MOF derived RuO2/V2O5 nanoneedles for robust and stable water oxidation in acid

Qing Li Yumei Feng Yuhua Xie Qi Xu Yifei Li Yingjie Yu Fang Luo Zehui Yang

Citation:  Qing Li, Yumei Feng, Yuhua Xie, Qi Xu, Yifei Li, Yingjie Yu, Fang Luo, Zehui Yang. MOF derived RuO2/V2O5 nanoneedles for robust and stable water oxidation in acid[J]. Chinese Chemical Letters, 2025, 36(7): 111074. doi: 10.1016/j.cclet.2025.111074 shu

MOF derived RuO2/V2O5 nanoneedles for robust and stable water oxidation in acid

English

  • The exploration of renewable energy sources is vital to relieve the environmental pollution and the increasing energy demands. Hydrogen gains much attention due to its sustainability, which can be electrochemically generated from splitting of water actuated by inconsecutive powers, wind and solar energies [1,2]. Proton exchange membrane water splitting (PEMWS) has been considered as the most encouraging water splitting technology for hydrogen generation attributing to its commercially available components [3,4]. As hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) are the uphill reactions, requiring efficient electrocatalyst to drive, especially OER catalysis, rate limiting step for overall water splitting [5,6]. Besides, the stability also should be significantly weighed because of the harsh working condition, strong acid, high temperature and voltage, leading to a formidable challenge [7]. Ruthenium oxides (RuO2) and iridium oxides (IrO2) are the commercial OER electrocatalysts [8]; however, the stability as well as catalytic performance should be substantially enhanced to forward the industrialization of PEMWS. Additionally, more attention should be paid to RuO2 because of its lower cost [911].

    The formation of *OOH specie is deemed as the rate-limiting step for acidic OER process, which forms via the nucleophilic attack by H2O; thereby, the electrophilic property is highly recommended for the electrochemical absorbed *O species on active sites. Thus, the oxidation state of Ru should be lower than +4, favorable for the conversion of *O to *OOH leading to a boosted acidic OER performance [1214]. Apart from catalytic activity, the low stability of RuO2 originates from the low oxidation potential for forming high valance state of Ru, dissolvable in electrolyte. In this regard, the simultaneous promotion in catalytic activity and stability normally achieved by formation heterostructure [15,16], creating new material phase [17,18] and decoration with heteroatoms [19] since the electronic configuration of active sites, regulation in electrochemical adsorption-desorption of reaction intermediates, are effectively modulated and the forming energy associated with cohesive energy, fingerprint of stability, are reasonably adjusted [20]. Formation of heterostructure has the merits of boosting the OER performance, stability and reducing the dosage of noble metals, making this methodology a hopeful strategy for RuO2.

    In this work, we have initially synthesized MIL-88B as template for the construction of RuO2/V2O5 heterostructure with nanoneedle structure. More active sites were exposed in the nanoneedle structure and diffusion resistance was reduced; moreover, tip effect also contributed to a boosted catalytic activity. The overpotential at 10 mA/cm2 was 216 mV for RuO2/V2O5, which was decreased by 27 mV with relative to commercial RuO2 due to more oxygen vacancies in RuO2/V2O5 triggering a lower valance state of Ru, beneficial for the nucleophilic attack by H2O, confirmed by the methanol. Besides, the OER performance tested under various pH indicated that the OER activity of RuO2/V2O5 was more sensitive to hydrogen concentration indicating a better deprotonation capability. Moreover, the electronic interaction between RuO2 and V2O5 efficiently increased the oxidation potential for Ru4+ resulting in a better stability, proved by the cyclic voltammetry test. The theoretical calculation revealed that the demanded energy for the formation of *OOH was decreased from 2.43 eV to 2.06 eV with incorporation of V2O5.

