Preparation of a Fe₂O₃/MIL-53(Fe) composite by partial thermal decomposition of MIL-53(Fe) nanorods and their photocatalytic activity

Abstract

A Fe₂O₃/MIL-53(Fe) composite (89% MIL-53(Fe), 11% Fe₂O₃) and Fe₂O₃ were prepared by partial and complete thermal decomposition of MIL-53(Fe). The catalysts were characterized by XRD, BET, FTIR, UV-vis absorbance spectroscopy, SEM, EDS and TG analysis. The derived Fe₂O₃/MIL-53(Fe) composite was found to be effective for visible light photocatalytic degradation.

---

Metal–organic Frameworks are a category of porous and crystalline materials consisting of infinite lattices made up of inorganic secondary building units (SBU) and organic linkers. Nowadays MOFs are extensively used for gas storage, separation, carbon dioxide capture, drug delivery and molecular sensing due to their high surface area and well defined porous structure.^1–5 Beyond these promising applications, MOFs can also be used as catalysts and photocatalysts due to well organized open channels, uniformly distributed metal ions, easy recovery and structure designability.^6–8 MOFs have been recognized as semiconductors based on their optical transitions, electrochemical and photochemical properties. Most MOFs synthesized by using terepthalic acid as a ligand generally show a good response in the UV region, so to extend their use in the visible region is a challenge.⁹ Zhang et al. synthesized Fe₃O₄/MIL-53(Fe) using solvothermal method. In this case, pure MIL-53(Fe) shows absorption edge around 470 nm whereas absorption edge of the composite is shifted to higher values due to presence of Fe₃O₄ nanoparticles.¹⁰ Sha et al. prepared a Zirconium based MOF (UiO-66) incorporated with bismuth tungstate by a simple hydrothermal method. UiO-66 has an absorption edge around 350 nm but absorption edge of each composites shift to 450 nm, which is the absorption edge of pure Bi₂WO₆.¹¹ Mukoyoshi et al. recently prepared Ni nanoparticles incorporated inside Ni-MOF-74 by partial thermal decomposition of Ni-MOF-74.¹² Till date so many efforts were made to improve the photocatalytic properties of MOFs by incorporating different materials but very few efforts were made to prepare a composite of MOFs and metal oxide from a single MOFs template. In this work, composite of Fe₂O₃/MIL-53(Fe) was prepared by partial thermal decomposition of MIL-53(Fe) keeping in mind that each component will compensate the limitations of its counterpart i.e. light absorption properties of MIL-53(Fe) can be improved by using Fe₂O₃ and high recombination rate of electrons and holes of Fe₂O₃ can be backed by the high degree of delocalization of MIL-53(Fe). Methylene blue was used as a representative organic pollutant to study the photocatalytic activity of MIL-53(Fe), FM-1 and FM-2. Photodegradation results of the samples were compared with TiO₂.

Thermal stability of MIL-53(Fe) was studied by means of TG analysis. The samples in this study were synthesized with respect to TG curve (ESI†). By looking at the multistep decomposition curve (Fig. S1, ESI†) one can say that the MOFs is stable up to 350 °C and the loss in weight of the sample is due to free DMF occupied inside the pores. A sharp decrease in weight was noted after 350 °C and the loss was continued till 500 °C, because of the decomposition of organic ligands. A low rate of decomposition was observed between 500 °C to 750 °C followed by a sharp decrease in weight due to complete decomposition of MIL-53(Fe). The low rate of decomposition in this case may be due to the formation of Fe₂O₃ particles surrounding MIL-53(Fe) particles forming a barrier to convective heat transfer. FM-1 contains 89% MIL-53(Fe) and 11% Fe₂O₃ and FM-2 contains about 100% Fe₂O₃ which is calculated by using TG analysis of MIL-53(Fe) (ESI†).

