Luminescence properties and Judd–Ofelt analysis of TiO₂:Eu³⁺ nanofibers via polymer-based electrospinning method

Abstract

One-dimensional TiO₂:xEu³⁺ nanofibers were fabricated via electrospinning and subsequent calcination. The as-spun nanofibers were calcined at 600, 700, 800 and 900 °C for 5 h at a heating rate of 1 °C min⁻¹ and the concentrations of Eu³⁺ dopants were varied from 17 mol% to 20 mol%. The TiO₂:19 mol% Eu³⁺ nanofibers which calcined at 700 °C (optimum condition) were investigated by thermogravimetric-differential thermal analysis (TG-DTA), X-ray diffraction (XRD), X-ray photo-electronic spectroscopy (XPS), scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FT-IR), UV-vis diffuse reflectance spectroscopy (UV-vis DRS), photoluminescence (PL) excitation and emission spectra. In this article, we have discussed the effect of different calcination temperature on fiber diameter and photoluminescence properties of europium doped titania (TiO₂:xEu³⁺) nanofibers. The possible formation mechanism of TiO₂:x mol% Eu³⁺ nanofibers was also discussed. The spectral characteristics and Eu–O ligand behavior were discussed through Judd–Ofelt parameters such as radiative transition probability (A_RAD), radiative lifetime (τ_rad), branching ratio (β_0J) and intensity parameters (Ω₂, Ω₄). Furthermore, the TiO₂:19 mol% Eu³⁺ nanofibers exhibit strong red luminescence that corresponds to the ⁵D₀–⁷F₂ transition (612 nm) of the Eu³⁺ ions under the excitation of ultraviolet light.

---

1. Introduction

Recently, many studies have been focused on the preparation and characterization of one-dimensional (1D) nanostructured architectures including nanobelts,^1,2 nanowires,³ nanotubes,⁴ and nanofibers^5,6 because of their unique physical and chemical properties. Among these various 1D nanomaterials, nanofibers have been demonstrated to be excellent candidates for the advanced applications, such as efficient electrode materials,⁷ nanocomposites,⁸ anode materials,⁹ photovoltaic and photocatalytic materials.¹⁰ Several techniques have been developed for the fabrication of one-dimensional nanostructures, such as oxidation,¹¹ in situ polymerization,¹² solvothermal reactions,¹³ solution-growth process,¹⁴ and steam explosion method.¹⁵ Compared with the above processes, electrospinning technique has been found to be a simple and cost-effective approach for manufacturing 1D nanomaterials. In the electrospinning process, nanofibers of metal oxides are obtained from a polymer solution ejected through electrostatic force between a polymer solution kept in a syringe and a metal electrode kept at a suitable distance, followed by evaporation of the polymer and solvent, calcination, and sintering. The nanomaterials diameter and morphology could be adjusted by controlling the spinning rate, electric field and the distance between the spinneret and collector. 1D nanomaterials fabricated by electrospinning possess many unique and fascinating properties such as uniform diameter, long length and high surface to volume ratio.

Among all the nanomaterials, rare-earth-doped compounds are important type of phosphors which have been widely used in the fields of high-performance luminescent devices,^16,17 sensors,¹⁸ catalysts,¹⁹ and other-functional materials based on their electronic, optical, and chemical characteristics arising from their 4f electrons. Some wide band-gap semiconductors including Y₂O₃,²⁰ ZnO,²¹ TiO₂ (ref. 22–24) have been used as host materials of rare earth in order to yield intense luminescence because of its low cost, high transparency in the visible-light region, good thermal, chemical, and mechanical properties. Therefore, TiO₂ is considered as host material in our research. As the most frequently used activator ions in luminescent materials, Eu³⁺ mainly shows emission due to transitions of ⁵D₀–⁷F_J (J = 1,2,3,4) in the orange-red regions.

Currently, the synthesis of Eu³⁺-doped TiO₂ nanostructures through electrospinning method and relevant luminescence properties have been reported. For example, Xie et al. have reported the morphology and PL intensity of samples with different doping concentration of Eu³⁺,²⁵ their team have also discussed the effect of temperature on the photoluminescence of Eu³⁺ doped TiO₂ nanofibers.²⁶ Thermal behaviour and luminescence properties of Eu-doped titania nanofibers have been investigated by Alessandra et al.²⁷ However, there were no report on the decay time and Judd–Ofelt analysis of Eu³⁺ doped TiO₂ nanofibers by electrospinning technique. Accordingly, here we employed electrospinning method to prepared 1D TiO₂:Eu³⁺ nanofibers and the structural and luminescent properties of products have been investigated in detail. Judd–Ofelt theory was adopted to calculate the intensity parameters (Ω₂, Ω₄) and other radiative properties such as radiative transition probability (A_RAD), radiative lifetime (τ_rad), and branching ratio (β_0J).

2. Experimental

2.1 Materials

Titanium sulfate (Ti(SO₄)₂), polyvinylpyrrolidone (PVP, M_w=1 300 000), N,N-dimethylformamide (DMF), Eu₂O₃ (99.99%) were analytical-grade reagents and used directly without further purification. The Eu(NO₃)₃ aqueous solution was obtained by dissolving Eu₂O₃ (99.99%) in dilute HNO₃ solution under heating with vigorously agitation. Deionized water was used for all treatment processes.

