Hongliang
Yu‡
,
Jiwei
Wang‡
,
Xianming
Wang
,
Ningbo
Yu
,
Xunxing
Zuo
,
Yanan
Xue
,
Ning
Cai
,
Jie
Liu
,
Jianzhi
Wang
* and
Faquan
Yu
*
Key Laboratory for Green Chemical Process of Ministry of Education, Hubei Key Laboratory for Novel Reactor and Green Chemistry Technology, Hubei Engineering Research Center for Advanced Fine Chemicals, School of Chemical Engineering and Pharmacy, Wuhan Institute of Technology, LiuFang Campus, No.206, Guanggu 1st road, Wuhan 430073, Hubei, China. E-mail: fyu@wit.edu.cn; fyuwucn@gmail.com; Fax: +86 27-87194465; Tel: +86 27-87194980
First published on 17th November 2022
Designing nonprecious but efficient electrocatalysts for the hydrogen evolution reaction (HER) in alkaline solution still remains a significant challenge. Herein, we report the NiMoP nanosheets on hollow Cu3P nanotube arrays on the Cu foam (NiMoP@Cu3P/CF) via solution immersion, hydrothermal reactions and phosphorization. By virtue of its unique core–shell open tubular structure, NiMoP@Cu3P/CF shows large specific surface area and mesoporous, short diffusion paths, which facilitates mass transport and easy accessibility of water, resulting in a superior HER activity in 1.0 M KOH. NiMoP@Cu3P/CF only requires 81 mV and 226 mV of overpotential when the current density is 10 mA cm−2 and 100 mA cm−2 for HER in 1.0 M KOH, respectively, with a low Tafel slope (133.1 mV dec−1) and as well as superior durability. Our work provides a new idea to design a powerful electrocatalyst for HER, which will be highly expected for various practical applications.
In the past few years, a lot of research has being conducted to develop dual-functional electrocatalysts to enhance the catalytic performance of HER,8,9 such as transition metal chalcogenides,10,11 oxides,12,13 phosphides,14,15 nitrides,16–18etc. Transition metal phosphide (TMP) is another potential non-noble metal catalyst similar to transition metal sulfide for HER. Experiments confirmed that the charge density is transferred from the transition metal to P in transition metal phosphides, resulting in the formation of two active sites of positively charged metal and negatively charged P, thereby enhancing the catalytic activity of the phosphide.19,20 Currently, Ni2P, CoP, FeP, WP, MoP and other TMPs have been continuously reported in the field of catalysis.21–24 Compared with oxides, hydroxides, and chalcogenide compounds, TMPs have low cost, nontoxicity, good chemical stability, and catalytic activity toward HER in electrolytes with a wide range of pH values, high electrical conductivity, and therefore often exhibit excellent catalytic activity.25,26 Compared with single-metal phosphides, bimetallic phosphides show excellent catalytic activities due to the coordination of two metals can provide bi-directional rich active centers and improved electronic conductivity. Tian et al.27 used carbon quantum dots to prepare NiMoP hollow nanoparticles and exhibited excellent electrocatalytic hydrogen evolution performance, in terms of a low overpotential of 183 mV at 10 mA cm−2, high conductivity, and remarkable long-term stability in acidic media for HER. However, the improvement of the conductivity in bimetallic phosphides is still limited, which often leads to instability of strong alkaline electrolytes and high overpotential under long-time operation. To solve this problem, the catalysts can be deposited on the conductive substrates to obtain accessibility of HER and OER relevant reactants with catalytic centers. It is also a good strategy by constructing hybrid electrocatalysts with hollow micro/nanostructures with a large number of exposed active sites and reduced diffusion length for both mass and charge transport.26,28
The electrochemical performance of electrocatalysts is strongly dependent on their morphologies and electrochemically active surface areas. Particularly, tubular Cu3P materials with both hollow and one-dimensional open structures, which provide high specific surface areas, low densities and fast diffusion of electrolyte/ionic species, have been used as effective electrocatalysts because they can better contact the electrolyte and have lower resistance along their longitudinal direction.29 Here, we have prepared a Cu3P hollow structure electrocatalyst on the copper foam through in situ reaction and Kirkendall effect,15 which has a higher specific surface area, more active sites and better chemical stability. This inspired us to explore a new type of electrocatalyst with core–shell structure by assembling NiMoP nanosheets on the surface of Cu3P. High performance electrocatalysts will be prepared by using the synergistic effect of the two components. The composite electrocatalyst is expected to have strong electron coupling and enhance the interface electron transport ability, prevent electrolyte erosion, reduce the dissolution of phosphorus, and improve the catalytic stability and durability.
