PAW) method was used to describe the interaction between core and valence electrons, 35 and generalized gradient approximation (GGA) in the Perdew–Burke–Ernzerhof (PBE) form was employed to account for exchange-correlation effects. 36 Grimme’s DFT-D3 method with Becke–Johnson damping was included in all calculations to capture dispersion interactions. 37 The plane-wave kinetic energy cutoff was set to 500 eV, and the electronic energy convergence criterion was set to 10 –5 eV. Structural relaxation was performed until the forces on all atoms were below 0.02 eV/Å. A vacuum layer of 15 Å was introduced along the z -direction to avoid spurious interactions between periodic images. The Brillouin zone was sampled using a Γ-centered Monkhorst–Pack k-point mesh of 3×3×1 for the (7×7) supercell models. 38 Monolayer, bilayer, and graphene-supported monolayer MoS 2 systems were constructed to evaluate the effect of support and vacancy concentration. Sulfur vacancies (1-S and 2-S) were created by removing one or two surface S atoms, and the thermodynamic stability of various dissociated intermediates (H 2 S*, HS* + H*, and H* + H* + S*) was evaluated. Adsorption energies were calculated relative to isolated gas-phase species and referenced to the pristine surface. The sulfur vacancy formation energy (ΔEs vac ) was determined in the absence and presence of hydrogen co-adsorption to capture the thermodynamic impact of reductive conditions. The filling energy of vacancies with sulfur atoms (i.e., S-vacancy healing) was assessed to understand the post-dissociation stability of active sites. The adsorption energy ( \(\:{E}_{ads}\) ) of a molecule or intermediate on the MoS₂ surface was calculated as follows: $$\:{E}_{ads}={E}_{slab+adsorbate}-{E}_{slab}-{E}_{adsorbate\left(g\right)}$$ 2 where \(\:{E}_{slab+adsorbate}\) is the total energy of the adsorbed system, \(\:{E}_{slab}\) is the energy of the clean surface, and \(\:{E}_{adsorbate\left(g\right)}\) is the total energy of the isolated gas-phase molecule optimized in a large periodic box. The sulfur-vacancy formation energy ( \(\:{\Delta\:}{E}_{Svac}\) ) was computed as follows: $$\:{\Delta\:}{E}_{Svac}-{E}_{vacant}-{E}_{pristine}+n{\mu\:}_{S}$$ 3 where \(\:{E}_{vacant}\) and \(\:{E}_{pristine}\) are the total energy of the defective and pristine MoS₂ slabs, respectively, n is the number of sulfur atoms removed, and \(\:{\mu\:}_{S}\) is the sulfur chemical potential derived from gas-phase S 2 : $$\:{\mu\:}_{S}=\frac{1}{2}{E}_{{S}_{2}\left(g\right)}$$ 4 Under reductive conditions, the hydrogen-assisted sulfur vacancy formation energy ( \(\:{\Delta\:}{E}_{Svac}^{{H}_{2}}\) ) was determined according to the following equation: $$\:{\Delta\:}{E}_{Svac}^{{H}_{2}}={E}_{vacant}-{E}_{pristine}+n\left({E}_{{H}_{2}S\left(g\right)}-{E}_{{H}_{2}\left(g\right)}\right)$$ 5 which represents the following process: MoS₂ + n H₂ (g) → MoS₂₋ₙ + n H₂S (g) (6) Eq. ( 5 ) therefore accounts for the removal of sulfur atoms from the surface through their conversion to H₂S gas in the presence of H₂. To obtain temperature-corrected Gibbs free energies, the gas-phase DFT energies in Eq. ( 5 ) were replaced by temperature-dependent chemical potentials from the NIST-JANAF thermochemical tables: $$\:{\Delta\:}{G}_{Svac}^{{H}_{2}\left(T,P\right)}=\left({E}_{vacant}-{E}_{pristine}\right)+n\left[{\mu\:}_{{H}_{2}S\:}\left(T,{p}_{{H}_{2}S}\right)-{\mu\:}_{{H}_{2}}\left(T,{p}_{{H}_{2}}\right)\right]$$ 7 where the temperature-dependent chemical potential of each gas is defined as shown in Eq. ( 8 ): $$\:{\mu\:}_{i}\left(T,{p}_{i}\right)={\mu\:}_{i}^{^\circ\:}\left(T\right)+{k}_{B}T\text{ln}\left(\frac{{p}_{i}}{{p}^{^\circ\:}}\right)$$ 8 and \(\:{\mu\:}_{i}^{^\circ\:}\left(T\right)={H}_{i}^{^\circ\:}\left(T\right)-T{S}_{i}^{^\circ\:}\left(T\right)\) = H_i°(T) − TS_i°(T) was computed from NIST Shomate equations. For example, at 700°C (973 K) and 800°C (1073 K), \(\:{\mu\:}_{{H}_{2}S}^{^\circ\:}\left(T\right)-{\mu\:}_{{H}_{2}}^{^\circ\:}\left(T\right)\) equals − 0.80 eV and − 0.89 eV, respectively. DATA AVAILABILITY The data supporting the findings of this article are available in the paper and the Supplementary Information. Additional data is available from the corresponding author on request. Source data are provided with this paper. Declarations AUTHOR INFORMATION Authors and Affiliations Multiscale Reaction Engineering, KAUST Catalysis Platform (KCP), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955 − 6900, Saudi Arabia Hend Mohamed, Vijay K. Velisoju, Muhammad Arief Shafarifky, Gontzal Lezcano, and Pedro Castaño Water Desalination and Reuse Platform (WDRP), Division of Biological and Environmental Science and Engineering (BESE), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955 − 6900, Saudi Arabia M. Obaid and Noreddine Ghaffour KAUST Catalysis Platform, King Abdullah University of Science and Technology (KAUST), Thuwal, 23955 − 6900, Saudi Arabia Rafia Ahmed, Ildar Mukhambetov, Huabin Zhang, Polina Tolstova, and Luigi Cavallo Department of Materials Engineering Science, Graduate School of Engineering Science, The University of Osaka, 1–3 Machikaneyama, Toyonaka, Osaka 560–8531, Japan Yuma Hirayama and Takato Mitsudome KAUST Core Labs, King Abdullah University of Science and Technology (KAUST), Thuwal, 23955 − 6900, Saudi Arabia Mohamed Ben Hassine and Omar El Tall Research and Development Center, Dhahran 31311, Saudi Aramco Zainab Alaithan, Ali Almofleh, and Hassan Aljama Chemical Engineering Program, Physical Science and Engineering (PSE) Division , King Abdullah University of Science and Technology (KAUST), Thuwal, Saudi Arabia Pedro Castaño Contributions H.M., V.K.V., H.A., and P.C. led the project and designed the experiments. H.M., M.O., and N.G. designed the materials and synthesis protocols using electrospinning. V.K.V. and H.M. performed and analyzed the basic characterization and contributed to the optimization of the thermal treatment of the catalysts. V.K.V. performed and analyzed the catalytic testing, kinetic analysis, and advanced characterization. G.L. and V.K.V. contributed to kinetic model development. R.A., L.C., Z.A., P.T., and H.A. contributed to the theoretical calculations. V.K.V., H.M., and I.M. contributed to reactor operation, testing protocols, and analytical methodologies. M.B.H., O.E.T., and H.Z., contributed to material characterization, results interpretation, and manuscript reviewing. Y.H., M.A.S., and T.M. have conducted and analyzed the XAFS measurement. V.K.V., H.M., and P.C. wrote the manuscript with input from all authors. P.C. and H.A. were responsible for funding acquisition and project supervision. Corresponding author Correspondence to Pedro Castaño. ETHICS DECLARATIONS The authors declare no competing interests. ACKNOWLEDGMENTS All authors thank Saudi Aramco and King Abdullah University of Science and Technology (KAUST) for funding this work (RGC/3/5222–01 and BAS/1/1403) and for the resources in the supercomputing laboratory at KAUST. Y.H and T.M. received JSPS KAKENHI grant 23H01761 and JST PRESTO grant JPMJPR21Q9 for sources and funding of the synchrotron experiments (2025A1813). References Bligaard T et al (2016) Toward Benchmarking in Catalysis Science: Best Practices, Challenges, and Opportunities. ACS Catal 6:2590–2602 Vogt C, Weckhuysen BM (2022) The concept of active site in heterogeneous catalysis. Nat Rev Chem 6:89–111 Reverberi A, Pietro, Klemeš JJ, Varbanov PS, Fabiano B (2016) A review on hydrogen production from hydrogen sulphide by chemical and photochemical methods. J Clean Prod 136:72–80 Almofleh A et al (2025) Designing sulfide catalysts for H 2 S dissociation to H 2 based on reaction descriptors and microkinetics. Appl Catal B Environ Energy 361:124605 Aljama H, Alaithan Z, Almofleh A (2023) Catalytic conversion of H 2 S to H 2 : Challenges and catalyst limitations. J Phys Chem C 127:9022–9029 Kraia T, Kaklidis N, Konsolakis M, Marnellos GE (2019) Hydrogen production by H 2 S decomposition over ceria supported transition metal (Co, Ni, Fe and Cu) catalysts. Int J Hydrogen Energy 44:9753–9762 Deng D et al (2016) Catalysis with two-dimensional materials and their heterostructures. Nat. Nanotechnol. 11, 218–230 (2016) Mao J, Wang Y, Zheng Z, Deng D (2018) The rise of two-dimensional MoS 2 for catalysis. Front Phys 13:1–19 Norskov JK, Christensen CH (2006) Toward efficient hydrogen production at surfaces. Sci (80-) 312:1322–1323 Voiry D et al (2013) Enhanced catalytic activity in strained chemically exfoliated WS 2 nanosheets for hydrogen evolution. Nat. Mater. 12, 850–855 (2013) Hu J et al (2021) Sulfur vacancy-rich MoS 2 as a catalyst for the hydrogenation of CO 2 to methanol. Nat Catal 2021 43(4):242–250 Mao J et al (2023) Direct conversion of methane with O 2 at room temperature over edge-rich MoS 2 . Nat Catal 2023 611(6):1052–1061 Hu J et al (2023) Edge-rich molybdenum disulfide tailors carbon-chain growth for selective hydrogenation of carbon monoxide to higher alcohols. Nat. Commun. 14, 1–11 (2023) Zhou S et al (2023) Strained few-layer MoS 2 with atomic copper and selectively exposed in-plane sulfur vacancies for CO 2 hydrogenation to methanol. Nat. Commun. 14, 1–15 (2023) Bojarska Z, Mazurkiewicz-Pawlicka M, Mierzwa B, Plocinski T, Makowski L (2022) Effect of the carbon support on MoS 2 hybrid nanostructures prepared by an impinging jet reactor for hydrogen evolution reaction catalysis. J Environ Chem Eng 10:108038 Obaid M et al (2020) Breakthroughs in the fabrication of electrospun-nanofiber-supported thin film composite/nanocomposite membranes for the forward osmosis process: A review. Crit Rev Environ Sci Technol 50:1727–1795 Pinilla JL, Purón H, Torres D, Suelves I, Millan M (2015) Ni-MoS 2 supported on carbon nanofibers as hydrogenation catalysts: Effect of support functionalisation. Carbon N Y 81:574–586 Cui J et al (2024) Carbon-anchoring synthesis of Pt1Ni1@Pt/C core-shell catalysts for stable oxygen reduction reaction. Nat. Commun. 15, 1–9 (2024) Jagadale A, Zhou X, Blaisdell D, Yang S (2018) Carbon nanofibers (CNFs) supported cobalt- nickel sulfide (CoNi 2 S 4 ) nanoparticles hybrid anode for high performance lithium ion capacitor. Sci Rep 8:1602 Zhou S et al (2024) Highly active single-layer 2H-MoS 2 for CO 2 hydrogenation to methanol. Appl Catal B Environ Energy 349:123870 Xu J et al (2023) Solution combustion synthesis of CoS x , MoS x , CoS x -MoS x and their catalytic activity in H 2 production from H 2 S decomposition. Int J Hydrogen Energy 48:8807–8818 Hussain S et al (2016) Large-area, continuous and high electrical performances of bilayer to few layers MoS 2 fabricated by RF sputtering via post-deposition annealing method. Sci Rep 2016 61(6):1–13 Xu Q et al (2019) Unsaturated Sulfur Edge Engineering of Strongly Coupled MoS 2 Nanosheet–Carbon Macroporous Hybrid Catalyst for Enhanced Hydrogen Generation. Adv Energy Mater 9:1802553 Zheng Z et al (2020) Boosting hydrogen evolution on MoS 2 via co-confining selenium in surface and cobalt in inner layer. Nat. Commun. 