Insight into Mo2C Nanoplates Growth by Means of Scanning Electrochemical Cell Microscopy for High-Resolution Electrocatalytic Hydrogen Evolution Activity Mapping

Insight into Mo 2 C Nanoplates Growth by Means of Scanning Electrochemical Cell Microscopy for High-Resolution Electrocatalytic Hydrogen Evolution Activity Mapping Geovane Arruda de Oliveira, David E. Sanchez, Alexander J. Sredenschek, Andres Fest, Moonjoo Kim, Carla Santana Santos, Mauricio Terrones, Wolfgang Schuhmann,* and Daniel Grasseschi* 1. Introduction Transition metal carbides (TMCs) are a group of interstitial compounds formed between group IV–VI transition metals and carbon. TMCs exhibit a combination of physical properties typically found in metals and ceramics, such as high electrical conductivity, melting point, and hardness. [1] Molybdenum carbides (Mo 2 C) have garnered significant attention due to their remarkable physicochemical properties, [2] making them suitable for various applications, including energy storage, [3] sensing, [4] and catalysis. [5–14] Mo 2 C is a promising electrocatalyst for the hydrogen evolution reaction (HER) due to its high catalytic activity and stability in both acidic and alkaline environments. [5,6,8,15] Various studies have shown that Mo 2 C nanomaterials, especially those with engineered surfaces and structures, exhibit enhanced HER performance. For example, Bang et al. demonstrated that a structure with multiple step-edges of G. A. de Oliveira, M. Kim, C. Santana Santos, W. Schuhmann, D. Grasseschi Analytical Chemistry-Center for Electrochemical Sciences (CES) Faculty of Chemistry and Biochemistry Ruhr University Bochum Universit¨ atstr. 150, 44780 Bochum, Germany E-mail: [email protected]; dgrasses[email protected] D. E. Sanchez, A. Fest, M. Terrones Department of Materials Science and Engineering The Pennsylvania State University University Park, PA 16802, USA The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/smsc.202500220. © 2025 The Author(s). Small Science published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. DOI: 10.1002/smsc.202500220 A. J. Sredenschek, M. Terrones Department of Physics The Pennsylvania State University University Park, PA 16802, USA M. Terrones Center for 2-Dimensional and Layered Materials The Pennsylvania State University University Park, PA 16802, USA M. Terrones Department of Chemistry The Pennsylvania State University University Park, PA 16802, USA D. Grasseschi Department of Inorganic Chemistry Chemistry Institute Federal University of Rio de Janeiro Av. Pedro Calmon, 550, Cidade Universitária, Rio de Janeiro 21941-901, Brazil Molybdenum carbides have emerged as promising catalysts for the hydrogen evolution reaction (HER). While numerous studies have investigated synthesis methods, structural properties, and their application, the understanding of their local electrochemical behavior and the correlation between particle size and activity remains elusive. This study addresses this gap by carrying out a comprehensive investigation of the HER activity of well-defined morphologies and sizes of α-Mo 2 C nanoplates, grown via chemical vapor deposition. Scanning electrochemical cell microscopy (SECCM) is employed for high-resolution HER mapping on flakes with dimensions ranging from 1 μmto40μm in lateral size, using SECCM capillaries with 130 nm tip diameter. Our findings reveal a significant variability in the HER activity at the subparticle level, suggesting that the heterogeneous activity observed in pristine flakes larger than 10 μm is due to an addition of effects caused by the long-term growth, such as step-edge formation, Mo 2 C oxidation, and the presence of residual graphene. This study underscores the importance of local characterization of individual Mo 2 C nanoplates, shedding light on the impact of size-dependence on the HER activity. RESEARCH ARTICLE www.small-science-journal.com Small Sci. 2025,5, 2500220 2500220 (1 of 13) © 2025 The Author(s). Small Science published by Wiley-VCH GmbH α-Mo 2 C nanosheets achieved efficient HER activity across a wide pH range. [9] In addition to HER, Mo 2 C is also reported as a catalyst for the oxygen reduction reaction (ORR). [10–12] Zhang et al. pointed to the potential of quasi-paired Pt atomic sites on Mo 2 C for running ORR with high selectivity for the four-electron transfer reaction to water, [10] with valuable performance for fuel cell applications. Furthermore, Mo 2 C has also been explored for electrochemical CO 2 reduction reaction. [13,14] Likith also discussed the increased thermodynamic stability of orthorhombic Mo 2 C due to the formation of molybdenum oxycarbides during the CO 2 reduction reaction. [14] These studies have shed light on the exploration of Mo 2 C as a promising catalyst for energy conversion reactions. Structurally, the stability and performance of Mo 2 C can be enhanced through precise control of the material size and phase composition. [16–18] However, there is a lack of a deeper understanding of the relationship between material size and electrochemical activity. Generally, thinner and larger 2D Mo 2 C nanoplates exhibit a higher surface area-to-volume ratio, leading to a higher number of active sites on the basal plane. Recent efforts have focused on implementing a bottom-up or top-down approach to synthesize 2D Mo 2 C. The former are known as nonlayered ultrathin TMCs (UThTMCs) and the latter are known as layered MXenes. In this context, Wu et al. implemented a bottom-up salt-assisted, methane carburization route to synthesize ultrathin Mo 2 C. [7] Microcell devices probed selected regions of the Mo 2 Cflakes and a high HER performance was measured on the edge and on thin flakes (0.36 nm). [7] Xu et al. developed a bottom-up, liquid-metal-assisted chemical vapor deposition (LMCVD) process to synthesize faceted, highly crystalline, UThTMC nanoplates with a thickness down to a few nanometers (3.4 nm) and tens of micrometers lateral area. [2] Geng et al. grew ultrathin Mo 2 C, on