Graphene controlled solid-state growth of oxygen vacancies riched V2O5 catalyst to highly activate Fenton-like reaction

Mengxiang Zhu Tao Ding Yunzhang Li Yuanjie Peng Ruiping Liu Quan Zou Leilei Yang Shenglei Sun Pin Zhou Guosheng Shi Dongting Yue

Citation:  Mengxiang Zhu, Tao Ding, Yunzhang Li, Yuanjie Peng, Ruiping Liu, Quan Zou, Leilei Yang, Shenglei Sun, Pin Zhou, Guosheng Shi, Dongting Yue. Graphene controlled solid-state growth of oxygen vacancies riched V2O5 catalyst to highly activate Fenton-like reaction[J]. Chinese Chemical Letters, 2024, 35(12): 109833. doi: 10.1016/j.cclet.2024.109833 shu

Graphene controlled solid-state growth of oxygen vacancies riched V2O5 catalyst to highly activate Fenton-like reaction

English

  • Advanced oxidation processes (AOPs), with advantages of high efficiency, universality and thoroughness oxidation on organic pollutants, have become one of the promising approaches for refractory organic wastewater treatment [1-6]. As one of the most mature and diverse strategies, transition metal-based Fenton-like catalysis have been studied extensively, such reactions require no additional energy and overcome the problems of homogeneous reactions [7-11]. To achieve the highly efficient generation of reactive oxygen species, many strategies are mainly focused on tailoring transition metal-based catalysts with desired active sites generation and exposure by controlling crystal facet growth and engineering geometry structural [12-17]. These traditional strategies are always conducted in the wet conduction, which needs the solvents such as water, ethanol or ethylene glycol [18,19]. Tedious multistep processes and harsh reaction conditions blocked the practical applications of these technologies [20-24]. Therefore, developing a facile scale-up method to create more active sites to boost transition metal-based catalysts performance still remains a major challenge.

    Notably, inducing chemical defects have emerged as a promising approach for improving catalytic performance by modulating the electronic structure of transition metal-based catalysts [25-28] and solid-state synthesis method with effectively introduce defects make it easy to scale-up production [29-32]. Hence, we develop a one-step all solid-state growth strategy to precisely engineer oxygen defects in V2O5 that are self-assembled during the process of selective crystallization, thereby boosting the Fenton-like performance in defective V2O5@graphene/H2O2 system (0.2012 min−1), which was 14.3, 28.2, and 17.3 times higher than that of graphene/H2O2 (0.0085 min−1), pure V2O5/H2O2 (0.0043 min−1) and mechanical mixed system (graphene+V2O5/H2O2 0.0070 min−1), respectively. Additionally, OVs-riched V2O5@graphene exhibits an outstanding catalytic stability and degradation ability for a variety of refractory organic pollutants. Our work provides a promising synthetic strategy for transition metals in Fenton-like catalytic degradation process to produce reactive oxygen species, paving the way for sustainable water purification and material design optimization.

    The graphene-controlled thermal decomposition reaction of VCl3 to defective V2O5@graphene by an in situ one-step, all-solid-state synthesis was showed in Fig. 1a. XRD patterns indicated that as the graphene addition was below 50%, VCl3 was mainly transformed to V2O5 at 400 ℃ for 120 min and increasing graphene can induce the further decomposition of V2O5 to V2O4 (Fig. 1b and Fig. S1 in Supporting information) [33,34]. The surface morphologies of defective V2O5@graphene composites were obtained by field-emission scanning electron microscopy (FESEM), which showed that V2O5 crystals were clearly observed on the graphene sheets (Fig. 1c). Transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) associated with the fast Fourier transform (FFT) showed a series clear lattice fringes of 2.501 Å, 2.143 Å, 2.250 Å and 2.076 Å corresponding to the (121), (012), (131) and (022) planes of V2O5, thereby confirming the formation of V2O5 crystals on graphene surface (Figs. 1dg, Figs. S2 and S3 in Supporting information). What is more, the magnified image in Fig. 1e clearly displayed that many dislocations and distortions (red circle) were distributed around the lattice fringes, which suggested that defective V2O5@graphene has a typical defect-rich structure. The X-ray photoelectron spectroscopy (XPS) analysis of V2O5 and V2O5@graphene was investigated. With the graphene regulation, the new peaks of 531.0 eV and 532.8 eV corresponding to OVs and -OH bonds were obtained in defective V2O5@graphene (Fig. S4 in Supporting information), which was absent in pure V2O5, thereby confirming the formation of OVs in V2O5@graphene.