11, 1–10 (2020) Ozaki F et al (2023) Hydrogen-induced sulfur vacancies on the MoS 2 basal plane studied by ambient pressure XPS and DFT calculations. ChemPhysChem 24, e202300477 Kibsgaard J, Chen Z, Reinecke BN, Jaramillo TF (2012) Engineering the surface structure of MoS 2 to preferentially expose active edge sites for electrocatalysis. Nat Mater 11:963–969 Zheng X, Zhu Y, Sun Y, Jiao Q (2018) Hydrothermal synthesis of MoS 2 with different morphology and its performance in thermal battery. J Power Sources 395:318–327 Liu S et al (2012) Preparation of MoS 2 nanofibers by electrospinning. Mater Lett 73:223–225 Duan Y et al (2022) Embedding of ultra-dispersed MoS2 nanosheets in N,O heteroatom-modified carbon nanofibers for improved adsorption of Hg2+. Compos Commun 31:101106 Guan J et al (2022) Electrospinning derivative fabrication of sandwich-structured CNF/Co 3 S 4 /MoS 2 as self-supported electrodes to accelerate electron transport in HER. Int J Hydrogen Energy 47:14930–14941 Newville M (2001) IFEFFIT : interactive XAFS analysis and FEFF fitting. urn:issn:0909 – 0495 8:322–324 Ravel B (2001) ATOMS: crystallography for the X-ray absorption spectroscopist. urn:issn:0909 – 0495 8:314–316 Ravel B, Newville MATHENA (2005) ARTEMIS, HEPHAESTUS: data analysis for X-ray absorption spectroscopy using IFEFFIT. urn:issn:0909 – 0495 12:537–541 Kresse G, Furthmüller J (1996) Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys Rev B 54:11169 Blöchl PE (1994) Projector augmented-wave method. Phys Rev B 50:17953 Perdew JP, Burke K, Ernzerhof M (1996) Generalized Gradient Approximation Made Simple. Phys Rev Lett 77:3865 Grimme S, Antony J, Ehrlich S, Krieg H (2010) A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J Chem Phys 132:154104 Monkhorst HJ, Pack JD (1976) Special points for Brillouin-zone integrations. Phys Rev B 13:5188 Additional Declarations There is NO Competing Interest. Supplementary Files 3SuppInfo.docx Supplementary Information Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8597696","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":584372207,"identity":"d39ae9ef-1fdb-448a-bf23-8c8f1bf789f3","order_by":0,"name":"Pedro Castaño","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABFUlEQVRIiWNgGAWjYHACAyR2RQKUwUaMFrCiMyAtzKRoYWwjQos5e/O2Dx8Y7OTl5zc/e/BxXpq8wfnzBxg+lB1mkG8/gFWLZc+x4pkzGJINNxxjMzecuS3HcMONZAbGGecOMxicScCqxeBGjjEzD8MBxg1sDGbSvNsqGDfcYGZg5m0DamHAoeX+G7AW+/lt7N+keedU2G84f5iB+S9Qi3z/Axy28IC1JDYc4wHa0pCTuOFAMgMzI1ALww3stlj2pBUzzjBITt5wLKdMcsaxtOSZN5INDvacS+cxuIHdFnP2w5sZPlTY2c5vPr5N4kNNsm3f+YMPH/wos5aT78fhFyQSAQ4AMQ9W9VgUj4JRMApGwSjABADQI10lIwE1ngAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-6454-9321","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":true,"prefix":"","firstName":"Pedro","middleName":"","lastName":"Castaño","suffix":""},{"id":584372208,"identity":"0622b6e7-ceb7-4a02-9184-c95a412db5af","order_by":1,"name":"Hend Mohamed","email":"","orcid":"","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":false,"prefix":"","firstName":"Hend","middleName":"","lastName":"Mohamed","suffix":""},{"id":584372209,"identity":"5996eb82-e656-4e96-b791-a0377b5629aa","order_by":2,"name":"Vijay Velisoju","email":"","orcid":"https://orcid.org/0000-0003-1182-8332","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":false,"prefix":"","firstName":"Vijay","middleName":"","lastName":"Velisoju","suffix":""},{"id":584372210,"identity":"511e4ed8-9e13-4c65-a077-09f72054e772","order_by":3,"name":"Mohammed Obaid","email":"","orcid":"","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":false,"prefix":"","firstName":"Mohammed","middleName":"","lastName":"Obaid","suffix":""},{"id":584372211,"identity":"9c984a83-6489-4c2d-98cc-ac8210e5186f","order_by":4,"name":"Rafia Ahmed","email":"","orcid":"","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":false,"prefix":"","firstName":"Rafia","middleName":"","lastName":"Ahmed","suffix":""},{"id":584372212,"identity":"94f75dbd-a89f-499e-ae2f-de060e0820f4","order_by":5,"name":"Yuma Hirayama","email":"","orcid":"","institution":"University of Osaka","correspondingAuthor":false,"prefix":"","firstName":"Yuma","middleName":"","lastName":"Hirayama","suffix":""},{"id":584372213,"identity":"c81b91d1-40f8-49d8-8ac7-73bc5f440514","order_by":6,"name":"Mohamed Hassine","email":"","orcid":"","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":false,"prefix":"","firstName":"Mohamed","middleName":"","lastName":"Hassine","suffix":""},{"id":584372214,"identity":"37c5c8ed-7805-4ddd-ac03-ecba7198cad8","order_by":7,"name":"Gontzal Lezcano","email":"","orcid":"","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":false,"prefix":"","firstName":"Gontzal","middleName":"","lastName":"Lezcano","suffix":""},{"id":584372215,"identity":"301e6aa8-831c-4035-a41e-d8484c6e12dc","order_by":8,"name":"Ildar Mukhambetov","email":"","orcid":"","institution":"KAUST","correspondingAuthor":false,"prefix":"","firstName":"Ildar","middleName":"","lastName":"Mukhambetov","suffix":""},{"id":584372216,"identity":"8221d9e4-899e-41ed-85f1-f36b3e93bdfa","order_by":9,"name":"Zainab Alaithan","email":"","orcid":"","institution":"Saudi Aramco","correspondingAuthor":false,"prefix":"","firstName":"Zainab","middleName":"","lastName":"Alaithan","suffix":""},{"id":584372217,"identity":"fcee879a-01cf-4642-88e0-80fdea1e0636","order_by":10,"name":"Muhammad Shafarifky","email":"","orcid":"","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":false,"prefix":"","firstName":"Muhammad","middleName":"","lastName":"Shafarifky","suffix":""},{"id":584372218,"identity":"d22d8a32-3fb3-45a2-9674-aa34b931344a","order_by":11,"name":"Ali Almofleh","email":"","orcid":"","institution":"Saudi Aramco","correspondingAuthor":false,"prefix":"","firstName":"Ali","middleName":"","lastName":"Almofleh","suffix":""},{"id":584372219,"identity":"2bf53cff-559e-478f-9b13-8a9ef94a0c09","order_by":12,"name":"Polina Tolstova","email":"","orcid":"https://orcid.org/0000-0002-3646-6211","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":false,"prefix":"","firstName":"Polina","middleName":"","lastName":"Tolstova","suffix":""},{"id":584372220,"identity":"2836b85b-679c-4ac9-9ef7-ee3da13cb81d","order_by":13,"name":"Omar El Tall","email":"","orcid":"","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":false,"prefix":"","firstName":"Omar","middleName":"El","lastName":"Tall","suffix":""},{"id":584372221,"identity":"b86c5441-fe51-4cfc-b4da-a3249f6396df","order_by":14,"name":"Takato Mitsudome","email":"","orcid":"https://orcid.org/0000-0003-3071-8243","institution":"Osaka University","correspondingAuthor":false,"prefix":"","firstName":"Takato","middleName":"","lastName":"Mitsudome","suffix":""},{"id":584372222,"identity":"0eb497ec-06ae-45fb-9bd2-1f736a5bb013","order_by":15,"name":"Huabin Zhang","email":"","orcid":"https://orcid.org/0000-0003-1601-2471","institution":"King Abdullah University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Huabin","middleName":"","lastName":"Zhang","suffix":""},{"id":584372223,"identity":"bba3ea1c-cd53-4e41-9239-628be9fa87d2","order_by":16,"name":"Luigi Cavallo","email":"","orcid":"https://orcid.org/0000-0002-1398-338X","institution":"King Abdullah University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Luigi","middleName":"","lastName":"Cavallo","suffix":""},{"id":584372224,"identity":"38881455-e22c-470c-a538-a3f1fba63ef4","order_by":17,"name":"Noreddine Ghaffour","email":"","orcid":"","institution":"King Abdullah University of Science and Technology (KAUST)","correspondingAuthor":false,"prefix":"","firstName":"Noreddine","middleName":"","lastName":"Ghaffour","suffix":""},{"id":584372225,"identity":"29969273-817a-4153-8edb-c387bd57515f","order_by":18,"name":"Hassan Aljama","email":"","orcid":"","institution":"Saudi Aramco","correspondingAuthor":false,"prefix":"","firstName":"Hassan","middleName":"","lastName":"Aljama","suffix":""}],"badges":[],"createdAt":"2026-01-14 05:30:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8597696/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8597696/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103913883,"identity":"72b24dee-db2b-4cbb-ac8a-e424b464786c","added_by":"auto","created_at":"2026-03-04 12:48:22","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":161892,"visible":true,"origin":"","legend":"\u003cp\u003eSchematic diagrams of catalyst selection and synthesis for bulk and supported MoS\u003csub\u003e2\u003c/sub\u003e catalysts using (a) electrospinning (ES), (b) hybrid, and (c) hydrothermal (HT) methods. (d) Powder X-ray diffraction (XRD) patterns and (e) Raman spectra for the catalysts after thermal treatment.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8597696/v1/bf13ede91e8a0ea378aa8865.jpg"},{"id":103913880,"identity":"656e6f6c-4b14-4e9a-b2f7-6ff4467ab639","added_by":"auto","created_at":"2026-03-04 12:48:22","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":5713,"visible":true,"origin":"","legend":"\u003cp\u003eHR-TEM and HAADF-STEM-EDX analysis of and (a) MoS\u003csub\u003e2\u003c/sub\u003e|CNF-ES, (b) MoS\u003csub\u003e2\u003c/sub\u003e-HT/CNF, and (c) MoS\u003csub\u003e2\u003c/sub\u003e–HT catalysts prepared using electrospinning, a hybrid route, and a hydrothermal method, respectively.\u003c/p\u003e","description":"","filename":"placeholderimage.png","url":"https://assets-eu.researchsquare.com/files/rs-8597696/v1/55eaef761682f61a88be05be.png"},{"id":104401299,"identity":"031633cd-b206-4986-a5ba-dc873d630b88","added_by":"auto","created_at":"2026-03-11 12:12:19","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":162570,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Mo 3d spectra and (b) in-situ DRIFTS spectra for bulk and supported MoS\u003csub\u003e2\u003c/sub\u003e catalysts after oxygen adsorption at 313 K. (c) Time-resolved in-situ DRIFT spectra for MoS\u003csub\u003e2\u003c/sub\u003e|CNF–ES after oxygen adsorption at 313 K. (d) XANES and (e) XAFS spectra for all catalysts. (f) In-situ XANES spectra for MoS\u003csub\u003e2\u003c/sub\u003e|CNF–ES under a flow of H\u003csub\u003e2\u003c/sub\u003e.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8597696/v1/10ccfc055aab909430dca915.jpg"},{"id":104401763,"identity":"7b0ab4c1-4bca-4df2-821c-c5ecd951e4a8","added_by":"auto","created_at":"2026-03-11 12:13:29","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":146776,"visible":true,"origin":"","legend":"\u003cp\u003eActivity of MoS\u003csub\u003e2\u003c/sub\u003e|CNF–ES with different (a) PAN concentrations in DMF at 973 K and (b) reaction temperatures. Reaction conditions: Pre-treatment: 773–973 K in H\u003csub\u003e2\u003c/sub\u003eS/H\u003csub\u003e2\u003c/sub\u003e for 2 h; H\u003csub\u003e2\u003c/sub\u003eS feed: 2 vol% H\u003csub\u003e2\u003c/sub\u003eS in N\u003csub\u003e2\u003c/sub\u003e; reaction temperature: 973 K; and space time: 1.5 g\u003csub\u003ecat\u003c/sub\u003e h mol\u003csub\u003eH2S\u003c/sub\u003e\u003csup\u003e-1\u003c/sup\u003e. (c) Parity plots, (d) activation energy, and (e) rate constants obtained from the kinetic model using the test data presented in Table S2. (f) Correlation between the H\u003csub\u003e2\u003c/sub\u003eS decomposition rate and NO adsorption capacity obtained from NO chemisorption experiments. (g) H\u003csub\u003e2\u003c/sub\u003eS conversion using the MoS\u003csub\u003e2\u003c/sub\u003e|CNF–ES catalyst with a time on stream of 8 h and different concentrations of H\u003csub\u003e2\u003c/sub\u003eS. (h) Time-on-stream analysis for 50 h for MoS\u003csub\u003e2\u003c/sub\u003e|CNF–ES. Reaction conditions: Pre-treatment: 973 K in H\u003csub\u003e2\u003c/sub\u003eS/H\u003csub\u003e2\u003c/sub\u003e for 2 h; H\u003csub\u003e2\u003c/sub\u003eS feed: 2 vol% H\u003csub\u003e2\u003c/sub\u003eS in N\u003csub\u003e2\u003c/sub\u003e; reaction temperature: 973 K; and space time: 1.8 g\u003csub\u003ecat\u003c/sub\u003e h mol\u003csub\u003eH2S\u003c/sub\u003e\u003csup\u003e-1\u003c/sup\u003e.(i) Mo 3d XPS spectra for MoS\u003csub\u003e2\u003c/sub\u003e|CNF–ES before and after stability testing (50 h).\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8597696/v1/016242e11dc337ed5e305100.jpg"},{"id":104779299,"identity":"9dd62f20-f817-4b75-8eb4-eb9ae4d17725","added_by":"auto","created_at":"2026-03-17 07:38:30","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":78145,"visible":true,"origin":"","legend":"\u003cp\u003eH\u003csub\u003e2\u003c/sub\u003eS decomposition reaction profiles for monolayer, bilayer MoS\u003csub\u003e2\u003c/sub\u003e, and graphene-supported monolayer MoS\u003csub\u003e2 \u003c/sub\u003emodel catalysts with both one and two sulfur vacancies: (a) catalyst models simulated using DFT calculations and reaction profiles for (b) one sulfur vacancy, and (c) two sulfur vacancies. The energies are reported at 0 K.