In this work, Cu(OH)2 nanorod arrays were in situ grown on the surface of CF by chemical oxidation, then NiMo LDH nanosheets were directly grown on the surface of Cu(OH)2 nanotubes by hydrothermal method, and finally NiMoP@Cu3P nanotubes with hollow layered structure were prepared by reaction with sodium hypophosphite in a tube furnace. The nano-array of the core–shell structure is conducive to the full contact between the active material and the electrolyte. The nano-sheet structure of NiMoP provides a large specific surface area and is beneficial to expose more active sites and reduce the resistance. These unique structures make NiMoP@Cu3P nanoarrays have excellent HER performance in alkaline conditions.
By contrast, the preparation of NiMo LDH/CF electrode was similar to the above-mentioned method except for the use of bare CF instead of Cu(OH)2/CF.
ERHE = ESCE + 0.241 + 0.059 × pH |
Linear sweep voltammetry (LSV) plots were recorded in the potential range of −1.668 to −0.868 V vs. RHE for HER at a scan rate of 2 mV s−1 with 85% iR compensation.
The Tafel plots were directly derived from the corresponding LSV curves according to the Tafel equation:
η = a + blogj |
To measure double-layer capacitance (Cdl) of the resulting electrocatalysts, the cyclic voltammetry measurement (CV) at different scan rates (20 to 100 mV s−1) was carried out in the potential range without faradaic reactions. There is a positive correlation between electrochemically active surface (ECSA) and Cdl. The higher the Cdl, the larger the electrochemically active specific surface area.31,32
Electrochemical impedance spectroscopy (EIS) analyses were researched at a potential of −0.2 V vs. RHE for HER in the frequency range of 10 Hz to 100000 Hz with AC amplitude of 5 mV. The durability test was measured by both CV and chronopotentiometric response.
Fig. 1 Schematic illustration of the synthesis process of hollow hierarchical structure of NiMoP@Cu3P nanorod arrays on copper foam substrate. |
Crystalline structure and phase purity of the NiMoP@Cu3P/CF was analyzed using X-ray diffraction (XRD) studies (Fig. 2). The intensity of the overall XRD peak of NiMoP@Cu3P/CF is shown in Fig. S2.† XRD results show that three strong signal peaks of NiMoP@Cu3P/CF are located at 43.4° (111), 50.5° (200) and 74.2° (220), which are attributed to the 3D Cu foam (JCPDS No.04-0836).33 In addition, the sample has obvious diffraction peaks at 36.1°, 39.3°, 41.7°, 45.1°, 46.2°, 47.3°, 53.4°, 66.6° and 77.3°, corresponding to (112), (202), (211), (300), (113), (212), (104), (214) and (412) crystal planes of Cu3P (JCPDS No. 71-2261),34 indicating the exposed copper framework and Cu(OH)2 nanotubes are converted to Cu3P during the phosphating process. The diffraction peaks of the sample at 40.8°, 44.6°, 47.3° and 74.7°, correspond to the (111), (201), (210) and (400) crystal planes of Ni2P (JCPDS NO.03-0953).34 The diffraction peaks at 27.6°,31.9° and 43.2°, correspond to the (001), (100) and (101) crystal plane of MoP (JCPDS NO.24-0771).35 These results indicate that Cu3P, Ni2P and MoP were successfully prepared on the CF substrate. The diffraction peaks of Ni2P and MoP shifted slightly, indicating that there is an interaction between Ni2P and MoP and Cu3P, which affects the positions of the diffraction peaks of Ni2P and MoP.
