Articles
Development of Iridium Clusters-Loaded Tungsten Disulfide Composite Catalyst with Poisoning Tolerance for Electrochemical Ammonia Oxidation
2025 年 93 巻 2 号 p. 027013
詳細
2025 年 93 巻 2 号 p. 027013
The electrochemical oxidation of ammonia (AOR) is a critical reaction for energy conversion and environmental remediation applications. However, the efficiency of AOR is significantly hindered by catalyst poisoning caused by adsorbed atomic nitrogen species (N,ad), which are generated during the reaction. In this study, a novel composite catalyst consisting of iridium clusters loaded onto tungsten disulfide (Ir/WS2) was synthesized and applied to AOR. Morphological and structural analyses confirmed the successful loading of metallic Ir clusters with an average diameter of 1.5 nm on WS2 nanobelts. Electrochemical measurements demonstrated that the Ir/WS2 catalyst exhibited superior activity and durability for AOR compared to conventional Ir clusters loaded onto activated carbon. This enhancement is attributed to the stable adsorption of NHx intermediate species on WS2, which suppresses N,ad formation via complete dehydrogenation and facilitates the dimerization of NHx intermediates. These findings highlight the potential of WS2-based composite materials for the development of efficient and durable AOR catalysts.
The electrochemical ammonia (NH3) oxidation reaction (AOR) has attracted significant attention due to its relevance in diverse applications, including energy conversion and environmental remediation.1–4 In the field of energy conversion, AOR serves as an anode reaction for the development of NH3 fuel cells without CO2 emission, which utilize NH3 as a hydrogen carrier and energy-dense fuel.5 Compared to hydrogen, NH3 is easier to store and transport, making it an appealing candidate for clean energy systems.6 In environmental applications, AOR provides an effective method for removing NH3 from aqueous solutions, addressing critical challenges in wastewater treatment and water purification.4 NH3 contamination in water sources is a significant environmental concern, contributing to eutrophication and posing toxicity risks to aquatic ecosystems. Electrochemical approaches enable controlled and efficient oxidation of NH3 to N2, presenting a sustainable solution for mitigating NH3 pollution.
Since AOR is more likely to occur in alkaline media than in acidic media, most AOR research has been conducted in alkaline media. The development of highly active catalysts is crucial to improve the efficiency of AOR.7–14 Among the catalysts studied for AOR, platinum (Pt) is widely recognized as the most active catalyst for AOR, serving as a benchmark due to its superior catalytic activity.15 However, Pt exhibits a large overpotential, which limits the efficiency of AOR. Iridium (Ir) is known as the second most active catalyst for AOR, exhibiting lower current densities compared to Pt but achieving a reduced overpotential. Theoretical studies suggested that excellent AOR activities can be obtained when the bond strength between the metal and the atom is moderate, i.e. not too weak to activate the reactant and not too strong to hinder product desorption.16 According to the Sabatier principle, the metal–nitrogen bond strength decreases in the order Ru > Rh > Pd > Ir > Pt ≫ Au, Ag, Cu. The moderate binding energies of intermediates on Pt and Ir allow them to show catalytic performances for AOR among single-component catalysts. Although the detailed mechanism of AOR is still under debate, the widely accepted mechanism in alkaline media, initially proposed by Maurer and Gerischer,17 involves the following sequence of reactions:
| \begin{equation} \text{NH$_{3}$ (aq)}\to \text{NH$_{\text{3,ad}}$} \end{equation} | (1) |
| \begin{equation} \text{NH$_{\text{3,ad}}$} + \text{OH$^{-}$}\to \text{NH$_{\text{2,ad}}$} + \text{H$_{2}$O} + \text{e$^{-}$} \end{equation} | (2) |
| \begin{equation} \text{NH$_{\text{2,ad}}$} + \text{OH$^{-}$}\to \text{NH$_{\text{,ad}}$} + \text{H$_{2}$O} + \text{e$^{-}$} \end{equation} | (3) |
| \begin{equation} \text{NHx$_{\text{,ad}}$} + \text{NHy$_{\text{,ad}}$}\to \text{N$_{2}$Hx+y$_{\text{,ad}}$} \end{equation} | (4) |
| \begin{equation} \text{N$_{2}$Hx+y$_{\text{,ad}}$} + (\mathrm{x} + \mathrm{y})\text{OH$^{-}$}\to \text{N$_{2}$} + (\mathrm{x} + \mathrm{y})\text{H$_{2}$O} + (\mathrm{x} + \mathrm{y})\text{e$^{-}$} \end{equation} | (5) |
