Flower-like CoP coated NiMoO₄ nanorods as self-supported core-shell heterojunction electrode can facilitate oxygen evolution reaction at a low overpotential

2.1 Chemicals and reagents

The Co(NO₃)₂·6H₂O (≥98.5%), ammonium fluoride (NH₄F, ≥96%), ammonium molybdate ((NH₄)₆Mo₇O₂₄·4H₂O, ≥98%), potassium hydroxide (KOH, ≥85%) and ethanol (C₂H₅OH) were obtained from Sinopharm Co., Ltd. Sodium hypophosphite (NaH₂PO₂·H₂O, ≥99%) was obtained from Tianjin Beilian Fine Chemicals Development Co., Ltd. Nickel sulfate hexahydrate (NiSO₄·6H₂O, ≥98.5%) was obtained from Tianjin Zhiyuan Chemical Reagent Co., Ltd. All reagents and chemicals were used without further purification.

2.2 The preparation of NiMoO₄/NF

Normally, a piece of commercial nickel foam (1 × 5 cm²) was washed for 10 min under ultrasonic waves with 3 M HCl solution, deionized water (DI) and ethanol in turn. This step of pretreatment could remove nickel oxide, oil and various organic pollutants from the surface of nickel foam. Then NF was dried naturally at room temperature. After that, dissolved 8 mmol NiSO₄·6H₂O (2.103 g) and 2 mmol (NH4)₆Mo₇O₂₄·4H₂O (2.472 g) in 50 ml of DI water and mixed the solution continuously until a clear and transparent solution was formed, then transferred the solution into an 80 ml teflon-lined autoclave and put the washed NF in it. The autoclave was heated at 150 °C for 6 h after sealing. Until the system was at room temperature, the NF with yellow–brown sediment was rinsed repeatedly with DI water and ethanol. Afterwards, the sample was dried in an oven at 60 °C overnight.

2.3 The formation of CoP@NiMoO₄/NF with electrodeposition

The CoP@NiMoO₄/NF was prepared by electrodeposition. The solution of electrodeposition was constituted of 20 mmol Co(NO₃)₂·6H₂O (5.82 g), 20 mmol NaH₂PO₂·H₂O (2.12 g), 25 mmol NH₄F (0.925 g) and 50 ml DI water. The NiMoO₄/NF electrode and carbon rod played a role in the working and counter electrodes, respectively. Electrodeposition of CoP was prepared at a constant current density of 100 mA cm⁻². The samples were recorded as CoP-X@NiMoO_4/NF by changing the deposition time (X = 1, 6, 9, 12, 15 min). The obtained samples were washed with DI water and ethanol several times, and dried overnight in an oven at 60 °C. The mass loading of the CoP-12@NiMoO₄ was 0.5 mg cm⁻². And other relevant information was in Supporting Information.

3.1 Materials synthesis and characterization

Scheme 1 illustrated the preparation of self-supported CoP-12@NiMoO₄/NF electrode with core-shell heterojunction structures. Nickel foam (NF) with low cost and excellent conductivity was used as the base. NiMoO₄ nanorod arrays had grew successfully on the surface of NF under a hydrothermal route. Next, CoP nanosheets were prepared by electrodeposition on NiMoO₄ nanorod arrays. The color of NF changed from light silver to dark yellow and then to light blue during the above preparation process, indicating that the various products had been prepared successfully.

Close

Scheme 1

Schematic illustration for the preparation of the self-supported core-shell structure amorphous CoP-12@NiMoO₄/NF electrode.

