Volume 33 Issue 4
Sep.  2020
Turn off MathJax
Article Contents

Rong-jie Qi, Jun-ying Liu, Zhi-dong Wei, Wei-qi Guo, Zhi Jiang, Wen-feng Shangguan. In situ One-Pot Fabrication of MoO3-x Clusters Modified Polymer Carbon Nitride for Enhanced Photocatalytic Hydrogen Evolution[J]. Chinese Journal of Chemical Physics , 2020, 33(4): 491-499. doi: 10.1063/1674-0068/cjcp1912220
Citation: Rong-jie Qi, Jun-ying Liu, Zhi-dong Wei, Wei-qi Guo, Zhi Jiang, Wen-feng Shangguan. In situ One-Pot Fabrication of MoO3-x Clusters Modified Polymer Carbon Nitride for Enhanced Photocatalytic Hydrogen Evolution[J]. Chinese Journal of Chemical Physics , 2020, 33(4): 491-499. doi: 10.1063/1674-0068/cjcp1912220

In situ One-Pot Fabrication of MoO3-x Clusters Modified Polymer Carbon Nitride for Enhanced Photocatalytic Hydrogen Evolution

doi: 10.1063/1674-0068/cjcp1912220
More Information
  • Corresponding author: Zhi Jiang. E-mail:zhijiang@sjtu.edu.cn
  • Received Date: 2019-12-12
  • Accepted Date: 2020-02-19
  • Publish Date: 2020-08-27
  • Developing low-cost and high-efficient noble-metal-free cocatalysts has been a challenge to achieve economic hydrogen production. In this work, molybdenum oxides (MoO$_{3-x}$) were in situ loaded on polymer carbon nitride (PCN) via a simple one-pot impregnation-calcination approach. Different from post-impregnation method, intimate coupling interface between high-dispersed ultra-small MoO$_{3-x}$ nanocrystal and PCN was successfully formed during the in situ growth process. The MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, 4) photocatalyst without noble platinum (Pt) finally exhibited enhanced photocatalytic hydrogen performance under visible light irradiation ($\lambda$$>$420 nm), with the highest hydrogen evolution rate of 15.6 μmol/h, which was more than 3 times that of bulk PCN. Detailed structure-performance revealed that such improvement in visible-light hydrogen production activity originated from the intimate interfacial interaction between high-dispersed ultra-small MoO$_{3-x}$ nanocrystal and polymer carbon nitride as well as efficient charge carriers transfer brought by Schottky junction formed.
  • 加载中
  • [1] N. S. Lewis, Science 351, 1920 (2016). doi:  10.1126/science.aad1920
    [2] A. Fujishima and K. Honda, Nature 238, 37 (1972). doi:  10.1038/238037a0
    [3] L. Li, J. Yan, T. Wang, Z. J. Zhao, J. Zhang, J. Gong, and N. Guan, Nat. Commun. 6, 5881 (2015). doi:  10.1038/ncomms6881
    [4] J. Yu and A. Kudo, Adv. Funct. Mater. 16, 2163 (2006). doi:  10.1002/adfm.200500799
    [5] S. Ouyang, H. Tong, N. Umezawa, J. Cao, P. Li, Y. Bi, Y. Zhang, and J. Ye, J. Am. Chem. Soc. 134, 1974 (2012) doi:  10.1021/ja210610h
    [6] X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen, and M. Antonietti, Nat. Mater. 8, 76 (2008).
    [7] X. Li, G. Hartley, A. J. Ward, P. A. Young, A. F. Masters, and T. Maschmeyer, J. Phys. Chem. C 119, 14938 (2015). doi:  10.1021/acs.jpcc.5b03538
    [8] P. Zhou, J. P. Lai, Y. H. Tang, Y. G. Chao, F. Lin, and S. J. Guo, Appl. Catal. B 238, 161 (2018). doi:  10.1016/j.apcatb.2018.07.007
    [9] J. Yang, D. Wang, H. Han, and C. Li, 46, 1900 (2013).
    [10] G. D. Moon, J. B. Joo, I. Lee, and Y. Yin, Nanoscale 6, 12002 (2014). doi:  10.1039/C4NR03521F
    [11] Z. Zhao, Y. Sun, and F. Dong, Nanoscale 7, 15 (2015). doi:  10.1039/C4NR03008G
    [12] Q. Xiang, J. Yu, and M. Jaroniec, J. Am. Chem. Soc. 134, 6575 (2012). doi:  10.1021/ja302846n
    [13] K. Xia, Z. Chen, J. Yi, H. Xu, Y. Yu, X. She, Z. Mo, H. Chen, Y. Xu, and H. Li, Ind. Eng. Chem. Res. 57, 8863 (2018). doi:  10.1021/acs.iecr.8b01268
    [14] F. Gao, Y. Zhao, L. Zhang, B. Wang, Y. Wang, X. Huang, K. Wang, W. Feng, and P. Liu, J. Mater. Chem. A 6, 18979 (2018). doi:  10.1039/C8TA06029K