copper-molybdenum foils using LMCVD, with good HER performance. [19] While the HER performance of 2D Mo 2 C has been previously studied, the lateral size-activity relationship has not been considered since conventional electrochemical measurements typically involve measuring a relatively large sample area, where a wide size distribution of materials is considered. Therefore, we employed scanning electrochemical cell microscopy (SECCM) to locally characterize individually the electrochemical activity of diverse sizes of Mo 2 C. SECCM has emerged as a pivotal technique for investigating local electrochemical properties at the nanoscale [20–23] mainly applied for acquiring electrocatalytic activity mapping of nanoparticles [23–25] and 2D materials [26–32] such as graphene [26] and transition metal dichalcogenides. [33–35] SECCM can reveal differences in activity at the nanoscale and with subparticle measurements, being capable of resolving the electrochemical behavior within individual sections inside a single nanoparticle. [25,36] This technique enables the correlation of physical and chemical properties with catalytic performance, which is essential for optimizing catalysts at the nanoscale. Advanced microscopy techniques such as SECCM are crucial for gaining deeper insights into nanoscale electrochemical processes, guiding the optimization of synthesis methods for enhanced catalytic performance. Recently, SECCM was applied to explore the electrochemical behavior of a MXene. [37] This study employed cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) to investigate the properties of a Ti 3 C 2 T x MXene. The findings indicated that the pseudocapacitive behavior of Ti 3 C 2 T x is primarily attributed to its unique surface chemistry and structure, which facilitate fast and reversible redox reactions. The results demonstrated a significant increase in capacitance with higher scan rates, highlighting the dominance of surface-controlled processes. Additionally, the EIS data revealed low charge transfer resistance and high capacitance, further confirming the efficient pseudocapacitive performance of Ti 3 C 2 T x . To the best of our knowledge, no studies have used SECCM to understand the local electrochemical properties of Mo 2 C UThTMC nanoplates of different sizes. Therefore, SECCM was employed systematically to provide nanoscale mappings of the HER activity of Mo 2 C UThTMC nanoplates in acidic media. A comprehensive study of the HER activity of single hexagonal Mo 2 C nanoplates with sizes ranging from 2 to 50 μm in lateral size was performed using nanoscale SECCM probes with 130 nm tip diameter. Our findings reveal significant variability in HER activity at the subparticle level, with nanoplates smaller than 10 μm exhibiting higher HER activity than larger nanoplates (>10 μm). We attributed this behavior to a convolution of effects: size-dependent formation of step-edges and native Mo oxide during the CVD growth, and the presence of residual graphene. The correlation between the electrochemical activity of Mo 2 C with the material size was supported by atomic force microscopy (AFM), transmission electron microscopy (TEM), energy-dispersive X-ray spectroscopy (EDX), Raman spectroscopy, X-ray diffraction (XRD), and X-ray photoelectron spectroscopy (XPS). 2. Results and Discussion The size dependence of the electrochemical activity of Mo 2 C nanoplates at the subparticle level was systematically investigated by employing a high-resolution SECCM to map the local electrochemical activity towards HER. Mo 2 C nanoplates were grown via a controllable atmospheric pressure liquid metal chemical vapor deposition process, [2] as detailed in the supporting information. As depicted in Figure S1, Supporting Information, Mo atoms diffuse through the liquid copper foil and react with methane on the Cu surface. The copper foil thickness, methane-hydrogen concentration, and growth duration serve as the basis to control the growth of specific lateral sizes, nucleation density, and thickness of the Mo 2 C nanoplates. [38] Optical microscopy and scanning electron microscopy (SEM), were employed to evaluate the effect of copper thickness, methane concentration and growth time, and tune the synthesis parameters to achieve the desired nanoplates’lateral size and thickness. Optical microscopy and SEM imaging provided a distribution of lateral lengths for the nanoplates, as illustrated in Figure 1 and S2, Supporting Information. The predominant morphology of Mo 2 C appears to be hexagonal prisms with a lateral side length varying from 2 μmto50μm. Two samples were selected for the electrochemical study and described in the supporting information. Summarizing, Sample I was prepared with parameters to induce flakes with small size, resulting in Mo 2 C nanoplates with lateral size distribution of 5.3 2.4 μm (Figure 1A,B). While Sample II was grown by maintaining www.advancedsciencenews.com www.small-science-journal.com Small Sci. 2025,5, 2500220 2500220 (2 of 13) © 2025 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 2025, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202500220 by Ruhr-Universität Bochum, Wiley Online Library on [04/11/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License parameters to achieve large lateral-sized flakes. Sample II Mo 2 C flakes presented a size distribution of 14.0 12.8 μm (Figure 1C,D). Note that Sample II growth protocol provided a broad size distribution with the presence of small (lateral sizes <10 μm) and large flakes (>10 μm). Interestingly, the formation of step-edges in Sample II was observed frequently, which was probably provoked due to an excess of molybdenum on the surface due to the long growth time. [39] The thickness of the two samples of Mo 2 C nanoplates was determined by atomic force microscopy (AFM) on nanoplates transferred to thermally oxidized silicon (SiO 2 /Si) substrates, utilizing the wet chemical transfer method (as outlined in the Supporting Information). The nanoplate thickness ranged from 40 nm to 150 nm, demonstrating the formation of ultrathin nanoplates (Figure S3, Supporting Information), depending on the processing conditions. AFM data revealed a correlation between the lateral size and the thickness of the nanoplates, where the thickness was inversely proportional to the nanoplate’s size. Specifically, the Mo 2 Cflakes on Sample II, with lateral size >10 μm, were observed to be thinner than the nanoplates in Sample I, with lateral size <10 μm. This could be related to the vertical and lateral growth modes studied by Buke et al. with varying the copper thicknesses and growth time. [38] We note, however, that since the copper foils are stacked on to the molybdenum foil, there is a challenge in maintaining a uniform flux of molybdenum across the entire substrate which could lead to a distribution of nanoplate thicknesses. 