| \begin{equation} \text{NH$_{\text{,ad}}$} + \text{OH$^{-}$}\to \text{N$_{\text{,ad}}$} + \text{H$_{2}$O} + \text{e$^{-}$} \end{equation} | (6) |
In this mechanism, NH3 reacts with hydroxide ions adsorbed on the catalyst surface and undergoes progressive dehydrogenation to form surface-adsorbed NHx,ad species. These intermediates subsequently dimerize into N2Hx+y,ad, followed by further dehydrogenation to form N2 as the final product. In contrast, the fully dehydrogenated atomic nitrogen (N,ad) species produced in reaction (6) strongly adsorbs to the electrode surface, poisoning the catalyst and inhibiting further NH3 adsorption. Both experimental and theoretical studies have suggested that the direct dimerization of N,ad species to form N2 is thermodynamically unfavorable.18–28 Metals with strong adsorption of nitrogen species demonstrate low energy barriers for dehydrogenation reactions, resulting in a lower onset potential for NH3 oxidation. However, their high affinity for nitrogen also leads to irreversible adsorption of N,ad species, causing catalyst poisoning and a decline in reaction rate. Therefore, the development of highly active electrocatalysts for AOR requires a material with sufficient dehydrogenation ability to facilitate NHx formation from NH3 while effectively suppressing the formation of N,ad. Additionally, it is necessary to enhance the activity and stability by stably adsorbing NHx species and promoting the formation of N2Hx+y intermediates.18
Composite materials are promising for the development of highly active electrocatalysts because heterointerfaces between different components produce synergistic effects that cannot be obtained with individual materials.29–34 Metal chalcogenides such as WS2 and MoS2 are known as layered materials and used as component materials of highly active composite electrocatalysts.35,36 Recent studies have reported that NHx species can be stably adsorbed on WS2. By using WS2 as a material for AOR electrocatalysts, it is expected to promote the formation of N2Hx+y,ad by stable adsorption of NHx and suppress the formation of N,ad species, achieving highly efficient AOR.37–39 However, since WS2 has very low dehydrogenation ability, WS2 alone will show negligible catalytic activities for AOR. Thus, a combination of WS2 and highly active catalysts for dehydrogenation will develop excellent electrocatalysts for AOR.
In this study, we synthesize a novel composite catalyst of Ir cluster-loaded WS2 (Ir/WS2) and demonstrate efficient AOR with tolerance of N,ad poisoning by synergistic effects of the higher dehydrogenation ability of Ir and the stable adsorption of N2Hx+y,ad on WS2. Cyclic voltammetry (CV) and chronoamperometry (CA) clarify that the Ir/WS2 show high catalytic performances for AOR due to tolerance of N,ad poisoning by the WS2 support.
A powdered WS2 catalyst was prepared using the hydrothermal method, as reported previously.40,41 (NH4)10W12O41·5H2O (0.5 mmol, FUJIFILM Wako Pure Chemical Co.), CH4N2S (30 mmol, FUJIFILM Wako Pure Chemical Co.), and (COOH)2·2H2O (6 mmol, FUJIFILM Wako Pure Chemical Co.) were dissolved in 35 cm3 of ultrapure water in a Teflon container. The mixture was transferred to an autoclave and heated at 200 °C for 24 h to synthesize WS2 powder. The product was filtered and dried in a vacuum at 80 °C for 1 h.
2.1.2 Loading of iridium clusters onto WS2For the Ir/WS2 composite catalyst preparation, colloidal Ir clusters were prepared separately and then loaded onto WS2. For the preparation of the Ir clusters dispersion, IrCl3·3H2O (0.88 mmol, Kanto Chemical Co.) was dissolved in a mixed solvent of 120 cm3 of ethylene glycol and 80 cm3 of ultrapure water, followed by heating at 140 °C for 3 h under Ar atmosphere.42 Concurrently, 50 mg of WS2 powder was dispersed in a mixture of 15 cm3 ethylene glycol and 10 cm3 ultrapure water via ultrasonication for 15 min. Then, 50 cm3 of the synthesized Ir cluster dispersion was added to the WS2 suspension and stirred at room temperature for 24 h. The product was filtered, washed with ultrapure water, and dried in vacuo. For comparison, the Ir clusters-loaded activated carbon (Ir/C) was synthesized using the same procedure, except that WS2 was replaced with acetylene black (Strem Chemicals, Inc.).
