Wen-Jing
Li
a,
Xin-Jie
Tian
a,
Hai-Yi
Sun
a,
Xue-Ying
Yang
a,
Denghao
Ouyang
b,
Guodong
Li
b,
Bin
Liu
*a,
Yong-Ming
Chai
a and
Bin
Dong
*a
aState Key Laboratory of Heavy Oil Processing, College of Chemistry and Chemical Engineering, China University of Petroleum (East China), Qingdao 266580, PR China. E-mail: liubin@upc.edu.cn; dongbin@upc.edu.cn; Fax: +86-532-86981156; Tel: +86-532-86981156
bPetroChina Shenzhen New Energy Research Institute Co., Ltd, Shenzhen 518052, PR China
First published on 16th September 2024
A low-cost carrier is developed and the size of Ru nanoparticles is reduced to enhance the performance of Ru catalysts used in the hydrogen evolution reaction (HER). In this study, a nanorod array catalyst was synthesized by combining ruthenium (Ru) with amorphous cobalt–molybdenum–phosphorus (CoMoP/NF). Numerous studies have shown that amorphous structures provide structural flexibility and rich defects for the HER and aid in the uniform loading of Ru nanoparticles. Ru nanoparticles that are uniformly distributed offer a high density of active sites, increasing the electrochemical surface area. Ru-CoMoP/NF exhibited exceptional HER performance in alkaline and alkaline seawater solutions (122 mV @ 100 mA cm−2 and 127 mV @ 100 mA cm−2). Meanwhile, the catalyst can be stably exported for 50 h at 100 mA cm−2. Additionally, the unique structure and isotropic characteristics of the amorphous phase improve the corrosion resistance and make the catalyst maintain good activity in seawater. This study demonstrates the advantages of an amorphous phosphide, enriches the types of supported Ru-based nanomaterials, and provides insights into the development of highly efficient catalysts.
At present, while Pt and Ru are the most efficient electrocatalysts in terms of hydrogen evolution in acid–base solutions, their natural scarcity restricts their widespread use.8,9 To address this issue, scholars have proposed utilizing non-precious metal supports to disperse these noble metals in order to decrease the amount of noble metals used and achieve superior catalytic performance. Transition metal phosphides (TMPs) have been widely developed as advanced HER catalysts by reason of their good electrical conductivity and inherent metallic properties.10 P atoms on the surface of TMPs can both adsorb and trap protons as a “proton trap” in the electrocatalytic HER process, as well as function as a “booster” to increase the reaction's activation energy and enable the desorption of hydrogen molecules from the catalytic interface.11–13 However, compared with platinum group metals, TMPs still have some gaps in electrocatalytic performance. The introduction of an abundant active moiety with increased conductivity and heat diffusion can further improve their electrocatalytic performance.14–17
The least expensive metal in the platinum group is ruthenium (Ru), which has a lot of d-orbital electrons and hydrogen binding strength comparable to that of platinum (Pt), making it an ideal active component to be incorporated into TMPs.18,19 Ru also has excellent adsorption capacity for hydrogen-containing intermediates.20 In other words, as a catalyst component, Ru can significantly enhance HER activity while maintaining cost efficiency.21 In addition, the loading of Ru can aid in improving the electronic structure of active sites in TMPs by inducing charge transfer and adjusting the electron density distribution, thereby enhancing their intrinsic activity.22,23
In addition, compared with crystalline phase materials, amorphous materials with an irregular atomic arrangement do not have a strictly defined crystal structure and exhibit the benefits of a large specific surface area, low hardness, and strong chemical stability. Specifically, the free volume region of atoms containing loose bonds produces abundant “dangle bonds” and defects, which may significantly enhance the number of active sites and optimize reactant adsorption and desorption. Consequently, the application prospects of amorphous materials in the field of electrocatalysis are favorable.24 At the same time, the unique structure and isotropic characteristics of the amorphous phase provide better corrosion resistance.25 Moreover, due to the disordered structure, the amorphous phase can be combined with foreign atoms to synergistically improve the catalytic performance.26 Compared with the surface of a crystalline support, which has a specific crystal structure, the active components may preferentially adsorb on some specific crystal faces, leading to uneven dispersion. The structural disorder of the amorphous support makes the adsorption and dispersion of active components on the surface of the support more random, where it is not easy to form local aggregations, and a more uniform distribution can be achieved.