    [15] B. Ma, D. Li, X. Wang, and K. Lin, ChemSusChem 11, 3871 (2018). doi:  10.1002/cssc.201801481
    [16] Y. Guo and J. Robertson, Appl. Phys. Lett. 105, 222110 (2014). doi:  10.1063/1.4903538
    [17] X. Xiao, Z. Peng, C. Chen. C. Zhang, M. Beidaghi, Z. Yang, N. Wu, Y. Huang, L. Miao, Y. Gogotsi, and J. Zhou, Nano Energy 9, 355 (2014). doi:  10.1016/j.nanoen.2014.08.001
    [18] C. Gretener, J. Perrenoud, L. Kranz, C. Baechler, S. Yoon, Y. E. Romanyuk, S. Buecheler, and A. N. Tiwari, Thin Solid Films 535, 193 (2013). doi:  10.1016/j.tsf.2012.11.110
    [19] C. Liu, W. Li, J. Chen, J. Fan, Y. Mai, and R. E. I. Schropp, Nano Energy 41, 75 (2017). doi:  10.1016/j.nanoen.2017.08.048
    [20] J. Bullock, A. Cuevas, T. Allen, and C. Battaglia, Appl. Phys. Lett. 105, 232109 (2014). doi:  10.1063/1.4903467
    [21] G. A. Saenz and A. B. Kaul, Surf. Coat. Int. 382, 125031 (2019).
    [22] C. Steven, B. Corsin, A. Angelica, M. Stephen, J.S. Kang, X. Yin, T. Mahmut, K. Rehan, F. Hui, W. Robert, and J. Ali, Nano Lett. 14, 1337 (2014). doi:  10.1021/nl4043505
    [23] J. Wen, J. Xie, X. Chen, and X. Li, Appl. Surf. Sci. 391, 72 (2017). doi:  10.1016/j.apsusc.2016.07.030
    [24] H. Li, H. Zhou, K. Chen, K. Liu, S. Li, K. Jiang, W. Zhang, Y. Xie, Z. Cao, H. Li, H. Liu, X. Xu, H. Pan, J. Hu, D. Tang, X. Qiu, J. Fu, and M. Liu, Solar RRL. https://doi.org/10.1002/solr.201900416 (2019).
    [25] Z. Chen, K. Xia, X. She, Z. Mo, S. Zhao, J. Yi, Y. Xu, H. Chen, H. Xu, and H. Li, Appl. Surf. Sci. 447, 732 (2018). doi:  10.1016/j.apsusc.2018.03.226
    [26] Y. X. Zhang, P. Zeng, Y. X. Yu, and W. D. Zhang, Chem. Eng. J. 381, 122635 (2020). doi:  10.1016/j.cej.2019.122635
    [27] Y. Hou, Y. Zhu, Y. Xu, and X. Wang, Appl. Catal. B 156-157, 122 (2014). doi:  10.1016/j.apcatb.2014.03.002
    [28] C. Ma, J. Zhou, H. Zhu, W. Yang, J. Liu, Y. Wang, and Z. Zou, ACS Appl. Mater. Interf. 7, 14628 (2015). doi:  10.1021/acsami.5b01356
    [29] C. Ma, J. Zhou, Z. Cui, Y. Wang, and Z. Zou, Solar Energy Mater. Solar Cells 150, 102 (2016). doi:  10.1016/j.solmat.2016.02.010
    [30] J. Ran, T. Y. Ma, G. Gao, X. W. Du, and S. Z. Qiao, Energy Environ. Sci. 8, 3708 (2015). doi:  10.1039/C5EE02650D
    [31] J. Baltrusaitis, B. Mendoza-Sanchez, V. Fernandez, R. Veenstra, N. Dukstiene, A. Roberts, and N. Fairley, Appl. Surf. Sci. 326, 151 (2015). doi:  10.1016/j.apsusc.2014.11.077
    [32] H. Al-Kandari, F. Al-Kharafi, and A. Katrib, Catal. Commun. 9, 847 (2008). doi:  10.1016/j.catcom.2007.09.009
    [33] S. Bai, S. Chen, L. Chen, K. Zhang, R. Luo, D. Li, and C. C. Liu, Sensors Actuators B 174, 51 (2012). doi:  10.1016/j.snb.2012.08.015
    [34] O. G. Marin Flores and S. Ha, Appl. Catal. A 352, 124 (2009). doi:  10.1016/j.apcata.2008.09.036
    [35] J. Xu, Y. Chang, Y. Zhang, S. Ma, Y. Qu, and C. Xu, Appl. Surf. Sci. 255, 1996 (2008). doi:  10.1016/j.apsusc.2008.06.130
    [36] S. Hu, X. Chen, Q. Li, F. Li, Z. Fan, H. Wang, Y. Wang, B. Zheng, and G. Wu, Appl. Catal. B 201, 58 (2017). doi:  10.1016/j.apcatb.2016.08.002
    [37] Z. Qin, Y. Chen, Z. Huang, J. Su, and L. Guo, J. Mater. Chem. A 5, 19025 (2017). doi:  10.1039/C7TA04434H
    [38] C. Li, S. Yu, L. Gu, J. Han, H. Dong, Y. Wang, and G. Chen, Adv. Mater. Interfaces 5, 1801297 (2018). doi:  10.1002/admi.201801297
    [39] Y. Guo, J. Li, Y. Yuan, L. Li, M. Zhang, C. Zhou, and Z. Lin, Angew. Chem. Int. Ed. 55, 14693 (2016). doi:  10.1002/anie.201608453
    [40] C. Chang, Y. Fu, M. Hu, C. Wang, G. Shan, and L. Zhu, Appl. Catal. B 142-143, 553 (2013). doi:  10.1016/j.apcatb.2013.05.044
    [41] Y. Y. Cai, X. H. Li, Y. N. Zhang, X. Wei, K. X. Wang, and J. S. Chen, Angew. Chem. Int. Ed. 52, 11822 (2013). doi:  10.1002/anie.201304652
  • 加载中
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

Figures(14)  / Tables(1)

Article Metrics

Article views(8) PDF downloads(3) Cited by()