2.2 Characterization of samplesThe morphology and composition of the samples were analyzed using a low acceleration voltage scanning electron microscopy (SEM, Sigma 500, ZEISS) operated at 1.5 kV with energy dispersive spectroscopy (EDS). Powder samples were placed on carbon tape and vacuum-dried overnight prior to observation. The crystalline structures of the samples were determined using an X-ray Diffraction (XRD) system (Rigaku Ultima IV) with a Cu-Kα radiation source (λ = 0.15418 nm). Measurements were performed at 40 kV and 20 mA over a 2θ range of 5°–70°, with a scanning speed of 5° min−1. Slit configurations included a divergence slit of 1/3°, a scattering slit of 1°, and a receiving slit of 0.10 mm. The nanostructures of prepared samples were observed by transmission electron microscopy (TEM, JEM-2010, JEOL Ltd.) with 200 kV acceleration voltage. Samples were prepared by dispersing powders in ethanol via ultrasonication for 30 min, followed by drop-casting the suspension onto copper grids and drying in a vacuum desiccator. X-ray Photoelectron Spectroscopy (XPS, JEOL-JPS9010MC, JEOL Ltd.) analysis with Al-Kα radiation (hν = 1486.6 eV) was performed to measure the valence states of the prepared samples. Binding energies were corrected by referring to the C 1s binding energy of the carbon atoms of the ligand in the specimens at 284.5 eV. Spectrum fitting was conducted using XPSPEAK41 software.
2.3 Evaluation of NH3 oxidation reaction activitiesA dispersion containing 1 cm3 of 2-propanol (Kanto Chemical Co., Inc.), 1 cm3 of ultrapure water and 2 mg of the sample was prepared by ultrasonication for 30 minutes. A glassy carbon disk electrode with a diameter of 3 mm was used as a substrate. After polishing the glassy carbon disk electrode, a 10 mm3 aliquot of the dispersion was drop-cast onto a glassy carbon electrode and allowed to dry (catalyst loading of 0.14 mg cm−2). Electrochemical AOR activity was evaluated using a three-electrode system connected to a Vertex.1A potentiostat (IVIUM TECHNOLOGIES) in 1.0 mol dm−3 KOH at room temperature, with a rotating ring-disk electrode. A coiled Pt wire and Hg/HgO electrode (RE-6A, BAS Co., Ltd.) with filling 1 mol dm−3 KOH aqueous solution were used as a counter and reference electrode, respectively. Prior to measurements, the electrolyte was degassed by Ar bubbling for 30 minutes. CV measurements were performed by sweeping the potential from 0 to 1.0 V vs. RHE at a scan rate of 1 mV s−1 in an aqueous solution of 1 mol dm−3 KOH electrolyte (80 cm3) with or without 1 mol dm−3 NH3 under Ar atmosphere using a single compartment glass cell with the working, reference and counter electrodes. CA measurements were conducted by applying the constant potentials (0.6 V or 0.9 V vs. RHE) using the same measurement condition. The working electrodes were rotated at 1600 rpm during the measurements. CA measurements were also conducted to quantify N2 formed from NH3 on the Ir/WS2 using a carbon felt substrate with a large surface area. A 1.0 cm3 aliquot of the catalyst dispersion was drop-cast onto the carbon felt substrate (E-525, KUREHA Co., 8 cm2) and allowed to dry (catalyst loading of 0.14 mg cm−2). The other conditions were the same as above. The gaseous components in the headspace of the cell were analyzed using a gas chromatograph (Shimadzu Corporation, GC-2014). The potential was calibrated with respect to a reversible hydrogen electrode (RHE) in 1 mol dm−3 KOH solution using the following equation:
| \begin{align*} \text{E vs. RHE} &= \text{E vs. Hg/HgO/4}\,\text{mol$\,$dm$^{-3}$ KOH} \notag\\ &\quad + 0.098 + 0.059\times \text{pH} \end{align*} |
The electrochemical surface area (ECSA) was determined by measuring the double-layer capacitance using cyclic voltammetry in an Ar-saturated electrolyte. The potential range was 0.35–0.45 V vs. RHE, and the potential sweep rates were varied from 10 to 50 mV s−1.
