Nanometer-sized titania hosted inside MOF-5

Abstract

Nanoscale titania particles were synthesized inside the porous coordination polymer [Zn₄O(bdc)₃] (bdc = 1,4-benzene-dicarboxylate, MOF-5) by adsorption of titanium isopropoxide from the gas-phase and subsequent dry oxidation and annealing.

---

Nanoscale titania is a prominent quantum dot material for photovoltaics¹ and photocatalysis.² The wet-chemical synthesis of nano-TiO₂ particles with tailored morphology and size has reached a sophisticated state-of-the-art stage.^1–3 The imbedding of TiO₂ inside nanoporous (siliceous) matrices was suggested to increase the selectivity of photocatalytic reactions by caging effects.⁴ Thus, we were led to investigate the titania-loading of zeolite-like porous coordination polymers (PCPs, or so-called metal–organic frameworks, MOFs).⁵ The combination of tailored chemical function and spacial confinement of MOFs with the size, shape and surface-structure-dependent properties of TiO₂ is quite attractive. Herein we report the solvent-free two-step synthesis of nano-TiO₂ material inside [Zn₄O(bdc)₃] (bdc = 1,4-benzene-dicarboxylate; MOF-5) achieved by oxidative annealing of the precursor materials [Ti(OⁱPr)₄]_a@MOF-5.‡

The gas-phase loading of pure, fully desolvated MOF-5⁶ with various molar equivalents a of [Ti(OⁱPr)₄] as a liquid and volatile precursor for TiO₂ was performed in a sealed glass tube under static vacuum (10⁻³ mbar) at room temperature for 24 h. The resulting [TiOⁱPr₄]_a@MOF-5 (1) was then converted to [TiO₂]_x@MOF-5 (2; x < a) by thermal treatment in a dry oxygen stream (4.5 vol.% in Argon) at 220 °C for 8 h and further annealing under Ar in analogy to our previous work on [ZnO]_x@MOF-5.⁷ The measured Ti-loadings of 2 range from 4 to 12 wt.%, which correspond to 0.7 < x < 2.4 (±5%) for our series of loading experiments. A typical sample [TiO₂]_x@TiO₂ of 6.3 (±0.3) wt.% Ti, i.e.x = 1.13 (±0.06), was selected for characterization. This sample is close to a 1 : 1 molar ratio of titania and host material and is denoted as 2 from here onwards.

Fig. 1 displays the IR spectra of [Ti(OⁱPr)₄] (A), the empty MOF-5 (B), [Ti(OⁱPr)₄]_a@MOF-5 (C) and [TiO₂]_x@MOF-5 (D). The presence of physisorbed [Ti(OⁱPr)₄] for 1 is evidenced by the occurrence of the characteristic IR bands at 3364, 2969, 2930 1124, 1001 and 621 cm⁻¹. The oxidative decomposition of the adsorbed [Ti(OⁱPr)₄] at 220 °C within 8 h results in a material, with an IR spectrum that still exhibits absorption bands of residual ligand fragments. The ¹³C-MAS-NMR spectra of this as-synthesized material [TiO₂]_x@MOF-5 reveals signals attributed to hydrocarbon impurities of incompletely decomposed [Ti(OⁱPr)₄] (Fig. 2). These impurities were quantitatively removed upon annealing for two days at 250 °C under Ar as shown by IR and NMR (see Fig. 1C and D and Fig. 2B and C).

Fig. 1 IR spectra of pure liquid [Ti(OⁱPr)₄] (A), pure activated MOF-5 (B), [Ti(OⁱPr)₄]_a@MOF-5 (C) and the title material [TiO₂]_x@MOF-5 (D, with a Ti loading of 6.3 wt.%, x = 1.13, after annealing at 250 °C for 2 days). Characteristic absorption bands of [Ti(OⁱPr)₄] are marked with an asterisk (*).

Fig. 2 ¹³C-MAS-NMR spectra of [Ti(OⁱPr)₄]_a@MOF-5 (A), as synthesized [TiO₂]_x@MOF-5 (B) and annealed [TiO₂]_x@MOF-5 (C). Peaks marked with an asterisk (*) belong to the MOF-5 matrix.

