1 Introduction

Lanthanide ions, especially trivalent terbium and europium, have recently attracted considerable interest in the numerous fields such as full-color display devices, less-harmful labelling reagents, microscopy and luminescent probes, because of their unique line-like emission bands, substantial stokes shifts, very long luminescence lifetimes and so on [1–4]. But direct photoexcitation of lanthanide ions disfavored due to the spin and parity forbidden nature of the f–f transitions. Weissman et al. first observed that the use of the organic ligands β-diketone in europium complexes increased the luminescence intensity from the lanthanide ion when such complexes were irradiated with ultraviolet lights. In these systems, the organic ligands act as photosensitizers to absorb and transfer the energy of light to excite the lanthanide ions via an energy-transfer process, which is defined as antenna effect [5, 6]. In this area, most of the previous studies concern the sol–gel approach to synthesize a kind of organic–inorganic materials, which allows mechanically mixing of inorganic and organic components at the predicated scale and under mild temperature processing conditions [7–10]. However, since there are only interactions of hydrogen bonds or van der Waals forces between organic and inorganic parts, this kind of materials presents two separate phases leading to leaching of the complexes and obtaining in-homogenous systems [11–14]. Thus, some work focuses on grafting covalently the organic ligands to the inorganic backbone. The materials were mainly developed for silicate systems so far. For instance, Franville et al. synthesized a series of hybrid materials with different structures in which lanthanide complex luminescent centers were bonded with silica-based matrix through Si–C linkage by hydrolysis and condensation reactions. They found that the complexes containing europium(III) ions in the silica hybrid matrix had a much higher thermal stability than the complex with the organic part alone [15]. The enlargement of the chemistry to metal oxides other than silica would allow interesting new options for the development of inorganic–organic hybrid materials. Aimed at preparing non-silicate hybrid systems such as titania and aluminum hybrid networks, some approaches have been tried. Li et al. took advantage of modification of the titanium alkoxide by the ligand isonicotinic acid to develop new strategies to place lanthanide complex luminescent molecules in desired inorganic titanium framework. The ligands can both react with titanium alkoxide to form an amorphous Ti–O network and act as an antenna to absorb and transfer energy to the lanthanide ions [16]. Cuan et al. prepared alumina and titania xerogels encapsulated with high luminescent lanthanide polyoxometalates via a sol–gel process. These hybrid materials possess favorable luminescent performances, and it is discovered that hybrids of titania gels can be applied in close white-light integration [17].

On the basis of the former work, the key procedure to construct molecular-based materials is to design functional bridge molecule, which both coordinates lanthanide ions and constitutes the covalent organic–inorganic network by sol–gel processing [16–19]. In this work, 3,4-bis(pyridin-4-ylmethoxy)benzoic acid is selected to construct the linkage between inorganic matrix and rare earth ions. The carboxylic acid group of it can react with metallic alkoxide to moderate the reactivity toward hydrolysis and condensation, while the heterocyclic group can coordinate with rare earth ions as well as sensitize the luminescence of them. And the novel Tb3+ (Eu3+) luminescent hybrid materials having titania-/alumina-based host were prepared by 3,4-bis(pyridin-4-ylmethoxy)benzoic acid and titanium isopropylate/aluminum isopropylate through the sol–gel method. In addition, structural characterization and detail studies of photophysical properties of the resulting hybrid materials were thoroughly investigated.

2 Experimental

2.1 Materials

Two lanthanide(III) chloride LnCl3·6H2O (Ln = Tb3+, Eu3+) were prepared by the corresponding oxides (Tb4O7, Eu2O3) in the laboratory. Titanium isopropylate, aluminum isopropylate, 3,4-dihydroxybenzoic acid and 4-(chloromethyl)pyridine were from Aladdin Industrial Corporation. Concentrated sulfuric acid (98.0 %), hydrochloric acid (37 %), sodium hydroxide, N,N-dimethylformamide (DMF) and ethanol were purchased from Beijing Chemical Company Limited.

2.2 Synthesis of 3,4-bis(pyridin-4-ylmethoxy)benzoic acid

Four milliliters of concentrated sulfuric acid was added into 200 ml alcoholic solution of 50 mmol 3,4-dihydroxybenzoic acid, and the solution was mixed and refluxed for 6–7 h. The excess ethanol was removed under vacuum, and then, the reaction mixture was poured into ice cold water. The resulting precipitate was filtered off, washed with deionized water and dried. To 10 mmol solid product, 20 mmol 4-pycolyl chloride hydrochloride and 20 mmol sodium hydroxide and 50 ml DMF were added. The mixture was heated at 100 °C for 8 h and then acidified with dilute HCl, and the resulting precipitate was filtered, washed, dried and recrystallized from ethanol. A light brown precipitate was obtained, and the yield was about 65 %. 1H NMR (300 MHz, DMSO): δ (ppm) 8.57 (m, 4H), 7.55 (t, J = 5.7 Hz, 1H), 7.46 (m, 5H), 7.13 (d, J = 8.4 Hz, 1H), 5.25 (s, 4H). Elemental analyses calcd (%) for C19H16N2O4 (336.35): C, 67.84; H, 4.81; N, 8.33. Found C, 67.87; H, 4.79; N, 8.32. MS (EI) m/e = 336 (M+).