Proportional views
Related

In situ One-Pot Fabrication of MoO3-x Clusters Modified Polymer Carbon Nitride for Enhanced Photocatalytic Hydrogen Evolution

doi: 10.1063/1674-0068/cjcp1912220

Abstract: Developing low-cost and high-efficient noble-metal-free cocatalysts has been a challenge to achieve economic hydrogen production. In this work, molybdenum oxides (MoO$_{3-x}$) were in situ loaded on polymer carbon nitride (PCN) via a simple one-pot impregnation-calcination approach. Different from post-impregnation method, intimate coupling interface between high-dispersed ultra-small MoO$_{3-x}$ nanocrystal and PCN was successfully formed during the in situ growth process. The MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, 4) photocatalyst without noble platinum (Pt) finally exhibited enhanced photocatalytic hydrogen performance under visible light irradiation ($\lambda$$>$420 nm), with the highest hydrogen evolution rate of 15.6 μmol/h, which was more than 3 times that of bulk PCN. Detailed structure-performance revealed that such improvement in visible-light hydrogen production activity originated from the intimate interfacial interaction between high-dispersed ultra-small MoO$_{3-x}$ nanocrystal and polymer carbon nitride as well as efficient charge carriers transfer brought by Schottky junction formed.

Rong-jie Qi, Jun-ying Liu, Zhi-dong Wei, Wei-qi Guo, Zhi Jiang, Wen-feng Shangguan. In situ One-Pot Fabrication of MoO3-x Clusters Modified Polymer Carbon Nitride for Enhanced Photocatalytic Hydrogen Evolution[J]. Chinese Journal of Chemical Physics , 2020, 33(4): 491-499. doi: 10.1063/1674-0068/cjcp1912220
Citation: Rong-jie Qi, Jun-ying Liu, Zhi-dong Wei, Wei-qi Guo, Zhi Jiang, Wen-feng Shangguan. In situ One-Pot Fabrication of MoO3-x Clusters Modified Polymer Carbon Nitride for Enhanced Photocatalytic Hydrogen Evolution[J]. Chinese Journal of Chemical Physics , 2020, 33(4): 491-499. doi: 10.1063/1674-0068/cjcp1912220
  • Efficient utilization and conversion of solar energy has been a key task for the sustainable development of human society [1]. Photocatalytic technology is regarded as a vital approach to alleviate energy shortage and environmental pollution [2]. Hitherto, various photocatalytic materials, such as TiO$_2$ [3], BiVO$_4$ [4], SrTiO$_3$ [5] have been found to exploit solar energy for hydrogen generation and pollutants degradation. Polymer carbon nitride (PCN), a charming non-metal photocatalyst, was reported by Wang et al. one decade ago [6] and has attracted immense interest as an artificial conjugated polymer semiconductor with the merits like abundance, high thermal and chemical stability and a suitable bandgap [7]. However obstacles, such as the high recombination rate of photo-generated charges, still hamper the further improvement of the photocatalytic efficiency of pristine graphitic carbon nitride in the hydrogen production process [8].

    Fortunately, these disadvantages can be overcome by loading suitable noble cocatalysts onto surface of bulk polymer carbon nitride [9]. It is well known that these noble metals with high work function (such as Pt, Pd) facilitate electron trapping process and benefit efficient separation and transportation of charge carrier via metal-semiconductor junction formed (Schottky barrier) [10, 11], thus enabling efficient photocatalytic hydrogen evolution. However, the high cost of these noble metal cocatalyst inevitably limits their large-scale industrial application. Therefore, the noble-metal-free and highly efficient cocatalyst has been under increasing demand to solve the above problem.

    Recently, transition metal molybdenum and its derivative compounds (MoS$_2$ [12], MoN [13], MoC [14]) emerged as promising noble-metal-free cocatalysts for their merits in earth abundant and excellent performance [15]. However, few reports were found to utilize MoO$_{3-x}$ clusters as cocatalysts in photocatalysis. MoO$_{3-x}$ had metallic conductivity since the deficient oxygen (or oxygen vacancy) was shallow and could be easily formed via partial reduction of MoO$_3$ [16, 17]. Theoretically, metallic molybdenum oxides (MoO$_{3-x}$) possessed a relatively high work function (6.0 eV to 6.6 eV, even larger than noble metal Pd) [18-21] together with tunable high conductivity [16, 22]. Therefore, the Schottky junction is easy to form between metallic MoO$_{3-x}$ clusters cocatalyst and semiconductor photocatalysts from the aspect of physical electronic structure match. The formed Schottky junction will suppress the electron back and transfer from metal to semiconductor via internal interfacial electric field built, promoting photo-generated charge carrier separation greatly [23]. In previous reports, molybdenum oxides were usually loaded on the surface of polymer carbon nitride via the simple impregnation or self-assembly method [24, 25]. The resulting molybdenum oxide cocatalysts in these cases might suffer from aggregation and lack sufficient interfacial contact interaction, hindering efficient charge migration process. Therefore, it is highly expected that polymer carbon nitride semiconductor with high-dispersed ultra-small metallic MoO$_{3-x}$ clusters and intimate metal-semiconductor Schottky junction interface would be synthesized through a simple strategy for improved charge carrier dynamics and enhanced visible-light photocatalytic activity.

  • Melamine, chloroplatinic acid (H$_2$PtCl$_6$) and sodium sulfate (Na$_2$SO$_4$) were purchased from Sinopharm Chemical Reagent Co., Ltd. Ammonium molybdate tetrahydrate ((NH$_4$)$_6$Mo$_7$O$_{24}\cdot$4H$_2$O) was obtained from Aldrich Chemical Co. ITO glass was ordered from Suzhou Crystal Silicon Electronics & Technology Co. All reagents in this study were used without further purification.