2.1. Correlative Mapping of Topographic and Electrochemical Properties of the Sample I and Sample II Flakes The electrochemical activity SECCM mapping was performed by recording voltammograms to follow the local HER. The SECCM tip was filled with 10 mmol L 1 HClO 4 , and the map was performed on flakes of the Sample I (Figure 2) and Sample II (Figure 3). The HER was selected as an inner-sphere electron transfer reaction, since it is highly dependent on the physicochemical properties of the catalyst surface, making it a valuable probe for studying structure-activity relationships. High-resolution HER current maps of single Mo 2 Cflakes were obtained using nanoscale SECCM probes with 130 nm tip diameter, as seen in Figure S4, Supporting Information, according to the methodology detailed in the supporting information. AFM topography image performed after the SECCM measurements reveals that the SECCM droplet size had an effective spot diameter of 214 1 nm (Figure S5, Supporting Information), confirming that the SECCM resolution covered subparticle resolution. SECCM can also provide a topographical image of the sample during the acquisition of electrochemical activity maps. This feature allowed us to correlate topographical SECCM, morphological (SEM), topographical AFM data, and electrochemical (SECCM) information, respectively, as shown in Figure 2A–D. As depicted in Figure 2A–C by morphological and topographical imaging, two out of the three Mo 2 C nanoplates of Sample I Figure 1. Analysis performed on A,B) Sample I and C,D) Sample II. A,C) Morphology is seen from optical microscopy and B,D) distribution of Mo 2 C lateral size. www.advancedsciencenews.com www.small-science-journal.com Small Sci. 2025,5, 2500220 2500220 (3 of 13) © 2025 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 2025, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202500220 by Ruhr-Universität Bochum, Wiley Online Library on [04/11/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Figure 2. Morphological and local electrochemical characterization of Mo 2 C nanoplates grown on sample I. A) SECCM topographic mapping constructed from the z-piezo extension. B) SEM image using secondary electrons and electron acceleration voltage of 5 kV. C) AFM topography image acquired in the noncontact mode. D) SECCM mapping of HER current at 0.6 V versus RHE. E) Tafel slope mapping calculated between 0.40 and 0.55 V vs RHE. F) AFM height profile extracted from lines 1 and 2 in (C). Representative LSV curves for G) flake 1, H) flake 2, and I) flake 3 are numbered in (D). The (x,y) coordinates from where the LSV curves were extracted are shown in the legend. LSV curves were obtained in 10 mmol L 1 HClO 4 with scan rate of 2 V s 1 SECCM image was acquired with a hopping distance of 500 nm and SECCM tip of 130 nm diameter. Figure 3. SECCM mappings for Mo 2 C nanoplates in Sample II showing A) the HER current at 0.6 V versus RHE and C) the Tafel slopes calculated between 0.4 and 0.55 V versus RHE acquired with a hopping distance of 1 μm. B) Current and D) Tafel slope histograms were extracted for the big flake (flake 4, orange bars) and small flake (flake 5, blue bars), respectively. The inset in (A) shows the optical microscopy image of the measured flakes. HR-TEM of α-Mo 2 C nanoplates at E,F) the basal plane and G,H) the cross-section plane. www.advancedsciencenews.com www.small-science-journal.com Small Sci. 2025,5, 2500220 2500220 (4 of 13) © 2025 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 2025, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202500220 by Ruhr-Universität Bochum, Wiley Online Library on [04/11/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License displayed a hexagonal shape, while one appeared rectangular. These flakes were also visualized in the central bottom part of Figure 1A. Figure 2D,E demonstrates the capability of SECCM to locally measuring the HER electrochemical activity covering nanoplates and subparticle information, while differentiating the activity of the basal plane and the edges. Regardless of the nanoplate geometry, all three Mo 2 Cflakes measured in Sample I have lateral sizes below 10 μm and exhibited smooth basal planes with low surface roughness without step-edges. This observation was confirmed by SEM and AFM analyses (Figure 2B,C, respectively), revealing flat surfaces with edges “buried”in the Cu substrate. The AFM height profile (Figure 2F) illustrates that the edges of both hexagonal nanoplates (depicted by the red line) are notably more prominent than the basal plane, evident from the increased height at the edges, also visualized in the SECCM topographical mapping (Figure 2A). Additionally, the hexagonal Mo 2 C nanoplates’basal plane surface lies below the Cu surface because the growth involves Mo diffusion through Cu. In contrast, the behavior of the rectangular nanoplate is distinct, growing “out”of the copper surface, as shown by the black line in Figure 2F. AFM and SEM revealed particulate features at the nanoplate edges attributed to copper oxide particles. Raman spectroscopy (Figure S6, Supporting Information) shows the presence of