2.2 Synthesis of the TiO₂:Eu³⁺ nanorods

The TiO₂:xEu³⁺ nanofibers were prepared by electrospinning and successive calcination. Scheme 1 shows the laboratory setup for electrospinning. First, 1.5 mmol Ti(SO₄)₂ was added to 10 mL of N,N-dimethylformamide (DMF) under vigorous stirring, and the stoichiometric amount of Eu(NO₃)₃ (0.1 mol L⁻¹) was dropped under vigorous stirring for 20 min. The doping concentrations of Eu³⁺ ions are changed from 17 mol% to 20 mol%. Then 1.2 g polyvinylpyrrolidone (PVP, M_w =1 300 000) were dissolved into the above solution with thorough stirring to form a homogeneous solution for further electrospinning. Then the mixture was loaded in 5 mL plastic syringe with stainless steel needle which connected to a high voltage supply. The high voltage supply was maintained at 16 kV. An aluminium plate was used as a collector and the distance between the spinneret and the collector was set at 20 cm. The spinning rate was controlled at 0.1 mm min⁻¹. After dried for 12 h at room temperature under vacuum, the electrospinning fibers were calcined at 600, 700, 800 and 900 °C for 5 h with a rising rate of 1 °C min⁻¹, then were self-cooled to room temperature, forming the final products.

Scheme 1 Schematic representation for the preparation of as-spun Ti(SO₄)₂/Eu(NO₃)₃/PVP composite nanofibers by the electrospinning.

2.3 Characterization

The samples morphologies were characterized by scanning electron microscope (SEM, S-4800, Hitachi). Energy-dispersive X-ray (EDX) spectroscopy analysis was performed with an H JEOL JXA-840 EDX system attached to the SEM microscope. Based on the SEM images, the diameter and diameter distribution of fibers were analyzed using image visualization software Image J. The crystalline structure of the products was distinguished by an X-ray diffractometer (Rigaku D/max-B II) with Cu-Kα radiation (λ = 0.15405 nm), scans were made from 10° to 70° (2θ). Thermogravimetric analysis (TGA) was performed using a TGA/SDTA851e analyzer (Mettler Toledo) over a temperature range of 50–800 °C in air atmosphere with a heating rate of 10 °C min⁻¹. The X-ray photoelectron spectra (XPS) were taken using a VG ESCALAB 250 electron energy spectrometer with Mg Kα (1253.6 eV) as the X-ray excitation source. Fourier transform infrared spectroscopy (FTIR) was recorded with a Nicolette 5PC FTIR spectrophotometer by using the KBr pellet technique. The UV-vis diffuse reflectance spectrum (Shimazu, UV-3600) was employed to measure the samples. Photoluminescence (PL) excitation and emission spectra were recorded with a Jobin Yvon FluoroMax-4 fluorescence spectrophotometer equipped with a 150 W xenon lamp as the excitation source.

3. Results and discussion

3.1 Thermal properties

The thermal decomposition behavior of the as-spun precursor nanofibers is demonstrated by TG-DTA curves as shown in Fig. 1. There are four thermogravimetric steps. The first 16.6% weight loss before 300 °C can be attributed to the loss of moisture and trapped solvent (DMF) in the precursor nanofibers. The second is between 300 and 361 °C, with 32.8% weight loss, accompanied by a strong exothermic peak (351 °C) in the DTA curve. It is mainly due to the decomposition of titanium sulfate and the pyrolysis of PVP by a dehydration on the polymer side chain. When the temperature further increases, a sharp exothermic peak at 463 °C in the DTA curve is detected, corresponding to the decomposition of main chain of PVP and Eu(NO₃)₃. There is 32.3% weight loss in the range of 361–500 °C. From 500 to 730 °C, the TG curve goes down slowly. This owes to the decomposition of few PVP which is encysted in TiO₂ crystals. Above 730 °C, no further weight loss and thermal events are observed in the TG-DTA curves, indicating that the formation of pure inorganic oxide. And according to the TG-DTA curves, we can set up corresponding calcination temperatures.

Fig. 1 TG-DTA curves of as-spun Ti(SO₄)₂/Eu(NO₃)₃/PVP composite nanofibers.

3.2 Crystal structure

The composition and phase purity of the products were first examined by XRD. Fig. 2 shows the representative XRD patterns of the as-prepared precursor for TiO₂:19 mol% Eu³⁺ samples and those annealed at different temperatures as well as the JCPD cards (no. 21-1272, no. 21-1276), respectively. For the as-formed precursor sample (Fig. 2a), there are no diffraction peaks in the 2θ angle ranging from 10° to 70° for samples except for the broadband at 2θ = 22°, which is ascribed to the semi-crystalline PVP. For the sample annealed at 600 °C, the broad peak disappears and a series of peaks which can be indexed well to the anatase phase of TiO₂ (JCPDS 21-1272) begin to appear, as shown in Fig. 2b. The intensity of the peaks corresponding to the anatase structure increased as the calcination temperature increased from 600 °C to 700 °C due to the increase of crystallinity. When the calcination temperature was elevated to 800 °C, the weak diffraction peaks of Eu₂Ti₂O₇ can be detected, which is ascribed to enhancement of defects for high Eu content samples. And the formation of the Eu₂Ti₂O₇ may catalyze mass transport to the nucleation region of the rutile phase, promoting rutile nuclei growth which can explain the phenomenon of coexistence of anatase and rutile phase.^28,29 With further increasing to 900 °C, the diffraction peaks of Eu₂Ti₂O₇ became sharp and anatase has converted into rutile phase completely. The relative intensities and sharpness of the rutile peaks in the XRD patterns increase with the increase of calcination temperature, therefore, we adjust the ratio of anatase phase and rutile phase by changing the calcination temperature.