  • FIG. S1 (supplementary materials) illustrated the process of the in situ one-pot impregnation-calcination synthesis method of MoO$_{3-x}$-PCN-$X$ samples. Typically, a certain amount of (NH$_4$)$_6$Mo$_7$O$_{24}\cdot$4H$_2$O (10, 50, 100, 200 mg) and 5 g melamine were firstly dissolved in 100 mL deionized water. The obtained solution was then stirred using magnetic stirrer at 60 ℃ for 24 h to evaporate water. Subsequently, the resulting product was collected and put into a crucible which was covered with aluminum foil later. Afterwards, the crucible was heated in a muffle furnace to 550 ℃ with a moderate heating rate of 5 ℃/min and maintained for 2 h under air atmosphere. These products with different amount of (NH$_4$)$_6$Mo$_7$O$_{24}\cdot$4H$_2$O were denoted as MoO$_{3-x}$-PCN-$X$ (where $X$=1, 2, 3, 4 correspond to the amount of (NH$_4$)$_6$Mo$_7$O$_{24}\cdot$4H$_2$O used, i.e. 10, 50, 100, 200 mg, respectively). In addition, bulk PCN was obtained by the above steps in the absence of (NH$_4$)$_6$Mo$_7$O$_{24}\cdot$4H$_2$O. MoO$_{3-x}$ powder was synthesized by heating (NH$_4$)$_6$Mo$_7$O$_{24}\cdot$4H$_2$O at 550 ℃ for 2 h. MoO$_{3-x}$-PCN-mixed sample was fabricated by grinding the same ratio of MoO$_{3-x}$ to bulk PCN in MoO$_{3-x}$-PCN-2 sample (1.27wt%) for 10 min. The Pt-PCN sample was synthesized by in situ photodeposited 1.27 wt% Pt on the surface of bulk PCN with H$_2$PtCl$_6$ as the precursor.

  • Powder XRD patterns of MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, 4) samples were gained on a powder X-ray diffractometer (Cu K$\alpha$ radiation source, D8, Bruker). Specific surface areas of these samples were obtained through TriStar II 3020 Micromeritics porosimeter via Brunauer-Emmett-Teller (BET) method. The surface morphology and elemental analysis were revealed by a Telos F200X G2 (scanning) transmission electron microscope (STEM) operated at 200 kV. High-resolution transmission electron microscope images and element mapping were also acquired using Telos F200X G2. The molybdenum concentration in MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, 4) samples was measured by Thermo iCAP6300 inductively coupled plasma optical equipment. The XPS patterns were generated on an electronic energy spectrum (AXIS UltraDLD, Kratos group) at 300 W using Mg K$\alpha$ X-rays as the excitation source. The diffuse reflectance spectra (DRS) between 230 and 850 nm were measured by Shimadzu UV-2450 spectrophotometer using a BaSO$_4$ reference. The steady-state photoluminescence (PL) spectra were recorded on Perkin-Elmer Model LS 55 luminescence spectrometer under the 350 nm excitation. Time-resolved fluorescence spectrum was obtained via a photon technology international (PTI) laserstrobe fluorescence lifetime fluorometer system.

  • The photocatalytic hydrogen production test was carried out on a Pyrex reaction cell which was linked to a closed gas circulation and evacuation system. 0.1 g photocatalyst was dispersed uniformly in 100 mL aqueous solution with 20 vol% TEOA as the sacrificial agent. The reacting solution was evacuated via vacuum pump and the reaction temperature was maintained at 25 ℃ with circular cooling water system. The photocatalytic hydrogen production activity was evaluated under visible light irradiation by a 300 W Xenon lamp with a UV420 cut-off filter.

  • Photoelectrochemical experiment was performed on a typical three-electrode electrochemical cell with a quartz window through a PARSTAT 4000 electrochemical workstation. The working electrode was prepared by dig-coating photocatalyst slurry on ITO conductive glass substrate (the effective area was 1 cm$^2$). An Ag/AgCl 3.5 mol/L (+0.2046 V vs. SHE) was used as the reference electrode while platinum foil was utilized as a counter electrode. The 0.2 mol/L Na$_2$SO$_4$ solution (pH=6.6) was used as the electrolyte solution during the photoelectrochemical measurements. Electrochemical impedance spectroscopy (EIS) measurement was carried out with a 10 mV sinusoidal ac perturbation applied over the frequency range of 0.01-10$^5$ Hz. Mott-Schottky analyses were generated at 500, 1000, and 1500 Hz with the sweeping voltage ranging from $-$0.8 V to 0.8 V.

  • The AQY for the hydrogen production was measured through a 300 W Xe lamp and band-pass filters with central wavelength of 420, 450, 500, and 550 nm. Full width at half maximum of these band-path filters was 10 nm, which was adopted as an error margin of the irradiation wavelength. Hydrogen production activities were tested and utilized to calculate the number of H$_2$ molecules evolved. The AQY value was computed using the following equation: AQY=$N_{\rm{e}}$/$N_{\rm{p}}$=2$M/N_{\rm{p}}$, where $N_{\rm{e}}$ represents the number of reacted electrons, $N_{\rm{p}}$ represents the total number of incident photons and $M$ represents the number of produced hydrogen molecules.