particles on the edges of Mo 2 Cflakes which match the Raman signature of copper oxide. STEM-EDS mapping of transferred and isolated flakes in Figure S7, Supporting Information, shows copper oxide particles present on the basal plane which we attribute to the incompletely etched copper substrate. The absence of copper oxide particles on the edges of the Mo 2 Cflakes is consistent with the interpretation that these particles originate from the oxidized copper substrate and are not intrinsic features of the Mo 2 C flakes. The AFM and SEM images align with the topographical SECCM image in Figure 2A. The HER activity of Mo 2 C nanoplates in Sample I (Figure 2D), appeared relatively uniform, suggesting homogeneity across the basal plane. The current range for all three nanoplates was about 800 pA at 0.6 V vs RHE, indicating that differences in flake morphology do not significantly influence electrochemical activity. The short growth time used in Sample I also resulted also in a homogeneous copper surface, observed in the activity SECCM mapping (yellow color—current range <50 pA), indicating uniform but relatively poor HER activity, accentuating the contrast compared to the active small Mo 2 C nanoplates. Representative linear sweep voltammetry curves (LSVs) of each flake were selected and plotted in Figure 2G–I, to delve deeper into the local electrochemical behavior. Figure 2G–I illustrates findings for the rectangular-shaped Mo 2 C nanoplate (flake 1), the central hexagonal Mo 2 C nanoplate (flake 2), and the right hexagonal Mo 2 C nanoplate (flake 3), respectively. For all three flakes, a consistent pattern emerged: Both the edge and basal plane of the flakes demonstrated significantly enhanced performance, characterized by elevated currents and improved HER activity when compared with the Cu surface. As previously discussed, at the nanoplates’edges, there is an interface between the Mo 2 C nanoplates and the Cu substrate that exhibits a distinct topographical profile. The LSV curves extracted from the edges exhibit lower HER activity (Figure 2G–I), probably resulting from SECCM tip landing on the heterogeneity. This reduced activity is primarily attributed to the growth mechanism, where Mo diffuses through the Cu foil during synthesis. It is important to note that the copper oxide particles in the nanoplates’edge do not dominate the HER activity (Figure 2D). The rectangular Mo 2 Cflake presented in Figure 2 offers additional evidence that residual copper does not block the catalytically active sites at the edges. In this case, the flake clearly grows out of the Cu substrate, with more exposed edges, as confirmed by the corresponding AFM image in Figure 2B,C, which shows increased flake height and minimal particle coverage at the periphery. Notably, the electrochemical activity of this rectangular flake is comparable to that of neighboring hexagonal flakes, as shown in the SECCM current and Tafel slope maps (Figure 2D,E), suggesting that the edge activity is not significantly impeded by copper oxide clusters. Additionally, Tafel analysis conducted between 400 and 550 mV versus RHE (Figure 2E) revealed a Tafel slope close to 70 mV dec 1 for all three analyzed flakes, thus confirming uniformity irrespective of geometry. Notably, the difference between some spots at the edge became more apparent in the Tafel plot, displaying higher and random Tafel slope values in Figure 2E. All Tafel plots and their respective regions can be found in Figure S8, Supporting Information. It is important to note that Tafel analysis was utilized to enhance image contrast and flake visualization, facilitating the evaluation of the size-activity relationship, rather than for mechanistic investigation or quantitative kinetics analysis. This consideration arises because visualizing nanoplates smaller than 10 μm in lateral size poses challenges, particularly when displaying topographical/activity mapping. Figure S9–S11, Supporting Information, present SECCM mappings of the HER activity for several Mo 2 Cflakes of Sample I exhibiting similar behavior as discussed in Figure 2: namely, displaying a homogenous electrochemical activity of the basal plane with consistent and similar current values of individual flakes. These results indicate that the difference between flakes’morphology has no significant influence on the electrochemical activity for Sample I. Subsequently, LSV curves were conducted at different scan rates (1, 2, and 4 Vs 1 ) in a 10 mmol L 1 HClO 4 solution (see Figure S8I–K, Supporting Information). As expected, the limiting HER current remained at the same values, and variations in the scan rate did not affect the overpotential for different selected regions of the flake. Using the mass transport limited-current and the wetted footprint of the SECCM meniscus (Figure S5, Supporting Information) the expected current density is around 4.8 A cm 2 . Hence, our localized currents (800 pA) are indeed consistent with A cm 2 magnitude. This reaffirms that our localized currents reside in the A cm 2 range, comparable to current densities in macroscopic HER benchmarks, [5] confirming the capability of SECCM to study HER with high current density due to high mass transport. Conversely, when the electrolyte concentration was increased from 10 to 100 mmol L 1 HClO 4 , resulting in a decrease in pH, an increase in current was observed (see Figure S8L, Supporting Information). This increment was also noted for the Cu surface, which was attributed to the available H þ concentration. These results indicate that no other process than HER occurs on the surface of pristine small Mo 2 C nanoplates. www.advancedsciencenews.com www.small-science-journal.com Small Sci. 2025,5, 2500220 2500220 (5 of 13) © 2025 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 2025, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202500220 by Ruhr-Universität Bochum, Wiley Online Library on [04/11/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License It is noteworthy