Fig. 2 XRD patterns of (a) as-spun Ti(SO₄)₂/Eu(NO₃)₃/PVP composite nanofibers and (b) TiO₂:19 mol% Eu³⁺ calcined at 600 °C, (c) at 700 °C, (d) at 800 °C, (e) at 900 °C and the standard cards (JCPDS 21-1272, JCPDS 21-1276) of TiO₂.

3.3 XPS analysis

XPS spectroscopy is a material characterization technique widely used to investigate the chemical composition of materials. Typical XPS survey scans of materials over a large energy range are presented in Fig. 3A, which indicates that the material contains Ti, O, Eu elements and a trace amount of carbon. Carbon present in sample was considered to be a contamination of CO₂ in air and carbon on the substrate. It can be seen that the Ti 2p XPS spectrum consists of two major peaks at 464.8 eV and 458.8 eV, corresponding to Ti 2p_1/2 and 2p_3/2 binding energies, respectively (Fig. 3B), however, the binding energies of TiO₂ without Eu³⁺ modification are 464 eV and 458.1 eV for Ti-2p_1/2 and Ti-2p_3/2, respectively. This shifting may arise from the reduction in the density of the electron cloud around Ti atoms by introduction of Eu³⁺ modification.³⁰ The spectrum of O 1s (Fig. 3C), shows that the sharp O 1s peak appears at 530.6 eV, the shift of O 1s binding energy for the TiO₂:19 mol% Eu³⁺ as compared to the O 1s state of pure TiO₂ (530.0 eV) was similar to that observed for the Ti 2p state.^31,32 Which reflects the decrease in the density of electron cloud on the O orbital and the formation of Ti–O–Eu bond. As observed in Fig. 3D, there are two symmetric peaks in the Eu 4d region, of which the peak located at 141.7 eV originates from the Eu 4d_3/2, and the other centered at 136.2 eV attributes to the Eu 4d_5/2. From which one can infer a spin–orbit splitting of 5.5 eV, confirming the existence of Eu³⁺.³³ The O 1s spectrum of TiO₂:19 mol% Eu³⁺ can be deconvoluted into two peaks (Fig. 3E). The major peak centered at 530.6 eV corresponds to the oxygen in titania, which is in line with O 1s spectrum of pure TiO₂. And the minor peak appearing at 532.5 eV corresponds to the oxygen of Ti–O–Eu bond. These XPS peaks further prove that Eu³⁺ ions were incorporated into the TiO₂ lattice.

Fig. 3 XPS spectra of the (A) survey, (B) Ti 2p, (D) O 1s for TiO₂:19 mol% Eu³⁺ and pure TiO₂ and (C) Eu 4d, (E) devolution spectra of O 1s for TiO₂:19 mol% Eu³⁺ nanofibers.

3.4 Morphology and composition

The SEM photographs of as-spun Ti(SO₄)₂/Eu(NO₃)₃/PVP composite fibers and the fibers calcined at 700 °C were showed in Fig. 4a and b. And through software Image J, we can analyze the diameter distribution of the nanofibers (Fig. 4e and f). It can be seen that the as-formed precursor fibers are relatively smooth with diameters ranging from 110 to 250 nm. However, after annealing at 700 °C for 5 h, the samples consist of smooth and uniform fibers with diameters from 30 to 90 nm. The fibers diameters decrease significantly due to the decomposition of the polyvinylpyrrolidone and crystallization of titanium dioxide. Energy dispersive X-ray (EDX) analysis was used to analyse the composition of the obtained nanofibers. As shown in Fig. 4c, which confirms the presence of C, O, Ti, S, and Eu in the as-spun precursor, the carbon peak mainly originated from the polymer. After annealing at 700 °C (Fig. 4d), the atomic ratios of O, Ti, and Eu increased, which further confirms the decomposition of the polyvinylpyrrolidone. Therein, the peak belonging to C come from the conducted C films used for coating the samples for SEM measurement. The appearance of element S should be attributed to a small quantity of rudimental SO₄²⁻.

Fig. 4 SEM micrographs of (a) as-spun Ti(SO₄)₂/Eu(NO₃)₃/PVP composite nanofibers; (b)TiO₂:19 mol% Eu³⁺ calcined at 700 °C and their corresponding EDX spectra ((c) and (d)) and diameter distribution ((e) and (f)).

SEM micrographs of TiO₂:19 mol% Eu³⁺ nanofibers calcined at different temperatures of 600, 700, 800 and 900 °C were depicted in Fig. 5a–d, and corresponding diameter distribution images have been shown in Fig. 5e–h. We can notice that different calcination temperatures have remarkable influence on the size and morphology of the products. After calcination at 700 °C (Fig. 5b and f), the average diameter of the nanofibers is decreased to 58 nm compared with the fibers calcined at 600 °C (Fig. 5a and e), the shrinkage of the fiber diameter was ascribed to further decomposition of PVP and crystallization of TiO₂ phase. With the temperature increased to 800 °C, the nanofibers were fractured because of the decomposition of few PVP which is encysted in TiO₂ crystals (Fig. 5c and g). And with the temperature increased to 900 °C, which leading to the thermal instability, the melting fibers have high surface energy and aggregate gradually through assembly, and the mean diameters of products increased. In summary, we have proposed the possible mechanism as follows: when the calcination temperature increased to 700 °C, the fibers decreased with temperature increased because of the decomposition of PVP and crystallization of TiO₂ phase. However, the fibers were broken up which caused by the burn out of residual PVP and the particle size has been formed and increased with further increasing of calcination temperature because of further crystallization and coalescence of the grains.

Fig. 5 SEM images of TiO₂:19 mol% Eu³⁺ nanofibers calcined at (a) 600, (b) 700, (c) 800, and (d) 900 °C and their corresponding diameter distribution ((e)–(h)).