  • ICP-MS analysis was firstly performed to determine molybdenum contents among these MoO$_{3-x}$-PCN-$X$ samples. These MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, 4) samples had 0.49wt%, 1.27wt%, 3.12wt%, 11.93wt% of molybdenum from the lowest to highest contents, respectively. XRD pattern of MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, 4) samples with different MoO$_{3-x}$ loading amount is shown in FIG. 1. Two characteristic peaks at 13$^\circ$ and 27.5$^\circ$ can be ascribed to (100) in-plane repeated motif and (002) interplanar periodic stacking of aromatic systems. Notably, the intensity of polymer carbon nitride characteristic (100) and (002) peaks gradually decreased with increasing content of the loaded MoO$_{3-x}$. Similar decrease in PCN characteristic (002) peak intensity was also reported after loading of noble-metal-free cocatalyst in previous studies [26, 27]. One possible reason is that the attachment of noble-metal-free cocatalyst onto PCN might destruct periodic graphitic stacking between its layers. Interestingly, no signals related to molybdenum species were detected from the XRD patterns above. This outcome might originate from the high dispersion of MoO$_{3-x}$ clusters in these MoO$_{3-x}$-PCN-$X$ samples.

    Figure 1.  XRD patterns of bulk PCN and MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, 4) samples.

    Morphology and microstructure of MoO$_{3-x}$-PCN-2 sample were observed via high-resolution (scanning) transmission electron microscope. Bulk PCN showed typical thick sheet-like structure (FIG. S2 in supplementary materials). This characteristic two-dimensional planar microstructure was mainly retained after in situ loading of MoO$_{3-x}$ clusters (FIG. 2(a)). Small clusters could be observed to anchor on the surface of PCN (marked in white circles). These well-dispersed ultra-small clusters were better displayed under 5 nm dimension in FIG. 2(b). High angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) was utilized to further investigate the microstructure and elements distribution of MoO$_{3-x}$-PCN-2 sample. It is well known that heavier atom will be presented as brighter in a typical HAADF image. Thus, the bright white pots in FIG. 2(c) could be safely attributed to in situ grown molybdenum-related clusters, confirming the high-dispersion state of the clusters again. Besides, the HRTEM images of MoO$_{3-x}$ clusters in FIG. 2(d) show the particle's resolved lattice fringes with spacing of 0.324 nm, which is in compliance with the (011) plane information of MoO$_{3-x}$ in a previous report [24]. In addition, the relatively homogeneous distribution of MoO$_{3-x}$ on PCN's surface was confirmed by HAADF-STEM mapping images (FIG. 2(e-i)). The possible synthesis mechanism in regard to in situ process is as follows briefly. (NH$_4$)$_6$Mo$_7$O$_{24}\cdot$4H$_2$O first converted to H$_2$MoO$_4$ during the mixing and drying process and then decomposed into MoO$_3$ at 190 ℃ during the polymerization process [28, 29]. Melamine in the synthesis of MoO$_{3-x}$-PCN-$X$ samples acted as a molten flux to homogenize the reactant system and help the dispersion of MoO$_{3-x}$ (resulted from partial reduction of MoO$_3$ by melamine). BET surface area of MoO$_{3-x}$-PCN-$X$ composite photocatalysts was measured via nitrogen adsorption-desorption method. The surface area of bulk PCN was 11.96 m$^2$/g. And the surface areas of MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, and 4) sample were 25.33, 40.42, 34.47, and 40.42 m$^2$/g as the Mo content increased from 0.49wt% to 1.27wt%. Larger surface area might provide more active sites, thus improving photocatalytic activity [30]. However, surface area of MoO$_{3-x}$-PCN-$X$ samples dropped with further increase in MoO$_{3-x}$ loading. The above result may be derived from several reasons. On one hand, the release of gas during decomposition of (NH$_4$)$_6$Mo$_7$O$_{24}\cdot$4H$_2$O might assist the formation of porous MoO$_{3-x}$-PCN-$X$ composite photocatalyst. On the other hand, the aggregation of MoO$_{3-x}$ nanoparticles may decrease the surface area of composite photocatalyst, hence leading to the volcano shape of BET surface area with increased doping amount of Mo precursors.

    Figure 2.  TEM images of MoO$_{3-x}$-PCN-2 sample (a) under 100 nm dimension and (b) under 5 nm dimension (MoO$_{3-x}$ clusters were encircled). (c) HAADF image of MoO$_{3-x}$-PCN-2 sample (the bright white pots were marked as MoO$_{3-x}$ clusters). (d) HRTEM image of MoO$_{3-x}$ clusters anchored on the polymer carbon nitride. (e) HAADF image, as well as element mapping of (f) C, (g) N, (h) O, and (i) Mo of selected area in MoO$_{3-x}$-PCN-2 sample.