The XRD patterns of MIL-53(Fe), FM-1 and FM-2 are shown in Fig. 1a. The well-defined peaks of MIL-53(Fe) reveal that it is highly crystalline in nature. XRD patterns of FM-2 is in clear match with JCPDS card no-840310, in which well-defined peaks at 24.30°, 33.40°, 35.84°, 41.11°, 49.80°, 54.47°, 62.86° and 64.41° corresponds to (012), (104), (110), (113), (024), (116), (214) and (300) reflections of orthorhombic Fe₂O₃. All major peaks of calculated XRD pattern of CCDC-690314 are in perfectly match with MIL-53(Fe), confirming the formation of MIL-53(Fe).^13,14 So, a conclusion can be drawn that FM-2 mainly consists of Fe₂O₃ as no major peaks of MIL-53(Fe) found in XRD pattern of FM-2. Major peaks of FM-1 clearly match with peaks of FM-2 and MIL-53(Fe) indicating absence of impurities in the composite. However, peak at 13.5° of MIL-53(Fe) shifted to 12.6° in case of FM-1 suggesting volume expansion of lattice due to formation of internal strain at higher temperature during composite formation. FTIR spectroscopic studies were performed for MIL-53(Fe), FM-1 and FM-2 in the wave range of 500–2000 cm⁻¹ (Fig. 1b). In case of MIL-53(Fe) and FM-1, two strong vibrational bands around 1539 and 1387 cm⁻¹ confirms the presence of dicarboxylate linker. Two strong bands at 748 cm⁻¹ and 540 cm⁻¹ are attributed to Ar–C–H and Fe–O vibrations respectively.^10,15–17 So no conclusive evidence of decomposition of organic linker is observed for FM-1. For FM-2, band vanishes around 1550 cm⁻¹ and very weak bands are observed at 1400 cm⁻¹ and 750 cm⁻¹, indicating decomposition of MIL-53(Fe). A strong peak is observed at 558 cm⁻¹ for Fe–O vibration which is initially presented at 538 cm⁻¹ in case of MIL-53(Fe). The unnecessary bands in the range of 1100 cm⁻¹ to 1200 cm⁻¹ are due to presence of KBr, which was used as binder for FTIR analysis. High resolution transmission electron microscopic (HRTEM) analysis was performed and the images shown in Fig. 2a confirms that MIL-53(Fe) nanorods are average of 110 nm diameter. Corresponding SAED pattern shows d spacing of 6.4 Å which is well satisfied with XRD peak at 13.5° (Fig. 2b). Morphology of FM-1 has changed due to composite formation and rods are creating a spiral network with Fe₂O₃ particle (Fig. 2c). Rods are bended in FM-1 which is obvious for the presence of tensile stretch in lattice. Corresponding SAED pattern also depicts crystallinity and d spacing of 7.0 Å which is well satisfied with the XRD peak at 12.6° (Fig. 2d). EDS results also confirm that FM-1 has higher Fe and O% than MIL-53(Fe), which is due to formation of little amount of Fe₂O₃. The higher Fe₂O₃% of FM-2 is also reflected in the results of EDS, indicating very high Fe and O% when compared to MIL-53(Fe) (Fig. S2, ESI†).

Fig. 1 (a) XRD patterns of calculated MIL-53(Fe), prepared MIL-53(Fe), FM-1 and FM-2; (b) FTIR spectra of (i) MIL-53(Fe), (ii) FM-1 and (iii) FM-2.

Fig. 2 (a) HRTEM micrograph of MIL-53(Fe) and (b) corresponding SAED pattern, (c) HRTEM micrograph of FM-1 and (d) corresponding SAED pattern.