3.5 FT-IR spectra analysis

Fig. 6 shows the FT-IR spectra of pure PVP nanofibers, as-spun TiO₂:19 mol% Eu³⁺ nanofibers and those calcined at 700 °C. For the spectra of pure PVP nanofibers (Fig. 6a), the broad band located at about 3438 cm⁻¹ is assigned to the O–H stretching vibrations of water absorbed on PVP nanofibers. The peaks at 2930, 1430, 1650, 1290 and 1100 cm⁻¹ can be attributed to the stretching vibrations of CH₂ groups, bending vibrations of CH₂ groups, C=O stretching vibrations,³⁴ the absorption of tertiary amine groups³⁵ and the asymmetric stretching vibration of the C–N groups of PVP, respectively. And it can be noticed that the peak of 472 cm⁻¹ which belong to the Ti–O stretching vibrations of titanium sulfate appears in the as-spun Ti(SO₄)₂/Eu(NO₃)₃/PVP composite nanofibers (Fig. 6b). When the nanofibers were annealed at 700 °C (Fig. 6c), the typical peaks of PVP disappeared gradually which confirmed that the thermal treatment was reasonably effective for removing PVP. At the same time, the main peak at 400–800 cm⁻¹ attributed to Ti–O stretching and Ti–O–Ti bridging stretching modes of TiO₂,³⁶ confirming the formation of TiO₂ nanofibers. Simultaneously, the band at ∼590 cm⁻¹ correspond to Ti–O–Eu vibrational modes.³⁷ The result provides additional evidence that the precursor has converted to the final sample after the annealing process.

Fig. 6 FT-IR spectra of different samples (a) pure PVP nanofibers, (b) as-spun Ti(SO₄)₂/Eu(NO₃)₃/PVP composite nanofibers, (c)TiO₂:19 mol% Eu³⁺ nanofibers calcined at 700 °C.

3.6 UV-vis spectra

The UV-vis diffuse reflectance spectra, in the range of 300–800 nm, were used to investigate the optical absorption properties of modified titania samples. Fig. 7a shows the UV-vis DRS spectra of TiO₂:x mol% Eu³⁺ samples (x = 17, 18, 19, 20). As displayed, all samples exhibit a wide optical absorption in the range below 400 nm, corresponding to the transition of electrons from the O 2p orbital to the Ti 3d orbital. The band gap for optical absorption is defined as the minimum energy required to excite an electron from the ground state (HOMO, at the top of the valence band) to the exciting state (LUMO, at the bottom of the conduction band). The band gap energy is usually calculated by the optical absorption spectrum using the following equation:

(αhν)ⁿ = B(hν − E_g)

where α is absorption coefficient, hν is the photo energy, B is a constant relative to the material, E_g is the band gap, and n is either 2 for direct transition or 1/2 for an indirect transition.³⁸ As we know, anatase is generally an indirect transition semiconductor, therefore, the band gap of different Eu doped concentrations can be obtained by extrapolating the linear portion of the (αhν)^1/2 versus hν curve to zero, as shown in Fig. 7b–e. The derived band gap energy (E_g) of TiO₂:x mol% Eu³⁺ are calculated to be about 2.88 eV, 2.90 eV, 2.95 eV, 3.13 eV for x = 17, 18, 19, 20, respectively. Which is lower than that of the bulk anatase TiO₂ (3.2 eV), this is due to the chemical defects or vacancies present in the intergranular regions, forming a new energy level to reduce the band gap energy. In addition, the band gap shifts to high energy as the amount of Eu³⁺ increases (Fig. 7f). A possible explanation can be attributed to the transformation of an n-type semiconductor into a degenerate semiconductor caused by Eu doping that decreases the fraction of light absorbed due to the Burstein–Moss effect.³⁹ Many reports have similar results.^40,41 However, Yu et al. have reported that band gap shifts to low energy as the amount of Fe³⁺ increases in TiO₂ matrix which has been explained through the creation of dopant levels near the valence band of TiO₂ on Fe³⁺ ion incorporation.⁴² Therefore, the band gap of semiconductor depends strongly on several factors including doping effects, structural modifications, defect chemistry, and so forth.

Fig. 7 (a) UV-vis diffuse reflectance spectra of different doped concentration, (b–e) Kubelka–Munk plots and bandgap energy estimation of Eu-doped TiO₂ for indirect transition, (f) x vs. energy gap.

3.7 Formation mechanism

On the basis of the aforementioned experiments, a possible formation mechanism of TiO₂:x mol% Eu³⁺ nanofibers is illustrated in Fig. 8. We proposed the whole evolution process of the nanofibers as follows: (1) hydrolysis and condensation; (2) sol process; (3) gelation; (4) spinning; (5) calcination. Firstly, solution have converted sol particles via hydrolysis and condensation reactions, the Ti(OH)₄ particles have been formed. And Eu³⁺ ions were incorporated into matrix through formation of Ti–O–Eu bond. Secondly, PVP was added to assist spinning by tuning the viscoelastic properties through the hydrogen bond between the C=O groups of PVP and –OH groups of Ti(OH)₄ network. After stirring for several hours, all PVP were dissolved and connected around the sol. In the electrospinning process, under applied electrical force, the Ti(OH)₄–PVP sol was elongated and transformed into gel accompanied with the evaporation of the solvent (DMF), and the gel fibers interlinked to form a large nonwoven web. The linear structure of PVP also helped the supramolecular arrange orderly and form a fibrous morphology. Ultimately, PVP were gradually decomposed and residual solvent eventually evaporated from the composite nanofibers, and after the dissolution of the Ti(OH)₄ gel, the TiO₂:x mol% Eu³⁺ nanofibers have been formed during the calcination process. In the experiment, it is noticeable that the morphology and diameter of nanofibers can be tuned via variation of properties of spinning solution which include the electrical conductivity, the type of polymer, surface tension of solution and electrospinning operating parameters which include voltage, spinning rate and the distance between the needle and the collector. For example, the appropriate amount of PVP can improve the spinnability of precursor sol and the high quality fibers can be easily achieved, however, with the amount of PVP decreased, more and more beads instead of fibers have appeared. So in order to obtain desired morphology, it is crucial to search for an optimum condition by varying some of the electrospun parameters in electrospinning process.