Export to PPT

The image in Fig. 1a showed that a large number of NiMoO₄ nonorod arrays with smooth surfaces grew uniformly on the surface of NF. The self-supported NiMoO₄ nanorod arrays could expose more active sites and also facilitate mass diffusion to further improve the catalytic activity under a high current density. The images in Fig. 1b and Fig. S1a showed that the flower-like CoP nanosheets had been electrodeposited successfully on NF. Plentiful lamellar structures were conducive to the formation of active sites and the enhancement of electron transport. As the images in Fig. 1c and Fig. S1b shown, it was observed that abundant CoP nanosheets grew on the surface of NiMoO₄ nanorod arrays, which strongly demonstrated that CoP-12@NiMoO₄/NF was core-shell structure, with NiMoO₄ nanorods acting as the core and CoP nanosheets as the shell. CoP nanosheets with crumpled morphology were observed on the gaps between rods and laminas, confirming that NiMoO₄ nanorods were bridged and entangled by CoP nanosheets. In such a structure, CoP nanosheets could act as a conductive network of shuttled electrons. It was worth noting that the prepared unique core-shell structure had plentiful rough surfaces, which contributed to the oxygen adsorption, thus promoting the reaction with electrolyte. Also, the adequate interlamellar space provided a release channel for the generated oxygen, consequently enhancing the stability of OER under a high current density (Liu et al., 2020). TEM analysis further confirmed the SEM results. Fig. 1d showed the prepared NiMoO₄ nanorods with obvious lattice distance of 0.668 nm, which correspond to PDF#12–0348. From Fig. S2a, the lattice distances of NiMoO₄ phase was 0.336 nm, which correspond to PDF#45–0142. These two phases were both proved by XRD. And in Fig. 1e, it was found that CoP nanosheets grew densely on NiMoO₄ nanorods to form the core-shell structure, which were exactly shown as the SEM image. As shown in Fig. 1f, there obviously existed heterojunction interface between NiMoO₄ phase and CoP phase. It demonstrated the lattice spacing of 0.335 nm and 0.371 nm, which correspond to NiMoO₄(2 2 0) and CoP (1 0 1) lattice plane respectively. Fig. 1g was the HRTEM image of CoP nanosheets, in which amorphous substances could be observed in the red region. Furthermore, the crystal lattice was clearly found with the planes spacing of 0.244 nm, which was consistent with (1 0 2) planes of CoP (PDF#29–0497). Obviously, CoP contained both crystalline and amorphous substances, where the boundary between the crystalline substances and the amorphous substances provided more active sites of Co³⁺ to accelerate the OER process. The formation of amorphous phosphates was very favorable for producing large active regions and active sites which were deeply exposed (Guo et al., 2018). The amorphous structure of high energy state can give low activation energy of electron transfer on the surface, because the long-range disordered structure lowered the energy barrier, so that it can significantly improve the catalytic activity. The distribution of Ni, Co, O, Mo and P elements were confirmed in the EDX mapping of Fig. 1h. From Fig. 1h, it can be seen obviously that Ni, Mo and O were mainly distributed on nanorods, while Co and P were mainly distributed on nanosheets, which indicated that CoP-12@NiMoO₄/NF had been successfully prepared.

Close

Fig. 1

SEM images of NiMoO₄/NF (a), CoP-12/NF (b), CoP-12@NiMoO₄/NF (c). HRTEM image of NiMoO₄/NF (d). TEM image of CoP-12@NiMoO₄/NF (e), HRTEM image of CoP-12@NiMoO₄/NF (f), CoP-12 /NF (g). EDS elemental mapping of CoP-12@NiMoO₄/NF (h).

Export to PPT

The crystal structure and composition of the products at different stages were analyzed by XRD. The powders for XRD characterization were obtained from the surface of NF. As shown in Fig. 2a, the diffraction peaks at 9.83°, 13.57° and 29.78° are indexed to (0 0 1), (1 0 0), (0 0 3) planes of NiMoO₄ (JCPDS #04–017–0338), respectively. And the typical (2 2 0), which was the intense diffraction peak and (1 3––3) diffraction reflections at 2θ = 26.59° and 48.92°, indicating the formation of β-phase NiMoO₄ (PDF#45–0142). All peaks were very consistent with standard diffraction cards and partly with Li (Yang et al., 2021). It was necessary to point out the prepared NiMoO₄ (PDF#45–0142) was metastable β-phase (Haetge et al., 2012), which showed higher OER reactivity than stable α-NiMoO4 (Ratha et al., 2017). It was reported that the coordination state of Mo cation caused the distinction of structure between β-NiMoO₄ and α-NiMoO₄. In α-NiMoO₄, the pseudo-octahedral symmetry is formed, which is attributed to the Mo ions coordinate with oxygen atoms. While in β-NiMoO₄, the Mo ions are coordinated with oxygen atoms in a distorted tetrahedral state (Kaddouri et al., 2004). Therefore, the prepared NiMoO₄ exists in α-NiMoO₄ and β-NiMoO₄ phase, which are also consistent with the lattice distances in HRTEM images of Fig. 1d and Fig. S2a. The presence of β-NiMoO₄ may greatly improve the electrochemical activity of the electrode. Fig. S3 showed that the CoP phase diffraction peak was weak, indicating that the CoP was mainly amorphous phase, which was also corresponding to the HRTEM image in Fig. 1g. The above evidence proved that amorphous CoP nanosheets had been electrodeposited on the surface of NiMoO₄ nanorod arrays, which can also be proved in Fig. S4.