    As shown in Fig. S1 (Supporting information), the XRD pattern of the prepared MIL-88B showed intensive diffraction peaks at 17.5°, 25.2° and 28.0°, similar to the previous reports indicating the successful construction of MIL-88B. With the calcination temperature increased to 450 ℃, the V2O5 phase was obtained confirmed by its XRD pattern, in which diffraction peaks at 15.4°, 20.3°, 26.2°, 31.1° and 34.3° assigned to the (200), (001), (110), (400) and (310) crystal planes of V2O5 according to JCPDS No. 09–0387. As shown in Fig. 1a, the utilization of RuCl3 as precursor, RuO2–450 was formed verified by the diffraction peaks at 27.5°, 35.0° and 54.3° stemming from the (110), (101) and (211) facets of RuO2 based on JCPDS No. 43–1027, similar to commercial RuO2 [21]. The XRD pattern of RuO2/V2O5 was resemble to that of V2O5, except the diffraction peak at 27.8° coming from the dominant (110) crystal plane of RuO2; therefore, RuO2 and V2O5 were both existed. To confirm the formation of heterostructure, XPS test has been carried out (Fig. S2 in Supporting information). The core-level Ru 3p peaks composed of Ru 3p1/2 at 462.53 eV and Ru 3p3/2 at 465.69 eV fitted to Ru3+ and Ru4+ species (Fig. 1b); interestingly, more Ru4+ species were noticed for RuO2/V2O5 compared to that of RuO2–450 demonstrating the electronic interaction between RuO2 and V2O5. Besides, the high valance state was also underlined by the positive shift in binding energy. As shown in Fig. 1c, the V 2p3/2 XPS peak was deconvoluted into two peaks centered at 517.11 and 516.60 eV, corresponding to V5+ and V4+ species, respectively [22]. A lower valance state of V atom was noted in RuO2/V2O5. Thus, a heterostructure was found for RuO2/V2O5 and Ru donated its electrons to oxygen atom from V2O5 resulting in a higher oxidation state compared to RuO2–450 [23]. The higher valance state of Ru atom would be favorable for promoting the stability since RuO2 was easily oxidized to Ru6+ species dissolved into electrolyte [24]. The presence of Ru valance state in RuO2 would show a high resistance towards electrochemical oxidation under OER condition. As shown in Fig. 1d, lattice oxygen, oxygen defects and adsorbed H2O composed of O 1s XPS spectrum [25,26]. It was found that oxygen defects in RuO2–450 was higher than V2O5 and RuO2, which would be harmful for the structural stability since the electrochemical oxidation occurred at defective sites [27]. Due to high percentage of oxygen vacancy, the O 1s peak of RuO2–450 was negatively shifted with relative to V2O5 and RuO2/V2O5.

    Figure 1

    Figure 1.  (a) XRD patterns of RuO2/V2O5, RuO2–450 and V2O5. (b) Ru 3p XPS spectra of RuO2/V2O5 and RuO2–450. (c) V 2p3/2 XPS spectra of V2O5 and RuO2/V2O5. (d) O 1s XPS spectra of RuO2–450, RuO2/V2O5 and V2O5.

    As shown in Fig. S3 (Supporting information), MIL-88B exhibited a pine needle structure. However, the RuO2–450 exhibited a microsphere structure. Due to the template of MIL-88B, nanoneedle structure was detected for RuO2/V2O5 shown in Fig. 2a. The formed nanoneedle structure originated from the MIL-88B, serving as template. Differently, as shown in Fig. S4 (Supporting information), nanoparticle structure was recorded for commercial RuO2. The TEM image shown in Fig. 2b also highlighted the nanoneedle structure of the formed RuO2/V2O5. From the HR-TEM image shown in Fig. 2c, the lattice spacing of 0.33 nm and 0.44 nm originated from the (110) and (001) facets of RuO2 and V2O5, respectively. Also, the relative Fast Fournier transfer (FFT) testified the presence of V2O5 (001) and RuO2 (110). The HR-TEM image clearly showed the heterointerface indicating the formation of heterostructure. TEM test also underscored the formation of heterostructure in RuO2/V2O5. The HAADF-STEM image shown in Fig. 2d implied the well-distributed Ru, O and V elements over the nanoneedle.

    Figure 2

    Figure 2.  SEM (a), TEM (b), HR-TEM (c), HAADF-STEM (d) and relative EDS mapping of RuO2/V2O5.