SEM images of MIL-53(Fe) (Fig. 3a) reveal that crystals of MIL-53(Fe) are uniformly well shaped hexagonal nanorods having crystal length around 500 nm. However, in case of FM-1 (Fig. 3b and c) distorted nanorods in the shape of leaf are observed along with Fe₂O₃ crystals. Due to sudden heating, distortion in the shape of MIL-53(Fe) was observed which leads to shift in XRD peak. Very packed Fe₂O₃ particles of crystal size around 20 nm are observed in case of FM-2 (Fig. 3d). The porous properties and pore structure of MIL-53(Fe), FM-1 and FM-2 were analysed by measuring N₂ adsorption–desorption isotherms using BET apparatus. The BET surface area and total pore volume of MIL-53(Fe), FM-1 and FM-2 are listed in Table 1 (ESI†). It is clear from the results that MIL-53(Fe) has highest surface it may be due to absence of closely packed Fe₂O₃. As the temperature increases up to 350 °C, the surface area and pore volume increases due to removal of foreign DMF molecules creating new pores and widening of the existing pores. The packed Fe₂O₃ (Fig. 3d) image suggests that formation of Fe₂O₃ leads to decrease in surface area and pore volume, whereas the pore volume for FM-1 is pumped due to heating at higher temperature and may be pulled back a little due to formation of Fe₂O₃ which leads to the highest pore volume amongst all. From N₂ adsorption desorption isotherms it is revealed that both MIL-53(Fe) and FM-1 (Fig. S3, ESI†) displays an intermediate mode between type I and type IV isotherm, which indicates presence of both micro and mesoporous materials.^18,19 By looking at the paths of N₂ adsorption and desorption isotherms, it can be said FM-1 has pores that are nearly cylindrical in nature.

Fig. 3 SEM images of (a) MIL-53(Fe), (b and c) FM-1 and (d) FM-2.

UV-vis absorbance spectrums of MIL-53(Fe), FM-1 and FM-2 (Fig. 4a) were used to analyse their optical properties. MIL-53(Fe) displays a sharp peak at 241 nm with an absorption edge around 473 nm (Fig. 4b). Peak at 241 nm is assigned to ligand-to-metal transfer, indicating the presence of Fe–O bonds (oxygen from linker). Absorbance in the region 300–500 nm is due to transition of Fe³⁺ in MIL-53(Fe) [⁶A_1g → ⁴A_1g + ⁴E_g(G)].^10,20,21 FM-2, mainly consisting of Fe₂O₃ does not have any sharp peak whereas, FM-1 exhibits a sharp peak at 241 nm and absorption edge is around 658 nm (Fig. 4c). Based on E_g = 1240/λ (E_g = band gap energy (eV) and λ = wavelength at absorption edge of the spectra (nm)), the optical band gap energy of MIL-53(Fe) and FM-1 are 2.62 and 1.88 eV respectively.^22–24

Fig. 4 (a) UV-vis absorbance spectra of MIL-53(Fe), FM-1 and FM-2 (b) absorbance edge of MIL-53(Fe) (c) absorbance edge of FM-1.

The photocurrent response experiment was performed to demonstrate the charge separation process. Catalysts (0.05 g) were dispersed into ethylene glycol by 1 h ultrasonication. A 300 W Xe light outfitted with a cutoff channel (λ > 350 nm) was utilized as the illumination source. Plating voltage is 5 V and time interval is 1 min. Photocurrent was measured on a CHI 643B electrochemical workstation using 1.0 M Na₂SO₄ solution by two compartment three-electrode framework having Ag/AgCl (3 M NaCl) as reference electrode, Pt wire as auxiliary electrodes and fluorine doped tin oxide (FTO) glass coated with as prepared catalyst as working electrode. Analysis depicts excellent photogenerated electron hole separation ability of FM-1 compare to both pure Fe₂O₃ and pure MIL-53 (Fe) by the plot of transient photocurrent density plot (Fig. 5). This could be explained by the fact that Fe₂O₃ and MIL-53(Fe) create a heterojunction which may enhance the electron hole separation ability.

Fig. 5 Photocurrent response experiment.