    Figure 1

    Figure 1.  (a) Schematic illustration of the sample synthesis process. (b) XRD pattern of annealed VCl3-graphene at different mass proportions. (c) SEM and magnified SEM image of defective V2O5@graphene. (d, e) TEM and HRTEM image of defective V2O5@graphene. (f) The associated FFT analysis taken from the regions outlined in (e). (g) Elemental mapping images of defective V2O5@graphene.

    Energy dispersive spectroscopy (EDS) analysis revealed that the molar ratio of V/O in defective V2O5@graphene, V2O4@graphene and pure V2O5 was 1:0.956, 1:1.487 and 2.128 (Table S1 in Supporting information), respectively, implying that the V atom content of defective V2O5@graphene was higher than that of pure V2O5, thereby confirming the OVs introduction in defective V2O5@graphene [35].

    To better reveal the positive effect of OVs on Fenton-like performance, degradation experiments with Rhodamine B (RhB, 10 mg/L) as a model dye were conducted. Fig. 2a showed that RhB cannot be decoloured in the presence of only H2O2 or defective V2O5@graphene. With both defective V2O5@graphene and H2O2 addition, the degradation of RhB was extremely active and conformed to quasi-first-order kinetics, and the reaction rate constant (kobs) was 0.1214 min−1, which was much higher than that of graphene/H2O2 (0.0085 min−1), pure V2O5/H2O2 (0.0043 min−1) and mechanical mixed system (graphene+V2O5/H2O2, 0.0070 min−1), respectively (Fig. 2b). The H2O2 dosages and solution pH values had an enormous effect on H2O2 activation and their influences on the RhB degradation efficiency were studied. To confirm the optimum addition of H2O2, the catalytic performance of V2O5@graphene/H2O2 system with different H2O2 addition was investigated and there was no major breakthrough in Fenton-like activity as H2O2 usage increasing to 200 µL (Fig. S5 in Supporting information). Thus the optimal 30 wt% H2O2 usage was selected as 200 µL. Notably, we found that defective V2O5@graphene had better catalytic activity under acidic and weak alkaline (pH range: 3–9, Fig. S6 in Supporting information). Under the optimal conditions, defective V2O5@graphene exhibits the highest intrinsic activity, which surpassed the most advanced heterogeneous catalysts (Fig. 2c and Table S2 in Supporting information). These findings demonstrated that graphene-controlled VCl3 thermal decomposition can in situ induced OVs in V2O5@graphene, which boosts Fenton-like performance for efficient complex organic wastewater treatment.

    Figure 2

    Figure 2.  (a) The degradation of RhB in pure graphene/H2O2, V2O5@graphene/H2O2, V2O5@graphene, graphene+V2O5/H2O2 and pure V2O5/H2O2 reaction systems. Reaction conditions: [catalyst] = 0.25 g/L, [H2O2] = 200 µL 30 wt%, [RhB] = 10 mg/L and pH 7.0. (b) Corresponding kinetic curves of the different catalytic systems. (c) Comparison of kobs of different pollutant degradation via Fenton-like oxidation in this work and the previous reports (details in Table S2 in Supporting information). (d) Mineralization efficiency of different pollutants including RhB (10 mg/L), MB (20 mg/L), MR (20 mg/L), BPA (20 mg/L) and TCP (10 mg/L) within 4 h. Reaction conditions: [catalyst] = 0.25 g/L, [H2O2] = 200 µL 30 wt% and pH 7.0.