Upon the titania loading, the Langmuir surface dropped from 3400 m² g⁻¹ for the empty MOF-5 to 2284 m² g⁻¹ (standard N₂-adsorption studies at 77 K) for 2. The above described oxidation and annealing procedures for conversion of 1 into 2 had no effect on the powder X-ray diffraction peak positions (pattern) related to the MOF-5 matrix (Fig. 3). However, the data show a gradual inversion of the relative intensities of the two first reflexes at 2θ = 6.9 and 9.7° as compared with the emptyMOF-5 as reference. This observation is a strong hint for inclusion of guest species inside the cavities of otherwise unchanged MOF-5.⁸ Similar effects were observed for solvent-loaded MOFs, precursor@MOF⁹ and metal@MOF¹⁰ as well as for the inclusion of guest molecules and nanoparticles inside mesoporous silica and zeolites in general. Upon prolonged annealing of 2 at 350 °C in vacuo for 2 days, additional peaks showed up which are attributed to some zinc carbonate species and hexagonal ZnO (typical fingerprint at 2θ =31.8° [100], 34.4° [002], 36.3° [101]). Even at those harsh conditions the PXRD did not provide any indication of crystalline TiO₂ phases.¹¹ Raman studies did not give any hint of rutile or anatase structures because of strong overshadowing by the PL background (see below).

Fig. 3 PXRD patterns of the samples [Ti(OⁱPr)₄]_a@MOF-5 and [TiO₂]_x@MOF-5 (annealed at different temperatures) in comparison to empty MOF-5 and anatase and rutile as the likely titania phases. Reflections assigned to zinc carbonate are marked with an arrow.

Transmission electron microscopy (TEM, Fig. 4) gave no further evidence for the inclusion of TiO₂nanoparticles, presumably because of the lack of contrast between the MOF-5 matrix and titania. However, the TEM images of 2 and the energy dispersive X-ray analysis (EDX) revealed a very homogenous material and confirmed the absence of any larger titania agglomerates outside the imaged MOF-5 specimens. The presence of TiO₂ was unequivocally confirmed by X-ray photoelectron spectroscopy (see ESI†).

Fig. 4 Transmission electron micrographs of empty MOF-5 material (A), annealed (350 °C, 1 day) [TiO₂]_x@MOF-5 (B) and as-synthesized [TiO₂]_x@MOF-5 (C).

The room-temperature UV-vis absorption and luminescence spectra of differently treated samples 2 are shown in Fig. 5. The onset of the absorption of 2 is red-shifted as compared with MOF-5, but still it is clearly well above the absorption edge of bulk titania (375 nm, 3.3 eV band gap for anatase phase). The UV-absorption of titania in a size regime of 1–2 nm, matching the MOF-5 cavities, is expected below 350 nm (3.54 eV bandgap).¹² Prolonged annealing at 250 °C for 2 days led to a further gradual red-shift of the absorption onset and by raising the temperature to 350 °C for 2 days, a structure evolves at 370–380 nm. These observations are assigned to sintering of the titania particles combined with some decomposition of the matrix and ZnO formation.¹³ The photoluminescence (PL) spectrum of 2 (annealed at 350 °C for 1 day, intact MOF-5) was recorded at an excitation wavelength of 365 nm and shows two strong emission bands at 417 and 435 nm, which are very characteristic for nano-TiO₂. The band at 417 nm is assigned to free exciton emission of TiO₂ particles of a few nm in size, whereas the peak at 435 nm is associated with surface defects.^12–14 The typical green luminescence of the (empty) MOF-5 matrix at 525 nm (Fig. 5, insert) is not observed. The Zn₄O₁₃clusters of MOF-5 are known to behave as quantum dots, while the bdc linkers act as photon antenna for energy transfer to the zinc oxo cluster, where the emission occurs.¹⁵ The PL properties of 2 are, however, dominated by the emission from the titania, possibly involving energy transfer from the host matrix to the guest titania particles.

Fig. 5 UV-Vis absorption spectra of [TiO₂]_x@MOF-5 (2) annealed at different conditions. The photoluminescence (PL) spectrum of 2 annealed at 350 °C for 1 day is shown as an inset (solid line). The PL spectrum of empty MOF-5 is shown as a dotted curve.