    Chemical states of the as-prepared MoO$_{3-x}$-PCN-$X$ sample were thoroughly analyzed by X-ray photoelectron spectroscopy (XPS) to obtain more information of anchored MoO$_{3-x}$ clusters. In Mo 3d spectra, the peaks located at 232.3 eV and 235.4 eV belonged to 3d$_{5/2}$ and 3d$_{3/2}$ of Mo$^{6+}$ while peaks at 230.8 eV and 233.9 eV could be ascribed to 3d$_{5/2}$ and 3d$_{3/2}$ of Mo$^{4+}$ (FIG. 3(a)) [31]. Besides, the O 1s spectra (FIG. S3 in supplementary materials) could be deconvoluted into peaks located at 531.5, 532.4, and 533.8 eV, respectively. The peak at 531.5 eV corresponded to lattice oxygen in MoO$_{3-x}$ clusters [32-35]. Based on the identification of Mo$^{4+}$, the peak located at 532.4 eV should be attributed to oxygen vacancy within MoO$_{3-x}$ clusters. In addition, the peak at 533.8 eV belonged to absorbed oxygen on the surface of the sample. The same peak could also be observed in bulk PCN sample. Therefore, these results clearly indicated the formation of defective MoO$_{3-x}$ clusters via partial reduction of MoO$_3$. The molar ratio of crystal lattice oxygen to molybdenum in MoO$_{3-x}$-PCN-2 sample was estimated to be 2.79 via XPS semi-quantitative analysis. This value was a bit lower than that of MoO$_3$, suggesting the existence of oxygen vacancy and unsaturated Mo-O chemical bonds within MoO$_{3-x}$ clusters. FIG. 3(b) demonstrates the C 1s XPS spectra of MoO$_{3-x}$-PCN-$X$ sample. It can be clearly observed that three deconvoluted peaks centered at 284.8, 286.2, and 288.1 eV, which represented C-C, C-NH$_x$, and N=C-N structures within the PCN's framework, respectively. Besides, the N 1s XPS spectra for MoO$_{3-x}$-PCN-2 sample consisted three deconvoluted peaks locating at 398.5, 399.9, and 401.1 eV. These peaks could be assigned to two-coordinated nitrogen, three-coordinated nitrogen and C-NH$_x$ structure, respectively. The N 1s XPS spectra of MoO$_{3-x}$-PCN-1 and MoO$_{3-x}$-PCN-2 samples were similar to that of bulk PCN. However, the characteristic nitrogen peaks of MoO$_{3-x}$-PCN-1 and MoO$_{3-x}$-PCN-2 samples shifted to high binding energy (FIG. 3(c)). Similar high angle drift of binding energy could also be observed in C 1s spectra in MoO$_{3-x}$-PCN-1 and MoO$_{3-x}$-PCN-2 samples, demonstrating the change of chemical environment in carbon nitride matrix after in situ MoO$_{3-x}$ loading. This outcome above indicated that decreased electron density around PCN framework in MoO$_{3-x}$-PCN-$X$ compared to that of bulk PCN, suggesting strong interaction between these metallic MoO$_{3-x}$ clusters and PCN substrate, which might be ascribed to the coordinative Mo-N bond [24, 36]. The possible mutual interaction between molybdenum and nitrogen after the growth of crystal molybdenum oxide clusters on surface of PCN, was expected to facilitate the formation of this unique architecture with compact interface between MoO$_{3-x}$ and PCN substrates. Therefore, an intimate Schottky junction interface existed between these metallic MoO$_{3-x}$ clusters and PCN, which was in favor of charge transfer process. Optical properties of MoO$_{3-x}$-PCN-$X$ samples were measured via UV-Vis DRS spectra (FIG. 4). The DRS spectra showed that the absorption edges of MoO$_{3-x}$-PCN-$X$ samples were nearly unchanged in comparison to bulk PCN counterpart. Their absorption edges were located at around 450 nm, which corresponded to bandgap of PCN (2.76 eV). In addition, the light absorption of these samples was enhanced with an increase in MoO$_{3-x}$ loading amount, which might be related to its dark brown color.

    Figure 3.  High-resolution XPS of (a) Mo 3d, (b) C 1s, and (c) N 1s spectra of MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, 4) samples.

    Figure 4.  UV-Vis DRS spectra for MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, 4) samples.

    To understand the efficiency of the Schottky junction interface between metallic MoO$_{3-x}$ clusters and semiconductor PCN, the charge carrier transfer dynamics of MoO$_{3-x}$-PCN-$X$ samples was investigated through various characterization methods. Steady-state photoluminescence (PL) spectra of bulk PCN and MoO$_{3-x}$-PCN-$X$ samples were carried out at excitation of 350 nm light (FIG. 5(a)). Both bulk PCN and MoO$_{3-x}$-PCN-2 samples exhibited a strong emission peak located at 445 nm. However, PL intensity of MoO$_{3-x}$-PCN-2 sample was much weaker than its bulk counterpart, indicating suppressed photo-induced charge recombination rate and higher charge transfer rate within MoO$_{3-x}$-PCN-2 system [30]. The steady-state PL spectra of other MoO$_{3-x}$-PCN-$X$ samples displayed in FIG. 5(a) also confirmed that the radiative charge recombination was successfully quenched after the loading of MoO$_{3-x}$ clusters. The above results indicated that loading of MoO$_{3-x}$ cocatalysts could facilitate charge transfer process through intimate Schottky junction interface formed between metallic MoO$_{3-x}$ clusters and semiconductor polymer carbon nitride to avoid fast charge combination rate within bulk PCN aromatic system [37]. Besides, lifetime of photo-induced charge carriers was significant for photocatalytic activity because it determined their probability of participating into photocatalytic reaction before recombination [30]. Therefore, the time-resolved photoluminescence (tr-PL) spectra of bulk PCN and MoO$_{3-x}$-PCN-$X$ samples are displayed in FIG. 5(b) in order to gain better understanding of the lifetime of photo-induced charge carriers. The fitting lifetime of photo-induced charge carrier is listed in Table I in detail. The calculated fluorescence lifetime of MoO$_{3-x}$-PCN-2 (5.917 ns) was much longer than that of the bulk PCN sample (5.017 ns). This result suggested that the efficient electron/hole pairs separation had been realized after the suitable amount of MoO$_{3-x}$ species loading [38]. Longer fluorescence lifetime than bulk PCN could also be observed for other MoO$_{3-x}$-PCN-$X$ samples (FIG. 5(b) and Table I). Specifically, transient photocurrent responses of bulk PCN and MoO$_{3-x}$-PCN-2 samples under visible light irradiation were also recorded to provide more information regarding separation and migration of photo-induced electrons and holes within system. FIG. 5(c) shows that the photocurrent response of MoO$_{3-x}$-PCN-2 modified electrode was much stronger than that of bulk PCN modified one. It was widely acknowledged that photocurrent could provide strong evidence for demonstrating charge separation ability of photocatalysts [39]. Higher photocurrent response usually suggested better separation efficiency of electron/hole pairs. Therefore, the above transient photocurrent response result indicated MoO$_{3-x}$ could act as a suitable cocatalyst to reduce the recombination rates and promote separation of photogenerated charge carriers via efficient Schottky junction formed.