    For the practical applications of defective V2O5@graphene/H2O2 system, we assessed the removal efficiency of methylene blue (MB), methyl red (MR), bisphenol A (BPA) and 2,4,6-trichlorophenol (TCP) which were universally discovered at acidic or weak alkaline environments. Fig. 2d showed that the TOC removal efficiency of RhB, MB, MR, BPA, and TCP reach up to 100%, 99.9%, 96.9%, 99.8%, and 99.6% within 6 h, confirming the excellent mineralization ability for persistent organic pollutants. In addition, catalyst stability was also tested and defective V2O5@graphene indicated that the defective V2O5@graphene/H2O2 system remained a high Fenton-like activity even after 12 cycles (Fig. S7 in Supporting information). During the cycle experiment, the leaching amount of V ions was investigated via inductively coupled plasma (ICP) and Fig. S8 (Supporting information) showed a low V ions concentration after each cycle (0.553, 0.573, 0.585 and 0.602 mg/L), confirming the stability of defects V2O5@graphene in Fenton-like system. Remarkably, no significant morphological changes were observed in TEM and HRTEM images (Fig. S9 in Supporting information). The XRD patterns of defective V2O5@graphene before and after reaction were basically identical, demonstrating that the structure was stable (Fig. S10 in Supporting information). What is more, Fig. S11 (Supporting information) showed that the removal efficiency of BPA was achieved higher than 90.0% within 60 min even in the presence of NaCl, Na2SO4 and Na2CO3, suggesting that such defective V2O5@graphene/H2O2 system provides an effective way for saline organic wastewater treatment. These results inferred that defective V2O5@graphene with outstanding H2O2 activation and stability would be a bright Fenton-like catalyst for removing environmental organic pollutants.

    On the basis of the above observations, in situ introduction of OVs in defective V2O5@graphene plays an important role for Fenton-like performance and the enhanced mechanism of defective V2O5@graphene was proposed as shown in Fig. 3a. Besides the redox cycle of V4+/V5+ in defective V2O5@graphene for H2O2 activation, the OVs of electron-donor nature can not only active H2O2 for the OH generation, but also facilitate the reduction of V5+ to V4+, thereby boosting the Fenton-like performance in defective V2O5@graphene/H2O2 system (Fig. 3a). This result was consistent with the previous research that the primary process of generating OH was the single electron transfer from surface V4+ to H2O2 (Eq. 1), and a chain reaction can be propagated as V4+ can be regenerated via H2O2 (Eq. 2) [36].

    (1)

    (2)

    Figure 3

    Figure 3.  (a) Schematic illustration of the Fenton-like catalysis mechanism via defective V2O5@graphene/H2O2 system. (b) EPR spectra of DMPO adducts in V2O5@graphene/H2O2, pure H2O2, pure graphene/H2O2, pure V2O5/H2O2 and V2O5@graphene reaction systems. Reaction conditions: [catalyst] = 0.25 g/L, [H2O2] = 200 µL 30 wt%, [DMPO] = 100 mmol/L, and pH 7.0. (c) High-resolution XPS spectrum of V 2p and O 1s of defective V2O5@graphene and pure V2O5 fresh and used. (d) The quenching experiments with existence of 1 mmol/L CH3OH, 10 mmol/L CH3OH in the defective V2O5@graphene/H2O2 catalytic system. Reaction conditions: [catalyst] = 0.25 g/L, [H2O2] = 200 µL 30 wt%, [RhB] = 10 mg/L and pH 7.0.

    The presence of OVs can maintain the electrostatic equilibrium, which facilitates the absorption of a large number of oxygen species, thereby promoting catalytic oxidation reactions and radical reactions [37]. Additionally, towards the oxides with surface OVs, interaction mode between OVs and H2O2 was a novel but overlooked reaction way for the H2O2 activation (Eq. 3) [38].

    (3)

    The electron paramagnetic resonance (EPR) analysis showed that DMPO−OH (1:2:2:1) signals produced in defective V2O5@graphene/H2O2 system was much higher than that in systems of pure H2O2, V2O5/H2O2 and graphene+V2O5/H2O2, which demonstrated the positive effect for H2O2 activation with the help of OVs (Fig. 3b).

