An Investigation of the Interface between Transition Metal Oxides (MnO_x, FeO_x, CoO_x and NiO_x)/MoO₃ Composite Electrocatalysts for Oxygen Evolution Reactions

Abstract

This study presents the synthesis of a multicomponent metal oxide electrocatalyst that increases the activity of the oxygen evolution reaction (OER). We synthesized transition metal oxides (MnO_x, FeO_x, CoO_x, and NiO_x) with MoO₃ heterostructures through a solid-state reaction approach at low cost. In comparison to the other compositions, CoO_x garnered higher attention and demonstrated superior performance on account of its large surface area and varied crystal facets. The MnO_x-MoO₃, FeO_x-MoO₃, CoO_x-MoO₃, and NiO_x-MoO₃ compositions attained an overpotential of 390 mV, 350 mV, 310 mV, and 340 mV, respectively, at a current density of 10 mA cm⁻² in alkaline solution. The performance of OER was enhanced in CoO_x-MoO₃ at 10 mA cm⁻², while FeO_x-MoO₃ exhibited a lower current density at 100 mA cm⁻² than other metal oxides. The CoO_x-MoO₃ material exhibited a favorable crystal interface transition due to the presence of MoO₃ oxide. For the first time, we report on the MoO₃-to-(MnO_x, FeO_x, CoO_x, and NiO_x) interface crystal transition and the active surface area for OER catalytic activity in water-splitting processes. This investigation intends to develop an electrocatalyst that is capable of producing hydrogen with the use of heterostructure metal oxides.

1. Introduction

Electrochemical energy storage provides a sustainable solution for generating green energy without harmful carbon emissions. One of the most significant non-carbon emission processes is the water-splitting process [1,2]. This process involves two main reactions, namely the hydrogen evolution reaction (HER) and the oxygen evolution reaction (OER), which produce hydrogen and oxygen, respectively, through the decomposition of water [3,4]. The OER reaction requires four electrons and has a high overpotential, making it a challenging issue in water decomposition. Therefore, many researchers have focused on developing OER electrocatalysts for electrochemical reactions. RuO₂ and IrO₂ are used in commercial applications [5]. However, these benchmark catalysts are lacking in resources and are costly. Therefore, electrocatalysts based on various transition metal composites, such as oxides, sulfides, nitrides, phosphides, hydroxides, and carbides, have been reported as potential replacements for benchmark catalysts [6,7]. These composite electrocatalysts, composed of transition metals, are facing challenges related to surface area, stability, and electrical conductivity. Therefore, research is needed to develop more innovative electrocatalysts in the transition metal composite by tuning composition, morphology, and size. The high surface area and good electrical conductivity of electrocatalysts boost their catalytic activity significantly [8]. Multi-metal oxide composites enhance surface area and electrical conductivity, boosting OER activity. This is because the OER reaction takes place on the surface of the electrode materials. The interface electrode materials play a significant role in improving electrokinetic reactions [9]. Recent reports indicate that molybdenum oxide (MoO₃), a key component in electrocatalytic systems, can exchange surface oxygen ions reversibly in a heterostructure interface. This property can aid in charge transfer and the production of O* during the OER process. MoO₃ has a higher number of active sites and easy charge transport during electrocatalysis compared to platinum. The RuO₂/MoO₃ composite heterostructure exhibited a low overpotential and 3.2 times greater activity [10]. Yi Liu et al. researched the integration of MoO₃ patches into Ni oxyhydroxide nanosheets [11]. At 10 mA cm⁻², this composite has an overpotential of 260 mV. According to the findings, the Ni-Mo heterostructure increases electrocatalytic activity. Yao Zhang et al. studied the multicomponent interface structure (Fe-S-NiMoO₄/MoO₃@NF) for OER activity. The catalyst worked at an overpotential of 271 at 500 mA cm⁻². In this composite, the electric conductivity improved the MoO₃ structure by doping Fe [12]. Li et al. developed a bifunctional catalyst of MoO₃-Ni-NiO via the electrodeposition method. It showed an OER overpotential of 347 mV at 100 mA cm⁻². The MoO₃-Ni-NiO interface composite revealed a low energetic barrier and boosted the catalytic activity [13]. Tariq et al. investigated low noble metal materials with the addition of MoO₃. The IrO₂/MoO₃ interface composite showed long-term stability up to 40,000 s in acidic conditions. The composite resulted in a two-fold enhancement at the current density and a seven-fold increase in mass activity [14]. The above studies indicate that MoO₃ exhibits good interface activity with other metal oxides. However, the heterostructure functions and electrocatalytic activity of MoO₃ with transition metal oxides (Mn, Fe, Co, and Ni) are not fully understood.