2.3 Synthesis of hybrids Ti–PYB–Ln, Al–PYB–Ln and Ti–Al–PYB–Ln (PYB = 3,4-bis(pyridin-4-ylmethoxy)benzoic acid, Ln3+ = Tb3+, Eu3+)

A typical synthesis procedure of Ti–PYB–Ln was as follows: 4 mmol 3,4-bis(pyridin-4-ylmethoxy)benzoic acid was dissolved in 40 ml DMF, and then, 4 mmol titanium isopropylate was slowly added. After refluxing and stirring for 1 h, 1 mmol LnCl3·6H2O dissolved in 5 ml DMF was added dropwise into the clear solution and refluxed for 3 h. Finally, 0.2 ml deionized water was dropped into the above solution in order to promote the hydrolysis of titanium isopropylate to form inorganic/organic hydrogels constructed by network. Then, the final materials Ti–PYB–Ln were obtained by centrifugation and washed with ethanol for three times and dried for 24 h at 100 °C. All the other hybrids were synthesized using the same procedure except that Ti–Al–PYB–Ln was prepared using 2 mmol titanium isopropylate and 2 mmol aluminum isopropylate, and Al–PYB–Ln was prepared using 4 mmol aluminum isopropylate.

For Ti–PYB–Eu (%) Anal. Calcd.: C, 44.20; H, 3.03; N, 5.43. Found: C, 44.20; H, 3.00; N, 5.33.

For Ti–PYB–Tb (%) Anal. Calcd.: C, 44.05; H, 3.02; N, 5.41. Found: C, 44.05; H, 2.99; N, 5.30.

For Al–PYB–Eu (%) Anal. Calcd.: C, 47.60; H, 3.26; N, 5.84. Found: C, 47.60; H, 3.15; N, 5.72.

For Al–PYB–Tb (%) Anal. Calcd.: C, 47.43; H, 3.25; N, 5.82. Found: C, C, 47.42; H, 3.17; N, 5.79. For Ti–Al–PYB–Eu (%) Anal. Calcd.: C, 45.84; H, 3.14; N, 5.63. Found: C, 45.84; H, 3.10; N, 5.52.

For Ti–Al–PYB–Tb (%) Anal. Calcd.: C, 45.68; H, 3.13; N, 5.61. Found: C, 45.68; H, 3.07; N, 5.55. In addition, the contents of lanthanide ions (Eu3+, Tb3+) in the hybrids are determined by EDTA titration.

For Ti–PYB–Eu (%) Anal. Calcd.: Eu, 7.36. Found: Eu, 7.33.

For Ti–PYB–Tb (%) Anal. Calcd.: Tb, 7.67. Found: Tb, 7.66.

For Al–PYB–Eu(%) Anal. Calcd.: Eu, 7.92. Found: Eu, 7.63.

For Al–PYB–Tb(%) Anal. Calcd.: Tb, 8.26. Found: Tb, 8.24.

For Ti–Al–PYB–Eu(%) Anal. Calcd.: Eu, 7.63. Found: Eu, 7.53.

For Ti–Al–PYB–Tb(%) Anal. Calcd.: Tb, 7.95. Found: Tb, 7.36.

2.4 Characterization

1H NMR spectra were recorded on a Bruker 510 spectrometer. Fourier transform infrared spectroscopy (FTIR) spectra were checked by Nicolette 5PC FTIR spectrophotometer. Powder X-ray diffraction (XRD) spectra were carried out on a XRD-6000 X-ray diffractometer (Shimadzu) with Cu Kα radiation (λ = 0.15405 nm) over the range of 10°–70°. Thermogravimetric analysis data were recorded with a thermal analysis instrument (TGA 1600 LF, METTLER TOLEDO, Switzerland) at a heating rate of 10 °C min−1 in an N2 flow. The morphology of the samples was inspected using a scanning electron microscope (SEM, S-4800, Hitachi). Photoluminescence (PL) excitation and emission spectra were recorded with a Jobin–Yvon FluoroMax-4 equipped with a 150-W xenon lamp as the excitation source at room temperature. The lifetime measurements were measured on an Edinburgh Instruments FS920P spectrometer, with a 450-W xenon lamp as the steady-state excitation source at room temperature. Rare earth contents of the complexes were determined by EDTA titration using xylenol orange as an indicator. Elemental analyses (C, H, N) were determined with an Elementar Cario EL elemental analyzer.