Close

Fig. 2

XRD patterns of NiMoO₄ (a). High-resolution XPS spectra for CoP-12@NiMoO₄/NF of O 1 s (b), Ni 2p (c), P 2p (d), Co 2p (e), Mo 3d (f).

Export to PPT

The species and chemical valence states of NiMoO₄/NF electrode before and after CoP electrodeposition were identified by X-ray photoelectron spectroscopy (XPS), as shown in Fig. S5, Fig. S6 and Fig. 2b-h. The spectra of all the detected elements were corrected by adventitious carbon which is at 284.80 eV (Zhang et al., 2016). It can be found from Fig. S5, there existed Ni, Mo and O elements in NiMoO₄/NF electrode, and the high resolution Ni 2p XPS spectra and Mo 3d XPS spectra were shown in Fig. S5b and Fig. S5c. After reasonable CoP electrodeposition, the presence of Co and P elements were appeared in CoP-12@NiMoO₄/NF electrode, which were well correlated with EDX mapping. In the CoP-12@NiMoO₄/NF electrode, the composition of Ni: Mo: O: Co: P: C was 1.88%: 1.53%: 51.85%: 20.03%: 5.97%: 18.74%. Fig. 2b showed the high resolution O 1 s XPS spectra for CoP-12@NiMoO₄/NF, which could be deconvolved into three peaks named as O1, O2 and O3. The O1 peak at 530.20 eV was attributed to the metal–oxygen (M−O) bond, the O2 peak at 531.12 eV was related to the oxygen atoms of hydroxyl group while the O3 peak at 532.41 eV was assigned to the adsorbed water near the surface. Fig. 2c showed the high resolution Ni 2p XPS spectra of CoP-12@NiMoO₄/NF electrode, the two peaks at 874.04 eV and 855.92 eV were attributed to Ni 2p _1/2 and Ni 2p _3/2 of Ni ²⁺, which were with a separation of 18.12 eV (Zhao et al., 2019). While peaks located at 872.38 eV and 854.66 eV were assigned to Ni 2p _1/2 and Ni 2p _3/2 of Ni ³⁺. According to the integral areas and the p spin–orbit splitting was 17.8 eV, it can be concluded that the Ni element in prepared catalyst was mainly in the Ni ²⁺ state. The atom ratio of Ni²⁺ state and Ni³⁺ state was about 85%:15% (Table S1). A little of Ni³⁺ was probably partially oxidized in the air. As shown in Fig. 2d, the peaks of 128.40 eV and 128.80 eV were attributed to P 2p _3/2 and P 2p _1/2. Another peak at 132.42 eV was attributed to the oxidation of the surface resulting from the exposure to air, which caused the oxidation of P element. In the Fig. 2e, the two peaks at 796.64 eV and 781.14 eV were attributed to Co 2p _1/2 and Co 2p _3/2 of Co ²⁺, respectively. The peaks at 795.40 eV and 779.75 eV were assigned to Co 2p _1/2 and Co 2p _3/2 of Co³⁺, the other two peaks at 802.41 eV and 786.31 eV were related satellite peaks. On the basis of the d-zone theory, transition metals with high d-zone centers can intensify the adsorption capacity of oxygen-containing intermediates, which contribute to enhancing OER performance (Bo et al., 2020). It was obviously that Ni and Co were both mainly in the Ni²⁺and Co²⁺ state. The coexistence of Co²⁺ and Co³⁺ led to abundant oxygen vacancy defects, which played a facilitating role during OER. Fig. 2f showed the high resolution Mo 3d XPS spectra, the peaks of 234.54 eV and 231.40 eV were attributed to Mo 3d _3/2 and Mo 3d _5/2, respectively. Since peak positions and the separation were 3.14 eV, according to the d spin–orbit splitting of 3.1 eV, one can find that the presence of Mo⁶⁺ in the prepared electrode (Chen et al., 2020). The XPS measurement analysis demonstrated that the states of Ni, Mo and O were + 2, +6 and − 2, respectively, which was consistent well with the chemical formula NiMoO₄. The results showed that after electrodeposition, the electron cloud density around the Mo atom decreased due to the existence of core-shell structure, which was evidenced by the small movement toward higher binding energy (from 234.45 to 234.55 eV). It was noteworthy that the binding energy of Ni atoms in CoP-12@NiMoO₄ shifted to the right compared with that of NiMoO₄. It was well known that P atoms were more electronegative than Ni atoms, but with the formation of heterojunctions, the electron density around Ni atoms still increased, leading to the decrease of BE. As proved by the theoretical analysis below, the formation of CoP-NiMoO₄ heterojunctions promoted electron redistribution and optimized the electronic structure. The strong electronic interaction between CoP and NiMoO₄ was conducive to regulating the adsorption capacity of the electrode for oxygen-containing intermediates, thereby improving the conductivity and catalytic ability (Sha et al., 2020).