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8597696/v1/0c0c36fdc8ba0b827211d002.jpg"},{"id":104785479,"identity":"5bb39974-6259-4752-b741-5b33b8ef0b64","added_by":"auto","created_at":"2026-03-17 08:11:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1874823,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8597696/v1/7b3cb21c-c0df-47ed-b853-88b042244ecf.pdf"},{"id":103913884,"identity":"ece34b14-778f-42d7-a19c-ff01bca1b081","added_by":"auto","created_at":"2026-03-04 12:48:24","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":80008591,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"3SuppInfo.docx","url":"https://assets-eu.researchsquare.com/files/rs-8597696/v1/8159707ade739cf662774b5f.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Engineering edge-rich, highly dispersed, and stable MoS₂ sites for the catalytic splitting of H₂S for H₂ production","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eCatalyst design for industrial hydrogen production requires precise control over the structure of the active sites, mechanistic insights derived under realistic conditions, and scalability without performance loss.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e These requirements are particularly important for the catalytic splitting of hydrogen sulfide (H\u003csub\u003e2\u003c/sub\u003eS),\u003csup\u003e3,4\u003c/sup\u003e which is a promising alternative to the energy-intensive Claus process.\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e Despite its simplicity, this endothermic reaction remains kinetically sluggish and prone to deactivation due to the high sulfur partial pressure and temperature required. Molybdenum disulfide (MoS\u003csub\u003e2\u003c/sub\u003e) is the most widely studied catalyst for the splitting of H\u003csub\u003e2\u003c/sub\u003eS due to its intrinsic tolerance to sulfur and its versatile chemistry.\u003csup\u003e\u003cspan additionalcitationids=\"CR7 CR8\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e However, the basal planes of layered MoS\u003csub\u003e2\u003c/sub\u003e prepared using the standard hydrothermal method are inaccessible,\u003csup\u003e5,10\u003c/sup\u003e and the edge sites responsible for S\u0026ndash;H bond activation are difficult to engineer, prone to sintering, and can be deactivated through vacancy \u0026ldquo;healing\u0026rdquo; under standard reaction conditions.\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eOvercoming these limitations requires the use of other supports and synthesis routes.\u003csup\u003e\u003cspan additionalcitationids=\"CR12 CR13\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e For example, carbon supports can enhance dispersion and stabilize sulfides,\u003csup\u003e15\u0026ndash;18\u003c/sup\u003e while electrospinning can be used to more efficiently graft sulfide sites.\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e In particular, the electrospinning\u0026ndash;reduction\u0026ndash;sulfidation route can be employed to transform MoS\u003csub\u003e2\u003c/sub\u003e into a hierarchical dual-structure catalyst consisting of (i) extended monolayer MoS\u003csub\u003e2\u003c/sub\u003e nanosheets conformally anchored to carbon nanofibers (CNFs) and (ii) atomically deposited MoS₂ stabilized by the CNFs. Molecularly dispersing the Mo\u0026ndash;S\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e2-\u003c/sup\u003e precursor in polyacrylonitrile (PAN) during electrospinning leads to sulfide grafting and limits nanoparticle growth. Subsequent sulfidation induces exfoliation, preventing restacking and seeding atomic-limit MoS\u003csub\u003e2\u003c/sub\u003e on the CNFs.\u003c/p\u003e \u003cp\u003eIn this work, we compare an electrospun catalyst (MoS\u003csub\u003e2\u003c/sub\u003e|CNF-ES) with a hydrothermal bulk analogue (MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT), a hybrid electrospinning\u0026ndash;hydrothermal material (MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF), and commercial bulk MoS\u003csub\u003e2\u003c/sub\u003e (MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;Comm). Combined spectroscopic, microscopic, and kinetic analyses demonstrate that electrospinning enabled more precise MoS\u003csub\u003e2\u003c/sub\u003e site engineering, enabling stable operation in highly concentrated H\u003csub\u003e2\u003c/sub\u003eS environments. This work establishes a method for the accurate engineering of transition-metal sulfides with a high dispersion and defect density for high-temperature H\u003csub\u003e2\u003c/sub\u003eS waste valorization and the production of C-free hydrogen.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e \u003cb\u003eSynthesis and structural evolution of MoS\u003c/b\u003e \u003csub\u003e \u003cb\u003e2\u003c/b\u003e \u003c/sub\u003e \u003cb\u003ecatalysts\u003c/b\u003e: To elucidate how synthesis dynamics regulate active-site accessibility and catalytic stability, we prepared a series of MoS\u003csub\u003e2\u003c/sub\u003e catalysts, including the proposed electrospun catalyst (MoS\u003csub\u003e2\u003c/sub\u003e|CNF-ES), a hydrothermal analogue (MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT), a hybrid electrospinning\u0026ndash;hydrothermal material (MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF), and a commercial reference (MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;Comm) (Methods and Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). In the electrospinning route, ammonium tetrathiomolybdate (ATTM) and PAN were dissolved in N,N-dimethylformamide (DMF) and subjected to electrohydrodynamic stretching under a high electric field, which immobilized the Mo\u0026ndash;S\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e2\u003c/sup\u003e⁻ complexes along the fiber axis. This restricted precursor mobility and suppressed plane growth and stacking during the subsequent transformation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). Under reductive carbonization at 723 K, ATTM decomposed within the CNF scaffold to form few-layered MoS\u003csub\u003e2\u003c/sub\u003e clusters (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). Subsequent sulfidation in H\u003csub\u003e2\u003c/sub\u003eS/H\u003csub\u003e2\u003c/sub\u003e/N\u003csub\u003e2\u003c/sub\u003e at 973 K induced exfoliation, yielding two types of MoS\u003csub\u003e2\u003c/sub\u003e sites: (i) monolayer MoS\u003csub\u003e2\u003c/sub\u003e slabs conformally anchored to the CNFs and (ii) atomically dispersed MoS\u003csub\u003e2\u003c/sub\u003e stabilized within the support (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). This synthesis route exploited growth control and support interaction to produce a monolayer, edge-enriched MoS\u003csub\u003e2\u003c/sub\u003e catalyst with unique features (see Supplementary Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ea for more details).\u003c/p\u003e \u003cp\u003eThe electrospinning parameters, including the polymer concentration, were systematically optimized to balance the CNF integrity, precursor dispersion, and final MoS\u003csub\u003e2\u003c/sub\u003e loading. PAN concentrations of 5\u0026ndash;15 wt/vol% were evaluated, with 10 wt/vol% PAN yielding the most uniform fibers (Figs. S1\u0026ndash;S3). In contrast, hydrothermal nucleation (MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec) produced densely stacked 2H-MoS\u003csub\u003e2\u003c/sub\u003e lamellae (Fig. S4), while the physically mixed MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb) exhibited intermediate behavior, with few-layered MoS\u003csub\u003e2\u003c/sub\u003e structures on the CNF surface. Together with MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;Comm, these materials constituted a structural continuum ranging from a bulk multilayer material to atomically thin, edge-rich MoS\u003csub\u003e2\u003c/sub\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003ePowder X-ray diffraction (XRD) analysis of MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES revealed only in-plane (100) and (110) reflections of 2H-MoS₂ at 2θ\u0026thinsp;=\u0026thinsp;32.3\u0026deg; and 57.0\u0026deg;, respectively, whereas the characteristic (002) basal-plane reflection at 14.5\u0026deg; was almost completely suppressed (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), confirming the absence of long-range stacking and the dominance of monolayer or highly exfoliated MoS\u003csub\u003e2\u003c/sub\u003e.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e This structural signature likely originated from the electrospinning environment, in which electrohydrodynamic stretching and rapid solvent evaporation immobilized Mo\u0026ndash;S₄\u0026sup2;⁻ complexes within the polymer jet. The resulting limited stacking dispersed the precursor at the nanoscale level, avoiding large flakes. During sequential reduction and sulfidation, ATTM decomposed, combined with gas-evolution-driven expansion and sulfur-mediated layer delamination, further suppressing plane growth and stacking, yielding exfoliated MoS\u003csub\u003e2\u003c/sub\u003e sheets firmly anchored to the CNFs. In contrast, both MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT and MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;Comm. exhibited intense (002) reflections, while MoS₂\u0026ndash;HT/CNF contained a (002) peak of moderate intensity, which was consistent with heterogeneous nucleation, controlled growth, and limited stacking (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ed and Figs \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u0026ndash;S4).\u003c/p\u003e \u003cp\u003eThese findings were in accordance with the Raman spectroscopy analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee and Figs. S3 and S4). In the commercial and bulk MoS\u003csub\u003e2\u003c/sub\u003e samples, pronounced E\u003csub\u003e2\u003c/sub\u003eg\u0026sup1; (~\u0026thinsp;380 cm⁻\u0026sup1;) and A\u003csub\u003e1\u003c/sub\u003eg (~\u0026thinsp;405 cm⁻\u0026sup1;) modes were observed. However, for the CNF-supported catalyst (MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES), these MoS\u003csub\u003e2\u003c/sub\u003e phonons were not detectable because the strong D- and G-bands of the PAN-derived CNFs dominated the spectra. The carbon features confirmed the partial graphitization and the formation of CNFs. In contrast, the sharper, more symmetric Raman modes in MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT reflected its highly ordered, defect-poor multilayer stacking, consistent with the prominent (002) reflection observed in the XRD results.\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e Corresponding nitrogen adsorption\u0026ndash;desorption isotherms (Fig. S5a) showed that MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES had a specific surface area of \u0026asymp;\u0026thinsp;180 m\u003csup\u003e2\u003c/sup\u003e g⁻\u003csup\u003e1\u003c/sup\u003e, nearly an order of magnitude higher than that of MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT and MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF (\u0026asymp;\u0026thinsp;40 m\u003csup\u003e2\u003c/sup\u003e g⁻\u003csup\u003e1\u003c/sup\u003e), highlighting the effectiveness of the reduction\u0026ndash;sulfidation sequence in generating hierarchical porosity and dispersing the active phases. The full results of the thermal analysis, including the BET surface area and TGA profiles, are provided in Figs. S5a,b and Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eIn addition, the morphology of the MoS\u003csub\u003e2\u003c/sub\u003e|CNF-ES, MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT, and MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF catalysts was investigated using scanning electron microscopy (SEM; Figs. S6 and S7) and high-resolution transmission electron microscopy (HR-TEM; Figs. S8\u0026ndash;S13). The SEM imaging revealed the continuous coverage of monolayer MoS\u003csub\u003e2\u003c/sub\u003e on the CNFs following sulfidation, whereas the hydrothermal samples retained a stacked morphology, and the hybrid material had an incomplete conformal coating. In the synthesis of the MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES sample, the as-spun ATTM\u0026ndash;PAN fibers formed uniform precursor filaments, which evolved into rigid CNFs decorated with MoS\u003csub\u003e2\u003c/sub\u003e when heated at 723 K under a H\u003csub\u003e2\u003c/sub\u003e atmosphere (Figs. S6a and S11a). Subsequent sulfidation in H\u003csub\u003e2\u003c/sub\u003eS/H\u003csub\u003e2\u003c/sub\u003e/N\u003csub\u003e2\u003c/sub\u003e at 973 K induced pore formation and exfoliation (Figs. S6b and S11b), producing a highly porous CNF network uniformly coated with a monolayer of MoS\u003csub\u003e2\u003c/sub\u003e. In contrast, MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT had a multilayer, densely stacked morphology (Fig. S7a), while the hybrid MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF sample (Fig. S7b) contained heterogeneously distributed few-layered MoS\u003csub\u003e2\u003c/sub\u003e platelets that did not fully conform to the carbon fibers. These differences indicated that the electrospinning strategy enabled optimized control, resulting in highly dispersed MoS₂ within the catalyst.