Photocatalytic response were enlightened by dissolving 1 mM solution of scavengers including EDTA, AgNO₃ and tert-butyl alcohol (TBA) which could trap photo-induced holes (H⁺), photo-induced electrons and ˙OH radical respectively. It is observed that degradation efficiency decreases with the addition of tert-butyl alcohol (TBA) and EDTA and increases with the addition of AgNO₃ (Fig. S5, ESI†) confirming the key role of ˙OH and H⁺ in this reaction. A possible charge transfer mechanism for Fe₂O₃/MIL-53(Fe) heterostructure photocatalytic system is illustrated in Fig. 6. The potentials at VB and CB and bandgap energies for Fe₂O₃ and MIL-53(Fe) are available in the earlier literatures.^25,26 At the initial stage, the light is absorbed by the valence band (VB) of MIL-53(Fe) and the photogenerated electrons are excited to the conduction band (CB) with simultaneous hole generation in the VB. First of all, the photogenerated electrons from CB of MIL-53(Fe) are migrated to CB of Fe₂O₃ due to difference of band gap resulting in, an electron-rich locale at the CB of Fe₂O₃. Essentially, the accumulation of photogenerated holes in the VB of Fe₂O₃ transfer to the valence band (VB) of MIL-53(Fe) creating a hole-rich locale. Subsequently, the combination of Fe₂O₃ and MIL-53(Fe) act as powerful photocatalytic oxidant and reductant respectively by composition of two narrow-bandgap semiconductors.

Fig. 6 Charge transfer mechanism of Fe₂O₃/MIL-53(Fe) heterojunction.

Photocatalytic activity of each catalyst were investigated by degradation of methylene blue in presence of both UV and visible light irradiation. A hollow cylindrical glass reactor of volume 250 mL with a light source (UV: 8W Philips (Poland); Visible: 8W Omex (India)) mounted inside the inner diameter of the cylinder. 40 mg of catalyst was thoroughly mixed with 200 mL MB (10 mg L⁻¹, pH = 7) solution inside the annular space of the reactor in dark condition for 30 min to reach adsorption equilibrium (Fig. S6, ESI†). For vigorous mixing and to maintain constant dissolved oxygen, air is bubbled from bottom of the reactor at 0.1 LPM throughout the reaction time.

It is observed that the degradation efficiency is maximum for FM-1 (Fig. 7a and b) which is about 71% in UV light and 52% in visible light. For MIL-53(Fe) the degradation efficiency is around 48% in UV light and around 30% in visible light. But in case of FM-2, the degradation efficiencies are very low in both UV and visible light. Hence it can be said that due to band gap narrowing and enhanced optical absorption, FM-1 shows better photocatalytic activity than MIL-53(Fe). But for FM-2, the degradation percentage is low, as in Fe₂O₃ recombination rate of electrons and holes are very high.^27,28 The degradation of MB solution can be described by a first-order reaction with equation ln(C₀/C) = kt, where C₀ corresponds to initial concentration after irradiation, C is concentration of MB at a particular time and k is the apparent first-order rate constant. From Fig. S7a and b (ESI†) k values for FM-1, MIL-53(Fe) and FM-2 in visible light were found to be 0.0040, 0.0017 and 0.0006 min⁻¹ respectively, whereas for UV light 0.0080, 0.0040 and 0.0010 min⁻¹ respectively. It is observed that FM-1 have a high rate constant which is two times of MIL-53(Fe), eight times of FM-2 and around three times of TiO₂ (LOBA Chemie, India, 99% A.R.) in UV light. The photocatalytic degradation of MB on P-25 TiO₂ was reported²⁹ as 4.5% degradation in 120 min of irradiation by visible light for a catalyst dose of 0.5 g L⁻¹ with an average rate constant of 0.00086 min⁻¹ which is much lower than the present data obtained by FM-1 (52% in 120 min of UV irradiation for a catalyst dose of 0.2 g L⁻¹ and rate constant of 0.004 min⁻¹).

Fig. 7 Photocatalytic degradation of MB in (a) UV irradiation (b) visible irradiation.