Fig. 8 A possible formation mechanism of TiO₂:x mol% Eu³⁺ nanofibers.

3.8 Luminescence properties

Fig. 9 shows the excitation and emission spectra of the as-prepared TiO₂:19 mol% Eu³⁺ nanofibers. The excitation spectrum (Fig. 9a), monitored with 612 nm emission of Eu³⁺ (⁵D₀–⁷F₂), consists of Eu³⁺ ions characteristic excitation peaks (383, 393, 414, 463, and 525 (533) nm), corresponding to the intra-⁴f₆ transitions from the ⁷F₀ to ⁵G₂, ⁵L₆, ⁵D₃, ⁵D₂, and ⁵D₁ levels of the Eu³⁺ ions. Upon excitation at 393 nm (maximum absorption line), the emission spectrum is composed of a group of lines peaking at 578, 589, 612 (618), 651 and 684 (696) nm. Which are ascribed to the ⁵D₀–⁷F_J (J = 0, 1, 2, 3, 4) transitions of the Eu³⁺ ions.⁴³ Among them, the strongest red emission, which splits into two peaks at 612 and 618 nm, arises from the forced electric-dipole ⁵D₀–⁷F₂ transitions of the Eu³⁺ ions. As we know, the ratio of integrated emission intensity of the ⁵D₀–⁷F₂ transition to the ⁵D₀–⁷F₁ transition is known as a red-to-orange fluorescence factor (R/O factor). According to the Judd–Ofelt theory, the intensity ratio (R/O) of the transitions ⁵D₀–⁷F₂ to ⁵D₀–⁷F₁ is a good probe for the symmetry of Eu³⁺ site.^44,45 If the Eu³⁺ ions occupy an inversion symmetry site in the crystal lattice, magnetic dipole transition ⁵D₀–⁷F₁ is the dominant transition. While Eu³⁺ situated low symmetries with no inversion center, the electric dipole transition ⁵D₀–⁷F₂ is the dominant transition. From Fig. 9b, it is clearly shown that the emission of the electric-dipole transition (612 nm) is much stronger than the magnetic dipole transition (589 nm), which indicates that Eu³⁺ ions prefer to hold a low symmetry site without an inversion center.⁴⁶ As is known, TiO₂ possesses a tetragonal structure with the I4₁/amd space group, and the site symmetries for the Ti⁴⁺ ions are D_2d in anatase. According to the branching rules of the 32 point groups, the substitution of Ti⁴⁺ (0.061 nm) with larger Eu³⁺ (0.098 nm) creates oxygen vacancies and caused the lattice distortions in the TiO₂ host, which leads to a descent of the intrinsic D_2d to a lower site symmetry.

Fig. 9 (a) Excitation and (b) emission spectra of the TiO₂:19 mol% Eu³⁺ nanofibers.

In order to investigate the effect of the dopant concentration on PL intensities, the concentrations of Eu³⁺ dopants were varied from 16 mol% to 21 mol%. As shown in Fig. 10, the peak positions and spectral shapes of emission spectra are not influenced by changing Eu³⁺ concentration, but the emission intensities increase with the increase of the concentration of Eu³⁺ from the beginning because of sufficient luminescent centers and reach the peak maximum at a dopant concentration of 19 mol% and then decrease after that due to the concentration quenching effect based on the energy transfer between adjacent luminescence centers. From the results discussed above, the optimum concentration for red emission of Eu³⁺ is 19 mol% in TiO₂:Eu³⁺ nanofibers. The relatively large quenching concentration has been observed, it is due to the interface effects of nanoscale materials that hinder the energy transfer between the activator ions.

Fig. 10 (a) Excitation and (b) emission spectra of TiO₂:x mol% Eu³⁺ (x = 16 17, 18, 19, 20, 21) nanofibers.

Fig. 11 shows the decay curves of the luminescence of the TiO₂:x mol% Eu³⁺ (x = 16, 17, 18, 19, 20, 21). The samples are excited by 393 nm and monitored by 612 nm. All the curves of samples exhibit multi-exponential feature that can be well-reproduced by a double-exponential functions as I(t) = I₀ + A₁ exp(−t/τ₁) + A₂ exp(−t/τ₂). Herein, I(t) and I₀ are the luminescence intensities at times t and 0, t is the time, A₁ and A₂ are constants, τ₁ and τ₂ are the double-exponential components of the decay time. The average fluorescence lifetime was defined as the following formula:

Fig. 11 Decay curves of (a) TiO₂:16 mol% Eu³⁺ (b) TiO₂:17 mol% Eu³⁺ (c) TiO₂:18 mol% Eu³⁺ (d) TiO₂:19 mol% Eu³⁺ (e) TiO₂:20 mol% Eu³⁺ (f) TiO₂:21 mol% Eu³⁺ nanofibers.