\u003c/p\u003e \u003cp\u003eHR-TEM, high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM), and energy-dispersive X-ray (EDX) mapping (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; Figs. S8\u0026ndash;S13) were used to investigate the nanoscale architecture of MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES. HR-TEM and HAADF-STEM analysis confirmed the presence of atomically thin MoS\u003csub\u003e2\u003c/sub\u003e nanosheets uniformly distributed across the CNFs, with no evidence of restacking or aggregation. In line with this, EDX mapping showed homogeneous dispersion of Mo and S, with no aggregated nanoparticles. In contrast, MoS₂\u0026ndash;HT contained well-stacked multilayer lamellae with 0.62 nm (002) fringes, which was reflective of equilibrium-controlled growth and minimal defect density. MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF exhibited few-layered MoS\u003csub\u003e2\u003c/sub\u003e clusters that assembled vertically and partially aggregated. The electron energy loss spectroscopy (EELS) spectra for MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES (Fig. S12) revealed broader S L₂,₃ and Mo M₄,₅ edges, indicating the presence of defects in the sulfide structure, with under-coordinated Mo/S atoms. These features were significantly weaker in the hydrothermal and commercial catalysts.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eX-ray photoelectron spectroscopy (XPS) analysis of Mo 3d spectra for all samples revealed dominant Mo\u003csup\u003e4+\u003c/sup\u003e species (229.1 and 231.9 eV) characteristic of MoS\u003csub\u003e2\u003c/sub\u003e,\u003csup\u003e21\u003c/sup\u003e with minor Mo\u003csup\u003e6+\u003c/sup\u003e contributions (233.6 and 235.3 eV) from surface oxidation (MoO\u003csub\u003e3\u003c/sub\u003e) and/or edge sites (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea and S14).\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e The S 2p spectrum confirmed the presence of S\u003csup\u003e2\u0026ndash;\u003c/sup\u003e species, while MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES exhibited an additional peak attributable to bridging disulfide (S\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e2\u0026ndash;\u003c/sup\u003e) ligands at the edges, consistent with an edge-rich structure that was also observed using in-situ diffuse-reflectance Fourier-transform infrared spectroscopy (DRIFTS) analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb).\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e The shift in the binding energy for Mo 3d and S 2p observed for MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES and MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF was indicative of strong sulfide\u0026ndash;support interactions.\u003c/p\u003e \u003cp\u003eQuantitative nitric oxide (NO) chemisorption analysis confirmed that electrospinning enhanced both metal dispersion and defect accessibility (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES had an adsorption rate of 132 \u0026micro;mol\u003csub\u003eNO\u003c/sub\u003e g\u003csub\u003ecat\u003c/sub\u003e\u003csup\u003e\u0026minus;1\u003c/sup\u003e, roughly six times that of MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT, while its O\u003csub\u003e2\u003c/sub\u003e uptake was also considerably higher, indicating a higher concentration of active sites associated with vacancies.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e In-situ O\u003csub\u003e2\u003c/sub\u003e DRIFTS spectra (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb) contained intense 960\u0026ndash;900 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e bands associated with terminal Mo\u0026thinsp;=\u0026thinsp;O groups at the edges for MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES, whereas MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT and MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF contained additional 860\u0026ndash;700 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e features corresponding to in-plane Mo\u0026ndash;O\u0026ndash;Mo linkages on the basal plane.\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e The absence of these basal-plane modes in MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES indicated that oxidation and vacancy formation were confined to the edges rather than distributed over the lattice. According to time-resolved DRIFTS analysis, these Mo\u0026thinsp;=\u0026thinsp;O bands remained stable with long-term O\u003csub\u003e2\u003c/sub\u003e exposure (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec).\u003c/p\u003e \u003cp\u003eX-ray absorption near-edge structure (XANES; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed) analysis of the MoS\u003csub\u003e2\u003c/sub\u003e-based catalysts revealed similar edge positions for MoO\u003csub\u003e3\u003c/sub\u003e and Mo\u003csub\u003e2\u003c/sub\u003eC, consistent with mixed\u0026thinsp;+\u0026thinsp;4 and minor\u0026thinsp;+\u0026thinsp;6 oxidation states and in accordance with the XPS results. Fourier transform\u0026ndash;extended X-ray absorption fine structure (FT\u0026ndash;EXAFS) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee) analysis revealed Mo\u0026ndash;S (1.8\u0026ndash;2.5 \u0026Aring;) and Mo\u0026ndash;Mo (2.6\u0026ndash;3.2 \u0026Aring;) coordination analogous to metallic Mo foil. In addition, quantitative EXAFS fitting (Table S2) suggested that MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES had the lowest CN\u003csub\u003eMo\u0026ndash;Mo\u003c/sub\u003e/CN\u003csub\u003eMo\u0026ndash;S\u003c/sub\u003e ratio (0.70), indicating that it contained the highest fraction of coordinatively unsaturated Mo atoms. This reduced coordination likely originated from its monolayer structure and sulfide\u0026ndash;support interactions, which stabilized the edge sites against sulfur replenishment.\u003c/p\u003e \u003cp\u003eTemperature-programmed in-situ XANES (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef) analysis under a H\u003csub\u003e2\u003c/sub\u003e atmosphere revealed a progressive shift at the edges to a lower energy, confirming the H\u003csub\u003e2\u003c/sub\u003e-induced reduction of Mo species and concomitant sulfur removal. These edge defects served as dual-function centers, providing adsorption sites for H\u003csub\u003e2\u003c/sub\u003eS and facilitating S\u0026ndash;H bond activation, leading to the superior intrinsic activity and long-term stability observed during catalytic H\u003csub\u003e2\u003c/sub\u003eS decomposition.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eCatalytic evaluation and structure\u0026ndash;performance relationships\u003c/strong\u003e \u003cp\u003eEvaluation of the catalysts under H\u003csub\u003e2\u003c/sub\u003eS decomposition conditions (973 K, 1 bar) revealed that the synthesis pathway and interfacial structure had a strong influence on intrinsic activity (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Space\u0026ndash;time adjustments were employed to account for variation in the bulk density, and equilibrium conversions were validated using Aspen Plus simulations (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea, Figs. S3 and S4). MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES achieved near-equilibrium H\u003csub\u003e2\u003c/sub\u003eS conversion at a lower contact time, significantly outperforming both its hydrothermal (MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT and MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF) and commercial (MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;Comm) counterparts (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea,b). Intrinsic kinetic parameters were obtained from a kinetic model that accounted for equilibrium limits (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec\u0026ndash;e and Tables S3\u0026ndash;S5), yielding a rate constant of 1.9 \u0026times; 10\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e mol H\u003csub\u003e2\u003c/sub\u003eS atm⁻\u0026sup1; g cat⁻\u0026sup1; s⁻\u0026sup1; for MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES, which was 9-fold higher than MoS₂\u0026ndash;Comm and 3-fold greater than MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT and MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF. The corresponding apparent activation energy (Eₐ \u0026asymp; 63 kJ mol⁻\u0026sup1;) was also markedly lower than that of the layered references, indicating facilitated H\u0026ndash;S bond activation at the edge sites (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed). Overall, MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES had the highest activity, followed by MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT, MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF, and MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;Comm, and it also compared favorably to previously reported catalysts (Table S6). The faster kinetics of MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES likely reflected the highly dispersed nature of the active sites, with a high edge density, and their interactions with the support.\u003c/p\u003e \u003c/p\u003e \u003cp\u003eThe intrinsic rate of H\u003csub\u003e2\u003c/sub\u003eS decomposition also scaled linearly with NO chemisorption (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ef), establishing a quantitative link between catalytic performance and the density of coordinatively unsaturated Mo atoms. Because NO selectively binds to exposed Mo edge sites and sulfur vacancies, this correlation confirmed that the activity of MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES was governed by accessible edge defects rather than the total Mo content. The similar proportionality between the rate constants and defect density derived from EXAFS coordination analysis and O\u003csub\u003e2\u003c/sub\u003e chemisorption (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) confirmed that the concentration of sulfur vacancies was the dominant kinetic descriptor. The lower activation barrier of MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES can be explained by its polarization at the sulfide\u0026ndash;support interface, as evidenced by the XPS and XANES results (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea,d\u0026ndash;f). This interaction with the support weakens the Mo\u0026ndash;S bonds at the edges, enhancing S\u0026ndash;H bond dissociation and facilitating H\u003csub\u003e2\u003c/sub\u003e evolution. This interaction enables MoS\u003csub\u003e2\u003c/sub\u003e to maintain a reaction turnover without deactivation.