Therefore, a multicomponent interface heterostructure was investigated in this research paper via a simple solid-state reaction method. The heterostructure was constructed using transition metal oxides (Mn, Fe, Co, and Ni) integrated with MoO₃. The performance of MnO_x-MoO₃, FeO_x-MoO₃, CoO_x-MoO₃, and NiO_x-MoO₃ for OER was analyzed and compared. The CoO_x-MoO₃ heterostructure exhibited a higher surface area (57.05 m² cm⁻¹) than the other composite electrocatalysts. The analysis of OER revealed that the CoO_x-MoO₃ composite exhibited a low overpotential (310 mV) and was more active than the other tested heterostructures at 10 mA cm⁻². However, the FeO_x-MoO₃ composite interface showed higher activity at 100 mA cm⁻² than cobalt oxide. Based on the results of this study, an electrocatalyst comprising MoO₃ coupled with CoO_x and FeO_x composites leads to a more significant interface transition for OER activity than other catalysts. The hybrid composite electrocatalyst of MoO₃ with a transition metal oxide enhances OER performance. This rational design of the heterostructure interface improves alkaline stable electrocatalysts for OER in water-splitting reactions.

2. Materials and Methods

2.1. Materials

The materials included nickel (II)acetate tetrahydrate (97%, Daejung, Siheung, Republic of Korea), Ferrous acetate, Cobalt acetate, manganese nitrate, ammonium molybdate tetrahydrate (97%, Daejung, Republic of Korea), carbon black–acetylene black 100% compressed (CA), specific surface area 75 m²/g (Alfa Aesar, Seoul, Republic of Korea), polyvinylidene fluoride (PVDF, Mw ~534,000, Sigma-Aldrich, Seoul, Republic of Korea), 1-methly 2 pyrrolidone (NMP, 99%, extra pure, Duksan chemicals, Ansan-si, Republic of Korea), potassium hydroxide (85%, KOH, Duksan chemicals, Republic of Korea), demineralized water (DI). All the chemicals used in this experiment were received as they were, without any additional purification process.

2.2. Synthesis of Active Metal Oxide Composites

The metal precursors were mixed with ammonium molybdate tetrahydrate with an equal molar ratio. The mixture was ground using a mortar and pestle for 20 min. Afterward, carbon (0.1 g) was added to the above reactant for further grinding and mixed well. Finally, the reactant was transferred into a ceramic crucible for thermal treatment. The crucible was placed in the air oven at 300 °C for 4 h with a temperature increase of 2 °C per minute. After the completion of the reaction, the product was ground for 20 min using a mortar and pestle again. Likewise, each metal oxide composite (MnO_x-MoO₃, FeO_x-MoO₃, CoO_x-MoO₃, and NiO_x-MoO₃) was synthesized separately.