3.2 Electrochemical performance

A standard three-electrode system was used to evaluate the electrocatalysts performance of OER in an alkaline solution (1 M KOH). Fig. 3a showed the LSV curves of the prepared samples. The overpotentials at 10 mA cm⁻² of NF, NiMoO₄/NF, CoP-12/NF, CoP-12@NiMoO₄/NF were 320 mV, 217 mV, 307 mV, 165 mV, respectively. The comparison of OER activity of CoP-12@NiMoO₄/NF with other reported OER electrocatalysts is shown in Table S2. Obviously, the prepared CoP-12@NiMoO₄/NF electrode had excellent electrochemical activity, which is because that CoP-NiMoO₄ heterojunctions promotes strong electronic interaction at the interface. In addition, the self-supported core-shell structure can increase specific surface area, minimize agglomeration and the synergies between core-shell components are also beneficial to the faster transfer of electrons to promote OER performance (Bai et al., 2018). The activity was greatly enhanced by the formation of core-shell structure with CoP-NiMoO₄ heterojunctions, which proved that our work was feasible and remarkable. The overpotentials of CoP-1@NiMoO₄/NF, CoP-6@NiMoO₄/NF, CoP-9@NiMoO₄/NF, CoP-12@NiMoO₄/NF, CoP-15@NiMoO₄/NF were 221 mV, 200 mV, 184 mV, 165 mV, 217 mV, respectively as shown in Fig. S7. The current density increases first and then decrease, and the overpotential increases after decreasing. Because long time electrodeposition led to excessive sediments, the large overlap area between the nanosheets resulted in less effective contact area between electrode and electrolyte. At the same time, the thick electrodeposition layer hindered the overflow of oxygen, which strongly reduced the OER activity. Therefore, the core-shell structure CoP-12@NiMoO₄/NF with the suitable electrodeposition time had abundant active sites, which were beneficial to the adsorption of OH-and the electron transfer. As the Fig. 3b shown, under the current density of 10 and 50 mA cm⁻², the overpotential of CoP-12@NiMoO₄/NF is much lower than that of other electrodes, which fully proved the active reaction kinetics of prepared CoP-12@NiMoO₄/NF and the feasibility of this work (Table S3).

Close

Fig. 3

The LSV curves of prepared electrodes (a). Comparison of overpotentials at 10 and 50 mA cm⁻² (b). The Tafel slopes of various electrodes (c-d). Linear fitting of double-layer capacitance for different materials (e-f). Nyquist plots of the CoP-12/NF, NiMoO₄/NF and CoP-12@NiMoO₄/NF electrodes (g). I-T curve of CoP-12@NiMoO₄/NF for 10 h (h).

Export to PPT

Usually, the smaller Tafel slope is associated with the smaller increase of overpotential during OER. As the Fig. 3c shown, the Tafel slopes of NF, CoP-12/NF, NiMoO₄/NF, CoP-12@NiMoO₄/NF were 112 mV dec^-1, 67 mV dec^-1, 68 mV dec^-1 and 45 mV dec^-1. It had been proved that the prepared CoP-12@NiMoO₄/NF electrode had the most rapid reaction kinetics. In Fig. 3d, the Tafel slopes of CoP-1@NiMoO₄/NF, CoP-6@NiMoO₄/NF, CoP-9@NiMoO₄/NF were 73 mV dec^-1, 68 mV dec^-1, 58 mV dec^-1. However, the Tafel slope of CoP-15@NiMoO₄/NF reached 186 mV dec^-1. The change trend of the Tafel slope was completely consistent with that of LSV curves. The increase of the Tafel slope was due to the decrease of active sites and the electron transfer channel. Too long electrodeposition time would cause thicker sediments, which impeded electron transfer and decline the reaction kinetics.