    To better insight into this mechanism, we used XPS analysis to analyze chemical state of catalysts before and after reaction (Fig. 3c). Two peaks located at 517.3 eV and 515.8 eV in the V 2p3/2 XPS spectra of pristine V2O5 and V2O5@graphene were assigned to the V5+ and V4+ signals, respectively. After Fenton-like reaction, the peak intensity of surface -OH groups (532.8 eV) were improved significantly, which was attributed to the H2O2 activation by OVs in V2O5@graphene [38,39]. Interestingly, the molar ratio of V5+/V4+ in V2O5@graphene before and after reaction was 1.27:1 and 1.80:1, while the increase amount of V5+ was much lower than that of pure V2O5 before and after reaction (3.51:1 and 5.71:1), suggesting that the electron-rich OVs can suppress high value vanadium (V5+) formation. Therefore, defective V2O5@graphene with rich OVs ensures the high efficiency and stable for Fenton-like catalysis. In addition, to determine the source of the electrons, we fabricated the V2O5@graphite, V2O5@CNT, and V2O5@graphene through the solid phase synthesis method. Interestingly, the catalytic performance of V2O5@graphene was much higher than that of V2O5@graphite and V2O5@CNT (Fig. S12 in Supporting information), which demonstrated that the formed electron-rich OVs gives electron to accelerate the high value vanadium (V5+) reduction. The H2O2 activation performance of defective V2O5@graphene and pure V2O5 was independent of the BET surface areas of the catalysts (Table S3 and Figs. S13a-d in Supporting information), indicating that the binding capacity of the catalysts with H2O2 was mainly related to their chemical structure. The radical quenching experiments were further conducted to explored the roles of the reactive species in pollutant degradation. The different concentrations of methanol (OH scavengers [40,41]) addition significantly inhibited the RhB degradation, which suggested that the OH-radical pathways dominated the RhB degradation in the V2O5/H2O2 catalytic system (Fig. 3d).

    In summary, the vacancy defective V2O5@graphene with rich OVs was successfully prepared by solid phase synthesis method. Defective V2O5@graphene provides sufficient OVs, which can not only active H2O2 for the OH generation, but also accelerate the electron transfer from V5+ to V4+, which make defective V2O5@graphene a bright Fenton-like catalyst for removing environmental organic pollutants. The defective V2O5@graphene/H2O2 system presents outstanding activity and stability for OH generation, which can degrade organic pollutants efficiently.

    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.

    This work was supported by the National Key R&D Program of China (No. 2019YFC1803900), the National Natural Science Foundation of China (Nos. U1932123, 22073069, 21773082, and 42107402), the National Science Fund for Outstanding Young Scholars (No. 11722548), and the University of Chinese Academy of Sciences (No. WIUCASOD2021014).

    Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.109833.


    1. [1]

      F. Yang, B. Sheng, Z. Wang, et al., J. Hazard. Mater. 406 (2021) 124774. doi: 10.1016/j.jhazmat.2020.124774

    2. [2]

      V. Sarria, M. Deront, P. Péringer, C. Pulgarin, Appl. Catal. B: Environ. 40 (2003) 231–246. doi: 10.1016/S0926-3373(02)00162-5

    3. [3]

      J.L. Wang, L.J. Xu, Cri. Rev. Environ. Sci. Technol. 42 (2012) 251–325. doi: 10.1080/10643389.2010.507698

    4. [4]

      Z.H. Xie, C.S. He, D.N. Pei, et al., Chem. Eng. J. 468 (2023) 143778. doi: 10.1016/j.cej.2023.143778

    5. [5]

      S. Feijoo, X. Yu, M. Kamali, L. Appels, R. Dewil, Rev. Environ. Sci. Biotechnol. 22 (2023) 205–248. doi: 10.1007/s11157-023-09645-4

    6. [6]

      H. Weng, Y. Yang, C. Zhang, et al., Chem. Eng. J. 453 (2023) 139812. doi: 10.1016/j.cej.2022.139812

    7. [7]

      L. Gu, N. Zhu, L. Wang, X. Bing, X. Chen, J. Hazard. Mater. 198 (2011) 232–240. doi: 10.1016/j.jhazmat.2011.10.035

    8. [8]

      M. Gągol, A. Przyjazny, G. Boczkaj, Chem. Eng. J. 338 (2018) 599–627. doi: 10.1016/j.cej.2018.01.049

    9. [9]

      F. Mo, C. Song, Q. Zhou, et al., Proc. Natl. Acad. Sci. U. S. A. 120 (2023) e2300281120. doi: 10.1073/pnas.2300281120

    10. [10]

      Y. Li, J. Xu, G. Shi, D. Yue, Chem. Commun. 59 (2023) 1341–1344. doi: 10.1039/D2CC05989D

    11. [11]