Although lacking the direct microscopic (TEM) evidence of imbedded TiO₂ particles, our data indicate that MOF-5 can be doped with titania species exhibiting optical properties comparable to surfactant-stabilized free-standing TiO₂nanoparticles.^16–18 Thus, PCPs¹⁹ and MOFs may be identified as attractive solid-state host matrices for nano-TiO₂ and other nanoscale oxide materials with properties beyond the established silica-based porous matrices.²⁰ A more detailed study of the microstructure of the titania particles and their distribution within the host matrix as well as the characterization of the photophysical interaction between the particles and the framework are in progress. In addition, we are aiming at the preparation of Au/TiO₂@MOFs with possible applications in CO-oxidation reactions.²¹

The authors acknowledge support within the Research Centre 558 “Metal Substrate Interactions in Heterogeneous Catalysis” of the German Research Foundation (DFG). The authors wish to thank Todor Hikov for very valuable help with UV-Vis and PL measurements. M. M. is grateful to the Ruhr-University Research School [DFG GSC 98/1] for supporting her doctoral thesis and to the Evangelische Studienwerk e.V. Villigst for a stipend.

Notes and references

1. M. Adachi, J. Jiu and S. Isoda, Curr. Nanosci., 2007, 3, 285–295.
2. (a) U. I. Gaya and A. H. Abdullah, J. Photochem. Photobiol., C, 2008, 9, 1–12; (b) M. Kitano, K. Tsujimaru and M. Anpo, Top. Catal., 2008, 49, 4–17.
3. (a) J. G. Lu, P. C. Chang and Z. Y. Fan, Mater. Sci. Eng., R, 2006, 52, 49–91; (b) B. L. Zhu, K. O. Li, S. M. Zhang, S. H. Wu and W. P. Huang, Prog. Chem., 2008, 20, 48–53.
4. (a) C. Aprile, A. Corma and H. Garcia, Phys. Chem. Chem. Phys., 2008, 10, 769–783; (b) D. G. Shchukin and D. V. Sviridov, J. Photochem. Photobiol., C, 2006, 7, 23–39.
5. J. L. C. Rowsell and O. M. Yaghi, Microporous Mesoporous Mater., 2004, 73, 3–14.
6. H. Li, M. Eddaoudi, M. O’Keeffe and M. Yaghi, Nature, 1999, 402, 276–279.
7. M. Müller, S. Hermes, K. Kähler, M. W. E. van den Berg, M. Muhler and R. A. Fischer, Chem. Mater., 2008, 20, 4576–4587.
8. J. Hafizovic, M. Bjørgen, U. Olsbye, P. D. C. Dietzel, S. Bordiga, C. Prestipino, C. Lamberti and K. P. Lillerud, J. Am. Chem. Soc., 2007, 129, 3612–3620.
9. S. Hermes, F. Schroeder, S. Amirjalayer, R. Schmid and R. A. Fischer, J. Mater. Chem., 2006, 16, 2464–2472.
10. (a) S. Hermes, M.-K. Schröter, R. Schmid, L. Khodeir, M. Muhler, A. Tissler, R. W. Fischer and R. A. Fischer, Angew. Chem., Int. Ed., 2005, 44, 6237–6241; (b) F. Schröder, D. Esken, M. Cokoja, M. W. E. van den Berg, O. I. Lebedev, G. Van Tendeloo, B. Walaszek, G. Buntkowsky, H.-H. Limbach, B. Chaudret and R. A. Fischer, J. Am. Chem. Soc., 2008, 130, 6119–6130.
11. (a) K. Fukuda, Y. Ebina, T. Shibata, T. Aizawa, I. Nakai and T. Sasaki, J. Am. Chem. Soc., 2007, 129, 202–209; (b) H. Zhang and J. F. Banfield, Chem. Mater., 2002, 14, 4145–4154.
12. L. Kavan, T. Stoto, M. Graetzel, D. Fitzmaurice and V. Shklover, J. Phys. Chem., 1993, 97, 9493–9498.
13. W. Li, C. Ni, H. Lin, C. P. Huang and S. I. Shah, J. Appl. Phys., 2004, 96, 6663–6668.
14. N. Serpone, D. Lawless and R. Khairutdinov, J. Phys. Chem., 1995, 99, 16646–16654.
15. S. Bordiga, C. Lamberti, G. Ricchiardi, L. Regli, F. Bonino, A. Damin, K.-P. Lillerud, M. Bjorgen and A. Zecchina, Chem. Commun., 2004, 2300–2301.
16. N. B. Lihitkar, M. K. Abyaneh, V. Samuel, R. Pasricha, S. W. Gosavi and S. K. Kulkarni, J. Colloid Interface Sci., 2007, 314, 310–316.
17. H. Parala, A. Devi, R. Bhakta and R. A. Fischer, J. Mater. Chem., 2002, 1625–1627.
18. T. Hikov, M.-K. Schröter, L. Khodeir, A. Chemseddine, M. Muhler and R. A. Fischer, Phys. Chem. Chem. Phys., 2006, 13, 1550–1555.
19. T. Uemura, D. Hiramatsu, K. Yoshida, S. Isoda and S. Kitagawa, J. Am. Chem. Soc., 2008, 130, 9216–9217.
20. S. Zheng, L. Gao, Q. Zhang, W. Zhang and J. Guo, J. Mater. Chem., 2001, 11, 578–583.
21. J. P. Gabaldon, M. Bore and A. K. Datye, Top. Catal., 2007, 44, 253–262.