Fig. S8 was the CV curves of CoP-12@NiMoO₄/NF with the scanning rate from 20 to 100 mV s⁻¹ in the no Faradaic interval, which was used as the double-layer capacitance (C_dl) to evaluate the electrochemical active surface area (ECSA) during OER (Hu et al., 2019). The CV curves of other prepared electrodes were in the Fig. S9. As shown in Fig. 3e, the double-layer capacitance (C_dl) values were 2.22 mF cm⁻² for NF, 2.44 mF cm⁻² for CoP-12/NF, 3.26 mF cm⁻² for NiMoO₄/NF and 4.69 mF cm⁻² for CoP-12@NiMoO₄/NF. Therefore, according to the correlation between ECSA and electrochemical OER performance, it can be inferred that CoP-12@NiMoO₄/NF had exposed more active sites than the single component sample, which could be due to the the hierarchical core-shell morphology, also facilitating more efficient adsorption of hydroxyl ions. In Fig. 3f, the C_dl values of CoP-6@NiMoO₄/NF, CoP-9@NiMoO₄/NF, CoP-12@NiMoO₄/NF, CoP-15@NiMoO₄/NF were 3.35 mF cm⁻², 3.82 mF cm⁻², 4.69 mF cm⁻², 2.25 mF cm⁻². Obviously, overmuch sediments led to a decrease in ECSA value, indicating that a lot of active sites had disappeared, which was mutually supported by overpotentials of LSV curves and Tafel slopes. In addition, we also calculated the roughness factors (RF) of CoP-12@NiMoO₄/NF and CoP-12/NF. The RF was 118 and 61, respectively. Obviously, the active surface area of CoP-12@NiMoO₄/NF was twice than that of CoP-12/NF, which demonstrated that CoP-12@NiMoO₄/NF core-shell structure had the stronger electrocatalytic activity.

Although ECSA was a key factor, it was not the only one to determine electrocatalyst activity. In this case, the charge transfer dynamics in the OER process were explored by recording electrochemical impedance spectroscopy (EIS) of various electrodes. As the Fig. 3g shown, the semicircle diameter of CoP-12@NiMoO₄/NF was smallest compared with CoP-12/NF and NiMoO₄/NF electrodes, indicating that prepared CoP-12@NiMoO₄/NF electrode had complete local electronic structure, the fastest charge transfer rate and the best conductivity during OER (Table S4). The lowest R_ct value of CoP-12@NiMoO₄/NF electrode was attributed to the enhanced electronic interaction at the heterojunction interface, which was beneficial to shorten the electron transport distance. Therefore, the kinetic barrier of OER process was reduced. The EIS results were consistent with the above data as well.

In the process of electrolysis of water, the bubbles that generated on the electrode surface rapidly nucleate, adsorb, aggregate and separate, which can produce small fluctuations with the change of voltage, thus decreasing the electrocatalytic activities (Hu et al., 2021). Therefore, it was necessary to test the stability of the electrocatalyst. After 5000 cycles CV, the overpotential of prepared CoP-12@NiMoO₄/NF electrode changed from 165 mV to 181 mV (Fig. S10), indicating that the electrode had little electrochemical activity degradation after 5000 cycles CV. In order to further explore the stability, CoP-12@NiMoO₄/NF electrode was also tested by chronoamperometry at a fixed potential in 1 M KOH. As the Fig. 3h and Fig. S11 shown, CoP-12@NiMoO₄/NF electrode showed steady durability under a certain voltage for 10 h long-term electrochemical process with negligible degradation and the good stability of different current densities. The excellent performance of CoP-12@NiMoO₄/NF electrode was attributed to the good growth of CoP nanosheets on NiMoO₄ nanorod arrays to form a core-shell structure under the appropriate conditions. The core-shell structure facilitated the adsorption of reactants, intermediate transformation and the product desorption, which avoided the material agglomerate and promoted the rapid release of O₂ in close contact with the electrolyte. In the Fig. 3h, the small current attenuation may be attributed to sectional oxidation of CoP during OER (Jin, 2017). It had been shown that although metal phosphates were easily oxidized during OER, they still had more electrochemical activity than the corresponding hydroxide or oxide, possibly because OER process could produce more catalytic active sites, making it promising for practical applications (Xu et al., 2018). And in Fig. S12, the high OER Faradic efficiency for CoP-12@NiMoO₄ has shown that it can achieve efficient catalysis while maintaining long-term stability. The morphology of CoP-12@NiMoO₄/NF before and after 5000 cycles CV was compared to research structural stability. As shown in Fig. S13, the structure of CoP nanosheets was still relatively complete, indicating the electrode had a good structural stability. The outstanding structure stability of CoP-12@NiMoO₄/NF was mainly attributed to the direct growth of the active material on NF without the polymer binder and additional conductive additives, which reduced the undesirable interface of the electrode and maintained the structure under high current density.