    Figure 5.  (a) Steady-state PL spectra of bulk PCN and MoO$_{3-x}$-PCN-$X$ samples at 350 nm excitation light. (b) Time-resolved PL spectra of bulk PCN and MoO$_{3-x}$-PCN-$X$ ($X$=1, 2, 3, 4) samples at $\lambda_{\textrm{exe}}$=368 nm. (c) Transient photocurrent response of bulk PCN and MoO$_{3-x}$-PCN-2 electrodes under visible light irradiation. (d) EIS Nyquist plots of bulk PCN and MoO$_{3-x}$-PCN-2 electrodes in 0.2 mol/L Na$_2$SO$_4$ aqueous solution.

    Table Ⅰ.  Kinetic analysis of emission decay for bulk PCN and MoO$_{3-x}$-PCN-$X$ sample. $\tau_1$: short decay lifetime, Rel$_1$: short decay amplitude, $\tau_2$: long decay lifetime, and Rel$_2$: long decay amplitude.

    The above techniques highlighted efficient charge carrier separation and transportation merits in MoO$_{3-x}$-PCN-2 sample under light irradiation. But the charge carrier dynamics of MoO$_{3-x}$-PCN-2 and bulk PCN sample under dark condition, which might reflect their intrinsic charge transfer capacity, was still unknown. Therefore, the electrochemical impedance spectroscopy (EIS) of bulk PCN and MoO$_{3-x}$-PCN-2 samples was studied under dark condition. It was well known that EIS spectra could demonstrate the interface charge separation efficiency and smaller radius of EIS Nyquist plots usually implied lower charge transportation resistance and better dynamic property [40]. FIG. 5(d) illustrates that the arc radius of MoO$_{3-x}$-PCN-2 sample was much smaller compared to bulk PCN sample, suggesting that there was highly efficient charge transportation within MoO$_{3-x}$-PCN-2 system via metal-semiconductor Schottky junction, in agreement with tr-PL and transient photocurrent outcomes above. These charge dynamics related properties confirmed that in situ loading of high-dispersed ultra-small MoO$_{3-x}$ clusters contributes to efficient separation, transportation and low-recombination of photo-generated charge carriers within photocatalysis by constructing sufficient Schottky junction interface between well-dispersed metallic MoO$_{3-x}$ clusters and two-dimensional polymer carbon nitride.

    Photocatalytic hydrogen evolution of the as-prepared MoO$_{3-x}$-PCN-$X$ samples was studied in the presence of TEOA under visible light irradiation ($\lambda$$>$420 nm). As displayed in FIG. 6(a), the bulk PCN sample showed limited photocatalytic hydrogen production ability in the absence of MoO$_{3-x}$ cocatalyst (4.5 μmol/h). The hydrogen generation rate increased to 11.8 μmol/h over MoO$_{3-x}$-PCN-1 sample after introduction of a small amount of MoO$_{3-x}$ cocatalysts (0.49wt%). Further increase in loading amount of MoO$_{3-x}$ cocatalyst (1.27wt%) leaded to optimal photocatalytic hydrogen production rate (15.6 μmol/h) over MoO$_{3-x}$-PCN-2 sample. And the AQY efficiency of the champion sample MoO$_{3-x}$-PCN-2 was calculated to be 1.25% at 420 nm monochromatic light irradiation (FIG. S4 in supplementary materials). However, a decrement in photocatalytic hydrogen production could be observed with increasing MoO$_{3-x}$ loading amount over MoO$_{3-x}$-PCN-3 and MoO$_{3-x}$-PCN-4 samples. This outcome could be explained by the fact that the over-loaded MoO$_{3-x}$ nanoparticles would act as charge carrier recombination centers and even cover the reaction active sites, deteriorating photocatalytic reactions. In addition, the photocatalytic hydrogen production yield of MoO$_{3-x}$-PCN-mixed sample (mechanically mixed MoO$_{3-x}$ and bulk PCN) was evaluated to emphasize the advantages of our in situ one-pot fabrication method. FIG. S5 (supplementary materials) demonstrates that lack in intimate coupling Schottky junction between mechanical-mixed MoO$_{3-x}$ and PCN finally leaded to limited enhanced photocatalytic performance compared to in situ fabricated MoO$_{3-x}$-PCN-2 sample. Furthermore, the comparison of MoO$_{3-x}$-PCN-2 and Pt-PCN samples in FIG. S6 (supplementary materials) showed the potential of MoO$_{3-x}$ clusters as efficient noble-metal-free cocatalysts in consideration of its costs and efficiency since noble-metal-free MoO$_{3-x}$ cocatalysts achieved nearly half hydrogen production amount of Pt-PCN sample under visible light irradiation. Moreover, the photocatalytic hydrogen production stability of MoO$_{3-x}$-PCN-2 was examined by five rounds photocatalytic hydrogen evolution tests (FIG. 6(b)) and structural characterizations (FIG. S7 and FIG. S8 in supplementary materials). No obvious change in photocatalytic activity or XPS spectra (FIG. S8) could be found, thus suggesting the good stability of well-dispersed ultra-small MoO$_{3-x}$ cocatalysts.