The morphology and structure of the synthesized sample were characterized by SEM, and the results are shown in Fig. 3 and S3.† The SEM image of Cu(OH)2/CF shows that the Cu(OH)2 nanorod array is densely and uniformly covered on the surface of CF (Fig. S1†). The diameter of Cu(OH)2 nanorods is about 300 nm. Cu(OH)2/CF has a highly smooth surface, which can be used as a hard template for subsequent synthesis. As shown in Fig. 3a, the SEM of NiMo LDH@Cu(OH)2/CF shows that hierarchical nanotubes are uniformly arranged on the CF substrate, and the size of the nanotubes is about 1 μm. From left to right, the surface of the nanotubes grows in a porous structure, formed by staggered stacked nanosheets. This structure can increase the specific surface area of the material, thereby improving its catalytic performance. The SEM image of NiMoP@Cu3P/CF is shown in Fig. 3b. It can be seen that the tube-like structure of the NiMo LDH@Cu(OH)2/CF still exists after phosphating. The diameter of the nanorods is about 800–900 nm, but the surface shrinks due to the high temperature annealing of the nanosheet structure. The smooth surface may be attributed to the volume shrinkage of NiMo LDH nanosheets during the phosphating process. Moreover, the STEM Energy dispersive spectrum demonstrate that the NiMoP@Cu3P/CF includes Cu, Ni, Mo and P (Fig. 3c). Through ICP-MS test, the result shows that the atomic ratio of Cu, Ni and Mo is 7.14:2.32:1. The energy-dispersive spectroscopy (EDS) mapping results of the NiMoP@Cu3P/CF in Fig. 3d clearly suggest the homogeneous distribution of the elements Cu, Ni, Mo, P, and O across the entire nanosheet. The presence of oxygen is mainly due to the oxidation of the sample when exposed to air. These results prove that the core–shell structure of NiMoP@Cu3P/CF was synthesized successfully.
Fig. 3 SEM images of NiMo LDH@Cu(OH)2/CF (a) and NiMoP@Cu3P/CF (b). Energy dispersive spectrum of NiMoP@Cu3P/CF (c). SEM and corresponding EDX elemental mapping images of NiMoP@Cu3P/CF (d). |
The TEM images of NiMo LDH@Cu(OH)2 and NiMoP@Cu3P are shown in Fig. 4a–d and S8.† It can be seen from Fig. 4a that during the growth of NiMo LDH ultra-thin nanosheets on the surface of Cu(OH)2 nanorods, the Kirkendall effect will cause the Cu(OH)2 nanorods to form a hollow structure.36 It can be seen from the inset in Fig. 4a that the width of the NiMo LDH ultra-thin nanosheet is about 260 nm, and the diameter of the hollow Cu(OH)2 nanotube cavity is about 230 nm. Fig. 4b is a TEM image of NiMoP@Cu3P. It can be seen that the nanosheets and hollow nanotubes still exist after phosphating, but the width of the nanosheets on the nanotubes decreases to 180 nm, which is mainly due to the shrinkage of the nanosheets during phosphating. The hollow layered structure has a large specific surface area, which accelerates the transfer of substances and enhances its activity. Fig. 4c and d is a HRTEM image of NiMoP@Cu3P. The lattice fringe with a lattice spacing of 0.502 nm and 0.228 nm can be clearly observed, which corresponds to the (010) and (111) crystal plane of Ni2P, which is consistent with the XRD results.