The morphology of WS2 was investigated by SEM and TEM observations (Fig. 1). The SEM image revealed that WS2 exhibits a two-dimensional belt-like morphology, with widths ranging from several tens to a few hundred nanometers and lengths extending to several micrometers. The surface of the WS2 nanobelts appeared smooth and uniform. SEM-EDX analysis confirmed a homogeneous distribution of tungsten (W) and sulfur (S) throughout the sample (Fig. S1), with an atomic ratio of W : S = 1 : 1.9, close to the stoichiometric composition. TEM analysis provided further insights into the structural characteristics of the WS2 nanobelts. The observed thickness of the nanobelts was approximately 10 to 30 nm (Fig. 1b). The high-resolution TEM image revealed a lattice fringe spacing of 0.26 nm, which corresponds to the (101) plane of hexagonal WS2 (Fig. 1c), consistent with previous reports.43
(a) SEM and (b, c) TEM images of the pristine WS2.
Figure 2 shows XRD patterns of the pristine WS2, activated carbon, Ir/WS2 and Ir/C. The pristine WS2 showed characteristic peaks at 2θ = 14.6°, 32.1°, and 35.7°, corresponding to the 002, 101, and 102 reflections of the hexagonal 2H-WS2 phase (JCPDS card no. 08-0237) and pristine activated carbon showed characteristic peaks at 2θ = 25.2° and 42.7°, corresponding to the carbon 002 and 10 reflections,44,45 respectively. The XRD patterns of Ir/WS2 and Ir/C basically matched those of their corresponding pristine samples, with no significant peak shifts, indicating a negligible change in crystalline structures during Ir loading. In addition, a broad peak at approximately 40° corresponding to the 111 reflection of face-centered cubic (fcc) Ir was also observed in the Ir/C sample, indicating loading metallic Ir cluster. For the Ir/WS2, multiple peaks of WS2 overlapped around 40°, so it is difficult to accurately identify the peak derived from the 111 reflection of fcc Ir. However, the increased intensity observed around 40° for the Ir/WS2 compared to the pristine WS2 probably corresponded to the 111 reflection of Ir, suggesting loading of Ir clusters.
XRD patterns of WS2, Ir/WS2, activated carbon and Ir/C.
Morphological analysis of the Ir/WS2 was conducted by SEM and TEM observation (Fig. 3). The SEM image revealed that the belt-like morphology of WS2 was almost preserved with partial fragmentation. The surface of the WS2 nanobelts was covered with particle-like structures. TEM observation clarified that Ir clusters with an average size of approximately 1.5 nm were loaded on the WS2. The Ir clusters were distributed on the entire surface of WS2, and aggregated Ir clusters were also observed in some areas, which is consistent with the SEM image. The Ir clusters show a lattice spacing of 0.22 nm corresponding to the (111) plane of fcc Ir. The Ir/C showed a similar situation of Ir loading, i.e., Ir clusters were distributed on the entire surface of spherical activated carbon (Fig. S2). The SEM-EDX analysis indicated that the amount of Ir clusters loaded was 15 wt% for Ir/WS2 and 16 wt% for Ir/C, confirming that composite catalysts have similar Ir loadings. The results of morphological and structural analysis indicate that the composite catalyst with metallic Ir clusters loaded on WS2 nanobelts was synthesized.
(a) SEM and (b, c) TEM images of the Ir/WS2. The inset in Fig. 3b shows size distribution of Ir clusters.