\u003c/p\u003e \u003cp\u003eMoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES exhibited exceptional long-term stability, maintaining the steady-state conversion of H\u003csub\u003e2\u003c/sub\u003eS feeds of varying concentrations (2\u0026ndash;15 vol%) over 8 h and 50 h of continuous operation (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eg,h). The reaction rate fully recovered when reverting to the initial feed conditions, demonstrating the reversibility of sulfur accumulation. Post-reaction analysis confirmed the preservation of the structure and oxidation state of the proposed catalyst. XRD and XPS analysis (Figs. S14 and 4i) revealed unchanged 2H reflections and Mo\u003csup\u003e4\u003c/sup\u003e⁺/Mo\u003csup\u003e6\u003c/sup\u003e⁺ ratios, while corresponding HR-TEM images contained intact monolayer sheets without stacking or sintering. The carbon framework remained structurally robust, underscoring that the sulfide\u0026ndash;support interaction suppressed both sulfur condensation and Mo migration, the principal deactivation pathways in conventional MoS\u003csub\u003e2\u003c/sub\u003e catalysts.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eDensity Functional Theory calculations\u003c/b\u003e: To evaluate how the graphene support influences H\u003csub\u003e2\u003c/sub\u003eS activation, three basal-plane models were constructed: (i) a freestanding MoS\u003csub\u003e2\u003c/sub\u003e monolayer (ML\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e), (ii) a freestanding bilayer (BL\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e) representing a bulk-like configuration, and (iii) a graphene-supported MoS₂ monolayer (ML\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e/graphene) simulating the experimentally observed CNF-supported catalytic domains (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). For each of these models, single sulfur vacancy (1-Vs) was generated by removing a top-layer S atom, while a second adjacent vacancy (2-Vs) was introduced to model multi-vacancy environments known previously as more catalytically active sites for S\u0026ndash;H scission.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e The corresponding optimized structures for the pristine, 1-Vs, and 2-Vs versions of the models are presented in Figs. S17\u0026ndash;22.\u003c/p\u003e \u003cp\u003eThe H\u003csub\u003e2\u003c/sub\u003eS adsorption and dissociation energetics for all three systems are summarized in Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea,b. In the 1-Vs models, H\u003csub\u003e2\u003c/sub\u003eS bound weakly to BL\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e (\u0026minus;\u0026thinsp;0.02 eV) and slightly stronger to ML\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e (\u0026minus;\u0026thinsp;0.20 eV) and ML\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e/graphene (\u0026minus;\u0026thinsp;0.24 eV), indicating that both reduced dimensionality and the graphene interface enhance initial H\u003csub\u003e2\u003c/sub\u003eS binding. The first S\u0026ndash;H bond cleavage to form HS* + H* was mildly endothermic for all three 1-Vs systems (0.19\u0026ndash;0.38 eV) together with the second cleavage leading to H* + H *+ S* formation (0.31\u0026ndash;0.43 eV, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb), though this was lowest for ML\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e/graphene.\u003c/p\u003e \u003cp\u003eIn contrast, all of the dissociation steps became exergonic at the 2-Vs sites. Across all models, the adsorption (\u0026minus;\u0026thinsp;0.32 to \u0026minus;\u0026thinsp;0.47 eV) and stepwise dissociation (\u0026minus;\u0026thinsp;0.34 to \u0026minus;\u0026thinsp;0.15 eV) of H\u003csub\u003e2\u003c/sub\u003eS were thermodynamically favorable, demonstrating that multi-vacancy ensembles stabilize sulfur-containing intermediates and enable spontaneous S\u0026ndash;H cleavage.\u003c/p\u003e \u003cp\u003eGraphene also promoted hydrogen-assisted vacancy formation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ec). For 1-Vs, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\Delta\\:}{\\text{E}}_{\\text{v}\\text{a}\\text{c}}\\)\u003c/span\u003e\u003c/span\u003e decreased from +\u0026thinsp;0.01 eV (ML\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e) and +\u0026thinsp;0.17 eV (BL\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e) to \u0026minus;\u0026thinsp;0.03 eV for ML\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e/graphene, confirming the stabilizing influence of the conductive support. Similarly, 2-Vs formation was most favorable for ML\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e/graphene (\u0026minus;\u0026thinsp;0.62 eV), indicating that graphene promotes sulfur extraction under reductive conditions. Including finite-temperature corrections (973 K) from NIST thermochemical data reduced \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\Delta\\:}{\\text{G}}_{\\text{v}\\text{a}\\text{c}}\\)\u003c/span\u003e\u003c/span\u003e by ~\u0026thinsp;0.8 eV per sulfur atom, meaning that vacancy formation was thermodynamically feasible under the experimental activation conditions (Table S7).\u003c/p\u003e \u003cp\u003eOverall, these results demonstrate that graphene not only enhances H\u003csub\u003e2\u003c/sub\u003eS binding but also lowers the energetic penalty for vacancy formation and sulfur dissociation. The ML\u0026ndash;MoS\u003csub\u003e2\u003c/sub\u003e/graphene system had the lowest energy barriers among the three models, providing a basis for accelerated H\u003csub\u003e2\u003c/sub\u003eS decomposition at supported basal planes, consistent with our experimental findings.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eCollectively, the operando, kinetic, and theoretical analyses produced a coherent mechanistic overview of H\u003csub\u003e2\u003c/sub\u003eS decomposition on MoS\u003csub\u003e2\u003c/sub\u003e-based catalysts. Within the electrospun MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES catalyst, MoS₂ is grafted onto CNFs and interacts strongly with the support, resulting in active, stable S\u0026ndash;H scission sites. As indicated by pulse chemisorption analysis, the electrospun catalyst has a far higher density of accessible Mo edge sites than either the hydrothermal or the bulk analogues. This enrichment likely arises from the electrospinning-induced grafting of MoS\u003csub\u003e2\u003c/sub\u003e precursors within polymer fibers, yielding uniformly dispersed monolayer MoS\u003csub\u003e2\u003c/sub\u003e domains following sulfidation. The resulting CNF architecture prevents sintering while introducing abundant defect sites. The sulfide-support interactions polarize the Mo\u0026ndash;S bonds and weaken the S\u0026ndash;H activation barrier, a phenomenon directly observed using in-situ spectroscopy. O\u003csub\u003e2\u003c/sub\u003e-DRIFTS and XANES\u0026ndash;EXAFS analyses suggest that sulfur vacancies form and heal dynamically under alternating H₂S/H₂ environments. The downshift of the Mo \u003cem\u003eK\u003c/em\u003e-edge and the reduction in Mo\u0026ndash;S and Mo\u0026ndash;Mo coordination are indicative of the formation of undercoordinated MoS₂ centers. XPS corroborates this by identifying mixed Mo\u003csup\u003e4+\u003c/sup\u003e/Mo\u003csup\u003e6+\u003c/sup\u003e species stabilized by the support interface. These support interactions allow the edge vacancies to persist and remain catalytically active under conditions where unsupported MoS₂ rapidly deactivates.\u003c/p\u003e \u003cp\u003eDFT calculations provided atomistic support for this mechanism. The graphene-supported MoS\u003csub\u003e2\u003c/sub\u003e monolayer, which is an analogue of the CNF-grafted catalyst, has the lowest energetic cost for hydrogen-assisted sulfur removal (ΔE\u003csub\u003eSvac\u003c/sub\u003e = \u0026minus;\u0026thinsp;0.03 eV for 1-V_S and \u0026minus;\u0026thinsp;0.62 eV for 2-V_S) and exergonic H\u003csub\u003e2\u003c/sub\u003eS dissociation for all intermediates. The carbon substrate stabilizes these vacancies through support interactions, reducing the S-vacancy formation barrier and promoting continuous reactive turnover. According to finite-temperature corrections, vacancies form at 973 K, which was in line with the experimental observations of sustained activity and vacancy regeneration. The CNFs interact with the MoS\u003csub\u003e2\u003c/sub\u003e and stabilize the active sites. The faster kinetics of the MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES catalyst arise from a high density of edge and vacancy sites and support interactions that stabilize these highly dispersed active sites.\u003c/p\u003e"},{"header":"CONCLUSIONS","content":"\u003cp\u003eThe edge-enriched MoS\u003csub\u003e2\u003c/sub\u003e|CNF catalyst developed in the present study delivered a 3-fold increase in H\u003csub\u003e2\u003c/sub\u003eS decomposition activity and remarkable stability over 50 h of operation compared to bulk and hydrothermally synthesized MoS\u003csub\u003e2\u003c/sub\u003e, while maintaining long-term stability under harsh conditions. This exceptional performance was due to the unique combination of (i) highly dispersed monolayer MoS\u003csub\u003e2\u003c/sub\u003e domains anchored on CNFs, (ii) strong sulfide\u0026ndash;support interactions that suppress sintering and stabilize undercoordinated edge atoms, and (3) a dense population of sulfur vacancies. In situ spectroscopy and DFT calculations jointly confirmed that the stabilized edge vacancies serve as active sites for H\u003csub\u003e2\u003c/sub\u003eS dissociation and hydrogen evolution. The proposed electrospinning\u0026ndash;sulfidation route establishes a material-design paradigm for engineering metastable, edge-rich sulfides through confinement and stabilization. The grafting of sulfides onto CNFs and the resulting defect stabilization offer a transferable blueprint for the synthesis of robust catalysts for other high-temperature, sulfur-tolerant hydrogen production processes.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eCatalyst synthesis\u003c/h2\u003e \u003cp\u003eThree MoS\u003csub\u003e2\u003c/sub\u003e-based model catalysts were prepared using hydrothermal and electrospinning methods reported in previous studies, with slight modifications and different pre-treatment protocols.\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eSynthesis of MoS\u003csub\u003e2\u003c/sub\u003e-HT\u003c/strong\u003e \u003cp\u003eIn the typical hydrothermal synthesis route, 0.8 mmol of ammonium heptamolybdate tetrahydrate (AMT; (NH\u003csub\u003e4\u003c/sub\u003e)\u003csub\u003e6\u003c/sub\u003eMo\u003csub\u003e7\u003c/sub\u003eO\u003csub\u003e24\u003c/sub\u003e. 4H\u003csub\u003e2\u003c/sub\u003eO; Sigma-Aldrich; 99.98%) was dissolved in 30 mL of deionized (DI) water at room temperature and stirred for 30 min before the addition of citric acid (CA) or oxalic acid (OA); (0.05 g; Sigma-Aldrich; 99.5%) as a chemical agent under vigorous magnetic stirring for 20 min. The sulfide source (10 mmol of thiourea; CH\u003csub\u003e4\u003c/sub\u003eN\u003csub\u003e2\u003c/sub\u003eS; Sigma-Aldrich; 99.99%) was dissolved in another 30 mL of DI water at room temperature under vigorous magnetic stirring. The molybdenum solution was then added slowly to the sulfur solution, and the resulting mixture was transferred to a 100 mL Teflon-lined stainless-steel autoclave and heated at 473 K. After cooling to room temperature, the black product was filtered and washed several times with distilled water and absolute ethanol. The final product was dried at 333 K overnight. As a comparison, a commercial MoS\u003csub\u003e2\u003c/sub\u003e catalyst (98%, Alfa Aesar) was purchased from Thermo Scientific Chemicals and used as received for testing.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eSynthesis of MoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES\u003c/strong\u003e \u003cp\u003eMoS\u003csub\u003e2\u003c/sub\u003e|CNF\u0026ndash;ES was prepared using a modified electrospinning method.