that commonly 0.5 mol L 1 H 2 SO 4 is employed as electrolyte solution to evaluate the HER activity of Mo 2 C nanoplates, with reported high current densities and overpotentials ranging between 200 and 400 mV versus RHE. [5,8] However, we deliberately chose to work with a diluted HClO 4 solution to decrease the material’s activity, accentuating the local activity differences in the basal plane and between flakes of different sizes. The adopted condition also improved the stability of SECCM measurements by mitigating evaporation and ionic strength effects. [40] To investigate the effect of size on the electrochemical activity of Mo 2 C nanoplates, SECCM measurements were conducted on Sample II (Figure 3). A contrasting trend emerged in SECCM experiments on Sample II compared to Sample I. Figure 3 depicts two distinct-sized flakes in the same SECCM activity mapping: flake 4 on the left side of Figure 3A with 45 μm width and flake 5 (right side of Figure 3A) with 15 μm size. The SECCM map revealed that flake 4 exhibited heterogeneous and lower HER activity than the smaller flake 5. The basal plane of flake 4, which exhibits non-uniform activity as illustrated in Figure 3A, can be segmented into three distinct regions based on the observed current according to the histogram in Figure 3B. The yellow regions (region 1—red square in Figure 3A), scattered randomly across the basal plane, evidencing the activity heterogeneity, with currents <|50|pA. Analyzing the behavior of the light and dark blue regions of the basal plane, representing regions 2 and 3 respectively, a transition from a non-uniform to a more uniform current range is observed. Specifically, region 3 demonstrates a more uniform current at 200 pA at 0.6 V versus RHE. This is an increase of around 170 pA at 0.6 V versus RHE compared to the yellow areas in region 1. The representative LSV curves of each region are depicted in Figure S12, Supporting Information. Tafel analysis was employed to enhance comprehension and visualization of the Mo 2 C nanoplates activity. The hexagonal structure of the nanoplates is readily discernible for both flakes 4 and 5 (Figure 3C). However, flake 4 exhibited non-uniform Tafel slopes spanning a wide range from 50 to 150 mV dec 1 (Figure 3D), indicating the heterogeneity of the local activity. It is important to note that for flake 5 in Figure 3, the activity exhibited higher levels of inhomogeneity compared to flakes of similar lateral size in Sample I (Figure 2 and S8–S11, Supporting Information). This observation was subsequently confirmed by SECCM mapping and corroborated by the Tafel slope mapping in Figure 3A,C, respectively. Specifically, the activity of flake 5 demonstrated a Tafel slope of 80 mV dec 1 for the basal plane. This finding underscores the significant influence of growth conditions on the Mo 2 C surface characteristics. It is worth noting that the copper substrate in Sample II seems to exhibit a non-uniform and rough surface, as evidenced by two colored regions with distinct HER activity (Figure 3A). However, the scale bar in Figure 3A indicates a similar current range of the copper surface obtained in Figure 2. The low HER activity of the Mo 2 C nanoplates in Sample II, the high subparticle heterogeneity, and the high roughness of the copper surface hampers the clear visualization of Mo 2 C nanoplates of Sample II in the SECCM electrochemical mapping. Moreover, as showed in the AFM analysis in Figure S3, Supporting Information, larger nanoflakes resulted in a reduced thickness, making it difficult to visualize the nanoplates in the topographic SECCM image. Figure S13, Supporting Information shows AFM topography and height profile of Mo 2 Cflakes of Sample II, revealing the presence of step-edges, which vary in height between 15 nm and 25 nm. Interestingly, reduced HER activity was discernible at the center of flake 4, indicated by the red arrow in Figure 3A. Tafel plot imaging explicated the well-shaped-low activity region (see Figure 3C), indicating the presence of the step-edge on the center of the flake, as seen in the optical microscopy image in the inset of Figure 3A. When the SECCM tip landed on the stepedge, the majority of the SECCM meniscus possibly contacted either the basal plane of the terrace and bottom layer, as well as the crystalline plane of the step-edge. In other words, SECCM mapping for flake 4 has a higher heterogeneity, probably due to the presence of the step-edges that expose different planes. Planar HR-TEM images in Figure 3E,F reveal the Mo 2 C (021) planes with an interplanar distance of 0.257 nm, which are perpendicular to the (200) planes, as shown in Figure 3F. This observation confirms the presence of the (200) planes on the Mo 2 C basal surface. Additionally, cross-sectional TEM images in Figure 3G,H show the (200) planes with an interplanar distance of 0.242 nm, demonstrating that the (001) planes, which are perpendicular to the (200) planes, are exposed laterally in the nanoplate. Figure S14, Supporting Information, reveals that the observed copper oxide particles were removed during the sample transfer to the TEM grid. Consequently, the edge structures observed in TEM represent the clean, intrinsic morphology of the highly crystalline Mo 2 C nanoplates. Therefore, the HER activity at the step-edges reflects the average response of both (200) and (001) planes. The (001) surface exhibits Mo terminations with the C atoms in a sublayer, while the (200) surface features mixed Mo and C terminations. Wang et al. calculated a more negative hydrogen adsorption energy for the (001) planes at low H coverage compared to the (200) planes using density functional theory. [15] Conversely, at high H coverage, both surfaces showed nearly thermoneutral H adsorption. [15] Hence, we anticipate that at low H þ concentrations, the (001) and (200) planes would exhibit distinct HER activity, with the (200) being more active than the (001), thus elucidating the low activity sites at the step-edges observed