    The oxygen evolution reaction (OER) performance was estimated in 0.5 mol/L H2SO4. As shown in Fig. 3a, the synthesized RuO2–450 exhibited an inferior OER performance with overpotential of 272 mV to reach 10 mA/cm2, which was higher than the commercial RuO2 (243 mV overpotential for 10 mA/cm2). RuO2–450 showed an inferior OER activity to commercial RuO2 due to its large particle size with less exposed active sites. However, with the introduction of V2O5, the overpotential at 10 mA/cm2 was decreased to 216 mV for RuO2/V2O5. Moreover, OER activity of RuO2/V2O5 outperformed the benchmarked RuO2. V2O5 performed an ultralow acidic OER performance with overpotential of 365 mV for 10 mA/cm2, which was strong evidence for the boosted OER performance of RuO2/V2O5 attributed to the reasonable modulation of electronic configuration of RuO2. The OER performance of RuO2+V2O5 was inferior to the counterpart of RuO2/V2O5 emphasizing the heterostructure contributing to the boosted OER activity. The mass activity reached 11.6 A/mgRu for RuO2/V2O5 at 1.53 V vs. RHE, boosted by 12- and 6-fold than RuO2–450 and benchmark RuO2 (Fig. S5 in Supporting information). As shown in Fig. 3b, Tafel slope was 45.87 mV/dec for RuO2/V2O5, lower than RuO2–450 (114.86 mV/dec) and RuO2 (118.17 mV/dec). The lower Tafel slope indicates a faster acidic OER process [28]. Based on the Tafel slope, Eley-Rideal-like (ER-like) OER mechanism was observed for RuO2/V2O5; while, the OER catalysis on commercial RuO2 and RuO2–450 followed Langmuir−Hinshelwood mechanism. As shown in Fig. S6 (Supporting information), charge transfer resistance (Rct) was 102.6, 77.0 and 65.6 Ω for RuO2–450, RuO2 and RuO2/V2O5, respectively. The lower Rct was due to the formation of heterostructure with different work function for RuO2 and V2O5 [29]; therefore, a built-in electric field was formed and electronic conductivity was promoted. The temperature dependent OER performance was tested (Fig. S7 in Supporting information) to calculate activation energy. As shown in Fig. 3c, the activation energy for RuO2/V2O5 was dramatically decreased to 13.82 kJ/mol; while, this value was 19.53 kJ/mol for commercial RuO2; thus, a better OER activity was achieved for RuO2/V2O5. To quantitatively analyze the active center, double layer capacitance (Cdl) was estimated from cyclic voltammetry curves recorded under various scan rate (Fig. S8 in Supporting information). The Cdl values were 22.43, 7.46 and 49.07 mF/cm2 for RuO2, RuO2–450 and RuO2/V2O5 (Fig. 3d), respectively. Compared to RuO2–450, Cdl of RuO2/V2O5 was boosted by 6.6-fold by comparison with RuO2–450. It is generally accepted that the adsorption of *OH species generated from water dissociation is important for the acidic OER, which can be detected by the nucleophilic methanol since *OH is considered as electrophilic reactant [30]. As shown in Fig. 3e, it was found that the current density was sharply increased at initial stage (below 5 mA/cm2) indicating the superb water dissociation capability of RuO2/V2O5 to generate *OH at active center favorable for acidic OER catalysis; while, with the potential increasing, the net current density was lower than the counterpart of RuO2 (Fig. S9 in Supporting information) due to the weakened *OH binding strength facilitating its conversion. Besides, the nucleophilic attacked by H2O is involved in acidic OER catalysis; therefore, methanol oxidation reaction (MOR) is a competitive reaction. With the increment in potential, the current density gap was narrowed for RuO2/V2O5 illustrated the active sites were inconspicuously affected by methanol; while, the RuO2 was significantly affected by methanol demonstrating H2O was more susceptible to electrochemically adsorb on active center of RuO2/V2O5. As well known that the acidic OER is a dehydrogenation process; thus, the OER performance was tested under various pH (Fig. S10 in Supporting information). As shown in Fig. 3f, the plot of the current density vs. pH indicated that the reaction order was 8.27 for RuO2/V2O5, which was higher than commercial RuO2 (5.02) demonstrating the OER performance of RuO2/V2O5 relied on the hydrogen concentration [31]. It was known that the deprotonation is customarily depended on the adjacent atoms. Due to the formation of heterostructure, Ru-Obri-V structure was detected in RuO2/V2O5 and the unbalanced electron density caused by different electronegativity of Ru (2.20) and V (1.63) led to high hydrogen binding capability for Obri site. A lower deprotonation capability was observed for commercial RuO2 due to the electronic symmetry of Ru-Obri-Ru structure. Thereby, a superior acidic OER performance was observed for RuO2/V2O5 [32].

    Figure 3

    Figure 3.  OER performance (a), Tafel slope (b), activation energy (c) and double layer capacitance (d) for RuO2/V2O5, RuO2–450 and commercial RuO2. (e) OER activity of RuO2/V2O5 with and without methanol. (f) Plot of pH vs. log∣j (mA/cm2)∣ of RuO2/V2O5, RuO2–450 and commercial RuO2.