3 Results and discussion

3.1 Structure of the ligand and hybrids

As shown in Scheme 1, a two-step route was used to synthesize the ligand 3,4-bis(pyridin-4-ylmethoxy)benzoic acid. Firstly, ethyl 3,4-dihydroxybenzoate was readily obtained by the esterification reaction of commercially available 3,4-dihydroxybenzoic acid with ethanol. Finally, the ligand 3,4-bis(pyridin-4-ylmethoxy)benzoic acid was prepared by NaOH-mediated nucleophilic reaction of ethyl 3,4-dihydroxybenzoate with 4-(chloromethyl)pyridine after acidification. The ligand is designed as the connection between organic rare earth ion complexes and inorganic matrix. The carboxylic acid group of it can react with titanium isopropylate and aluminum isopropylate via ester exchange reaction, while the di-pyridine groups can coordinate with rare earth ions as well as sensitize the luminescence of them. In this way, organic rare earth complexes can be covalently anchored within the network of inorganic–organic hybrid systems. Theoretically, eight-coordination positions of Ln3+ (Tb3+, Eu3+) could be occupied by the ligands and water molecules to form an eight-coordination number structure.

Scheme 1
scheme 1

Synthesis process and predicted structure of the hybrids

Full size image

3.2 FTIR

The Fourier transform infrared spectra of Al–PYB–Tb, Ti–PYB–Tb and Ti–Al–PYB–Tb hybrids in the 4000–500 cm−1 range are shown in Fig. 1, and all the spectra present are almost similar. It is observed that a broad band locates at about 3340 cm−1, which is attributed to ν(O–H) of H2O molecule in the hybrids. The two peaks at about 3061 and 1603 cm−1 correspond to the stretching vibration of C=N in pyridine ring. The peaks located at 1653 cm−1 can be assigned to the typical stretching vibration of the carbonyl group, and no absorption bands from the acid can be observed comparing the spectra of PYB, indicating that the corresponding acid was successfully connected to the inorganic frameworks. The broad band in the range of 500–991 cm−1 has proved the formation of the group Ti–O–Ti, Al–O–Al and Al–O–Ti framework network in the hybrids [20, 21].

Fig. 1
figure 1

FTIR spectra of the hybrids containing Tb3+ and PYB

Full size image

3.3 XRD

The diffractograms of the hybrid materials from 10° to 70° presented in Fig. 2 show that all the structures of the hybrids are amorphous or non-periodic at long range. It might be attributed that the inorganic backbones possess less orderliness. The broad peak centered around 23.5° in the XRD patterns of Al–PYB–Tb may be ascribed to the coherent diffraction of the titanic backbone of the hybrids. The structural unit distance, calculated using the Bragg law, is approximately 3.78 Å. In Ti–PYB–Tb system, their broad peaks center at about 24.6° and structural unit distance is about 3.61 Å. In Ti–Al–PYB–Tb system, their broad peaks center at about 23.9° and structural unit distance is about 3.72 Å (Fig. 3).

Fig. 2
figure 2

XRD spectra of the hybrids containing Tb3+

Full size image
Fig. 3
figure 3

Thermogravimetry trace of the hybrids containing Tb3+

Full size image

3.4 TGA

Thermo gravimetric analysis (TGA) is performed on the hybrids in N2 atmosphere from 100 to 1000 °C. As shown in Fig. 5, it can be observed that these three samples behave the similar change trends in weight loss and three main degradation steps from the TGA curves. The first weight loss procedure about 2–2.5 % before 200 °C can be attributed to the loss of the adsorbed water and residual solvent DMF evaporated. All the curves exhibit sharp weight loss in the temperature from the beginning 200 °C to about 375 °C, and the weight loss is about 17.6 % for Ti–PYB–Tb, 20.5 % for Al–PYB–Tb and 23 % for Ti–Al–PYB–Tb. It is related to the decomposition of terbium complex moieties linking between Tb3+ and inorganic matrixes. The last mass loss about 24 % for Ti–PYB–Tb, 22 % for Al–PYB–Tb and 23.7 % for Ti–Al–PYB–Tb is from 375 to 1000 °C, probably corresponding to the decomposition of the residual organic groups covalently bonded to the inorganic matrixes. Comparing the three curves, Ti–PYB–Tb and Al–PYB–Tb possess higher thermal stability than composite Ti–Al–PYB–Tb. Besides, it also reveals that Ti–O–Ti inorganic matrix has more excellent thermal stabilities compared to the other two inorganic matrixes.