XPS technology was used to investigate the chemical composition and the oxidative state change of CoP-12@NiMoO₄/NF electrode before and after 5000 CV cycles in 1 M KOH. The image in Fig. 4a showed that Ni, Mo O, Co and P elements still existed but the intensity of Mo peak and P peak decreased significantly. In the high resolution O 1 s XPS spectra of Fig. 4b, there were three peaks appeared at 530.00 eV, 531.12 eV and 528.20 eV. Obviously, the M−O bond and the oxygen atoms of hydroxyl groups still existed, the peak at 532.41 eV of the adsorbed water on the surface disappeared. But there was a new peak at 528.20 eV, which was attributed to the new hydroxyl species. In the high resolution Ni 2p XPS spectra of Fig. 4c, the two peaks at 873.72 eV and 855.77 eV were assigned to Ni 2p _1/2 and Ni 2p _3/2 of Ni ²⁺, respectively. While peaks located at 871.83 eV and 854.31 eV were assigned to Ni 2p _1/2 and Ni 2p _3/2 of Ni ³⁺. As Fig. 4d shown, the peak at 127.55 eV was assigned to P 2p _3/2, the other peak at 131.79 eV was assigned into P-O bond. In Fig. 4e, the two peaks at 795.14 eV and 780.94 eV were attributed to Co 2p _1/2 and Co 2p _3/2 of Co ²⁺, respectively. The peaks at 793.94 eV and 778.83 eV were assigned to Co 2p _1/2 and Co 2p _3/2 of Co³⁺. In the high resolution Mo 3d XPS spectra of Fig. 4f, the peaks of 234.01 eV and 230.92 eV were attributed to Mo 3d _3/2 and Mo 3d _5/2, respectively. And the peak at 227.30 eV was assigned to Mo⁵⁺. It showed that Mo had more metallic properties. After 5000 cycles CV, the composition of Ni: Mo: O: Co: P: C was 9.49%: 0.52%: 41.97%: 7.28%: 1.36%: 39.39%. It can be simply found that the content of Ni increased significantly, while the content of Mo decreased greatly. Mo was dissolved or leach out, maybe resulting in the formation of nickel hydroxide. At the same time, the bond shape also demonstrated a large increase in nickel hydroxide. And the atom ratio of Ni²⁺ state and Ni³⁺ state was changed from 85%: 15% to 40%: 60% (Table S1). The Co 2p related peaks also shifted slightly towards lower BE, but the band shape was the same as that before 5000 cycles CV, thus confirming the presence and excellent stability of the CoP compound. Meanwhile a little Co species were converted to a higher oxidation state of Co³⁺. Particularly, the rapid decay of Co-P bond signal had also indicated that the phosphide had been oxidized to form a hydroxyl oxide. The increase of the valence was conducive to the combination with oxygen-containing species in the reaction, which had produced more active sites and enhanced the OER performance (Lv et al., 2020; Feng et al., 2020). Since the anodic potential further oxidized the metals on the catalyst surface, an increase in the oxidation state should have been observed. However, for our experiment, after 5000 cycles CV, the BE of Mo, Ni, and Co atoms all shifted to a lower value. Therefore, it is reasonable to infer that the metal cations (Ni, Co, and Mo) participate in the interface electron transfer process in the OER process. It was also proved that the heterojunction which was formed at the interface promoted the electron transfer, optimized the local electron structure and enhanced the OER performance.