Fig. 4 Low-resolution (a) TEM image of NiMo LDH@Cu(OH)2. Low-resolution (b) and high-resolution (c and d) TEM images of NiMoP@Cu3P. |
X-ray photoelectron spectroscopy (XPS) was used to test and analyze the elemental composition and valence of NiMoP@Cu3P/CF, thereby revealing the electronic interaction between Cu3P and NiMoP (Fig. 5). The scanning spectrum of NiMoP@Cu3P/CF (Fig. 5a) confirms the presence of Cu, Ni, Mo, P, C and O elements. The presence of the O element may be caused by exposure of the sample to the air. In the Cu 2p spectrum (Fig. 5b), the binding energy peaks at 932.26 and 934.81 eV are attributed to Cu 2p2/3 of Cu+ and Cu2+, respectively, and the peaks at 952.01 and 954.95 eV are also attributed to Cu 2p1/2 of Cu+ and Cu2+. The peaks of 942.32, 944.02, 961.63 and 962.71 eV are characteristic satellite peaks of Cu2+.37,38Fig. 5c is the XPS spectrum of NiMoP@Cu3P/CF material Ni 2p. The peaks at 854.79 and 872.33 eV correspond to the Ni 2p3/2 and Ni 2p1/2 spin orbits of Ni2+, respectively, and the peaks at 856.57 and 874.18 eV correspond to the Ni 2p3/2 and Ni 2p1/2 spin orbitals of Ni3+ valence, respectively.39 In the spin orbit, the peak at 851.75 eV may be attributed to Ni+ or metallic nickel. The peaks at 860.48 and 878.54 eV are satellite peaks of Ni 2p.40Fig. 5d is the XPS spectrum of Mo 3d. It can be clearly seen that Mo 3d has four high-intensity peaks. The peaks at 230.4 and 233.5 eV are attributed to Mo4+ 3d5/2 and Mo4+ 3d3/2 of spin orbits of Mo4+, respectively. The peaks at 232.1 and 235.25 eV are assigned to the Mo6+ 3d5/2 and Mo6+ 3d3/2 of spin orbits of Mo6+, respectively. The positive shift of the Mo element may be due to the existence of the O element.39,41Fig. 5e is the XPS spectrum of P 2p. The strong peak at 132.05 eV can be attributed to phosphorus oxide (P–O), while the features at 127.45 eV and 128.26 eV can be attributed to P 2p3/2 and P 2p1/2 of metal phosphide (Mo–P).42,43 These results indicate that the electron density is transferred from Ni and Mo to Cu and P. This electron density transfer is conducive to exposing more active sites and enhancing the electron transfer between interfaces, thereby increasing the HER activity of NiMoP@Cu3P/CF.
Fig. 5 XPS survey spectrum of NiMoP@Cu3P/CF (a), high-resolution scans of Cu 2p (b), Ni 2p (c), Mo 3d (d), and P 2p electrons for NiMoP@Cu3P/CF (e). |
In order to further understand the electrocatalytic HER kinetics of the prepared materials, the Tafel slope was studied, as shown in Fig. 6b It can be seen that the Tafel slope value of NiMoP@Cu3P/CF is 133.1 mV dec−1, which is lower than the Tafel slope of NiMo LDH@Cu(OH)2/CF (188.1 mV dec−1). The lower Tafel slope value of NiMoP@Cu3P/CF catalyst indicates that the Volmer–Heyrovsky reaction pathway is operative with the superior HER kinetics of desorption of hydrogen (Volmer step) and rapid increase of catalytic current towards the HER catalytic activity.
Electrochemical impedance spectroscopy (EIS) is utilized to probe the resistance and kinetics of electrocatalytic reaction that occurs at the interface between electrode and electrolyte (Fig. 6c and d). In the Nyquist plots measured under open-circuit potential (Fig. 6c), the solution resistance (Rs) of all synthetic materials is very small (<5 Ω), indicating that the rod-shaped structure is in good contact with the CF substrate, thereby reducing electrical resistance. NiMoP@Cu3P/CF electrode exhibits the outstanding HER performance with small charge transfer resistance (Rct) (12.1 Ω) at the overpotential at −200 mV versus RHE (Fig. 6d), which is lower than other samples. The results show that the electron transfer between nanosheet structure NiMoP and hollow structure Cu3P reduces the Rct and improves the conductivity of the composite.