\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e PAN (1.0 g) and ATTM ((NH\u003csub\u003e4\u003c/sub\u003e)\u003csub\u003e2\u003c/sub\u003eMoS\u003csub\u003e4\u003c/sub\u003e; 0.5 g) were dissolved in dimethylformamide to obtain PAN concentrations of 5, 10, or 15 wt/vol%. The precursor solutions were stirred for 24 h at 60\u0026deg;C until fully homogenized, and the resulting solutions were subsequently loaded into a syringe equipped with a 23-gauge stainless-steel needle. Electrospinning was conducted at a voltage of 15 kV with a flow rate of 1 mL h\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e and a tip-to-collector distance of 15 cm. The as-spun composite mats were collected on grounded aluminum foil and dried under a vacuum at 60\u0026deg;C. The obtained ATTM\u0026ndash;PAN nanofibers were pre-treated at 723 K for 5 h in a H\u003csub\u003e2\u003c/sub\u003e (5 vol%)/N\u003csub\u003e2\u003c/sub\u003e (95 vol%) atmosphere, followed by further treatment at 973 K in 2 vol% H\u003csub\u003e2\u003c/sub\u003eS/N\u003csub\u003e2\u003c/sub\u003e for 2 h and 50 vol% H\u003csub\u003e2\u003c/sub\u003e/N\u003csub\u003e2\u003c/sub\u003e for 1 h with heating rates of 5\u0026deg;C min\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e for each treatment stage to obtain a porous MoS\u003csub\u003e2\u003c/sub\u003e/CNF structure. As a comparison, after the first-stage heat treatment at 723 K for 5 h under H\u003csub\u003e2\u003c/sub\u003e (5 vol%)/N\u003csub\u003e2\u003c/sub\u003e (95 vol%), the resulting material was subjected to heat treatment at 973 K for 5 h under a N\u003csub\u003e2\u003c/sub\u003e atmosphere (Fig. S2).\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eSynthesis MoS\u003csub\u003e2\u003c/sub\u003e\u0026ndash;HT/CNF\u003c/strong\u003e \u003cp\u003eCNFs were synthesized via electrospinning, followed by thermal treatment and the hydrothermal deposition of MoS\u003csub\u003e2\u003c/sub\u003e. Initially, 1.0 g of PAN was dissolved in 7.0 g of DMF (Sigma-Aldrich, 99.8%) under continuous stirring for 24 h at room temperature. The resulting homogeneous solution was subjected to electrospinning to prepare the mats using the previously described process. The collected fibrous mats were pre-treated at 723 K for 5 h under a H\u003csub\u003e2\u003c/sub\u003e (5 vol%)/N\u003csub\u003e2\u003c/sub\u003e (95 vol%) atmosphere, followed by treatment at 973 K in 2 vol% H\u003csub\u003e2\u003c/sub\u003eS/N\u003csub\u003e2\u003c/sub\u003e for 2 h and 50 vol% H\u003csub\u003e2\u003c/sub\u003e/N\u003csub\u003e2\u003c/sub\u003e for 1 h with heating rates of 5 K min\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e for each treatment stage to obtain CNFs without MoS\u003csub\u003e2\u003c/sub\u003e.\u003c/p\u003e \u003c/p\u003e \u003cp\u003eA hydrothermal method was employed for MoS\u003csub\u003e2\u003c/sub\u003e deposition, similar to the synthesis process for MoS\u003csub\u003e2\u003c/sub\u003e-HT and a previously reported method.\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e Ammonium molybdate (15 wt% Mo) was dissolved in 30 mL of DI water and stirred for 30 min, followed by the addition of 0.05 g of OA under vigorous stirring for another 20 min. In parallel, 10 mmol of thiourea was dissolved in 30 mL of DI water. The molybdate solution was slowly added to the thiourea solution under stirring. The mixture was then transferred to a 100 mL Teflon-lined stainless-steel autoclave, and the required quantity of the CNF mat was added. The mixture was heated at 493 K for 18 h. After cooling, the product was filtered, washed with distilled water and ethanol, and dried overnight at 333 K.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eCatalyst characterization\u003c/h3\u003e\n\u003cp\u003eCrystallographic analysis was conducted via powder XRD in a Bruker D8 advanced diffractometer with a Bragg\u0026ndash;Brentano geometry fitted with a Cu tube operating at 40 kV and 40 mA. Diffractograms were acquired over a 2θ range of 10\u0026ndash;80\u0026deg; with a step size of 0.1\u0026deg; and a scan speed of 0.5 s per step. PDF-4+ (2019) was the database used for phase identification.\u003c/p\u003e \u003cp\u003eTo investigate the textural properties of the samples, N\u003csub\u003e2\u003c/sub\u003e adsorption\u0026ndash;desorption isotherms were obtained at 77 K using an ASAP 2040 instrument (Micromeritics). Before the measurements were taken, the samples were degassed for 10 h at 523 K. The specific surface area (S\u003csub\u003eBET\u003c/sub\u003e) was determined from the isotherms using the Brunauer\u0026ndash;Emmett\u0026ndash;Teller (BET) equation.\u003c/p\u003e \u003cp\u003eThe surface oxidation states of Mo and S were determined using XPS analysis in a K-ALPHA spectrometer (Thermo Scientific) with an Al-Kα (1486.6 eV) radiation source at room temperature under an ultra-high vacuum at 3 mA \u0026times; 12 kV. The powder samples were pressed, mounted on the holder, and placed in the vacuum chamber. The C 1s peak at 284.5 eV was used to calibrate and correct the binding energy for the element, and charge compensation was achieved using a system flood gun, providing low-energy electrons. In addition, Raman spectra for the MoS\u003csub\u003e2\u003c/sub\u003e catalysts were obtained using a Renishaw Via Reflex confocal spectrometer with 532 nm laser excitation. The laser power was set to 5\u0026ndash;10 mW, and the sample was irradiated for 10 s with six accumulations.\u003c/p\u003e \u003cp\u003eThe catalyst morphology was investigated using HR-TEM with a Titan Themis-Z microscope (Thermo Fisher Scientific) at an accelerating voltage of 300 kV and a beam current of 0.5 nA. Dark-field (DF) imaging employed an STEM and an HAADF detector. The STEM-HAADF data were collected at a convergence angle of 29.9 mrad and an HAADF inner angle of 30 mrad. Furthermore, an X-ray energy dispersive spectrometer (EDS) was utilized in conjunction with DF-STEM imaging to acquire STEM-EDS spectrum-imaging datasets and a post-column EELS instrument (GIF-Quantum 966; Gatan, Inc.).\u003c/p\u003e \u003cp\u003eNO chemisorption measurements were taken using a Micromeritics AutoChem II 2920 instrument.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e In a typical fabrication route, 0.1 g of the catalyst sample was pretreated in situ under a flow of H\u003csub\u003e2\u003c/sub\u003e at 573 K for 3 h, followed by purging with He at the same temperature for 60 min. Subsequently, NO chemisorption was conducted at 313 K by pulsing 2% NO in He through the catalyst every 30 min. The NO signal was monitored via mass spectrometry and observed to increase until reaching a steady state, indicating saturation of the catalyst surface. The total NO uptake was then determined from the cumulative amount adsorbed.\u003c/p\u003e \u003cp\u003eO\u003csub\u003e2\u003c/sub\u003e pulse chemisorption experiments were performed with in-situ DRIFTS to quantify the active sites in the catalysts.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e O\u003csub\u003e2\u003c/sub\u003e pulse titration was conducted using a Micromeritics Autochem II 2930. Typically, ~\u0026thinsp;0.1 g of the catalyst was pretreated by H\u003csub\u003e2\u003c/sub\u003e at 573 K for 2 h and purged with He at 573 K for another 1 h. O\u003csub\u003e2\u003c/sub\u003e chemisorption was conducted at 313 K by pulsing 1 mL of 5% O\u003csub\u003e2\u003c/sub\u003e/He over the catalyst every 15 min until peak saturation. Effluent O\u003csub\u003e2\u003c/sub\u003e was detected using a mass spectrometer until its level reached a constant value, representing O\u003csub\u003e2\u003c/sub\u003e saturation of the catalyst, after which the total uptake of O\u003csub\u003e2\u003c/sub\u003e was calculated.\u003c/p\u003e \u003cp\u003eIn-situ DRIFTS measurements were taken on a Nicolet 6700 FTIR spectrometer with a liquid-nitrogen cooled MCT detector using a high-temperature reaction cell (Harrick) equipped with a temperature programmer and connected to a gas-dosing device with mass flow controllers (Bronkhorst). The catalyst was first reduced by pure H\u003csub\u003e2\u003c/sub\u003e at 573 K with a gas flow of 30 ml min\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e at 1 bar for 1 h and then cooled with He to 313 K for another 1 h. Spectra obtained at 313 K under a H\u003csub\u003e2\u003c/sub\u003e atmosphere were used as background spectra. Subsequently, the catalyst samples were treated with 5% O\u003csub\u003e2\u003c/sub\u003e/He at a rate of 30 mL min\u003csup\u003e\u0026minus;\u0026thinsp;1,\u003c/sup\u003e followed by spectra collection throughout in-situ DRIFTS.\u003c/p\u003e \u003cp\u003eMo \u003cem\u003eK\u003c/em\u003e-edge X-ray absorption spectra were recorded at room temperature using a Si (311) monochromator in the BL01B1 line of Spring-8 at the Japan Synchrotron Radiation Research Institute (JASRI) in Harima, Hyogo, Japan. The obtained spectra were analyzed and fitted using Athena and Artemis, which were equipped with ATOMS and FEFF6 in the software package Demeter 0.9.26.\u003csup\u003e31\u0026ndash;33\u003c/sup\u003e After normalization at the edge height, the \u003cem\u003ek\u003c/em\u003e\u003csup\u003e3\u003c/sup\u003e-weighted χ spectra were extracted. Subsequently, these spectra in the k range of 3\u0026ndash;14 \u0026Aring;\u003csup\u003e\u0026minus;1\u003c/sup\u003e were Fourier-transformed into the \u003cem\u003eR\u003c/em\u003e-space. Curve-fitting analysis was conducted in an R range of 1.8\u0026ndash;3.4 \u0026Aring; for Mo foil and 1.4\u0026ndash;3.3 \u0026Aring; for all MoS\u003csub\u003e2\u003c/sub\u003e catalysts using the back-scattering amplitude and phase shift functions of Mo\u0026ndash;Mo and Mo\u0026ndash;S.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eCatalyst testing\u003c/h2\u003e \u003cp\u003eCatalytic H\u003csub\u003e2\u003c/sub\u003eS splitting experiments were conducted using a PID testing unit with a fixed-bed quartz reactor (8 mm ID, length 340 mm) in down-flow mode (Fig. S23). The reactor was placed in a furnace, and the flow was controlled using a mass flow controller. In the reactor, ~\u0026thinsp;0.03 g of sieved catalyst particles (100\u0026ndash;150 \u0026micro;m) were loaded onto a 0.15 g SiC (particle grit 40) bed to ensure the catalyst bed rested in the isothermal zone of the reactor. The catalyst was pre-treated in situ first with a mixture of H\u003csub\u003e2\u003c/sub\u003eS and H\u003csub\u003e2\u003c/sub\u003e (50:50) at 973 K for 1 h, followed by testing with a mixed feed containing 2 vol.% of H\u003csub\u003e2\u003c/sub\u003eS at 1 bar in the temperature range of 773\u0026ndash;973 K. A flow of 1 mL min\u003csup\u003e\u0026ndash;\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e of He was added to the mixed feed as an internal standard. Reactor outputs were then analyzed using a Varian micro gas chromatograph with TCD detectors. The two-channel configuration and columns of this analytical system were used to quantify H\u003csub\u003e2\u003c/sub\u003e, He, H\u003csub\u003e2\u003c/sub\u003eS, and N\u003csub\u003e2\u003c/sub\u003e. The conversion (X) is expressed as follows:\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:{X}_{i}\\left(\\%\\right)=\\frac{{F}_{i,0}-{F}_{i}}{{F}_{i,0}}\\times\\:100$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{F}_{i,0}\\)\u003c/span\u003e\u003c/span\u003e is the molar flow rate at the inlet of the reactor of species \u003cem\u003ei\u003c/em\u003e (e.g., H\u003csub\u003e2\u003c/sub\u003eS), and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{F}_{i}\\:\\:\\)\u003c/span\u003e\u003c/span\u003e is the outlet molar flow rate of species \u003cem\u003ei\u003c/em\u003e.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSimulations, kinetic model, rate constant, and turnover frequency calculations\u003c/h3\u003e\n\u003cp\u003eThe thermodynamic equilibrium conversion of H\u003csub\u003e2\u003c/sub\u003eS under each tested condition was simulated using Aspen Plus with an R-Gibbs reactor and the WILSON base method. A kinetic model was used to describe the intrinsic reaction rate by integrating the rate expression over the catalyst weight. The rate constant (k) and activation energy (\u003cem\u003eEₐ\u003c/em\u003e) were determined by minimizing the difference between the experimentally observed and theoretically predicted rates, ensuring an accurate fit to the data.