The fitting parameters of the decay time for the materials are given in Table 1. The average lifetimes are 0.0151, 0.0192, 0.0090, 0.0463, 0.0174 and 0.0055 ms, corresponding to the Eu³⁺ concentration of 16 mol%, 17 mol%, 18 mol%, 19 mol%, 20 mol% and 21 mol%, respectively. Some values of τ₁ and τ₂ with x = 16–21 mol% are different, which indicates that Eu³⁺ ions were located in two kinds of lattice environments in TiO₂ nanofibers. The Eu³⁺ ions with a short decay time may exist near the surface, whereas the Eu³⁺ ions with long decay time may exist at ordered lattice sites. However, because of the difference of ionic radius between Eu³⁺ and Ti⁴⁺ and according to the spectra discussion above, we can speculate that there are more Eu³⁺ ions exist on the surface rather than in the ordered lattice sites. We can notice that nanofibers with the strongest fluorescence (TiO₂:19 mol% Eu³⁺) exhibit the longest decay time. As we know, luminescence intensity is proportional to radiative transition probability, and the decay time is the inverse of the sum of the radiative transition probability and the non-radiative transition probability. Therefore, the non-radiative transition probability of Eu³⁺ from the ⁵D₀ energy level in the TiO₂:19 mol% Eu³⁺ is the smallest and lead to the longest decay time. However, decay times with different concentration (16–21 mol%) are shorter than other TiO₂:Eu³⁺ phosphors in previous reports.^47,48 This phenomenon can be explained through the quenching effect of the overmuch defects. The TiO₂:Eu³⁺ nanofibers have some quenching centers because of the defects which come from surface states.⁴⁹ When the excited luminescent centers are near quenching centers which exist in the products, the excited energy will be transferred to these quenching centers. As a result, decay times become shorter.

Table 1 Fitting parameters of the decay time for TiO₂:xEu³⁺

x	A1	τ1	A2	τ2	tav
16 mol%	204.19	0.16561	91 256.26	0.00640	0.0151
17 mol%	281.40	0.09329	21 631.34	0.00807	0.0192
18 mol%	193.07	0.09031	81 419.05	0.00613	0.0090
19 mol%	229.72	0.12898	5591.42	0.01326	0.0463
20 mol%	246.96	0.08749	19 989.85	0.00829	0.0174
21 mol%	244.25	0.12026	77 9034.57	0.00450	0.0055

---

3.9 Judd–Ofelt analysis

The detailed information of the site symmetry and luminescence behavior of Eu³⁺ ions in TiO₂ host was obtained by Judd–Ofelt parameter which can be calculated through PL spectra. The integrated emission intensities of the radiative emission of the transition between two manifolds ⁵D₀ and ⁷F_J (J = 2, 4) were associated with the radiative emission rates and can be written as:
where hν₁ and hν_J are energies for ⁵D₀–⁷F₁ and ⁵D₀–⁷F_J transitions and I₀₁ and I_0J are integral intensities, respectively. The ⁵D₀–⁷F₁ of Eu³⁺ ion is a magnetic dipole transition, which is independent of the environment and can be used as a reference. The magnetic dipole transition rate (A₀₁) of ⁵D₀–⁷F₁ transition of Eu³⁺ ion is
where S_md refers to the strength of the magnetic dipole ⁵D₀–⁷F₁ transition, which is a constant and independent of the medium, being equal to 9.6 × 10⁻⁴² units,⁵⁰ the value of A₀₁ are taken to be 50 s⁻¹,⁵¹ hence the electric dipole transition probability can be obtained from the above relation. The radiative emission rates A_0J were related to forced electric dipole transitions can be written as a function of the J–O intensity parameters:
where A_0J is the coefficient of spontaneous emission, e is the elementary charge, ν_J is the wavenumber of electric dipole transition luminescence, J equaling zero for ⁵D₀ transitions, h is the Planck's constant, and n is the refractive index of the nanophosphor sample. The |〈⁵D₀‖U^(λ)‖⁷F_J〉|² are the square reduced matrix elements whose values are independent of the chemical environment of the Eu³⁺ ion. The values are 0.00324, 0.00229 for J = 2 and 4, respectively.⁵² Since the transition rate of each energy level is in direct proportion to integral intensity of emission spectrum, and hence the ratio of electric dipole transition to magnetic dipole transition rate can be expressed as:
hence, A₀₂, A₀₄, Ω₂ and Ω₄ can be calculated. Some important radiative properties such as radiative transition probability (A_RAD), radiative lifetime (τ_R) and branching ratio (β_R) are useful in predicting the lasing potentiality for the electric dipole transitions between an excited level to its lower lying levels. The radiative transition probability (A_RAD) can be calculated using the equation below:

The radiative lifetime τ_rad of an excited state in terms of A_RAD, is given by:

The branching ratio β_0J corresponding to the emission from an excited level to its lower levels is given by:

The relevant Judd–Ofelt intensity parameters and radiative transition probability of TiO₂:Eu³⁺ phosphors are displayed in Fig. 12. The relative intensity of the hypersensitive electric dipole transition (⁵D₀–⁷F₂) depends on the local symmetry of the Eu³⁺ ions. This transition was mainly responsible for the Ω₂ value which is expected to increase with decrease in site symmetry, an increase in coordination number, an increase in covalence and a decrease in bond length. According to the ED selection rule, the ⁵D₀–⁷F₀ (0–0) transition is only allowed in the following 10 site symmetries: C_s, C₁, C₂, C₃, C₄, C₆, C_2v, C_3v, C_4v, and C_6v (ref. 53) and combined with the branching rules of the 32 point groups and relatively low Ω₂ values, we speculated that Eu³⁺ ions may occupy the C_2v, C₂ and C_s symmetry sites in tetragonal anatase titania. The parameter Ω₄ was not directly related to the symmetry of the Eu³⁺ ion but to the electron density on the surrounding ligands. The higher the Ω₄ value, the lower the electron density on the ligands. In the ⁵D₀–⁷F₂ emission, the β value are higher than the other transitions, indicating that the ⁵D₀–⁷F₂ transition is the main emission of Eu³⁺. The relative intensity ratio (R) of electric dipole transition (⁵D₀–⁷F₂) to magnetic dipole transition (⁵D₀–⁷F₁) can be used to understand the variation of symmetry and coordination environment around Eu³⁺ ion doped in TiO₂ matrix. The decrease in asymmetry ratio indicates the decrease in covalency in Eu³⁺–O²⁻ bond and increase in site symmetry.