    Figure 6.  (a) Photocatalytic hydrogen evolution rate of MoO$_{3-x}$-PCN-$X$ samples, and (b) stability test of photocatalytic hydrogen evolution over MoO$_{3-x}$-PCN-2 sample under visible light irradiation ($\lambda$$>$420 nm).

    Compared to bulk PCN sample, the photocatalytic hydrogen evolution activity of MoO$_{3-x}$-PCN-2 sample was greatly enhanced, as mentioned above. Such enhancement in photocatalytic hydrogen evolution activity was derived from its unique constructed metal-semiconductor Schottky junction. The loading of high-dispersed ultra-small MoO$_{3-x}$ clusters greatly facilitated its charge carrier dynamics through intimate coupling Schottky junction interface between well-dispersed metallic MoO$_{3-x}$ clusters and polymer carbon nitride semiconductor. According to the above experiment results, a possible photocatalytic mechanism in MoO$_{3-x}$-PCN-$X$ systems is proposed and shown in FIG. 7. For the Schottky junction to be constructed, electrons would transfer via the metal-semiconductor interface until Fermi equilibrium was reached. This process leaded to the bending of conduction band (FIG. 7) [41]. The electrochemical Mott-Schottky analysis (FIG. S9 in supplementary materials) indeed confirmed that the conduction band position of polymer carbon nitride descended after loading of MoO$_{3-x}$ clusters. The flat band position of MoO$_{3-x}$-PCN-2 was calculated to be $-$0.34 V while flat band position of bulk PCN was estimated to be $-$0.86 V. At the same time, the bandgap of these samples was almost unchanged (from the UV-Vis DRS outcome). Therefore, these observations revealed that Schottky junction was successfully constructed at the interface of polymer carbon nitride and metallic MoO$_{3-x}$ clusters. In this MoO$_{3-x}$-PCN-$X$ system, electrons and holes were simultaneously excited and separated under visible-light irradiation. Then, the photo-generated electron could efficiently transfer from PCN to metallic MoO$_{3-x}$ clusters via intimate Schottky junction interface, as demonstrated by the enhanced transient photocurrent response, compressed PL intensity and longer lifetime of charge carriers. Afterwards, the absorbed H$^+$ would be reduced to produce hydrogen via electrons captured by MoO$_{3-x}$ cocatalyst. While the holes on the valance band of polymer carbon nitride would be consumed by TEOA sacrificial agents. The hydrogen production rate of bulk PCN sample was quite low, emphasizing the Schottky junction between MoO$_{3-x}$ clusters and PCN played a vital role in accelerating charge dynamics and improving photocatalytic activity.

    Figure 7.  Schematic illustration of the possible photocatalytic mechanism in MoO$_{3-x}$-PCN-$X$ system.

  • Noble-metal-free MoO$_{3-x}$ clusters were successfully in situ loaded as an efficient cocatalyst over polymer carbon nitride through a simple one-pot impregnation-calcination approach. These high-dispersed ultra-small MoO$_{3-x}$ clusters significantly enhance photocatalytic activity under visible light irradiation. The visible-light hydrogen evolution rate over champion MoO$_{3-x}$-PCN-2 sample without noble metal Pt was 15.6 μmol/h, more than three times of bulk PCN sample. Such enhancement in hydrogen evolution under visible light irradiation could be ascribed to the intimate coupling interface between high-dispersed ultra-small MoO$_{3-x}$ cluster and two-dimensional PCN as well as transportation of efficient charge carriers brought by the formed Schottky junction. This work may provide a strategy to design molybdenum-based noble-metal-free cocatalysts for efficient visible light hydrogen evolution in the future.

    Supplementary materials: The in-situ one-pot impregnation-calcination synthesis of MoO$_{3-x}$-PCN-$X$ samples, TEM image of bulk PCN sample, XPS O 1s spectra of PCN-samples, AQY of MoO$_{3-x}$-PCN-2, Photocatalytic hydrogen evolution rate of different PCN samples, XRD patterns of MoO$_{3-x}$-PCN-2 sample before and after photocatalytic reaction, High-resolution XPS of Mo 3d, N 1s and C 1s spectra of MoO$_{3-x}$-PCN-2 sample before and after photocatalytic reaction, Mott-Schottky analysis for different PCN sample are available in the Supplementary materials.

  • This work was supported by the National Natural Science Foundation of China (No.21872093), the National Key Research and Development Program of China (No.2018YFB1502001), the Center of Hydrogen Science of Shanghai Jiao Tong University.

    Figure S1.  In-situ one-pot impregnation-calcination synthesis method of MoO3-x-PCN-x samples

    Figure S2.  TEM image of bulk PCN sample

    Figure S3.  XPS O 1s spectra of (a) bulk PCN (b) MoO3-x-PCN-2 samples

    Figure S4.  Photocatalytic hydrogen evolution rate of bulk PCN, MoO3-x-PCN-2 and MoO3-x-PCN-mixed samples

    Figure S5.  Photocatalytic hydrogen evolution rate of bulk PCN, MoO3-x-PCN-2 and Pt-PCN samples

    Figure S6.  Mott-Schottky analysis for (a) bulk PCN sample (b) MoO3-x-PCN-2 sample

Reference (41)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return