2.3. Physiochemical Characterization

BET (Brunauer–Emmett–Teller) surface areas (m²/g) were determined through nitrogen adsorption–desorption analysis. The nature of crystal structure type and composition of metal oxide was investigated using X-ray diffraction (XRD) with an Xpert Pro device. The device was equipped with a Cu Kα radiation source with a wavelength of 1.5406 Å and a step size of 0.02°. The oxidation state of as-prepared materials was analyzed by X-ray photoelectron spectroscopy (XPS) using Thermo Scientific (Waltham, MA, USA) K-α surface analysis. To analyze high-resolution transmission electron microscopic (HRTEM) images, we utilized a transmission electron microscope with JEM-2100 (Tokyo, Japan). For distinguishing different particles, high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) and energy dispersive X-ray (EDX, Bruker Nano GmbH, Berlin, Germany) element mapping images were used. Carbon hybridization was confirmed by the G and D bands in Raman spectroscopy using an XploRA plus device, HORIBA Jobin Yuon S.A.S, France. The scanning electron microscopy (SEM) analysis was conducted using a HITACHI S-4800, Tokyo, Japan, instrument equipped with an energy-dispersive X-ray spectroscopy (EDAX) system to examine the surface microstructures and particle size.

2.4. Preparation of Working Electrodes

The electrode was prepared using carbon cloth (CC). The CC was cleaned by thermal treatment at 400 °C. Next, it was immersed in ethanol, acetone, and DI water for 20 min each. Finally, the CC was dried overnight at 60 °C. The electrocatalyst slurry, consisting of 90% active catalyst, 10% PVDF, and NMP solvent, was applied to a pretreated CC substrate measuring 1 cm × 1 cm. The slurry was then dried overnight at 60 °C.

2.5. Electrochemical Analysis

All electrochemical measurements were conducted at room temperature in a 1 M KOH aqueous solution using a three-electrode system. The system comprised a working electrode, a Pt sheet as the counter electrode, and a Hg/HgO electrode as the reference electrode. The measurements were carried out using a Corretest electrochemical workstation. Electrochemical impedance spectroscopy (EIS) was studied at an open-circuit potential at frequencies ranging from 1 Hz to 100 kHz. Chronopotentiometry curves were recorded at a constant potential without iR compensation. Cyclic voltammetry (CV) analyses were carried out at different scan rates (5, 10, 15, 20, and 25 mV s⁻¹) at room temperature. The electrochemical double-layer capacitance (Cdl) and electrochemical surface area (ECSA) were calculated from different scan rates of CV [15] as follows:

$$\mathrm{E}\mathrm{C}\mathrm{S}\mathrm{A}={\displaystyle \frac{\mathrm{C}\mathrm{d}\mathrm{l}}{Cs}}$$

(1)

where Cs is a specific capacitance of 1 M KOH solution (0.040 mF cm⁻²).

Linear sweep voltammetry (LSV) was recorded at 5 mV/S scan rate. All recorded potentials were converted into reversible hydrogen electrodes (RHEs) using Equation (1).

$${\mathrm{E}}_{\mathrm{R}\mathrm{H}\mathrm{E}}={\mathrm{E}}_{{\mathrm{H}}_{\mathrm{g}}/{\mathrm{H}}_{\mathrm{g}}\mathrm{O}}+0.0596 \mathrm{X} \left({\mathrm{P}}^{\mathrm{H}}\right)+{\mathrm{E}}_{{\mathrm{H}}_{\mathrm{g}}/{\mathrm{H}}_{\mathrm{g}}\mathrm{O}}^0$$

(2)