\u003c/p\u003e \u003cp\u003eThe turnover frequency (TOF) was calculated based on the initial reaction rate per active site. While the standard TOF expression assumes differential operation, we employed an alternative formulation suitable for near-equilibrium conditions, based on the initial rate at low conversion. Further details of model derivation, the fitting process, and TOF equations are provided in the Supporting Information.\u003c/p\u003e\n\u003ch3\u003eComputational details\u003c/h3\u003e\n\u003cp\u003eAll spin-polarized density functional theory (DFT) calculations were performed using the Vienna Ab initio Simulation Package (VASP 5.4.4).\u003csup\u003e34\u003c/sup\u003e The projector-augmented wave (PAW) method was used to describe the interaction between core and valence electrons,\u003csup\u003e35\u003c/sup\u003e and generalized gradient approximation (GGA) in the Perdew\u0026ndash;Burke\u0026ndash;Ernzerhof (PBE) form was employed to account for exchange-correlation effects.\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e Grimme\u0026rsquo;s DFT-D3 method with Becke\u0026ndash;Johnson damping was included in all calculations to capture dispersion interactions.\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eThe plane-wave kinetic energy cutoff was set to 500 eV, and the electronic energy convergence criterion was set to 10\u003csup\u003e\u0026ndash;5\u003c/sup\u003e eV. Structural relaxation was performed until the forces on all atoms were below 0.02 eV/\u0026Aring;. A vacuum layer of 15 \u0026Aring; was introduced along the \u003cem\u003ez\u003c/em\u003e-direction to avoid spurious interactions between periodic images. The Brillouin zone was sampled using a Γ-centered Monkhorst\u0026ndash;Pack k-point mesh of 3\u0026times;3\u0026times;1 for the (7\u0026times;7) supercell models.\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eMonolayer, bilayer, and graphene-supported monolayer MoS\u003csub\u003e2\u003c/sub\u003e systems were constructed to evaluate the effect of support and vacancy concentration. Sulfur vacancies (1-S and 2-S) were created by removing one or two surface S atoms, and the thermodynamic stability of various dissociated intermediates (H\u003csub\u003e2\u003c/sub\u003eS*, HS* + H*, and H* + H* + S*) was evaluated. Adsorption energies were calculated relative to isolated gas-phase species and referenced to the pristine surface.\u003c/p\u003e \u003cp\u003eThe sulfur vacancy formation energy (ΔEs\u003csub\u003evac\u003c/sub\u003e) was determined in the absence and presence of hydrogen co-adsorption to capture the thermodynamic impact of reductive conditions. The filling energy of vacancies with sulfur atoms (i.e., S-vacancy healing) was assessed to understand the post-dissociation stability of active sites.\u003c/p\u003e \u003cp\u003eThe adsorption energy (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{ads}\\)\u003c/span\u003e\u003c/span\u003e) of a molecule or intermediate on the MoS₂ surface was calculated as follows:\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$$\\:{E}_{ads}={E}_{slab+adsorbate}-{E}_{slab}-{E}_{adsorbate\\left(g\\right)}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{slab+adsorbate}\\)\u003c/span\u003e\u003c/span\u003e is the total energy of the adsorbed system, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{slab}\\)\u003c/span\u003e\u003c/span\u003e is the energy of the clean surface, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{adsorbate\\left(g\\right)}\\)\u003c/span\u003e\u003c/span\u003e is the total energy of the isolated gas-phase molecule optimized in a large periodic box.\u003c/p\u003e \u003cp\u003eThe sulfur-vacancy formation energy (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\Delta\\:}{E}_{Svac}\\)\u003c/span\u003e\u003c/span\u003e) was computed as follows:\u003cdiv id=\"Equ3\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ3\" name=\"EquationSource\"\u003e\n$$\\:{\\Delta\\:}{E}_{Svac}-{E}_{vacant}-{E}_{pristine}+n{\\mu\\:}_{S}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e3\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{vacant}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{pristine}\\)\u003c/span\u003e\u003c/span\u003e are the total energy of the defective and pristine MoS₂ slabs, respectively, \u003cem\u003en\u003c/em\u003e is the number of sulfur atoms removed, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\mu\\:}_{S}\\)\u003c/span\u003e\u003c/span\u003e is the sulfur chemical potential derived from gas-phase S\u003csub\u003e2\u003c/sub\u003e:\u003cdiv id=\"Equ4\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ4\" name=\"EquationSource\"\u003e\n$$\\:{\\mu\\:}_{S}=\\frac{1}{2}{E}_{{S}_{2}\\left(g\\right)}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e4\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eUnder reductive conditions, the hydrogen-assisted sulfur vacancy formation energy (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\Delta\\:}{E}_{Svac}^{{H}_{2}}\\)\u003c/span\u003e\u003c/span\u003e) was determined according to the following equation:\u003cdiv id=\"Equ5\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ5\" name=\"EquationSource\"\u003e\n$$\\:{\\Delta\\:}{E}_{Svac}^{{H}_{2}}={E}_{vacant}-{E}_{pristine}+n\\left({E}_{{H}_{2}S\\left(g\\right)}-{E}_{{H}_{2}\\left(g\\right)}\\right)$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e5\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhich represents the following process:\u003c/p\u003e \u003cp\u003eMoS₂ + n H₂ (g) \u0026rarr; MoS₂₋ₙ + n H₂S (g) (6)\u003c/p\u003e \u003cp\u003eEq.\u0026nbsp;(\u003cspan refid=\"Equ5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) therefore accounts for the removal of sulfur atoms from the surface through their conversion to H₂S gas in the presence of H₂.\u003c/p\u003e \u003cp\u003eTo obtain temperature-corrected Gibbs free energies, the gas-phase DFT energies in Eq.\u0026nbsp;(\u003cspan refid=\"Equ5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) were replaced by temperature-dependent chemical potentials from the NIST-JANAF thermochemical tables:\u003cdiv id=\"Equ6\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ6\" name=\"EquationSource\"\u003e\n$$\\:{\\Delta\\:}{G}_{Svac}^{{H}_{2}\\left(T,P\\right)}=\\left({E}_{vacant}-{E}_{pristine}\\right)+n\\left[{\\mu\\:}_{{H}_{2}S\\:}\\left(T,{p}_{{H}_{2}S}\\right)-{\\mu\\:}_{{H}_{2}}\\left(T,{p}_{{H}_{2}}\\right)\\right]$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e7\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere the temperature-dependent chemical potential of each gas is defined as shown in Eq.\u0026nbsp;(\u003cspan refid=\"Equ7\" class=\"InternalRef\"\u003e8\u003c/span\u003e):\u003cdiv id=\"Equ7\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ7\" name=\"EquationSource\"\u003e\n$$\\:{\\mu\\:}_{i}\\left(T,{p}_{i}\\right)={\\mu\\:}_{i}^{^\\circ\\:}\\left(T\\right)+{k}_{B}T\\text{ln}\\left(\\frac{{p}_{i}}{{p}^{^\\circ\\:}}\\right)$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e8\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eand \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\mu\\:}_{i}^{^\\circ\\:}\\left(T\\right)={H}_{i}^{^\\circ\\:}\\left(T\\right)-T{S}_{i}^{^\\circ\\:}\\left(T\\right)\\)\u003c/span\u003e\u003c/span\u003e = H_i\u0026deg;(T) \u0026minus; TS_i\u0026deg;(T) was computed from NIST Shomate equations. For example, at 700\u0026deg;C (973 K) and 800\u0026deg;C (1073 K), \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\mu\\:}_{{H}_{2}S}^{^\\circ\\:}\\left(T\\right)-{\\mu\\:}_{{H}_{2}}^{^\\circ\\:}\\left(T\\right)\\)\u003c/span\u003e\u003c/span\u003e equals \u0026minus;\u0026thinsp;0.80 eV and \u0026minus;\u0026thinsp;0.89 eV, respectively.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eDATA AVAILABILITY\u003c/h2\u003e \u003cp\u003eThe data supporting the findings of this article are available in the paper and the Supplementary Information. Additional data is available from the corresponding author on request. Source data are provided with this paper.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eAUTHOR INFORMATION\u003c/h2\u003e \u003cp\u003eAuthors and Affiliations\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eMultiscale Reaction Engineering, KAUST Catalysis Platform (KCP), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955\u0026thinsp;\u0026minus;\u0026thinsp;6900, Saudi Arabia\u003c/strong\u003e \u003cp\u003eHend Mohamed, Vijay K. Velisoju, Muhammad Arief Shafarifky, Gontzal Lezcano, and Pedro Casta\u0026ntilde;o\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eWater Desalination and Reuse Platform (WDRP), Division of Biological and Environmental Science and Engineering (BESE), King Abdullah University of Science and Technology (KAUST), Thuwal, 23955\u0026thinsp;\u0026minus;\u0026thinsp;6900, Saudi Arabia\u003c/strong\u003e \u003cp\u003eM. Obaid and Noreddine Ghaffour\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eKAUST Catalysis Platform, King Abdullah University of Science and Technology (KAUST), Thuwal, 23955\u0026thinsp;\u0026minus;\u0026thinsp;6900, Saudi Arabia\u003c/strong\u003e \u003cp\u003eRafia Ahmed, Ildar Mukhambetov, Huabin Zhang, Polina Tolstova, and Luigi Cavallo\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eDepartment of Materials Engineering Science, Graduate School of Engineering Science, The University of Osaka, 1\u0026ndash;3 Machikaneyama, Toyonaka, Osaka 560\u0026ndash;8531, Japan\u003c/strong\u003e \u003cp\u003eYuma Hirayama and Takato Mitsudome\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eKAUST Core Labs, King Abdullah University of Science and Technology (KAUST), Thuwal, 23955\u0026thinsp;\u0026minus;\u0026thinsp;6900, Saudi Arabia\u003c/strong\u003e \u003cp\u003eMohamed Ben Hassine and Omar El Tall\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eResearch and Development Center, Dhahran 31311, Saudi Aramco\u003c/strong\u003e \u003cp\u003eZainab Alaithan, Ali Almofleh, and Hassan Aljama\u003c/p\u003e \u003c/p\u003e\u003ch2\u003e \u003cb\u003eChemical Engineering Program, Physical Science and Engineering (PSE) Division\u003c/b\u003e,\u003c/h2\u003e \u003cp\u003e \u003cstrong\u003eKing Abdullah University of Science and Technology (KAUST), Thuwal, Saudi Arabia\u003c/strong\u003e \u003cp\u003ePedro Casta\u0026ntilde;o\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eContributions\u003c/strong\u003e \u003cp\u003eH.M., V.K.V., H.A., and P.C. led the project and designed the experiments. H.M., M.O., and N.G. designed the materials and synthesis protocols using electrospinning. V.K.V. and H.M. performed and analyzed the basic characterization and contributed to the optimization of the thermal treatment of the catalysts. V.K.V. performed and analyzed the catalytic testing, kinetic analysis, and advanced characterization. G.L. and V.K.V. contributed to kinetic model development. R.A., L.C., Z.A., P.T., and H.A. contributed to the theoretical calculations. V.K.V., H.M., and I.M. contributed to reactor operation, testing protocols, and analytical methodologies. M.B.H., O.E.T., and H.Z., contributed to material characterization, results interpretation, and manuscript reviewing. Y.H., M.A.S., and T.M. have conducted and analyzed the XAFS measurement. V.K.V., H.M., and P.C. wrote the manuscript with input from all authors. P.C. and H.A. were responsible for funding acquisition and project supervision.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eCorresponding author\u003c/strong\u003e \u003cp\u003eCorrespondence to Pedro Casta\u0026ntilde;o.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eETHICS DECLARATIONS\u003c/strong\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eACKNOWLEDGMENTS\u003c/h2\u003e \u003cp\u003eAll authors thank Saudi Aramco and King Abdullah University of Science and Technology (KAUST) for funding this work (RGC/3/5222\u0026ndash;01 and BAS/1/1403) and for the resources in the supercomputing laboratory at KAUST. Y.H and T.M. received JSPS KAKENHI grant 23H01761 and JST PRESTO grant JPMJPR21Q9 for sources and funding of the synchrotron experiments (2025A1813).