    As shown in Fig. 4a, the overpotentials at 10 mA/cm2 and 50 mA/cm2 were 221 mV and 307 mV for RuO2/V2O5 after 1000 cycles, which was similar to its initial performance implying the superb stability. To deeply understand the stability, Cdl was calculated to 33.23 mF/cm2 after 1000 cycles (Fig. S11 in Supporting information); moreover, Rct was 68.1 Ω (Fig. S12 in Supporting information), which was only increased by 1.8%. Thus, it was concluded that the superior stability of RuO2/V2O5 originated from the high structural stability. The HR-TEM image shown in Fig. S13 (Supporting information) suggested the superior structural stability of RuO2/V2O5 stemming from the strong electronic interaction between RuO2 and V2O5. However, as shown in Fig. S14 (Supporting information), commercial RuO2 showed an inferior stability during 1000 cycles ascribed to the electrochemical decomposition reflected by decrement in Cdl (Fig. S15 in Supporting information). Also, as shown in Fig. 4b, RuO2/V2O5 showed a stable current density within 100 h; in contrast, the OER activity of the commercial RuO2 was decayed by 74% after merely 2 h. The voltage was also seriously increased for RuO2–450, indicating the importance of heterostructure for enhancing stability. The OER performance RuO2/V2O5 was well maintained after stability test underscoring the superior stability (Fig. S16 in Supporting information). To certificate the high compositional stability, the cyclic voltammetry curve was recorded. As shown in Fig. 4c, the obvious redox peak was observed for RuO2/V2O5, signal of Ru3+/Ru4+ conversion [33]. It was found that the reversibility was better than commercial RuO2 because the potential was 0.08 V for RuO2/V2O5, lower than RuO2 (0.12 V). Thus, the structural stability is higher for RuO2/V2O5. Also, the in-situ electrochemical impedance spectroscopy was carried out. As shown in Fig. 4d, with the same change in applied potential, a larger deviation in phase angle was found for RuO2/V2O5 below 100 Hz than commercial RuO2 (Fig. 4e) and RuO2–450 (Fig. S17 in Supporting information) suggesting a better OER performance. Moreover, the Rct of RuO2/V2O5 was also lower than the counterparts of RuO2–450 and commercial RuO2 (Fig. S18 in Supporting information) [34]. It was important to identify the OER mechanism since lattice oxygen mechanism (LOM) and adsorption evolution mechanism (AEM) occur during OER process [35,36]. O2 and O22− are the important reaction intermediates for AEM and LOM; moreover, tetramethylammonium cations are easily to coordinate with O22− triggering a serious deterioration in OER performance [37]. As shown in Fig. 4f, the OER catalytic activity was almost stable with TMA+ implying RuO2/V2O5 followed AEM pathway. The OER performance was inconspicuously decayed with additional TMA+ suggesting the AEM occurred on commercial RuO2 (Fig. S19 in Supporting information).

    Figure 4

    Figure 4.  (a) Cyclic stability of RuO2/V2O5. Stability test (b), cyclic voltammetry curves (c) and bode plots (d, e) of commercial RuO2 and RuO2/V2O5. (f) OER performance of RuO2/V2O5 with and without TMA+.

    As shown in Fig. 5a, the charge density of RuO2/V2O5 suggested the electronic interplay between RuO2 and V2O5; moreover, it was noticed that electric density of Ru atom was decreased; in contrast, an increment in electronic density was detected for V atom at heterointerface, in accordance to XPS analysis. Consequently, the density of state (DOS) of RuO2/V2O5 and RuO2 was analyzed (Fig. 5b). The d band center of Ru was −1.19 eV and −1.91 eV for RuO2 and RuO2/V2O5, respectively. The downshifted d band center suggested a lower binding strength with reaction intermediates, which was also reflected by the increment in eg filling of RuO2/V2O5 shown in Fig. 5c, well matched with methanol detection [38,39]. The d band composes of eg and t2g orbitals and eg occupies the high energy state, which hybridizes with the reaction intermediates; therefore, the eg filling is vital for predicting the binding strength between active site and reaction coordinates. As shown in Fig. S12 (Supporting information), the free energy for various reaction intermediates was calculated based on AEM mechanism. As listed in Fig. 5d, the free energy for rate limiting step, formation of *OOH moiety, was 2.06 eV and 2.43 eV for RuO2/V2O5 and RuO2, respectively. Theoretical calculation indicated that RuO2/V2O5 performed a better OER performance with relative to RuO2.

    Figure 5

    Figure 5.  (a) Charge density of RuO2/V2O5. Density of state (b), eg filling (c) and OER energy barrier (d) of RuO2 and RuO2/V2O5.