The valence states of the constituent elements for the prepared catalysts were investigated using XPS (Fig. 4). In the pristine WS2 spectrum, the deconvoluted W 4f7/2 peak at 31.5 eV and the S 2p3/2 peak at 161.3 eV corresponded to W4+ and S2−, respectively. The Ir/WS2 showed the different W 4f spectrum from that of the pristine WS2. The W 4f7/2 spectrum for the Ir/WS2 were deconvoluted into multiple peaks, and W 4f7/2 peaks at 32.1 eV and at 35.3 eV were observed, which were attributed to W4+ and W6+ derived from tungsten oxides, in addition to W4+ derived from WS2.46 The results indicated that the surface of WS2 was partially oxidized during the Ir loading. Since the XRD pattern of Ir/WS2 did not show any peaks attributable to tungsten oxides, tungsten oxides were formed only on the surface of WS2. The Ir/WS2 showed the Ir 4f spectrum characteristic of both the metallic and oxidative states of Ir. The deconvoluted Ir 4f7/2 peaks were observed at 60.9 eV and 62.0 eV, which are similar to the peak energy of metallic Ir (60.8 eV) than that of IrO2 (62.5 eV), respectively. Since the XRD measurement and TEM observation revealed that Ir clusters have the fcc structure of metallic Ir, the Ir 4f spectrum reflected the internal metallic states and the surface oxidative states of Ir clusters. The Ir/C also showed deconvoluted Ir 4f7/2 peaks at 60.7 eV and 61.6 eV derived from metallic and oxidative states of Ir, respectively. The Ir 4f7/2 peaks for the Ir/WS2 were observed at slightly higher energy region than those of the Ir/C. Concurrently, the W 4f7/2 peak derived from WS2 (31.4 eV) and the S 2p3/2 peak (161.0 eV) for the Ir/WS2 slightly shifted to a lower energy region compared to the pristine WS2. The results indicate that charge transfer from Ir to WS2 occurred in the Ir/WS2, leading to changes in the valence states of the constituent elements. Previous studies reported that charge transfer occurs between components when the components of a composite catalyst are closely combined.32,47 Therefore, in the Ir/WS2, Ir clusters and WS2 were closely combined.
XPS spectra for (a) W 4f, (b) S 2p and (c) Ir 4f of the pristine WS2, Ir/WS2 and Ir/C.
Figure 5 shows the CV curves of the Ir/C, pristine WS2, and Ir/WS2 catalysts in 1 mol dm−3 KOH aqueous solution, with and without 1 mol dm−3 NH3 in the potential range of 0.0–1.0 V vs. RHE. Under blank conditions (in the absence of NH3), the Ir/C catalyst showed anodic and cathodic peaks around 0.05 V, 0.16 and 0.22 V vs. RHE, corresponding to hydrogen adsorption/desorption on Ir.48 However, in the presence of NH3, distinct anodic current attributable to NH3 oxidation was observed within the potential range of 0.45–1.0 V vs. RHE. The current was increased from the onset potential of 0.45 V vs. RHE and steeply decreased above 0.61 V vs. RHE, which would correspond to dehydrogenation of NH3 and subsequent poisoning of active sites by N,ad, respectively. In contrast, pristine WS2 showed negligible differences between the CV curves in the absence and presence of NH3. The results indicate that the pristine WS2 is inactive for AOR probably due to its very low dehydrogenation ability. The Ir/WS2 also showed anodic and cathodic peaks due to hydrogen adsorption/desorption under the blank condition, and showed anodic current due to NH3 oxidation in the presence of NH3. Since the pristine WS2 showed no activities for AOR, Ir clusters in the Ir/WS2 work as the active species for NH3 oxidation. The Ir/WS2 showed the similar onset potential of NH3 oxidation at 0.45 V vs. RHE to that of Ir/C. However, the decrease in current density above the peak potential at 0.66 V vs. RHE was quite moderate for the Ir/WS2, which was different from the Ir/C. The results suggest that poisoning of active sites by N,ad was suppressed for the Ir/WS2 compared to the Ir/C.
Cyclic voltammograms of (a) Ir/C, (b) WS2 and (c) Ir/WS2 in 1 mol dm−3 KOH aqueous solution in the presence (color) and absence (black) of 1 mol dm−3 NH3. (d) Cyclic voltammograms of (blue) Ir/C and (red) Ir/WS2 in 1 mol dm−3 KOH aqueous solution in the presence of 1 mol dm−3 NH3 for comparison.