3.5 SEM

The transmission electron microscopy is introduced to investigate the morphology of the hybrid materials. The images of the hybrid materials Ti–PYB–Tb, Al–PYB–Tb and Ti–Al–PYB–Tb are given in Fig. 4. The morphology of hybrids is aggregates of small particles owing to sol–gel procedure. The dimension of the particles is about 0.2–0.5 μm in Ti–PYB–Tb system, while much larger particles in Al–PYB–Tb and Ti–Al–PYB–Tb systems can be observed. There is difference between the three kinds of hybrid materials which may be assigned to the difference of the metallic alkoxide matrix.

Fig. 4
figure 4

Selected scanning electron images for the hybrid materials a Ti–PYB–Tb, b Ti–Al–PYB–Tb, c Al–PYB–Tb

Full size image

3.6 Photophysical properties

The luminescence property of these resulting hybrid materials have been measured at room temperature. As shown in Fig. 5a, the excitation spectra of Tb-containing hybrid materials are obtained by monitoring the emission of Tb3+ at 545 nm and dominated by a broad band centered at about 300 nm for these hybrid materials which are ascribed to the characteristic absorption of the rare earth complexes arising from the efficient transition based on pyridine ring. The intensity of the excitation spectrum of Al–PYB–Tb is stronger than that of the other Tb-containing hybrid materials. From the emission spectra in Fig. 6a, the emission lines are assigned to the 5D4 → 7F J transitions located at 489, 542, 583 and 620 nm, for J = 6, 5, 4, and 3, respectively, where the most striking green luminescence (5D4 → 7F5) located at 542 nm is observed. The emission intensities of the hybrids are in the order: Al–PYB–Tb > Ti–Al–PYB–Tb > Ti–PYB–Tb, which corresponds to the results of excitation spectra.

Fig. 5
figure 5

Excitation spectra of a the hybrids containing Tb3+ and b the hybrids containing Eu3+

Full size image
Fig. 6
figure 6

Emission spectra of a the hybrids containing Tb3+ and b the hybrids containing Eu3+

Full size image

The excitation spectra of Eu-containing hybrid materials in Fig. 5b were recorded by monitoring the 5D0 → 7F2 transition at 613 nm. The spectra are dominated by a broad excitation band centered at about 250–375 nm in the ultraviolet region. It is assigned to the absorption of the ligands to their π/π* transition. Transitions at 363, 383 and 394 nm were observed and were attributed to 7F0 → 5D4, 7F0 → 5G2 and 7F0 → 5L6 transitions of Eu3+ ion, respectively. The emission luminescence behavior of the Eu-containing hybrid materials has been investigated by the direct excitation of the ligand (293 nm for Ti–PYB–Eu, 276 nm for Ti–Al–PYB–Eu, 282 nm for Al–PYB–Eu). The emission spectra are given in Fig. 6b. It can be clearly observed that characteristic Eu3+ ion emission bands in the 450–700 nm range, which are assigned to the 5D0 → 7F J (J = 0–4), transition at 578, 590, 613, 650 and 695 nm. Further, the 5D0 → 7F2 emission around 613.0 nm is the most predominant transition among these transitions, so the strong red luminescence can be observed in the emission spectra. It is well known that the 5D0 → 7F2 transition is a typical electric dipole transition, which is sensitive to the local symmetry of the coordination sphere of Eu3+ ions, while 5D0 → 7F1 is practically independent of the host material and therefore can be used as internal standard to explain the ligand differences. Moreover, the intensity ratios of the luminescence transitions (5D0 → 7F2/5D0 → 7F1, the red/orange ratio) have been widely taken as an indicator of the local environment of europium ions [22–24].

According to the intensity ratio listed in Table 1, it is concluded that the chemical environment around the Eu3+ ions is in low symmetry. The luminescence lifetime decay curves of the hybrid materials are measured at the room temperature, and all of the luminescence decays fit a double-exponential law, suggesting that Tb3+ and Eu3+ occupy two different environments in the hybrid systems. The resulting average luminescent lifetimes of hybrids are in Table 1. It appears that the lifetimes of Eu-containing hybrid materials are slightly lower than those of Tb-containing hybrid materials.

Table 1 Intensity ratios of 5D0 → 7F2/5D0 → 7F1 and lifetimes of the samples
Full size table

4 Conclusion

In summary, modifying PYB with alumina and titania gels through the sol–gel process results in the formation of new inorganic/organic rare earth hybrid materials. These obtained materials display connection between organic and inorganic parts on a molecular level, and they are amorphous and heat stabilized. Measurements of the photoluminescent properties show that the final Tb/Eu-containing hybrid materials present the typical green and red emissions, respectively, suggesting that organic ligand can efficiently sensitize rare earth ions in the hybrid materials.