In light of the aforementioned factors, we designed and constructed a catalyst of Ru-loaded amorphous CoMoP (Ru-CoMoP/NF), which grows uniformly on a nickel foam mesh (NF). The uniform distribution of Ru nanoparticles, amorphous properties, and the nanorod structure provide Ru-CoMoP/NF with plentiful active sites and excellent electrical conductivity. Under both alkaline and seawater conditions, the produced Ru-CoMoP/NF demonstrated outstanding HER electrocatalytic activity and durability, needing overpotentials of just 122 and 127 mV to reach 100 mA cm−2, respectively, and it remained stable for 50 h with only minor decay in performance. As anticipated, Ru-CoMoP/NF has shown outstanding stability when catalyzing hydrogen evolution reactions under alkaline and seawater conditions. Lastly, we integrated solar and wind power generation with the water splitting mechanism. The application potential of this integrated green energy hydrogen production system has been thoroughly shown.27,28
To explore the optimal doping amount of Ru, Rux-CoMoP/NF (x = 30 mg, 40 mg, 50 mg, and 60 mg) samples were prepared using a similar procedure, but RuCl3 was added at 30 mg, 40 mg, 50 mg, and 60 mg, respectively.
XRD was used to examine the structural characteristics of the Ru-CoMoP/NF catalyst. The XRD results depicted in Fig. 2 reveal strong diffraction peaks at 13.386°, 26.811°, 27.811°, 29.303°, and 33.584° for CoMoO4/NF, corresponding to the (001), (002), (−202), (301), and (−222) crystal planes of PDF #21-0868, respectively.29 Additionally, the diffraction peak of CoMoP/NF at 26.472° corresponds to the (002) crystal plane of PDF #21-0868. Comparing the XRD patterns of CoMoO4/NF and CoMoP/NF, a slight low-angle shift can be observed in the diffraction pattern of CoMoP/NF, suggesting that the lattice constant of CoMoP/NF is larger than that of pure CoMoO4/NF, confirming that P is doped into the CoMoO4 lattice and destroys the crystalline phase structure, and CoMoO4 is converted into amorphous CoMoP. Furthermore, the presence of a diffraction peak at 44.287° for Ru-CoMoP/NF aligns with the (101) crystal plane of PDF #01-1256, suggesting successful Ru doping.30
The SEM images clearly show that the Ru-CoMoP/NF nanorod arrays grew uniformly on the NF substrate (Fig. 3a and b). The morphologies of CoMoO4/NF and CoMoP/NF (Fig. S3 and 4†) show that the surface of CoMoO4/NF nanorods is coated with a nanosheet structure after phosphation, but after Ru doping by the solvothermal method, the nanosheet is almost completely dissolved and the nanorod structure recovered, and some Ru nanoparticles are attached (Fig. 3c and d), indicating the successful loading of Ru. According to the XRD results of CoMoP/NF and Ru-CoMoP/NF, P has been successfully doped and has disrupted the precursor phase, and the structure of amorphous CoMoP is still maintained after nanosheet dissolution. The EDS mapping images of SEM (Fig. S5†) show uniform distribution of Co, Mo, P, and Ru elements on the Ru-CoMoP/NF nanorods with atomic percentages of 31.58, 31.49, 31.22, and 5.71, respectively (Fig. S6†), indicating that low-content Ru was successfully loaded into CoMoP/NF. To further explore the nanostructures of Ru-CoMoP/NF, we carried out an extensive investigation using scanning transmission electron microscopy (TEM). Fig. S7 and S8† display TEM, HRTEM, and SAED images of CoMoO4/NF and CoMoP/NF samples, respectively. With a determined lattice spacing of 0.438 nm, the CoMoO4/NF catalyst displays clear lattice stripes on its surface corresponding to the (−201) crystal plane of CoMoO4. The lattice spacing of 0.286 nm in the HRTEM image of CoMoP/NF is consistent with the (−131) plane of CoMoO4. However, the diffraction spot appears very faint in the SAED image of CoMoP/NF, indicating that the crystal phase structure of CoMoO4/NF is disrupted by the addition of P, leading to the formation of amorphous CoMoP/NF. The HRTEM image of Ru-CoMoP/NF in Fig. 3d–f shows a lattice spacing of 0.203 nm, which agrees with the Ru (101) crystal plane that the XRD data indicate. Ru nanoparticles are present on the surface of the nanorods, as shown by analysis of the TEM and related EDS images of Ru-CoMoP/NF. Co, Mo, P, and Ru elements were uniformly dispersed throughout the nanorods (Fig. 3g). Fig. S9† demonstrates the mass fractions of Co, Mo, P and Ru elements, which indicate that low amounts of Ru are loaded onto Ru-CoMoP/NF. The above characterization results indicate that Ru-CoMoP/NF was successfully synthesized.