To further estimate the electrochemical activity surface area (ECSA) of CF, NiMo LDH/CF, Cu(OH)2/CF, NiMo LDH@Cu(OH)2/CF, NiMoP/CF, Cu3P/CF and NiMoP@Cu3P/CF, electrochemical double-layer capacitance (Cdl) was measured via cyclic voltammetry (CV) at various scan rates in the non-Faraday region (Fig. 6e). Fig. S4a–h† shows the CV curves of CF, NiMo LDH/CF, Cu(OH)2/CF, NiMo LDH@Cu(OH)2/CF, NiMoP/CF, Cu3P/CF and NiMoP@Cu3P/CF in the voltage range of 0.268–0.368 V versus RHE at different scanning speeds. It can be observed that NiMoP@Cu3P/CF has the largest closed curve at each scanning speed, indicating that it has the largest electric double layer capacitance. Fig. 6e depicts the CV curves of various electrodes at scan rates of 20, 40, 60, 80, and 100 mV s−1. After calculation, the Cdl value of NiMoP@Cu3P/CF is 40.06 mF cm−2, which is significantly higher than CF (2.05 mF cm−2), NiMo LDH/CF (3.19 mF cm−2), Cu(OH)2/CF (11.75 mF cm−2), NiMo LDH@Cu(OH)2/CF (16.59 mF cm−2), NiMoP/CF (27.2 mF cm−2) and Cu3P/CF (31.68 mF cm−2). The electrochemically active specific surface area of the material is positively correlated with the electric double layer capacitance. The higher the electric double layer capacitance corresponds to the larger the electrochemically active specific surface area. The large electrochemically active specific surface area of NiMoP@Cu3P/CF indicates that it has more efficient active sites, which can be attributed to its unique hollow composite structure and electronic effects. Therefore, these parameters such as hollow composite structure, low charge transfer resistance, high active surface area and synergistic effects are all conducive to the progress of HER.
From the perspective of practical application, stability or durability is one of the important properties of the catalyst, and it is the ability of the catalyst to maintain the original activity for a long period of time. The stability of the catalyst can be evaluated by recording the change in cathode current density under the applied overpotential, or by measuring the continuous CV curve. We have also tested the HER stability of NiMoP@Cu3P/CF by continuous cyclic voltammetry scan between −0.2 and 0.2 V vs. RHE at a 100 mV s−1 scan rate. Fig. 7a is the polarization curve after the first and 1000 continuous CV scanning experiments. It can be seen that the hydrogen evolution performance of NiMoP@Cu3P/CF has basically not changed after 1000 continuous CV scanning. Besides, the electrocatalytic durability of NiMoP@Cu3P/CF during HER was further examined in 1.0 M KOH by the chronopotentiometry method at the current density of 10 mA cm−2 for 12 h under alkaline conditions. As shown in Fig. 7b, the overpotential still demonstrates a surprisingly stable HER behavior with a nearly unchanged overpotential for 12 h of continuous electrolysis, confirming the excellent durability of NiMoP@Cu3P/CF. To further investigate the possible morphological or composition changes after long-term HER, we performed XRD (Fig. S5†), SEM (Fig. 7c), and XPS (Fig. S6†) analyses on the post-HER NiMoP@Cu3P/CF. The XRD patterns in Fig. S5† show that the bulk phase of NiMoP@Cu3P/CF remains unchanged after electrolysis. As observed by SEM, the morphology of NiMoP@Cu3P/CF is also well-maintained. We noted that the Cu 2p, Ni 2p, Mo 3d, and P 2p spectra of NiMoP@Cu3P after electrolysis are consistent with the fresh ones, and there is no shift in binding energy. All of the studies shown above confirm that our NiMoP@Cu3P heterostructure is a highly efficient and stable electrocatalyst toward the HER.
Footnotes |
† Electronic supplementary information (ESI) available. See https://doi.org/10.1039/d2se00942k |
‡ These authors contributed equally to this work and should be considered co-first authors. |
This journal is © The Royal Society of Chemistry 2023 |