3. Results

XRD analysis and Raman spectroscopy were used to establish the crystal structure and content. The MoO₃ diffraction line in Figure 1a matches the as-synthesized MnO_x-MoO₃, FeO_x-MoO₃, CoO_x-MoO₃, and NiO_x-MoO₃. The diffraction angles indexed as 12.7 (001), 23.3 (100), 25.7 (002), and 27.7 (011) on JCPDS Card Nos. 85-2407 identify the MoO₃ crystal planes [14]. The diffraction pattern verifies that the MoO₃ crystal plane overlaps with that of the composite metal oxide. The diffractogram does not reveal the diffraction peaks for the nanostructures of MnO_x, FeO_x, CoO_x, and NiO_x because of overlaps with the MoO₃ peak and small concentration, as also confirmed by XPS [16]. The intensity of MoO₃ diffraction peaks is decreased in the composite metal oxide diffraction pattern, which is determined by the metal oxide composite combined with MoO₃. However, HRTEM, XPS, and SAED analyses confirm the presence of this metal oxide composite. The carbon substance’s dislocation was not detected by XRD. Carbon and chemical alterations in MoO₃ samples were examined using Raman spectroscopy, as shown in Figure 1b. The carbon structural properties and dislocations were largely determined by the maximum intensity and position matching. The D and G band intensity ratios of the samples were calculated at 1.2, 1.0, 1.1, and 0.91 for MnO_x-MoO₃, FeO_x-MoO₃, CoO_x-MoO₃, and NiO_x-MoO₃, respectively [17,18]. These intensity ratios indicate that the carbon surface was delocalized by metal oxide. Mn, Co, and Ni metal oxides have a strong affinity with the edge group of carbon. However, the Fe metal oxide’s affinity with carbon was very low. Figure 1c shows N₂ adsorption–desorption isotherms for specific surface area (SSA) analysis. It was found that the SSA of CoO_x-MoO₃ (57.05 m² g⁻¹) was higher than that of MnO_x-MoO₃ (37.42 m² g⁻¹), FeO_x-MoO₃ (47.04 m² g⁻¹), and NiO_x-MoO₃ (45.44 m² g⁻¹).

The porosity of the active catalyst could determine the surface area of the metal oxide composition. As a result, the high surface area obtained in CoO_x-MoO₃ was due to the multi-crystal facet morphology, which was confirmed in the TEM image [19]. Furthermore, we calculated the sample’s pore size using the Barrett–Joyner–Halenda (BJH) plots, which were created using the data obtained from the BET surface area technique [20,21]. Figure 1d presents the mesopore in the range of 12 to 14 nm for all samples (Table S1). Compared to other samples, the higher surface area of CoO_x-MoO₃ led to a higher pore volume (0.1805 cm³) and pore diameter (14.9 nm). This pore diameter facilitated the formation of an activation site during an electrocatalytic reaction. CoO_x-MoO₃ had a higher SSA and cumulative volume in the mesopores than other metal oxide composites [22].

The XPS analysis was used to characterize the surface chemical composition of the as-synthesized materials. The Mn 2p, Fe 2p, Co 2p, and Ni 2p spectra of MnO_x-MoO₃, FeO_x-MoO₃, CoO_x-MoO₃, and NiO_x-MoO₃, respectively, reveal the elemental composition of the metal species present in the composite. Figure 2a displays two peaks located at the binding energy of 653.37 and 641.63 eV, which correspond to Mn 2p1/2 and Mn 2p3/2, respectively. The presence of Mn³⁺ was confirmed with a binding energy difference of 11.7 eV between Mn 2p1/2 and Mn 2p3/2 [23]. However, satellite peaks at 646.5 and 657.3 eV for Mn 2p3/2 and Mn 2p1/2, respectively, are deconvoluted [24,25]. In Figure 2b, the XPS spectrum of Fe is split into two energy bands, namely Fe 2p3/2 (lower energy band) and Fe 2p1/2 (higher energy band). The peaks are observed at 711.7 eV (Fe 2p3/2) and 724.6 eV (Fe 2p1/2) in Fe³⁺ and Fe²⁺ core lines, respectively [26,27]. Additionally, the peaks located at approximately 718.9 and 714.5 eV are the satellite peaks associated with them [23,28]. Figure 2c illustrates that Co 2p is deconvoluted into two major peaks, and two small satellite peaks are observed. The two major peaks are located at 780.6 eV and 796.2, corresponding to the Co³⁺ core line in Co 2p3/2 and Co 2p1/2, respectively. The satellite peaks are observed at ~785.1 eV (Co 2p3/2) and ~803.6 eV (Co 2p1/2), which corresponds to the core line of Co²⁺ [29,30]. Figure 2d shows two component peaks in Ni 2p spectra. The first main peak matches with Ni 2p3/2 and is located at 854.4 eV, whereas the second main peak matches with Ni 2p1/2 spin-orbit levels of NiO and is centered at 871.6 eV. The Ni²⁺ state is represented by the peaks that match with Ni 2p3/2 [30]. Furthermore, at 880.7, 877.7, 867.6, and 863.9 eV, the characteristic satellite peaks are detected as deconvoluted [18,31].