Fig. 12 Spectral parameters of Eu³⁺ doped TiO₂ nanofibers.

4. Conclusions

In summary, one-dimensional TiO₂:x mol% Eu³⁺ nanofibers have been successfully synthesized through electrospinning and subsequent calcination techniques. The as-spun Ti(SO₄)₂/Eu(NO₃)₃/PVP composite nanofibers are relatively smooth with diameters ranging from 110 to 250 nm, the calcination process at 700 °C for 5 h induced the formation of TiO₂:x mol% Eu³⁺ nanofibers and resulted in smooth and uniform fibers with diameters ranging from 30 to 90 nm. The possible formation mechanism including hydrolysis and condensation, sol process, gelation, spinning and calcination has been discussed in detail. We have studied the photoluminescence (PL) properties of europium doped titania (TiO₂:x mol% Eu³⁺) nanofibers with different concentration of Eu³⁺ and confirmed the concentration quench effect. In order to investigate the nature of the luminescence behavior of Eu³⁺ in TiO₂, the Judd–Ofelt intensity parameters were calculated from the emission data. Under the excitation of ultraviolet light, the TiO₂:19 mol% Eu³⁺ nanofibers exhibit strong red luminescence that corresponds to the ⁵D₀–⁷F₂ transition (612 nm) of the Eu³⁺ ions. These studies reveal that electrospinning is a facile and novel route for the development 1D luminescent materials that are useful in the fields of optics and electronics in the future.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Grant No. 21171066 and 51272085).

