Mechanistic investigation on up and down conversion of Er³⁺ and Gd³⁺ co-doped YTiNbO₆ phosphors

Abstract

YTiNbO₆ phosphors doped with Er³⁺, Gd³⁺ were synthesized via a facile sol–gel and combustion approach. The crystal structure and particle morphology were characterized and the phonon energy was investigated in detail. The up-conversion emission excited at 980 nm and down-conversion emission excited at 378 nm of Er³⁺ with different concentrations were analysed. They both have strong green emission peaks at 522 and 552 nm but the red emission peak at 668 nm only appears in the up-conversion. Furthermore, the Er³⁺ quenching concentration for up-conversion is higher than that for down-conversion because the up-conversion is a complex multiple stepwise process which includes energy transfer, phonon-assisted energy transfer and excited state absorption that can cover a wider range of Er³⁺ doping concentrations. Moreover, the influence of co-doped Gd³⁺ on the morphology and particle size was also explored. Gd³⁺ had a great effect on the up and down conversion emission intensity as well as the decay time of up-conversion by changing the structure distortion, vibration frequencies and surface defects.

---

Introduction

Rare-earth doped phosphors, which not only can effectively convert ultraviolet light into visible light, but also have the ability to absorb near-infrared light and emit fluorescence have attracted great attention due to their potential use in fundamental and technological applications.^1–3 The emission band of rare-earth (RE) metals is sharp and narrow, and its position is weakly dependent on the surrounding environment or the crystal field since the unique 4f–4f transitions are shielded by the outer orbitals. Besides, the up-conversion (UC) of RE has large anti-Stokes shifts up to 500 nm.⁴ With these excellent and unique photoluminescence features, RE doped UC phosphors have a large number of potential applications such as in biomedicine, solar cells, luminescent labels, photochemical catalysis, detectors, and temperature sensors.^5–11 In the energy transfer of UC, in order to achieve multi-photon UC, the RE ions should have an energy level scheme with equally spaced intermediate states. Meanwhile RE co-doping such as Yb–Er, Yb–Tm, Yb–Ho, Yb–Tb, Yb–Er–Tm has been explored in diverse host materials since Yb³⁺ can act as a sensitizer to absorb near-infrared wavelengths of 980 nm efficiently and transfer it to the luminescence center.^12,13 Yb³⁺ sensitized materials have been well investigated in bulk and nanoscale materials.¹⁴ Therefore, we focused our work on the less-studied single Er³⁺ doped UC phosphors, where Er³⁺ is not only a sensitizer but also acts as the luminescence centre since Er³⁺ in the 4f¹¹ configuration has a unique energy level structure to facilitate IR-to-visible multi-photon up-conversion and multi-colour emission. Phosphors doped with Er³⁺ exhibited preferential green and red emission.

Inorganic oxide compounds doped with RE ions are a class of phosphor available for many photonic applications due to their exceptional thermal and chemical stability, luminescence properties and environmental friendliness.¹⁵ As a typical euxenite-type compound, YTiNbO₆ with space group Pbcn has attracted much attention attributed to its potential application as a novel host material since its properties were first reported in 1974.^16,17 What is more, our previous work has demonstrated that RETiNbO₆ (RE = Y, La) doped with Eu³⁺, Er³⁺, Dy³⁺ and Ho³⁺ has great down-conversion (DC) emission as the RE³⁺ doped into the YNbTiO₆ host may offer more probability of gaining many multi-stage transition pathways and broadly distributed emission spectra.^18–21 However, to the best of our knowledge, there are few reports about the UC of YTiNbO₆. On the one hand, Y³⁺ can be easily replaced by RE³⁺, which induced some novel up-conversion luminescence (UCL) phenomena without any charge compensation problems. On the other hand, the low phonon energy as well as the crystal symmetry and microstructure of YTiNbO₆ is also beneficial to the UCL.