The composite structure of MnO_x-MoO₃ and FeO_x-MoO₃ in Figure 3 was confirmed by the HRTEM image and elemental mapping. In Figure 3a,b, MnO_x and MoO₃ crystal facets are observed on the carbon surface. The C, O, Mn, and Mo elemental mapping in Figure 3c–f confirms the elemental and atomic composition present in MnO_x-MoO₃. Figure 3g,h show the image of FeO_x-MoO₃. The elemental composition was confirmed by the mapping image presented in Figure 3i–l. The crystallite metal oxides were dispersed in the amorphous carbon network. Figure 4a,b show images of CoO_x-MoO₃, while Figure 4g,h show images of NiO_x-MoO₃. In the TEM image, CoO_x-MoO₃ appears as a multi-faceted crystal, and its surface area is increased compared to other materials. This increase was confirmed in the BET analysis. The elemental composition of C, O, Co, and Mo is confirmed in the elemental mapping images in Figure 4c–f. Similarly, the mapping images in Figure 4i–l show the composition of NiO_x-MoO₃. The interplanar spacing diagram and chemical composition for MnO_x-MoO₃, FeO_x-MoO₃, CoO_x-MoO₃, and NiO_x-MoO₃ are shown in Figures S1–S4. The FE-SEM image in Figure S5 shows the spherical shape of particles in nanometers.

A study was conducted to analyze the electrochemical performance of an all-composite nanostructure and compare its performance with various other similar nanostructures. Using linear sweep voltammetry (LSV) experiments, the OER activity performance was measured in a 1 M KOH medium at a scan rate of 5 mV s⁻¹. The performance of the all-composite nanostructure regarding OER is presented in Figure 5a. The CoO_x-MoO₃ LSV curve indicates that the overpotential required to obtain a current density of 10 mA cm⁻² is 310 mV. This value is significantly lower than its counterparts such as MnO_x-MoO₃ (390 mV), FeO_x-MoO₃ (350 mV), and NiO_x-MoO₃ (340 mV). Electrochemical impedance spectroscopy (EIS) was used to gain a better understanding of the catalytic activity of all catalysts in terms of OER, as shown in Figure 5b. The CoO_x-MoO₃ heterostructure has a much smaller semicircular area than the MnO_x-MoO₃, FeO_x-MoO₃, and NiO_x-MoO₃ heterostructures. This suggests that the CoO_x-MoO₃ heterostructure has a lower impedance (Rct). It was predicted that the CoO_x-MoO₃ heterostructure would have a much smaller Rct value (0.205 Ω) than MnO_x-MoO₃ (0.285 Ω), FeO_x-MoO₃ (0.911 Ω), and NiO_x-MoO₃ (0.259 Ω). This means that it conducts electricity better and moves charges faster. The equivalent circuit pattern and fitting curves are given in Figure S6. As shown in Figure 5c, the OER kinetic was examined using Tafel slope analysis. The CoO_x-MoO₃ catalyst exhibited a Tafel slope of 146 mV dec⁻¹, indicating superior OER activity compared to its constituents MnO_x-MoO₃, FeO_x-MoO₃, and NiO_x-MoO₃, which exhibited Tafel slopes of 258, 103, and 125 mV dec⁻¹, respectively. Figure 5d shows the comparative analysis of OER overpotential with all active catalysts. Figure 6a shows the Cdl value obtained from CV analysis at different scan rates (5, 10, 15, 20, and 25 mV s⁻¹). The ECSA calculated from the Cdl of a specific metal oxide composite is illustrated in Figure 6b. A high ECSA value enhances the electrokinetic interaction between the electrode and the electrolyte, leading to improved OER activity. Therefore, the OER activity of Co and Fe was enhanced with the MoO₃ composite electrocatalyst.