      D. Yue, X. Yan, C. Guo, X. Qian, Y. Zhao, J. Phys. Chem. Lett. 11 (2020) 968–973. doi: 10.1021/acs.jpclett.9b03597

    12. [12]

      Y. Xue, Q. Sui, M.L. Brusseau, et al., Chem. Eng. J. 353 (2018) 657–665. doi: 10.1016/j.cej.2018.07.124

    13. [13]

      P. Salgado, V. Melin, M. Albornoz, H, et al., Appl. Catal. B: Environ. 226 (2018) 93–102. doi: 10.1016/j.apcatb.2017.12.035

    14. [14]

      Q. Yi, J. Ji, B. Shen, et al., Environ. Sci. Technol. 53 (2019) 9725–9733. doi: 10.1021/acs.est.9b01676

    15. [15]

      D. Yue, C. Guo, X. Yan, et al., Chem. Eng. J. 360 (2019) 97–103. doi: 10.1016/j.cej.2018.11.201

    16. [16]

      D.Q. He, Y.J. Zhang, D.N. Pei, et al., J. Hazard. Mater. 382 (2020) 121090. doi: 10.1016/j.jhazmat.2019.121090

    17. [17]

      Q. Ji, L. Bi, J. Zhang, H. Cao, X.S. Zhao, Energy Environ. Sci. 13 (2020) 1408–1428. doi: 10.1039/D0EE00092B

    18. [18]

      C. Li, X. Han, F. Cheng, et al., Nat. Commun. 6 (2015) 7345. doi: 10.1038/ncomms8345

    19. [19]

      P. Zhang, L. Wang, S. Yang, et al., Nat. Commun. 8 (2017) 15020. doi: 10.1038/ncomms15020

    20. [20]

      S.W.K. Andreas Stein, T.E. Mallouk, Science 259 (1993) 1558–1564. doi: 10.1126/science.259.5101.1558

    21. [21]

      M. Bianchini, J. Wang, R.J. Clement, et al., Nat. Mater. 19 (2020) 1088–1095. doi: 10.1038/s41563-020-0688-6

    22. [22]

      J. Diao, Y. Qiu, S. Liu, et al., Adv. Mater. 32 (2020) e1905679. doi: 10.1002/adma.201905679

    23. [23]

      R. Li, D. Speed, D. Siriwardena, et al., Chem. Eng. J. 425 (2021) 131785. doi: 10.1016/j.cej.2021.131785

    24. [24]

      K. Paździor, L. Bilińska, S. Ledakowicz, Chem. Engin. J. 376 (2019) 120597. doi: 10.1016/j.cej.2018.12.057

    25. [25]

      J.J. Ye, P.H. Li, H.R. Zhang, et al., Adv. Funct. Mater. 33 (2023) 2305659. doi: 10.1002/adfm.202305659

    26. [26]

      J. Chen, X.Y. Chen, Y. Liu, et al., Energ. Environ. Sci. 16 (2023) 792–829. doi: 10.1039/D2EE03015B

    27. [27]

      Y. Wen, J. Yan, B. Yang, Z. Zhuang, Y. Yu, J. Mater. Chem. A 10 (2022) 19184–19210. doi: 10.1039/D2TA02188A

    28. [28]

      W. Shao, M. Xiao, C. Yang, et al., Small 18 (2022) e2105763. doi: 10.1002/smll.202105763

    29. [29]

      J.R. Chamorro, T.M. McQueen, Acc. Chem. Res. 51 (2018) 2918–2925. doi: 10.1021/acs.accounts.8b00382

    30. [30]

      S.R. Bauers, S.R. Wood, K.M. Jensen, et al., J. Am. Chem. Soc. 137 (2015) 9652–9658. doi: 10.1021/jacs.5b04838

    31. [31]

      S. Ahlawat, K.R. Mote, N.A. Lakomek, V. Agarwal, Chem. Rev. 122 (2022) 9643–9737. doi: 10.1021/acs.chemrev.1c00852

    32. [32]

      O. Bubnova, C. Pastore, W. Sun, A. Weingarten, ACS Nano 13 (2017) 12772–12779.

    33. [33]