Optimum catalyst dose plays an important role in every catalysis reaction. To find the optimum dose of FM-1, 200 mL of MB (10 mg L⁻¹, pH = 7) was degraded by varying catalyst dose from 10 mg to 60 mg (Fig. S8, ESI†). As the catalyst dose is increased from 20 to 40 mg, degradation efficiency of MB increases from 50 to 71%. Further increase in catalyst does not show a significant increase in degradation of MB. This variation may be due to the formation of bulky solution which results in obstacle of light to reach active sights. So, 40 mg has been considered as an optimum dose of FM-1 for degradation of MB dye for each set of experiments. The effect of solution pH on MB degradation was studied for FM-1 by adjusting the pH in the range of 3–11 as shown in Fig. S9 (ESI†). The degradation of MB increases from 48% to 75% as the solution pH is varied from 3 to 11. FM-1 shows less degradation in acidic medium due to the electrostatic repulsion between cationic MB and positively charged catalyst surface.³⁰ In basic medium, the effect of positive charge on catalyst surface is neutralized and further making it negative charge as a result interaction with MB increases resulting higher catalytic degradation.^6,31,32 In basic medium, the rate of formation hydroxyl free radical (h_VB⁺ + OH⁻ → OH˙) increases as the interaction between excessive OH⁻ and holes on catalyst surface increases, therefore maximum degradation was observed at pH 11. Up to some extent, industrial application of catalysts depends on stability of the catalysts. Stability of FM-1 was determined by analysing FTIR spectrums of FM-1, both before and after use (Fig. S10, ESI†). No variation in peaks indicates that FM-1 is highly stable in nature.

The MIL-53(Fe)/Fe₂O₃ heterojunction (FM-1) prepared by partial thermal decomposition of MIL-53(Fe) shows very high photocatalytic degradation efficiency for methylene blue in both UV and visible light compared to other catalysts. This is due to band gap narrowing and enhanced photogenerated electron–hole separation capability of FM-1 which is confirmed by the transient photocurrent density plot. It was observed that the optical absorption edge was enhanced from 473 nm for MIL-53(Fe) to 658 nm for MIL-53(Fe)/Fe₂O₃ due to formation of heterojunction. The high current density obtained for FM-1 in the photocurrent response experiment also confirms the excellent photogenerated electron hole separation ability of FM-1 compare to both pure Fe₂O₃ and pure MIL-53(Fe). Optical absorption properties and electronic construction of MIL 53(Fe) has been changed through the electronic interaction with lower band gap material Fe₂O₃. The rate constant for FM-1 was higher in comparison to other catalysts which is 0.0080 min⁻¹ in UV light especially when compared to conventional photocatalyst like TiO₂ having rate constant of 0.0030 min⁻¹. The effect of various operating parameters such as solution pH, catalyst doses have a great influence on the rate of photo-reactions. It was observed that, the rate of photodegradation of MB increases with pH because of better interaction of MB and catalyst sites.