    In this work, we have synthesized RuO2/V2O5 heterostructured nanoneedle using MIL-88B as template. The formed RuO2/V2O5 performed a better acidic OER activity than commercial IrO2, demanding 216 mV and 243 mV overpotential to reach 10 mA/cm2, corresponding to a 6-time higher mass activity. From the pH-independent test a higher deprotonation capability was recorded for RuO2/V2O5 contributing to a boosted OER performance. Also, the cyclic voltammetry curve illustrated that the oxidation potential for the formation of Ru4+ was increased for RuO2/V2O5 leading to a superior resistance towards electrochemical oxidation; as a result, a robust stability was achieved for RuO2/V2O5. The DFT calculation indicated the introduction of V2O5 to RuO2 downshifted d band center and increased the eg filling of Ru resulting in a lower energy barrier for the formation of *OOH species.

    The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

    Qing Li: Data curation. Yumei Feng: Data curation. Yuhua Xie: Data curation. Qi Xu: Data curation. Yifei Li: Data curation. Yingjie Yu: Data curation. Fang Luo: Supervision, Resources, Funding acquisition. Zehui Yang: Writing – review & editing, Writing – original draft, Supervision.

    This work is supported by the National Natural Science Foundation of China (No. 22209126) and the Opening Project of Key Laboratory of Optoelectronic Chemical Materials and Devices, Ministry of Education, Jianghan University (No. JDGD-202314). The authors also thank Dr. Liu from the Analytical and Testing Center of Wuhan Textile University for her assistance with XPS test and helpful analysis.

    Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111074.


    1. [1]

      X. Li, W. Xu, Y. Fang, et al., SusMat 3 (2023) 160–179. doi: 10.1002/sus2.114

    2. [2]

      Y. Feng, Y. Xie, Y. Yu, et al., Angew. Chem. Int. Ed. 64 (2025) e202413417.

    3. [3]

      Y. Hao, S.F. Hung, L. Wang, et al., Nat. Commun. 15 (2024) 8015.

    4. [4]

      Y. Hao, S.F. Hung, W.J. Zeng, et al., J. Am. Chem. Soc. 145 (2023) 23659–23669. doi: 10.1021/jacs.3c07777

    5. [5]

      L. Deng, S.F. Hung, S. Liu, et al., J. Am. Chem. Soc. 146 (2024) 23146–23157. doi: 10.1021/jacs.4c05070

    6. [6]

      Q. Li, Y. Feng, Y. Yu, et al., Chin. Chem. Lett. 36 (2025) 110612.

    7. [7]

      J. Ying, J.B. Chen, Y.X. Xiao, et al., J. Mater. Chem. A 11 (2023) 1634–1650. doi: 10.1039/d2ta07196g

    8. [8]

      M. Kim, J. Park, M. Wang, et al., Appl. Catal. B 302 (2022) 120834.

    9. [9]

      Y. Xie, Y. Feng, S. Pan, et al., Adv. Funct. Mater. 34 (2024) 2406351.

    10. [10]

      Z. Chen, N. Han, W. Wei, et al., EcoEnergy 2 (2024) 114–140.

    11. [11]

      J. Ge, X. Wang, H. Tian, et al., Chin. Chem. Lett. 35 (2025) 109906.

    12. [12]

      N. Yao, H. Jia, J. Zhu, et al., Chem 9 (2023) 1882–1896.

    13. [13]

      W. Zhu, K. Yin, M. Ma, et al., Chin. Chem. Lett. (2024), doi: 10.1016/j.cclet.2024.110749.

    14. [14]

      Q. Wu, R. Zhou, Z. Yao, et al., Chin. Chem. Lett. 35 (2024) 109416.

    15. [15]

      H. Song, X. Yong, G.I.N. Waterhouse, et al., ACS Catal. 14 (2024) 3298–3307. doi: 10.1021/acscatal.3c06182

    16. [16]

      Y. Wu, R. Yao, K. Zhang, et al., Chem. Eng. J. 479 (2024) 147939.

    17. [17]

      W. Zhu, X. Song, F. Liao, et al., Nat. Commun. 14 (2023) 5365.

    18. [18]

      J. Zhu, Y. Guo, F. Liu, et al., Angew. Chem. Int. Ed. 60 (2021) 12328–12334. doi: 10.1002/anie.202101539

    19. [19]

      D. Zhang, M. Li, X. Yong, et al., Nat. Commun. 14 (2023) 2517.

    20. [20]