XPS was used to study the electronic states and surface chemical composition of Ru-CoMoP/NF and the comparison samples CoMoO4/NF and CoMoP/NF. In the Co 2p region of Ru-CoMoP/NF (Fig. 4a), there are two typical peaks at 781.5eV and 797.6 eV, representing Co 2p3/2 and Co 2p1/2, respectively, which indicate the existence of Co2+ in the Ru-CoMoP/NF sample.31 Additionally, there are two oscillatory satellite peaks at 803.1 eV and 786.5 eV, respectively. By comparing the Co 2p spectra of CoMoO4/NF (Fig. S10a†) and CoMoP/NF (Fig. S11a†), it is found that after the incorporation of P, the peak of Co 2p generally shifted to the direction of increasing binding energy, and the Co–P peak was generated, indicating the successful incorporation of P.32 The electronic structure of Co changes after P doping, and the electron density of Co decreases. However, comparing the Co 2p spectra of Ru-CoMoP/NF and CoMoP/NF, it is also found that after Ru doping by the solvothermal method, the peak of Co 2p is shifted in the direction of decreasing binding energy, attributed to the greater electronegativity of Ru than Co, resulting in electrons being transferred from Ru to Co. As demonstrated in Fig. S11b,† the 3d peak of Mo was deconvoluted into four peaks at 235.7/232.5 and 233.4/230.6 eV in CoMoP/NF, respectively, which belonged to Mo6+ and Mo5+.33 Mo5+ can be attributed to the decrease of MoO42− during phosphating. Ru-CoMoP/NF (Fig. 4b) exhibits a positive shift in binding energy to its peak at 233.8 eV when compared to CoMoP/NF, suggesting that Ru dopants influence the electronic structure of molybdenum atoms. The characteristic peaks of the P–O bond, P 2p3/2, and P 2p1/2 are situated at 133.7, 129.3, and 130.2 eV on the P 2p spectra of Ru-CoMoP/NF (Fig. 4c).33 The strong P–O bond can be attributed to inevitable surface oxidation. In contrast to CoMoP/NF, the P 2p spectra of Ru-CoMoP/NF shifted in the direction of decreasing the binding energy, resulting from the addition of Ru elements, which can enhance the electron transfer between CoMoP and Ru nanoparticles. The Ru 3p spectrum of Ru-CoMoP/NF (Fig. 4d) shows two peaks at 461.5 and 484.1 eV, corresponding to Ru 3p3/2 and 3p1/2 of zero-valent Ru,34 and the oxidation state peak was not found. In conclusion, the introduction of Ru nanoparticles can alter the electron transfer of the HER process and thus increase the conductivity and intrinsic activity of the catalyst, enhancing the hydrogen evolution performance of Ru-CoMoP/NF.
Furthermore, the durability of Ru-CoMoP/NF in terms of the HER is also a crucial factor. As depicted in Fig. 5f, the shift of the LSV curve of Ru-CoMoP/NF before and after 2000 cycles is almost negligible. In addition, the timing electrochemical test results (Fig. 5f) demonstrated that the catalyst was able to sustain stable catalysis for a period of 50 h, confirming the excellent long-term stability of Ru-CoMoP/NF.