Notes and references

1. J. D. Collins and H.-C. Zhou, J. Mater. Chem., 2007, 17, 3154.
2. X. Duan, R. Song, J. Yu, H. Wang, Y. Cui, B. Chen and G. Qian, RSC Adv., 2014, 4, 36419.
3. S. Seth, G. Savitha and J. N. Moorthy, Inorg. Chem., 2015, 54, 6829.
4. D. Fujita and M. Fujita, ACS Cent. Sci., 2015, 1, 352.
5. S.-S. Zhao, J. Yang, Y.-Y. Liu and J.-F. Ma, Inorg. Chem., 2016, 55, 2261.
6. H.-P. Jing, C.-C. Wang, Y.-W. Zhang, P. Wang and R. Li, RSC Adv., 2014, 4, 54454.
7. J. Liu, L. Chen, H. Cui, J. Zhang, L. Zhang and C.-Y. Su, Chem. Soc. Rev., 2014, 43, 6011.
8. J. Park, J.-R. Li, Y.-P. Chen, J. Yu, A. A. Yakovenko, Z. U. Wang, L.-B. Sun, P. B. Balbuena and H.-C. Zhou, Chem. Commun., 2012, 48, 9995.
9. M. D. Hardi, C. Serre, T. Frot, L. Rozes, G. Mavrin, C. Sanchez and G. Ferey, J. Am. Chem. Soc., 2009, 131, 10857.
10. C. Zhang, L. Ai and J. Jiang, J. Mater. Chem. A, 2015, 3, 3074.
11. Z. Sha, J. Sun, H. S. O. Chan, S. Jaenicke and J. Wu, RSC Adv., 2014, 4, 64977.
12. M. Mukoyoshi, H. Kobayashi, K. Kusada, M. Hayashi, T. Yamada, M. Maesato, J. M. Taylor, Y. Kubota, K. Kato, M. Takata, T. Yamamoto, S. Matsumura and H. Kitagawa, Chem. Commun., 2015, 51, 12463.
13. T. Devic, P. Horcajada, C. Serre, F. Salles, G. Maurin, B. Moulin, D. Heurtaux, G. Clet, A. Vimont, J. M. Greneche, B. L. Ouay, F. Moreau, E. Magnier, Y. Filinchuk, J. Marrot, J.-C. Lavalley, M. Daturi and G. Ferey, J. Am. Chem. Soc., 2010, 132, 1127.
14. F. Millange, N. Guillou, R. I. Walton, J.-M. Greneche, I. Margiolaki and G. Ferey, Chem. Commun., 2008, 4732.
15. R. Liang, F. Jing, L. Shen, N. Qin and L. Wu, J. Hazard. Mater., 2015, 287, 364.
16. L. Ai, L. Li, C. Zhang, J. Fu and J. Jiang, Chem.–Eur. J., 2013, 19, 15105.
17. W. Dong, X. Liu, W. Shi and Y. Huang, RSC Adv., 2015, 5, 17451.
18. L. Peng, J. Zhang, J. Li, B. Han, Z. Xue and G. Yang, Chem. Commun., 2012, 48, 8688.
19. T. A. Vu, G. H. Le, C. D. Dao, L. Q. Dang, K. T. Nguyen, Q. K. Nguyen, P. T. Dang, H. T. K. Tran, Q. T. Duong, T. V. Nguyena and G. D. Lee, RSC Adv., 2015, 5, 5261.
20. G.-T. Vuong, M.-H. Pham and T.-O. Do, Dalton Trans., 2013, 42, 550.
21. G.-T. Vuong, M.-H. Pham and T.-O. Do, CrystEngComm, 2013, 15, 9694.
22. J. He, J. Wang, Y. Chen, J. Zhang, D. Duan, Y. Wang and Z. Yan, Chem. Commun., 2014, 50, 7063.
23. C.-H. Li, M.-C. Li, S.-P. Liu, A. C. Jamison, D. Lee, T. R. Lee and T.-C. Lee, ACS Appl. Mater. Interfaces, 2016, 8, 9152.
24. H. Wang, X. Yuan, H. Wang, X. Chen, Z. Wu, L. Jiang, W. Xiong, Y. Zhang and G. Zeng, RSC Adv., 2015, 5, 95643.
25. R. Liang, F. Jing, L. Shen, N. Qin and L. Wu, J. Hazard. Mater., 2015, 287, 364–372.
26. T. K. Townsend, E. M. Sabio, N. D. Browning and F. E. Osterloh, Energy Environ. Sci., 2011, 4, 4270–4275.
27. J. Theerthagiri, R. A. Senthil, A. Priya, J. Madhavan, R. J. V. Michael and M. Ashokkumar, RSC Adv., 2014, 4, 38222.
28. J. Liu, S. Yang, W. Wu, Q. Tian, S. Cui, Z. Dai, F. Ren, X. Xiao and C. Jiang, ACS Sustainable Chem. Eng., 2015, 3, 2975.
29. Y. Zhang, G. Li, H. Lu, Q. Lv and Z. Sun, RSC Adv., 2014, 4, 7594.
30. R. Liang, L. Shen, F. Jing, N. Qin and L. Wu, ACS Appl. Mater. Interfaces, 2015, 7, 9507.
31. R. J. Tayade, T. S. Natarajan and H. C. Bajaj, Ind. Eng. Chem. Res., 2009, 48, 10262.
32. L. Xionga, Y. Yang, J. Mai, W. Suna, C. Zhang, D. Wei, Q. Chen and J. Ni, Chem. Eng. J., 2010, 156, 313.

---

Footnote

† Electronic supplementary information (ESI) available: Synthesis, characterization, composition calculation, Fig. S1–S10 and Table S1. See DOI: 10.1039/c6ra15792k

---