Notes and references

1. X. Zeng, X. Zhang, M. Yang and Y. Qi, Mater. Lett., 2013, 112, 87–89.
2. Q. Ma, W. Yu, X. Dong, J. Wang and G. Liu, Nanoscale, 2014, 6, 2945–2952.
3. J. Qiu, Y. Qiu, K. Yan, M. Zhong, C. Mu, H. Yan and S. Yang, Nanoscale, 2013, 5, 3245–3248.
4. T. Panda, T. Kundu and R. Banerjee, Chem. Commun., 2012, 48, 5464–5466.
5. J. Mu, B. Chen, Z. Guo, M. Zhang, Z. Zhang, P. Zhang, C. Shao and Y. Liu, Nanoscale, 2011, 3, 5034–5040.
6. Q. Li, L. Sun, H. Zhao, H. Wang, L. Huo, A. Rougier, S. Fourcade and J.-C. Grenier, J. Power Sources, 2014, 263, 125–129.
7. L.-F. Chen, X.-D. Zhang, H.-W. Liang, M. Kong, Q.-F. Guan, P. Chen, Z.-Y. Wu and S.-H. Yu, ACS Nano, 2012, 6, 7092–7102.
8. Z.-M. Huang, Y.-Z. Zhang, M. Kotaki and S. Ramakrishna, Compos. Sci. Technol., 2003, 63, 2223–2253.
9. C. T. Cherian, J. Sundaramurthy, M. Kalaivani, P. Ragupathy, P. S. Kumar, V. Thavasi, M. Reddy, C. H. Sow, S. G. Mhaisalkar and S. Ramakrishna, J. Mater. Chem., 2012, 22, 12198–12204.
10. X. Zhang, V. Thavasi, S. G. Mhaisalkar and S. Ramakrishna, Nanoscale, 2012, 4, 1707–1716.
11. A. Isogai, T. Saito and H. Fukuzumi, Nanoscale, 2011, 3, 71–85.
12. J. Li, H. Xie, Y. Li, J. Liu and Z. Li, J. Power Sources, 2011, 196, 10775–10781.
13. J.-Y. Liao, J.-W. He, H. Xu, D.-B. Kuang and C.-Y. Su, J. Mater. Chem., 2012, 22, 7910–7918.
14. W. Li, F. Zhang, Y. Dou, Z. Wu, H. Liu, X. Qian, D. Gu, Y. Xia, B. Tu and D. Zhao, Adv. Energy Mater., 2011, 1, 382–386.
15. B. Deepa, E. Abraham, B. M. Cherian, A. Bismarck, J. J. Blaker, L. A. Pothan, A. L. Leao, S. F. De Souza and M. Kottaisamy, Bioresour. Technol., 2011, 102, 1988–1997.
16. A. Stanulis, A. Katelnikovas, D. Enseling, D. Dutczak, S. Šakirzanovas, M. Van Bael, A. Hardy, A. Kareiva and T. Jüstel, Opt. Mater., 2014, 36, 1146–1152.
17. H. Li, K. Zheng, X. Xu, H. Zhao, Y. Song, Y. Sheng, Q. Huo and H. Zou, Powder Technol., 2012, 228, 277–283.
18. T. Wang, S. Ma, L. Cheng, J. Luo, X. Jiang and W. Jin, Sens. Actuators, B, 2015, 216, 212–220.
19. D. P. Dutta, M. Roy and A. Tyagi, Dalton Trans., 2012, 41, 10238–10248.
20. X. Wu, Y. Liang, R. Chen, M. Liu and Y. Li, J. Mater. Sci., 2011, 46, 5581–5586.
21. J.-z. Huang, S.-y. Liu, N.-n. Yao and X.-j. Xu, Optoelectronics Letters, 2014, 10, 161–163.
22. H. Li, H. Zhao, Y. Sheng, J. Huang, Y. Song, K. Zheng, Q. Huo and H. Zou, J. Nanopart. Res., 2012, 14, 1–10.
23. Y. Cao, Z. Zhao, J. Yi, C. Ma, D. Zhou, R. Wang, C. Li and J. Qiu, J. Alloys Compd., 2013, 554, 12–20.
24. W. Luo, C. Fu, R. Li, Y. Liu, H. Zhu and X. Chen, Small, 2011, 7, 3046–3056.
25. J. Zhao, C. Jia, H. Duan, Z. Sun, X. Wang and E. Xie, J. Alloys Compd., 2008, 455, 497–500.
26. J. Zhao, H. Duan, Z. Ma, L. Liu and E. Xie, J. Optoelectron. Adv. Mater., 2008, 10, 3029–3032.
27. A. Bianco, I. Cacciotti, M. E. Fragalá, F. R. Lamastra, A. Speghini, F. Piccinelli, G. Malandrino and G. Gusmano, J. Nanosci. Nanotechnol., 2010, 10, 5183–5190.
28. Y. Zhang, H. Zhang, Y. Xu and Y. Wang, J. Mater. Chem., 2003, 13, 2261–2265.
29. I. Cacciotti, F. R. Lamastra, A. Bianco, A. Speghini, F. Piccinelli, M. E. Fragalà, G. Malandrino and G. Gusmano, Ceram. Mater., 2010, 62, 516–520.
30. J.-H. Huang, P. Y. Hung, S.-F. Hu and R.-S. Liu, J. Mater. Chem., 2010, 20, 6505–6511.
31. S. Chenakin, G. Melaet, R. Szukiewicz and N. Kruse, J. Catal., 2014, 312, 1–11.
32. M. Yan, H. Zou, H. Zhao, Y. Song, K. Zheng, Y. Sheng, G. Wang and Q. Huo, CrystEngComm, 2014, 16, 9216–9223.
33. D. Chen, Q. Zhu, Z. Lv, X. Deng, F. Zhou and Y. Deng, Mater. Res. Bull., 2012, 47, 3129–3134.
34. O. Saensuk, S. Phokha, A. Bootchanont, S. Maensiri and E. Swatsitang, Ceram. Int., 2015, 41, 8133–8141.
35. Y. Zhao, H. Wang, X. Lu, X. Li, Y. Yang and C. Wang, Mater. Lett., 2008, 62, 143–146.
36. C. Song, W. Yu, B. Zhao, H. Zhang, C. Tang, K. Sun, X. Wu, L. Dong and Y. Chen, Catal. Commun., 2009, 10, 650–654.
37. N. Paul and D. Mohanta, Physical properties of nanoscale TiO₂ with mild rare earth ion doping, 2013.
38. A. G. Al-Sehemi, A. S. Al-Shihri, A. Kalam, G. Du and T. Ahmad, J. Mol. Struct., 2014, 1058, 56–61.
39. A. V. Emeline, Y. Furubayashi, X. Zhang, M. Jin, T. Murakami and A. Fujishima, J. Phys. Chem. B, 2005, 109, 24441–24444.
40. S. Kumar, Z. Jindal, N. Kumari and N. K. Verma, J. Nanopart. Res., 2011, 13, 5465–5471.
41. M. Pal, U. Pal, J. M. G. Y. Jiménez and F. Pérez-Rodríguez, Nanoscale Res. Lett., 2012, 7, 1–12.
42. J. Yu, Q. Xiang and M. Zhou, Appl. Catal., B, 2009, 90, 595–602.
43. X. Qi, H. Zou, Y. Song, H. Zhang, H. Zhao, Z. Shi and Y. Sheng, Ceram. Int., 2014, 40, 12993–12997.
44. B. Judd, Phys. Rev., 1962, 127, 750.
45. L. Żur, J. Janek, M. Sołtys, J. Pisarska and W. Pisarski, Phys. Scr., 2013, 2013, 014035.
46. S. Prashantha, B. Lakshminarasappa and B. Nagabhushana, J. Alloys Compd., 2011, 509, 10185–10189.
47. H. Zhang, Y. Sheng, K. Zheng, X. Zhou, Z. Shi, X. Xu and H. Zou, Eur. J. Inorg. Chem., 2014, 2014, 3305–3311.
48. Q. Zeng, Z. Zhang, Z. Ding, Y. Wang and Y. Sheng, Scr. Mater., 2007, 57, 897–900.
49. Y. Zheng, H. You, G. Jia, K. Liu, Y. Song, M. Yang and H. Zhang, Cryst. Growth Des., 2009, 9, 5101–5107.
50. M. Weber, T. Varitimos and B. Matsinger, Phys. Rev. B: Solid State, 1973, 8, 47.
51. G. De Sa, O. Malta, C. de Mello Donegá, A. Simas, R. Longo, P. Santa-Cruz and E. Da Silva, Coord. Chem. Rev., 2000, 196, 165–195.
52. L. Liu and X. Chen, Nanotechnology, 2007, 18, 255704.
53. W. Luo, R. Li, G. Liu, M. R. Antonio and X. Chen, J. Phys. Chem. C, 2008, 112, 10370–10377.

---