Since seawater makes up the majority of water on Earth and is a plentiful resource for people, there has been an increase in interest in the use of seawater electrolysis to produce hydrogen. Inspired by the excellent performance of Ru-CoMoP/NF in producing hydrogen in alkaline settings, we evaluated the electrocatalytic efficacy of these materials in alkaline seawater electrolytes from China's Yellow Sea.38 As demonstrated in Fig. 6a, the Ru-CoMoP/NF electrode still showed the best HER performance. In particular, Ru-CoMoP/NF only required overpotentials of 20 and 127 mV to achieve 10 and 100 mA cm−2 in 1 M KOH seawater. Comparatively, the overpotential required for Ru-CoMoO4/NF, CoMoP/NF, CoMoO4/NF, and NF to drive a current density of 100 mA cm−2 is large, being 173, 230, 349, and 369 mV, respectively (see Fig. 6a for details), indicating that Ru doping with amorphous CoMoP can significantly improve HER activity. Importantly, the Tafel slope of the prepared Ru-CoMoP/NF in 1 M KOH seawater (53.6 mV dec−1) was close to the slope in 1 M KOH, and is much lower than those of other samples, which indicates that the HER has rapid reaction kinetics in alkaline seawater (Fig. 6b). However, the presence of high concentrations of chloride ions, insoluble precipitates, bacteria, and microorganisms in seawater can block or inactivate the active sites, resulting in relatively poor catalytic performance in alkaline seawater compared to alkaline freshwater media.34 Similarly, the stability of Ru-CoMoP/NF in seawater electrolytes was investigated. As shown in Fig. 6c, the shift of the LSV curve of Ru-CoMoP/NF before and after 2000 cycles is almost negligible. Ru-CoMoP/NF remained in good condition for 50 hours during the stability test (Fig. 6d). Further comparison of overpotential and Tafel slopes showed that Ru-CoMoP/NF exhibited impressive excellent electrocatalytic properties compared to those of the recently reported HER electrocatalysts in 1 M KOH seawater (Table S3†). Excellent structural stability is exhibited by Ru-CoMoP/NF, as demonstrated by Fig. S14,† whereby the nanorod structure in 1 M KOH and 1 M KOH saltwater before and after stability shows no appreciable alterations.
In order to further explore how Ru and CoMoP influence the activity of the alkaline HER, we analyzed the H* intermediate adsorption/combination energy in the alkaline HER with the help of previous studies. Water absorption, the creation of H* intermediates (Volmer steps), and the ensuing H2 gas generation (Heyrovsky steps or Tafel steps) are the three fundamental processes in the water decomposition process in alkaline solution.39 At the same time, a pure ruthenium catalyst has poor hydrogen evolution performance in an alkaline environment due to its poor water decomposition ability.40 However, transition metal compounds can effectively break the HO–H bond in water. In addition, the recognized Sabatier principle states that the ideal hydrogen adsorption on the surface of the catalyst should be neither too strong nor too weak. Strong adsorption affects the overall hydrogen precipitation efficiency, while weak adsorption affects the formation of hydrogen adsorption in Volmer steps. Consequently, the hydrogen adsorption energy (ΔGH) of a perfect HER catalyst ought to be zero.41 Noble metals have the best hydrogen binding energy. The introduction of Ru can adjust the H* intermediate adsorption/combination energy on the catalyst surface such that ΔGH approaches 0. In summary, the excellent HER activity is due to the synergistic effect between Ru and CoMoP, in which water dissociates on the CoMoP and then mass transfers to the Ru sites to quickly complete the adsorption and desorption processes of H* intermediates.
Therefore, the exceptional electrocatalytic performance of Ru-CoMoP/NF can be ascribed to multiple factors: first, the incorporation of Ru increases the active site density and intrinsic activity, as well as promotes faster charge transfer of Ru-CoMoP/NF. Moreover, the distinct structure and isotropic characteristics of the amorphous phase improve corrosion resistance. Second, the distinctive structure deposited onto nickel foam exhibits strong electrical conductivity, facilitating proton and electron transfer during the electrolysis of water. Ultimately, the introduction of Ru and P can effectively improve HER dynamics, which leads to excellent HER performance.
Renewable energy has been widely used in human daily life and production, but its intermittency and instability have caused hundreds of millions of electrical energy losses every year.42 One of the best ways to solve this problem is to convert it in situ into small molecules and store it as chemical energy. Combined with this feature, the combination of renewable energy and hydrogen production from electrolyzed water can make use of renewable energy, further realizing the notion of “green hydrogen energy”.43 To achieve this, a system that combines a water electrolysis hydrogen production device and a photovoltaic power generation device is developed. Solar energy is then transformed into electricity by solar panels, which power the electrolytic water hydrogen production (Fig. 7b). The successful production of hydrogen and oxygen is demonstrated by observing bubbles on both sides of the cell, as showcased in Fig. 7b. Furthermore, an integrated system connecting a wind power generation device with an electrolytic water hydrogen evolution device is designed. By manually simulating the natural wind, the wind turbine is driven to generate electricity, and electrocatalytic hydrogen production is realized (Fig. 7c). Experiments have shown that a large amount of gas can be produced on both sides of the electrolyzer, revealing the high efficiency for hydrogen production and potential for wide applications, as depicted in Fig. 7d. In conclusion, it is possible to solve the problem of electric energy limitation by using renewable wind and solar energy to produce hydrogen and this shows great promise for the practical large-scale implementation of Ru-CoMoP/NF.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4qi01851f |
This journal is © the Partner Organisations 2024 |