in the SECCM activity mapping in Figure 3A. Furthermore, since both flake edges and step-edge features expose the (001) crystallographic planes, it is reasonable to conclude that the flake edges and the step edges would exhibit similar intrinsic HER activity, further supporting our interpretation that copper residues do not enhance or dominate the observed catalytic behavior of the flake’s edges. However, the presence of only one step edge in flake 4 depicted in Figure 3A, along with the observation that nanoplates in Sample II without step edges also exhibit a heterogeneous HER response (refer to Figure S15, Supporting Information), suggests that the distinct local electrochemical behavior of flakes synthesized with longer growth times (Sample II) cannot solely be attributed to the presence of step edges. Figure 4A presents an AFM topographic image, acquired on Sample II, showcasing three different Mo 2 C nanoplates, indicated as flakes 6, 7, and 8. Figure 4D displays the height profile extracted from the AFM topography image in Figure 4A for the www.advancedsciencenews.com www.small-science-journal.com Small Sci. 2025,5, 2500220 2500220 (6 of 13) © 2025 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 2025, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202500220 by Ruhr-Universität Bochum, Wiley Online Library on [04/11/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License three flakes. Flake 6 in Figure 4D has a lateral size of 20 μm and deeper growth into the copper surface, evident from a significant valley in the AFM height profile. In contrast, flake 7 exhibits a lateral size of 8μm and edges “buried”in the Cu surface, like the nanoplates in Sample I. Interestingly, flake 8 in Figure 4D reveals the presence of vertically aligned nanoplates, with heights between 400 nm and 600 nm above the copper surface, and a clear elevation of the Cu surface. The topographical SECCM image in Figure S16, Supporting Information was aligned with the AFM analysis, revealing a higher Z extension on the vertical flake. As anticipated, the 20 μmMo 2 C nanoplates (flake 6) exhibited poor and non-uniform activity, almost indistinguishable from the Cu surface (Figure 4B), with HER currents smaller than 70 pA (Figure 4F). The opposite trend was observed for flake 7, which displayed currents higher than 300 pA at 0.7 V versus RHE (Figure 4F). Conversely, flake 8, the vertical one, showed an intermediate trend with HER currents around 200 pA at 0.7 V versus RHE (Figure 4B,F). The Tafel slope image in Figure 4C shows that the three flakes can be distinguished from the Cu surface due to difference on Tafel slope versus the Cu surface. As discussed in Figure 3 and confirmed in Figure 4, Sample II exhibited a broad size distribution of Mo 2 C nanoflakes, with high heterogeneity in the electrochemical activity toward HER to the big flakes than to the small ones. Moreover, the small flake 7 presented a larger electrochemical activity than the big flake 6. The vertical flake (flake 8) displayed heterogeneous activity, with regions showing a similar overpotential as flake 7 (green curve in Figure 4E), and others with an overpotential of 550 mV versus RHE (light blue curve in Figure 4E). The vertical flake is expected to expose the (001) plane at the surface; thus, a similar activity as the step-edges observed in the horizontal larger flakes is anticipated. Figure S17–S19, Supporting Information, present SECCM HER mapping for vertically aligned Mo 2 Cflakes, showing a similar behavior. In these measurements, no Cu oxide nanoparticles are observed in either the AFM image (Figure 4A) or the corresponding SEM images (Figure S17 and S20, Supporting Information), allowing a clean assessment of the intrinsic activity of the exposed crystal planes. These measurements consistently demonstrate that the (001) planes exposed at the flake edges exhibit lower HER activity compared to the (200) basal planes, as revealed by the SECCM current and Tafel slope maps. This reinforces our interpretation that the decreased edge activity is not an artifact of surface contamination, but rather reflects the intrinsic electrocatalytic properties of the Mo 2 C crystal facets. However, we encountered challenges in mapping the electrochemical activity of the vertical flakes, such as: 1) Vertical nanoplates had small widths (500 nm— Figure S20, Supporting Information), making it easy to accidentally skip over them with a hopping distance of 500 nm or less, resulting in the loss of important information about their activity, and 2) visualization of vertical nanoplates was complicated due to more heterogeneous flakes of Sample II and copper surface (Figure S19, Supporting Information). Based on the SECCM measurements at the subparticle level in Sample II, it can be concluded that nanoplates larger than 20 μm in lateral size exhibit a heterogeneous and low HER activity, with currents lower than 70 pA (Figure 3 and 4). The behavior of flakes with sizes ranging between 10 and 20 μm in lateral size differs from larger flakes, showing current values around 200 pA across the entire basal plane, as depicted in the histogram in Figure 3B. This behavior is distinct from that of Sample I, where the shorter growth time led to flakes with sizes smaller than 10 μm and exhibiting currents around 1000 pA at 0.6 V versus RHE (see Figure 2). These findings underscore a clear size-electrochemical activity relationship for Mo 2 C nanoplates towards HER. Bulk electrochemical measurements performed in samples transferred to glassy carbon (GC) surface (Figure S21, Figure 4. A) AFM topography image, B) SECCM HER current, and C) Tafel slope mapping, acquired with a hopping distance of 1 μm, for Mo 2 C nanoplates in Sample II. D) Height profile for flakes 6, 7, and 8 extracted from the lines indicated in (A). E) Representative LSV curves of flakes 6, 7, and 7 extracted from the (x,y) coordinates of the current mapping in (B). F) Current histograms for flakes 6, 7, and 8 extracted from (B), only the pixels inside the flakes were considered. www.advancedsciencenews.com www.small-science-journal.com Small Sci. 2025,5, 2500220 2500220 (7 of 13) © 2025 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 2025, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202500220 by Ruhr-Universität Bochum, Wiley Online Library on [04/11/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Supporting Information) reveal no clear difference between the HER activity of different Mo 2 C samples, which does not reflect the size-dependent heterogeneity that is clearly resolved by the SECCM maps. This result is expected, as the low surface coverage of Mo 2 C and the ensemble averaging inherent to bulk measurements obscure the local variations in activity. More important, the LSV curves using standard methods did not achieve mass transport limited current, which is shown in the SECCM LSV curves (Figure 2G–I). 2.2. Comprehensive Structural Characterization of Mo 2 C Flakes The correlative AFM/SECCM mappings demonstrated a distinct correlation between the nanoflake size and the electrochemical activity of Mo 2 C. Furthermore, the observed heterogeneity in electrochemical activity for flakes synthesized with longer growth times (Sample II) could not be solely attributed to the presence of step edges. To elucidate the underlying differences in chemical composition between Sample I and Sample II, a comprehensive characterization was conducted, providing deeper insight into their respective physicochemical properties. XRD analysis was conducted in the Bragg–Brentano geometry on the as-grown Sample I and Sample II. The XRD diffractograms were plotted in logarithmic intensity scales to enhance the visualization of low-intensity peaks, as depicted in Figure 5A,B and S2C, Supporting Information. Our analysis revealed that the flakes consist of α-Mo 2 C (PDF 01-071-0242) with an orthorhombic unit cell, belonging to space group Pbcn, with a (200) surface orientation. This orientation is evidenced by the prominent peaks near 2θ=38.04° and 81.37° and 2θ=37.99° and 81.23° in Sample I and II, respectively. Additionally, flakes oriented along various planes were obtained, such as the (021), (121), (221), (023), and (321) planes, where the (200) plane is oriented perpendicular to the substrate surface. This orientation is facilitated by texturing between the (111) copper substrate and the vertically oriented flakes, enabling the observation of planes perpendicular to the (200) basal surface. In Sample II, the presence of an unidentified XRD peak at 2θ=37° was noted and additionally the presence of its second-order diffraction peak at 2θ=78.8° (Figure 5B and S2, Supporting Information). We hypothesized that the unknown XRD peaks correspond to crystalline (020) and (040) MoO 2 (PDF 01-073-1249) with a monoclinic unit cell belonging to space group P2 1 /c. The observation of MoO 2 from XRD indicates that it is likely textured along the (200) basal surface of α-Mo 2 C. Notably, the presence and intensity of MoO 2 exhibited a clear dependence on the synthesis parameters. Sample I, with Mo 2 C flakes with lateral sizes smaller than 10 μm lacked the attributed MoO 2 peak, as illustrated in Figure 5A. Conversely, Sample II, with nanoplates lateral sizes ranging from 2 μmto40μm, consistently exhibited MoO 2 peaks (Figure S2, Supporting Information). XRD diffractograms on a linear intensity scale in Figure S2, Supporting Information, evidenced the hypothesized MoO 2 peaks are of low intensity and low quantity, making it a challenge to detect additional XRD peaks corresponding to MoO 2 . Additionally, the linear scale highlights the sharpness of the Mo 2 C (200) peak more clearly, confirming the high crystallinity of the samples. Although the peaks at 2θ=37° and 78.8° Figure 5. XRD on as-grown substrates in log intensity scales for flakes in A) Sample I and B) Sample II. Chemical and structural characterization of Mo 2 C from C–F) Sample I with 9 μm lateral size and G–J) Sample II with 27 μm lateral width. HAADF-STEM and the corresponding EDS maps showing the distribution of D,H) molybdenum and E,I) carbon throughout the flakes. F,J) SAED acquired along the textured [200] of Mo 2 C. www.advancedsciencenews.com www.small-science-journal.com Small Sci. 2025,5, 2500220 2500220 (8 of 13) © 2025 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 2025, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202500220 by Ruhr-Universität Bochum, Wiley Online Library on [04/11/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License match with monoclinic MoO 2 peaks, we observed significant broadening, which could be related to the presence of molybdenum oxycarbides (MoO x C y ). To check the presence of a mixture of MoO x C γ , we simulated powder XRD patterns with x and y being 1:4 and 1:8 ratios between oxygen and carbon. The simulated diffractograms revealed that substituting oxygen into the Mo 2 C lattice increases the (200) and (400) interplanar distances, shifting the diffraction peaks to smaller angles, towards the extra peaks at 37° and 78.8° (Figure S22, Supporting Information). However, a high concentration of oxygen (1:4) results in interplanar distances larger than those associated with the unknown XRD peaks. This indicates that specific, low concentrations and favorable oxygen configurations within the Mo 2 C lattice are necessary to reproduce the experimentally observed XRD pattern. Based on both experimental and theoretical XRD analysis, Sample I likely contains a lower oxygen content, insufficient to produce a pronounced (010) MoO 2 peak. In contrast, Sample II exhibits a higher oxygen content, leading to the formation of a well-defined MoO 2 and low intensity peak in the XRD pattern. Mo 2 Cflakes were transferred onto a TEM grid utilizing the wet chemical transfer method (as outlined in the Supporting Information). Examination via low-magnification high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM), as depicted in Figure 