      T. Chirayil, P.Y. Zavalij, M.S. Whittingham, Chem. Mater. 10 (1998) 2629–2640. doi: 10.1021/cm980242m

    34. [34]

      N. Bahlawane, D. Lenoble, Chem. Vap. Deposition 20 (2014) 299–311. doi: 10.1002/cvde.201400057

    35. [35]

      Y. He, C. Chen, Y. Liu, et al., Nano Lett. 22 (2022) 4970–4978. doi: 10.1021/acs.nanolett.2c01666

    36. [36]

      G. Fang, Y. Deng, M. Huang, et al., Environ. Sci. Technol. 52 (2018) 2178–2185. doi: 10.1021/acs.est.7b05303

    37. [37]

      X. Liu, J. Wu, S. Zhang, et al., Appl. Catal. B: Environ. 320 (2023) 121994. doi: 10.1016/j.apcatb.2022.121994

    38. [38]

      H. Li, J. Shang, Z. Yang, et al., Environ. Sci. Technol. 51 (2017) 5685–5694. doi: 10.1021/acs.est.7b00040

    39. [39]

      L. Zhang, S. Wang, C. Lu, Anal. Chem. 87 (2015) 7313–7320. doi: 10.1021/acs.analchem.5b02267

    40. [40]

      Y. Yang, G. Banerjee, G.W. Brudvig, J.H. Kim, J.J. Pignatello, Environ. Sci. Technol. 52 (2018) 5911–5919. doi: 10.1021/acs.est.8b00735

    41. [41]

      X. Li, X. Liu, C. Lin, et al., Chem. Eng. J. 382 (2020) 123013. doi: 10.1016/j.cej.2019.123013

  • Figure 1  (a) Schematic illustration of the sample synthesis process. (b) XRD pattern of annealed VCl3-graphene at different mass proportions. (c) SEM and magnified SEM image of defective V2O5@graphene. (d, e) TEM and HRTEM image of defective V2O5@graphene. (f) The associated FFT analysis taken from the regions outlined in (e). (g) Elemental mapping images of defective V2O5@graphene.

    Figure 2  (a) The degradation of RhB in pure graphene/H2O2, V2O5@graphene/H2O2, V2O5@graphene, graphene+V2O5/H2O2 and pure V2O5/H2O2 reaction systems. Reaction conditions: [catalyst] = 0.25 g/L, [H2O2] = 200 µL 30 wt%, [RhB] = 10 mg/L and pH 7.0. (b) Corresponding kinetic curves of the different catalytic systems. (c) Comparison of kobs of different pollutant degradation via Fenton-like oxidation in this work and the previous reports (details in Table S2 in Supporting information). (d) Mineralization efficiency of different pollutants including RhB (10 mg/L), MB (20 mg/L), MR (20 mg/L), BPA (20 mg/L) and TCP (10 mg/L) within 4 h. Reaction conditions: [catalyst] = 0.25 g/L, [H2O2] = 200 µL 30 wt% and pH 7.0.

    Figure 3  (a) Schematic illustration of the Fenton-like catalysis mechanism via defective V2O5@graphene/H2O2 system. (b) EPR spectra of DMPO adducts in V2O5@graphene/H2O2, pure H2O2, pure graphene/H2O2, pure V2O5/H2O2 and V2O5@graphene reaction systems. Reaction conditions: [catalyst] = 0.25 g/L, [H2O2] = 200 µL 30 wt%, [DMPO] = 100 mmol/L, and pH 7.0. (c) High-resolution XPS spectrum of V 2p and O 1s of defective V2O5@graphene and pure V2O5 fresh and used. (d) The quenching experiments with existence of 1 mmol/L CH3OH, 10 mmol/L CH3OH in the defective V2O5@graphene/H2O2 catalytic system. Reaction conditions: [catalyst] = 0.25 g/L, [H2O2] = 200 µL 30 wt%, [RhB] = 10 mg/L and pH 7.0.

  • 加载中
计量
  • PDF下载量:  0
  • 文章访问数:  37
  • HTML全文浏览量:  1
文章相关
  • 发布日期:  2024-12-15
  • 收稿日期:  2023-11-09
  • 接受日期:  2024-03-25
  • 修回日期:  2024-02-26
  • 网络出版日期:  2024-03-26
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