The chronopotentiometry curve of the CoO_x-MoO₃ heterostructure was plotted at a current density of 10 mA cm⁻² for 24 h of electrolysis, which is presented in Figure 7a. It shows strong stability in the alkaline-mediated electrolysis. The XRD patterns before and after the stability test of CoO_x-MoO₃ are illustrated in Figure 7b. The mixed cobalt oxide peaks change into Co₃O₄ peaks at (31.3° and 36.7°) (JCPDS No. 48–1719). The characteristic diffraction pattern of MoO₃ is not observed by XRD clearly, demonstrating that the MoO₃ layer is amorphous. Its peak is merged with CC peaks at 25°.

4. Discussion

The electrocatalytic activity of FeO_x and CoO_x with MoO₃ exhibited significant OER activity due to the heterostructure of these metal oxides on a carbon substrate when compared to other metal oxides, as indicated in

Table 1. Comparison of the electrocatalytic activity of metal oxide composites with MoO₃ as OER catalysts.

Catalyst	Preparation Method	Electrolyte (M)	Over Potential @ 10 mA cm−2	Reference
MnO	Thermal annealing	1.0 KOH	394 mV	[32]
FeOx	Hydrothermal	1.0 KOH	379 mV	[33]
Co3O4	Thermal annealing	1.0 KOH	490 mV	[34]
NiO	Thermal annealing	1.0 KOH	330 mV	[35]
MnOx-MoO3	Thermal annealing	1.0 KOH	390 mV	This work
FeOx-MoO3	Thermal annealing	1.0 KOH	350 mV	This work
CoOx-MoO3	Thermal annealing	1.0 KOH	310 mV	This work
NiOx-MoO3	Thermal annealing	1.0 KOH	340 mV	This work

. As the atomic number increased in the transition metal 3d series from Mn to Ni, the high spin decreased. Together with the ligand, MnO_x(d⁵) and MoO₃(d⁰) produced a high-spin stable oxide composite. Because of Mn’s high-spin state, the interaction between Mn and Mo was quite weak in this instance. Consequently, the OER activity of the MnO_x-MoO₃ composite electrocatalyst decreased. However, the presence of a low-spin composite increased the interaction between FeO_x (d⁶) and CoO_x(d⁷) with MoO₃. At the electrode surface, an electrokinetic channel was created by the interaction of the low-spin d orbital electrons of Fe and Co with MoO₃. The intermediate performance of Co and Fe with MoO₃ was observed in NiO_x(d⁸). The electrical distribution of the metal oxide interface was altered by the addition of MoO₃ to the structure, which improved OER performance [19]. The metal oxide–metal oxide interface’s electron-transfer channel, which was made possible by orbital delocalization in the composite catalyst, significantly increased reaction kinetics. The unique interfacial contact and synergistic catalytic effects were found to ensure strong intrinsic performance and higher stability for electrochemical water splitting.

5. Conclusions

A synthesized composite catalyst demonstrated electrocatalytic activity in the OER for water electrolysis. The analysis using XRD, XPS, Raman spectroscopy, and HRTEM confirmed the formation of a metal oxide composite. The CoO_x-MoO₃ composite exhibited a high surface area, which enhanced its electrocatalytic activity. This resulted in a 310 mV overpotential at 10 mA cm⁻². The FeO_x-MoO₃ composite showed the best performance in OER at 100 mA cm⁻² compared to all other catalysts. The effectiveness of higher OER activity is directly related to the order of CoO_x-MoO₃, FeO_x-MoO₃, NiO_x-MoO₃, and MnO_x-MoO₃. This is explained by the synergy action of the conductive carbon support and the metal oxide–MoO₃ heterostructure. An increase in the number of accessible active sites, better charge transmission, and increased structural durability led to better OER performance. This study paves the way for exploring improved electrocatalyst activity in metal oxide composites via heterojunctions for OER in overall water splitting without using noble metals.