Furthermore, CA measurements were conducted under the application of 0.6 V vs. RHE and 0.9 V vs. RHE to evaluate the catalytic performances of the prepared catalysts using the glassy carbon electrode. Figure 6 shows time courses of current densities of the Ir/WS2 and Ir/C. Both Ir/WS2 and Ir/C showed a gradual decrease in current densities with time course, which is due to the accumulation of N,ad poisoning species on the catalyst surface. The Ir/WS2 exhibited higher current densities and maintained superior catalytic performances compared to the Ir/C under both potential conditions. Since the amounts of Ir clusters loaded and the electrochemical surface area of the Ir/WS2 and Ir/C were comparable (Fig. S3), the difference in activities originated from difference in the component materials, i.e., WS2 and activated carbon. Therefore, the results clearly indicate that the Ir/WS2 catalyst has excellent tolerance to N,ad poisoning and the WS2 support promoted the catalytic performances of NH3 oxidation. Additional CA measurements were conducted to quantify N2 formed from NH3 on the Ir/WS2 using a carbon felt substrate with a large surface area. Figure 7 shows time courses of current density, the amount of N2 formed and current efficiency (CE) of N2 formation under the application of 0.6 V vs. RHE and 0.9 V vs. RHE. During the CA measurements, N2 was produced continuously for 60 min under both potential conditions. The current efficiencies for N2 formation under the application of 0.6 V vs. RHE and 0.9 V vs. RHE were always more than 99 % and more than 98 %, respectively. The results indicate that the product of AOR on the Ir/WS2 was almost N2. The current efficiencies of N2 formation, which were slightly less than 100 %, may suggest the production of the trace amount of NO2− and NO3−. The standard electrode potentials for oxidation to N2, NO2−, and NO3− are shown below.49
| \begin{equation*} \begin{array}{ll} \text{2NH$_{3}$} + \text{6OH$^{-}$}\to \text{N$_{2}$} + \text{6H$_{2}$O} + \text{6e$^{-}$} &\quad{+}0.0567\,\text{V vs. SHE}\ (\text{pH}=0)\\ \text{2NH$_{3}$} + \text{6OH$^{-}$}\to \text{N$_{2}$} + \text{6H$_{2}$O} + \text{6e$^{-}$} &\quad{-}0.772\,\text{V vs. SHE}\ (\text{pH}=14) \end{array} \end{equation*} |
| \begin{equation*} \begin{array}{ll} \text{NH$_{3}$} + \text{2H$_{2}$O}\to \text{NO$_{2}{}^{-}$} + \text{7H$^{+}$} + \text{6e$^{-}$} &{+}0.792\,\text{V vs. SHE}\ (\text{pH}=0)\\ \text{NH$_{3}$} + \text{7OH$^{-}$}\to \text{NO$_{2}{}^{-}$} + \text{5H$_{2}$O} + \text{6e$^{-}$} &{-}0.174\,\text{V vs. SHE}\ (\text{pH}=14) \end{array} \end{equation*} |
| \begin{equation*} \begin{array}{ll} \text{NH$_{3}$} + \text{3H$_{2}$O}\to \text{NO$_{3}{}^{-}$} + \text{9H$^{+}$} + \text{8e$^{-}$} &{+}0.799\,\text{V vs. SHE}\ (\text{pH}=0)\\ \text{NH$_{3}$} + \text{9OH$^{-}$}\to \text{NO$_{3}{}^{-}$} + \text{6H$_{2}$O} + \text{8e$^{-}$} & {-}0.132\,\text{V vs. SHE}\ (\text{pH}=14) \end{array} \end{equation*} |
Chronoamperometry curves of Ir/WS2 and Ir/C at (a) 0.6 V vs. RHE and (b) 0.9 V vs. RHE in 1 mol dm−3 KOH aqueous solution in the presence of 1 mol dm−3 NH3. The insets show chronoamperometry curves in the high current density range.
(a) Chronoamperometry curves of the Ir/WS2 at 0.6 V vs. RHE and 0.9 V vs. RHE in 1 M KOH aqueous solution in the presence of 1 M NH3. Time courses of the amounts of N2 formed and current efficiencies of N2 formation on the Ir/WS2 at (b) 0.6 V vs. RHE and (c) 0.9 V vs. RHE.