5C,G, reveals the hexagonal nanoplate morphology of flakes with 9 μm (Sample I) and 27 μm (Sample II) in lateral length, respectively. Selected area electron diffraction (SAED) from both flakes, shown in Figure 5F,J, unequivocally confirms the high crystallinity of the synthesized α-Mo 2 C nanoplates, with the [200] orientation and the (200) plane prominently exposed at the basal surface. Notably, the SAED pattern reveals distinct strong and weak spots, both corresponding to the molybdenum sublattice, with the weak spots indicative of long-range periodic molybdenum lattice distortions induced by ordered carbon vacancies within α-Mo 2 C (Figure S23, Supporting Information). Furthermore, TEM analysis showed that the hexagonal nanoplate comprises six grains with facets corresponding to (001) planes, perpendicular to [001], or the zigzag plane of the crystal, as schematically depicted in Figure S23, Supporting Information. [41,42] Upon close inspection of SAED, we observed a notable difference in the spot pattern. In Sample II, extra spots were observed that could not be indexed to the forbidden reflections in α-Mo 2 C (Figure S24, Supporting Information). Furthermore, these extra spots appear to form a hexagonal pattern along with the primary spots. We hypothesize that these extra spots occur due to double diffraction and the presence of stacking faults within the structure. XPS was conducted directly on the as-grown samples. The survey scan reveals that Sample II contains molybdenum, carbon, and oxygen, as seen in Figure S25A, Supporting Information. The high-resolution Mo 3d spectrum, shown in Figure S25B, Supporting Information, consists primarily of an asymmetric doublet with Mo3d 5/2 at 228.13 eV, which corresponds to Mo 2 C. [13] A second contribution, with Mo3d 5/2 at 229.2 eV, is attributed to Mo 4þ presented as MoO 2[43] and was modeled considering the characteristic satellite features of MoO 2 . [43,44] The high-resolution C1s spectrum, shown in Figure S25C, Supporting Information, was deconvoluted into four components: 1) a peak at 284.7 eV, attributed to sp 3 carbon; 2) a peak at 284.4 eV, which corresponds to sp 2 hybridized carbon, originating from graphene present in the as-grown sample; 3) a peak at 283.3 eV, which is attributed to carbon in Mo 2 C; [13] and 4) a small peak at 288.3 eV, which corresponds to carbonyl groups. The total carbon/molybdenum ratio (C/Mo) was determined to be 10.8, due to the dominant presence of graphene. Importantly, when considering only the components assigned to Mo 2 C (Figure S25, Supporting Information) on each of the high-resolution regions, the C/Mo ratio is 0.57, in close agreement with the stoichiometry of Mo 2 C. We pointed out that prior to the SECCM measurements, the graphene formed on the Mo 2 C nanoplates was removed via mechanical exfoliation using a Kapton tape, as described in the supporting information (Figure S26, Supporting Information). However, certain flakes with larger lateral size could exhibit irregular surfaces with residual graphene, which may influence the electrochemical performance of these flakes (Figure S27, Supporting Information). To evaluate the possible composition variation with the lateral size, EDS mappings were acquired both in HAADF-STEM and SEM modes. STEM-EDS mapping shown in Figure 5 and S28–S29, Supporting Information reveals a uniform distribution of molybdenum and carbon in the nanoplates. In the SEM EDS mapping, due to low sensitivity and the effect of Cu substrate, just Mo and C were detected in the nanoplates mapped in the as-grown samples regardless of their sizes (Figure S30 and S31, Supporting Information). STEM-EDS profiles in Figure S29, Supporting Information, reveal a stronger oxygen signal near the edges of the nanoplates in Sample I, whereas in Sample II, the oxygen signal appears uniformly distributed across the basal plane. AFM data indicated a correlation between the lateral size and thickness of the nanoplates (Figure S3, Supporting Information). Specifically, for Sample I, grown with low methane concentrations and short growth times, the nanoplates have lateral sizes below 10 μm and thicknesses ranging from 80 to 150 nm, with oxide formation localized at the edges. Conversely, for Sample II, prepared with higher methane concentrations and longer growth times, the nanoplates exhibit a broader lateral size distribution (5–40 μm) but are thinner, with thicknesses below 50 nm, and oxide formation distributed uniformly across the basal plane. These findings suggest that native MoO 2 tends to form on the slower-growing surfaces. In short-growth-time samples, the growth predominantly occurs through the basal plane, resulting in smaller nanoplates with oxide localized at the edges. In contrast, longer growth times promote lateral growth, producing larger but thinner flakes with extensive oxide coverage across the basal plane. Spatially resolved Raman mapping on Sample II was conducted to directly correlate the dimensions and density of native oxide sites (Figure S32, Supporting Information). Characteristic Raman modes of orthorhombic Mo 2 C [45] were observed in Sample II. However, no Raman peaks related to MoO 2 in either the small or large lateral dimension nanoflakes were observed. Significant variations in the Mo 2 C peak intensity, position, and width were observed for the large flakes, indicating a heterogeneous crystalline structure. These variations were especially observed in the 143 cm 1 (B 3g ,A g ) Raman peak likely caused by structural defects or oxygen impurities, which can induce local strain in the crystalline structure. Defects such as stacking faults www.advancedsciencenews.com www.small-science-journal.com Small Sci. 2025,5, 2500220 2500220 (9 of 13) © 2025 The Author(s). Small Science published by Wiley-VCH GmbH 26884046, 2025, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smsc.202500220 by Ruhr-Universität Bochum, Wiley Online Library on [04/11/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License