      Y. Wang, K. Wei, Y. Song, et al., Chem. Eng. J. 497 (2024) 154724.

    21. [21]

      J. Wang, C. Cheng, Q. Yuan, et al., Chem 8 (2022) 1673–1687.

    22. [22]

      C. Hao, H. Lv, Q. Zhao, et al., Int. J. Hydrogen Energy 42 (2017) 9384–9395.

    23. [23]

      Y. Cheng, Y. Wang, Z. Shi, et al., EcoEnergy 3 (2025) 131–155. doi: 10.1002/ece2.79

    24. [24]

      L. Liu, Y. Ji, W. You, et al., Small 19 (2023) 2208202.

    25. [25]

      S. Li, G. Xing, S. Zhao, et al., Natl. Sci. Rev. 11 (2024) nwae193.

    26. [26]

      S. Zhao, S.F. Hung, L. Deng, et al., Nat. Commun. 15 (2024) 2728.

    27. [27]

      Y. Wang, R. Yang, Y. Ding, et al., Nat. Commun. 14 (2023) 1412. doi: 10.3390/cryst13101412

    28. [28]

      C. Liu, B. Sheng, Q. Zhou, et al., Nano Res. 15 (2022) 7008–7015. doi: 10.1007/s12274-022-4337-z

    29. [29]

      T. Xiong, X. Li, Z. Ma, et al., J. Colloid Interface Sci. 672 (2024) 170–178.

    30. [30]

      S. Xin, Y. Tang, B. Jia, et al., Adv. Funct. Mater. 33 (2023) 2305243.

    31. [31]

      H. Jia, N. Yao, Z. Liao, et al., Angew. Chem. Int. Ed. 63 (2024) e202408005.

    32. [32]

      S. Chen, S. Zhang, L. Guo, et al., Nat. Commun. 14 (2023) 4127.

    33. [33]

      Z. Shi, J. Li, Y. Wang, et al., Nat. Commun. 14 (2023) 843.

    34. [34]

      H. Su, W. Zhou, W. Zhou, et al., Nat. Commun. 12 (2021) 6118.

    35. [35]

      X. Wang, S. Xi, P. Huang, et al., Nature 611 (2022) 702–708. doi: 10.1038/s41586-022-05296-7

    36. [36]

      Z. Wu, Y. Wang, D. Liu, et al., Adv. Funct. Mater. 33 (2023) 2307010.

    37. [37]

      Z. Feng, C. Dai, P. Shi, et al., Chem. Eng. J. 485 (2024) 149992.

    38. [38]

      H. Jia, N. Yao, Y. Jin, et al., Nat. Commun. 15 (2024) 5419.

    39. [39]

      Y. Xie, Y. Feng, S. Zhu, et al., Adv. Mater. 37 (2025) 2414801.

  • Figure 1  (a) XRD patterns of RuO2/V2O5, RuO2–450 and V2O5. (b) Ru 3p XPS spectra of RuO2/V2O5 and RuO2–450. (c) V 2p3/2 XPS spectra of V2O5 and RuO2/V2O5. (d) O 1s XPS spectra of RuO2–450, RuO2/V2O5 and V2O5.

    Figure 2  SEM (a), TEM (b), HR-TEM (c), HAADF-STEM (d) and relative EDS mapping of RuO2/V2O5.

    Figure 3  OER performance (a), Tafel slope (b), activation energy (c) and double layer capacitance (d) for RuO2/V2O5, RuO2–450 and commercial RuO2. (e) OER activity of RuO2/V2O5 with and without methanol. (f) Plot of pH vs. log∣j (mA/cm2)∣ of RuO2/V2O5, RuO2–450 and commercial RuO2.

    Figure 4  (a) Cyclic stability of RuO2/V2O5. Stability test (b), cyclic voltammetry curves (c) and bode plots (d, e) of commercial RuO2 and RuO2/V2O5. (f) OER performance of RuO2/V2O5 with and without TMA+.

    Figure 5  (a) Charge density of RuO2/V2O5. Density of state (b), eg filling (c) and OER energy barrier (d) of RuO2 and RuO2/V2O5.

  • 加载中
计量
  • PDF下载量:  0
  • 文章访问数:  29
  • HTML全文浏览量:  3
文章相关
  • 发布日期:  2025-07-15
  • 收稿日期:  2024-12-31
  • 接受日期:  2025-03-12
  • 修回日期:  2025-03-10
  • 网络出版日期:  2025-03-13
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

/

返回文章