The oxidation potentials of NH3 to NO2− and NO3− are more positive than the oxidation potentials to N2. The dependence of the products on the applied potential in AOR has been studied.50 The NO2− and NO3− are mainly produced at potentials more positive than 1.0 V vs. RHE due to the high overpotential, and the ratio of NO2− and NO3− production increases as the potential becomes more positive. Almost exclusively N2 is produced at potentials more negative than 0.9 V vs. RHE. In fact, the products in the CA at 0.6 V vs. RHE and 0.9 V vs. RHE in this study were almost N2, which are similar to the results of previous studies.49,50 The CE for N2 formation in the CA at 0.9 V vs. RHE was slightly lower than that at 0.6 V vs. RHE, implying higher ratio of NO2− and NO3− production in the CA at 0.9 V vs. RHE. The results suggest that excellent activities of the Ir/WS2 for AOR corresponded to the promotion of N2 formation.
In the CV and CA measurements, the anodic current densities were obtained due to the dehydrogenation of NH3, and then the current densities decreased due to the poisoning of active sites by N,ad produced through the complete dehydrogenation of NHx. Experimental and computational studies suggested that N2 is not produced by the dimerization of N,ad, but rather by the dimerization of NHx. The milder decrease of current densities and superior catalytic performances on the Ir/WS2 than those on the Ir/C indicate that the formation of N,ad was suppressed and the dimerization of NHx species was promoted. Previous studies reported that WS2 materials can stably adsorb NxHy.37–39 We believe that NHx species, which are produced by the dehydrogenation of NH3 near the heterointerface between Ir clusters and WS support, can be stably adsorbed on WS2, resulting in suppression of formation of N,ad and promotion of dimerization of NHx species. Recently, it was reported that the WS2-WO3 heterostructure greatly enhances the adsorption of -NH2 and -NH species.37 XPS analysis revealed that the presence of surface tungsten oxides in the Ir/WS2 composite, suggesting that the heterostructure of surface tungsten oxides and bulk WS2 may further contribute to the stable adsorption of NHx species.
In this study, we developed the novel Ir/WS2 composite catalyst and evaluated its performance for AOR. Morphological and structural characterizations revealed that metallic Ir clusters with an average size of approximately 1.5 nm were distributed on two-dimensional WS2 nanobelts. Electrochemical measurements demonstrated that the Ir/WS2 exhibited a comparable onset potential to the Ir/C but significantly improved tolerance to poisoning by N,ad as evidenced by the more moderate decrease in current densities during CV and CA measurements. The results suggest that the WS2 support played a crucial role in stabilizing NHx intermediates while suppressing the formation of strongly adsorbed N,ad species, facilitating the dimerization of NHx. The synergistic effects of high dehydrogenation ability of Ir clusters and stable adsorption of NHx on WS2 contributed to high catalytic activities and durability for AOR. This study could highlight the potential of WS2-based composite catalysts for developing efficient and stable AOR electrocatalysts.
Part of this study was conducted at the Laboratory of XPS analysis, Joint-use facilities and the Multi-Quantum Beam High Voltage Electron Microscope Laboratory at Hokkaido University, supported by the Nanotechnology Platform Program of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan (A-21-HK-0035). We acknowledge funding by JSPS KAKENHI (20K15373), Japan.
The data that support the findings of this study are openly available under the terms of the designated Creative Commons License in J-STAGE Data at https://doi.org/10.50892/data.electrochemistry.28202216.
Sho Kitano: Conceptualization (Lead), Data curation (Lead), Formal analysis (Lead), Funding acquisition (Lead), Project administration (Lead), Resources (Supporting), Supervision (Lead), Writing – original draft (Lead), Writing – review & editing (Lead)
Reiko Tagusari: Data curation (Equal), Investigation (Lead)
Yuta Nagasaka: Investigation (Supporting)
Yoshitaka Aoki: Conceptualization (Supporting), Funding acquisition (Supporting), Resources (Supporting), Supervision (Supporting)
Hiroki Habazaki: Conceptualization (Equal), Funding acquisition (Lead), Resources (Lead), Supervision (Equal), Writing – review & editing (Lead)
The authors declare no conflict of interest in the manuscript.
Japan Society for the Promotion of Science: A-21-HK-0035
Japan Society for the Promotion of Science: 20K15373
S. Kitano and Y. Aoki: ECSJ Active Members
H. Habazaki: ECSJ Fellow