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBligaard T et al (2016) Toward Benchmarking in Catalysis Science: Best Practices, Challenges, and Opportunities. ACS Catal 6:2590\u0026ndash;2602\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVogt C, Weckhuysen BM (2022) The concept of active site in heterogeneous catalysis. Nat Rev Chem 6:89\u0026ndash;111\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReverberi A, Pietro, Klemeš JJ, Varbanov PS, Fabiano B (2016) A review on hydrogen production from hydrogen sulphide by chemical and photochemical methods. J Clean Prod 136:72\u0026ndash;80\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlmofleh A et al (2025) Designing sulfide catalysts for H\u003csub\u003e2\u003c/sub\u003eS dissociation to H\u003csub\u003e2\u003c/sub\u003e based on reaction descriptors and microkinetics. Appl Catal B Environ Energy 361:124605\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAljama H, Alaithan Z, Almofleh A (2023) Catalytic conversion of H\u003csub\u003e2\u003c/sub\u003eS to H\u003csub\u003e2\u003c/sub\u003e: Challenges and catalyst limitations. J Phys Chem C 127:9022\u0026ndash;9029\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKraia T, Kaklidis N, Konsolakis M, Marnellos GE (2019) Hydrogen production by H\u003csub\u003e2\u003c/sub\u003eS decomposition over ceria supported transition metal (Co, Ni, Fe and Cu) catalysts. Int J Hydrogen Energy 44:9753\u0026ndash;9762\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeng D et al (2016) Catalysis with two-dimensional materials and their heterostructures. \u003cem\u003eNat. Nanotechnol.\u003c/em\u003e 11, 218\u0026ndash;230 (2016)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMao J, Wang Y, Zheng Z, Deng D (2018) The rise of two-dimensional MoS\u003csub\u003e2\u003c/sub\u003e for catalysis. Front Phys 13:1\u0026ndash;19\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNorskov JK, Christensen CH (2006) Toward efficient hydrogen production at surfaces. Sci (80-) 312:1322\u0026ndash;1323\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVoiry D et al (2013) Enhanced catalytic activity in strained chemically exfoliated WS\u003csub\u003e2\u003c/sub\u003e nanosheets for hydrogen evolution. \u003cem\u003eNat. Mater.\u003c/em\u003e 12, 850\u0026ndash;855 (2013)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu J et al (2021) Sulfur vacancy-rich MoS\u003csub\u003e2\u003c/sub\u003e as a catalyst for the hydrogenation of CO\u003csub\u003e2\u003c/sub\u003e to methanol. Nat Catal 2021 43(4):242\u0026ndash;250\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMao J et al (2023) Direct conversion of methane with O\u003csub\u003e2\u003c/sub\u003e at room temperature over edge-rich MoS\u003csub\u003e2\u003c/sub\u003e. Nat Catal 2023 611(6):1052\u0026ndash;1061\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu J et al (2023) Edge-rich molybdenum disulfide tailors carbon-chain growth for selective hydrogenation of carbon monoxide to higher alcohols. \u003cem\u003eNat. Commun.\u003c/em\u003e 14, 1\u0026ndash;11 (2023)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou S et al (2023) Strained few-layer MoS\u003csub\u003e2\u003c/sub\u003e with atomic copper and selectively exposed in-plane sulfur vacancies for CO\u003csub\u003e2\u003c/sub\u003e hydrogenation to methanol. \u003cem\u003eNat. Commun.\u003c/em\u003e 14, 1\u0026ndash;15 (2023)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBojarska Z, Mazurkiewicz-Pawlicka M, Mierzwa B, Plocinski T, Makowski L (2022) Effect of the carbon support on MoS\u003csub\u003e2\u003c/sub\u003e hybrid nanostructures prepared by an impinging jet reactor for hydrogen evolution reaction catalysis. J Environ Chem Eng 10:108038\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eObaid M et al (2020) Breakthroughs in the fabrication of electrospun-nanofiber-supported thin film composite/nanocomposite membranes for the forward osmosis process: A review. Crit Rev Environ Sci Technol 50:1727\u0026ndash;1795\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePinilla JL, Pur\u0026oacute;n H, Torres D, Suelves I, Millan M (2015) Ni-MoS\u003csub\u003e2\u003c/sub\u003e supported on carbon nanofibers as hydrogenation catalysts: Effect of support functionalisation. Carbon N Y 81:574\u0026ndash;586\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCui J et al (2024) Carbon-anchoring synthesis of Pt1Ni1@Pt/C core-shell catalysts for stable oxygen reduction reaction. \u003cem\u003eNat. Commun.\u003c/em\u003e 15, 1\u0026ndash;9 (2024)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJagadale A, Zhou X, Blaisdell D, Yang S (2018) Carbon nanofibers (CNFs) supported cobalt- nickel sulfide (CoNi\u003csub\u003e2\u003c/sub\u003eS\u003csub\u003e4\u003c/sub\u003e) nanoparticles hybrid anode for high performance lithium ion capacitor. Sci Rep 8:1602\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou S et al (2024) Highly active single-layer 2H-MoS\u003csub\u003e2\u003c/sub\u003e for CO\u003csub\u003e2\u003c/sub\u003e hydrogenation to methanol. Appl Catal B Environ Energy 349:123870\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu J et al (2023) Solution combustion synthesis of CoS\u003csub\u003ex\u003c/sub\u003e, MoS\u003csub\u003ex\u003c/sub\u003e, CoS\u003csub\u003ex\u003c/sub\u003e-MoS\u003csub\u003ex\u003c/sub\u003e and their catalytic activity in H\u003csub\u003e2\u003c/sub\u003e production from H\u003csub\u003e2\u003c/sub\u003eS decomposition. Int J Hydrogen Energy 48:8807\u0026ndash;8818\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHussain S et al (2016) Large-area, continuous and high electrical performances of bilayer to few layers MoS\u003csub\u003e2\u003c/sub\u003e fabricated by RF sputtering via post-deposition annealing method. Sci Rep 2016 61(6):1\u0026ndash;13\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu Q et al (2019) Unsaturated Sulfur Edge Engineering of Strongly Coupled MoS\u003csub\u003e2\u003c/sub\u003e Nanosheet\u0026ndash;Carbon Macroporous Hybrid Catalyst for Enhanced Hydrogen Generation. Adv Energy Mater 9:1802553\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng Z et al (2020) Boosting hydrogen evolution on MoS\u003csub\u003e2\u003c/sub\u003e via co-confining selenium in surface and cobalt in inner layer. \u003cem\u003eNat. Commun.\u003c/em\u003e 11, 1\u0026ndash;10 (2020)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOzaki F et al (2023) Hydrogen-induced sulfur vacancies on the MoS\u003csub\u003e2\u003c/sub\u003e basal plane studied by ambient pressure XPS and DFT calculations. \u003cem\u003eChemPhysChem\u003c/em\u003e 24, e202300477\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKibsgaard J, Chen Z, Reinecke BN, Jaramillo TF (2012) Engineering the surface structure of MoS\u003csub\u003e2\u003c/sub\u003e to preferentially expose active edge sites for electrocatalysis. Nat Mater 11:963\u0026ndash;969\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng X, Zhu Y, Sun Y, Jiao Q (2018) Hydrothermal synthesis of MoS\u003csub\u003e2\u003c/sub\u003e with different morphology and its performance in thermal battery. J Power Sources 395:318\u0026ndash;327\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu S et al (2012) Preparation of MoS\u003csub\u003e2\u003c/sub\u003e nanofibers by electrospinning. Mater Lett 73:223\u0026ndash;225\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDuan Y et al (2022) Embedding of ultra-dispersed MoS2 nanosheets in N,O heteroatom-modified carbon nanofibers for improved adsorption of Hg2+. Compos Commun 31:101106\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuan J et al (2022) Electrospinning derivative fabrication of sandwich-structured CNF/Co\u003csub\u003e3\u003c/sub\u003eS\u003csub\u003e4\u003c/sub\u003e/MoS\u003csub\u003e2\u003c/sub\u003e as self-supported electrodes to accelerate electron transport in HER. Int J Hydrogen Energy 47:14930\u0026ndash;14941\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNewville M (2001) IFEFFIT : interactive XAFS analysis and FEFF fitting. urn:issn:0909 \u0026ndash; 0495 8:322\u0026ndash;324\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRavel B (2001) ATOMS: crystallography for the X-ray absorption spectroscopist. urn:issn:0909 \u0026ndash; 0495 8:314\u0026ndash;316\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRavel B, Newville MATHENA (2005) ARTEMIS, HEPHAESTUS: data analysis for X-ray absorption spectroscopy using IFEFFIT. urn:issn:0909 \u0026ndash; 0495 12:537\u0026ndash;541\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKresse G, Furthm\u0026uuml;ller J (1996) Efficient iterative schemes for \u003cem\u003eab initio\u003c/em\u003e total-energy calculations using a plane-wave basis set. Phys Rev B 54:11169\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBl\u0026ouml;chl PE (1994) Projector augmented-wave method. Phys Rev B 50:17953\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerdew JP, Burke K, Ernzerhof M (1996) Generalized Gradient Approximation Made Simple. Phys Rev Lett 77:3865\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrimme S, Antony J, Ehrlich S, Krieg H (2010) A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J Chem Phys 132:154104\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMonkhorst HJ, Pack JD (1976) Special points for Brillouin-zone integrations. Phys Rev B 13:5188\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"MoS2, H2S decomposition, electrospinning, hydrogen, nanofiber","lastPublishedDoi":"10.21203/rs.3.rs-8597696/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8597696/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eHydrogen sulfide (H\u003csub\u003e2\u003c/sub\u003eS) decomposition offers a carbon-neutral route for hydrogen production, but it remains limited by sluggish kinetics and catalyst deactivation. Here, we report an electrospinning\u0026ndash;sulfidation strategy to engineer a confined molybdenum disulfide (MoS\u003csub\u003e2\u003c/sub\u003e) catalyst with stable edge-rich active sites. During electrospinning, confined Mo\u0026ndash;S\u003csub\u003e4\u003c/sub\u003e\u0026sup2;⁻ complexes are dispersed within the nanofibers, which then nucleate into MoS\u003csub\u003e2\u003c/sub\u003e during calcination. The sulfidation stage induces exfoliation, forming highly dispersed MoS₂ domains that interact strongly with the carbon nanofibers. The prepared catalyst has a high surface area (~\u0026thinsp;180 m\u003csup\u003e2\u003c/sup\u003e g⁻\u003csup\u003e1\u003c/sup\u003e), abundant sulfur-vacancy edge sites, and strong support interactions, stabilized by coordinated Mo atoms. In-situ spectroscopy and ab-initio calculations reveal that these interfaces facilitate H\u003csub\u003e2\u003c/sub\u003eS dissociation, leading to a 3-fold higher intrinsic rate and improved long-term stability (\u0026gt;\u0026thinsp;50 h at 973 K) compared to bulk analogues. This work establishes design principles for fabricating grafted, stable, and highly dispersed sulfide catalysts.\u003c/p\u003e","manuscriptTitle":"Engineering edge-rich, highly dispersed, and stable MoS₂ sites for the catalytic splitting of H₂S for H₂ production","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-04 12:48:17","doi":"10.21203/rs.3.rs-8597696/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d1b8e173-9360-405c-8ec4-4f74e739e6ba","owner":[],"postedDate":"March 4th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":62165207,"name":"Physical sciences/Chemistry/Catalysis/Heterogeneous catalysis"},{"id":62165208,"name":"Physical sciences/Engineering/Chemical engineering"}],"tags":[],"updatedAt":"2026-03-04T12:48:17+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-04 12:48:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8597696","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8597696","identity":"rs-8597696","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}