---

Footnotes

† Electronic supplementary information (ESI) available: Fig. S1–S4. See DOI: 10.1039/b814241f

‡ All manipulations were carried out under an inert atmosphere. MOF-5 synthesis was performed according to the literature.^6,9,10Gas-phase loading: samples of 100–250 mg (0.130 – 0.325 mmol) of MOF-5 were exposed to various molar equivalents a of [Ti(OⁱPr)₄] at 25 °C at 10⁻³ mbar (static conditions) for several hours allowing quantitative adsorption. The obtained samples [Ti(OⁱPr)₄]_a@MOF-5 (1) were placed in a glass tube and treated with oxygen (4.5 vol.% in Ar, 1 sccm s⁻¹) at 220 °C for 8 h. After oxygenation, the samples were annealed in dynamic vacuo as described in the text. The obtained yellowish samples [TiO₂]_x@MOF-5 (2) were stored under Ar (x < a, because of some desorption of [Ti(OⁱPr)₄] during oxidation). Elemental analysis/atomic absorption spectroscopy (AAS): for determination of the titania and zinc contents an AAS apparatus by Vario of type 6 (1998) was employed; C and H analyses were carried out using a Vario CHNSO EL (1998) instrument. N₂-Adsorption: nitrogen-adsorption experiments were carried out with a Quantachrome Autosorb-1 MP. The specific surface areas were calculated by fitting the measured type I isotherms to the Langmuir surface model in a pressure range of P/P₀ = 0.1–0.3 at T = 77.36 K. Transmission electron microscopy (TEM): samples were prepared under Ar (glove box) by placing a droplet of a suspension onto a carbon-coated Cu grid which were then transferred into a Hitachi H-8100, operating at 200 kV, equipped with a LaB6-filament). Powder X-ray diffraction (PXRD): PXRD diffractogramms were recorded using a D8-Advance Bruker AXS diffractometer with Cu Kα radiation (λ = 1.5418 Å) in θ–2θ geometry and with a position-sensitive detector. The experimental setup was in the capillary mode (samples filled into capillaries in the glovebox and sealed). Photoluminescence: the optical properties were investigated by room-temperature photoluminescence measurements using a Xe-lamp (PL excitation wavelength: 365 nm; S.A. Instruments Fluoromax-2). Infrared spectroscopy: FT-IR-spectra were measured with an ATR setup using a Bruker Alpha FT-IR spectrometer. UV-Vis: spectra were measured with a Perkin-Elmer Lambda 9 UV/vis/NIR spectrometer in reflection mode. Solid-state NMR: spectra were measured with a Bruker DSX 400 MHz spectrometer under MAS conditions in 2.5 mm ZrO₂ rotors with a sample volume of 12 μL. The rotation frequency was 20 kHz. For the proton NMR measurements a ZG4PM pulse program was used. The carbon NMR measurements were carried out with the pulse program CP4C (cross polarization) and referenced to adamantine at 38.56 ppm.

---