Close

Fig. 4

XPS survey spectrum of CoP-12@NiMoO₄/NF after 5000 CV cycles (a). High-resolution XPS spectra after 5000 CV cycles of O 1 s (b), Ni 2p (c), P 2p (d), Co 2p (e), Mo 3d (f).

Export to PPT

Close

Fig. 5

Schematic image of OER (a). The differential charge density image of CoP-12@NiMoO₄/NF (b). The image of the density of states for CoP (c), NiMoO₄ (d) and CoP-12@NiMoO₄/NF (e).

Export to PPT

3.3 Theoretical investigation

To investigate the reason for OER activity of CoP-12@NiMoO₄/NF, precise DFT calculations were studied based on the HRTEM, XPS, XRD and theoretical calculations results comprehensively. The details of the calculations were described in the ‘‘Supporting Information’’. The configurational structure of CoP, NiMoO₄, and CoP-NiMoO₄ heterojunction were displayed in Fig. 5a. As shown in Fig. 5b, the differential charge density indicated that the heterojunction surfaces exhibited differential electron distribution. Yellow and cyan represented the accumulation and depletion of charge in space respectively, which evidently demonstrated the charge transfer on the boundary area from NiMoO₄ to CoP. Such a heterojunction can disrupt the periodicity of the surface by introducing an amorphous CoP structure, resulting in an uneven charge distribution around the CoP, which was consistent with the XPS analysis, proving the strong electronic interactions of CoP-NiMoO₄ heterojunction. In order to gain insight into the electron transfer at the CoP-NiMoO₄ heterojunction interface, the density of states was subsequently supplemented in Fig. 5c-e. The density of states for CoP from about 0 to −4 eV was highest, which was consisted with the energy band diagram in Fig. 6a. And the density for NiMoO₄ from about −8 to −18 eV was zero, which exhibited poor electrical conductivity as Fig. 6b shown. Compared with both of them, CoP-12@NiMoO₄/NF electrode displayed a smallest bandgap showing the improved conductivity. Furthermore, when CoP-NiMoO₄ heterojunction was formed, more Co elements entered the forbidden band, therefore, the electrons redistributed at the interface, where it can be seen that the CoP-NiMoO₄ (75 eV) heterojunction interface had higher DOS near the Fermi level than the single component sample of CoP (20 eV) and NiMoO₄ (28 eV), indicating that the electronic structure of CoP-NiMoO₄ heterojunction and the metallic properties were conducive to rapid electron transfer in OER, which can be supported by the smallest R_ct of CoP-12@NiMoO₄/NF electrode.

Close

Fig. 6

Band diagram and element projection of CoP (red is Co, black is P) (a) and NiMoO₄ (red is Mo, yellow is Ni and purple is O) (b). Free energy diagram of OER intermediates for CoP (c), NiMoO₄ (d) and CoP-NiMoO₄ (e). Free energy diagram of OER intermediates at U = 0 V (f).

Export to PPT

OER free energy for CoP, NiMoO₄ and CoP-NiMoO₄ heterojunction was calculated at Co sites. The reaction energy barrier was an important parameter for evaluating catalytic activity, which can be considered as the free energy of the rate-determining step (RDS) (Xing et al., 2022). As shown in Fig. 6c-d, the RDS at 0 V of CoP and NiMoO₄ was the fourth step of the OOH* decomposition reaction, which the reaction energy barrier was 8.743 eV and 7.685 eV, respectively. OOH* should have been more active than the intermediates of O* and OH*, but the reasons for this situation can attribute to the stable six-member ring structure of *OOH, which was forming hydrogen bonds with the exposed oxygen atoms on the surface of NiMoO₄. In the Fig. 6e-f, the exciting thing was that not only the RDS at 0 V for CoP-NiMoO₄ heterojunction was changed to the first step, but also the reaction energy barrier of RDS was reduced to only 2.519 eV, which was down to about 1/4 compared with CoP. And the RDS of CoP, NiMoO₄ and CoP-NiMoO₄ heterojunction were not changed at 1.23 V, however, extra voltage had reduced the reaction energy barrier of RDS. In addition, combined with the difference of electrostatic potential of CoP, NiMoO₄ and CoP-NiMoO₄ heterojunction in Fig. S14-15, it can be clearly concluded that the formation of CoP-NiMoO₄ heterojunction can increase electrostatic potential. And the high electrostatic potential was conducive to OER reaction.