In this study, Er³⁺ and Gd³⁺ co-doped YTiNbO₆ phosphors have been prepared through a facile sol–gel and combustion approach. The crystal structure, morphology and the concentration quenching effect on UC and DC emission were investigated. Furthermore, UC has multiple mechanisms of electronic transitions such as stepwise photo excitation, energy transfer, cross relaxation and non-radiative relaxation processes. The electronic transition and the energy transfer as well as the phonon energy of YTiNbO₆ for UC and DC processes were systematically studied. The difference in the concentration quenching mechanism for UC and DC was contrasted. What is more, the behaviour of co-doping Gd in YTiNbO₆: 4% Er affected the crystallinity and the symmetry of the local field around Er³⁺ and had a great influence on the UCL and down-conversion luminescence (DCL) intensity.

Experimental

Synthesis procedures

Pure and doped YTiNbO₆ phosphors were prepared by a facile sol–gel combustion process just like our previous work.^20,21 Erbium oxide (Er₂O₃), yttrium oxide (Y₂O₃), gadolinium oxide (Gd₂O₃), niobium oxide (Nb₂O₅), tetra-n-butyl titanate (Ti(C₄H₉O)₄), citric acid (C₆H₈O₇·H₂O) and ammonium nitrate (NH₄NO₃) were purchased from Sinopharm Chemical Reagent Beijing and used as the starting materials. Yttrium nitrate (Y(NO₃)₃), erbium nitrate (Er(NO₃)₃) and gadolinium nitrate (Gd(NO₃)₃) solutions were obtained by dissolving Er₂O₃, Y₂O₃, and Gd₂O₃ in diluted nitric acid at a certain mole ratio. All the regents were analytical grade and used without any further purification.

Typically, 0.5 mmol Nb₂O₅ was dissolved with excess hydrofluoric acid (HF, 40%) in a water bath at 90 °C, and the pH of the solution was regulated to 9.0 by adding aqueous ammonia solution. Then, the white precipitate of Nb₂O₅·nH₂O was filtered and washed with deionized water several times to make sure that the F⁻ ion was completely removed. Afterwards, the Nb₂O₅·nH₂O was dissolved with Ti(C₄H₉O)₄ and citric acid with a mole ratio of 1 : 1 : 4 at 80 °C. Then, the RE nitrate and ammonium nitrate were added into the solution. The mole ratio of citric acid and ammonium nitrate was 1 : 5. Citric acid was used as the complexant and ammonium nitrate used as fuel for the combustion process. The obtained solution was mixed homogeneously with continuous stirring at 85 °C until a sol formed. After the water was evaporated, the transparent sol turned into a yellow gel with high viscosity. The gel was dried at 110 °C for 24 h to form a yellow xerogel. Then, the xerogels were introduced into crucibles, and directly transferred into a muffle furnace. The xerogels obtained were kept at a calcining temperature of 550 °C for 15 min, then increased to 800, 900, 1000 and 1100 °C for 1 h, and then cooled down to room temperature in the furnace. Finally, all the samples were ground into powder for characterization.

Characterization

The phase composition and structure were examined using X-ray powder diffraction patterns (Germany Bruker Axs D8-Avance X-ray diffractometer with graphite monochromatized Cu Kα irradiation (λ = 1.5418 Å)), and the data were collected over the 2θ range of 10°–70°. The morphology and composition were characterized using scanning electron microscopy (SEM; Hitachi, S-4800) and energy dispersion X-ray spectroscopy (EDS, Horiba, EMAX, Energy, EX-350). Thermal analysis of the powder dried at 110 °C for 24 h was carried out from 30 to 1200 °C using a thermogravimetric-differential thermal analyser (TG-DTA) (PerkinElmer Corporation, Diamond TG-DTA) with a constant heating rate of 20 °C min⁻¹. Raman spectroscopy measurements were performed on a Lab Ram Aramis (Thermo Nicolet. American. 1064 nm laser). UCL spectra were recorded using a SpectroPro300i and a power tuneable 980 nm semiconductor laser (0–800 mW). UCL decay curves were measured with a customized phosphorescence lifetime spectrometer (FSP920-C, Edinburgh) equipped with a digital oscilloscope (TDS3052B, Tektronix) and a tuneable mid-band Optical-Parametric Oscillator (OPO) pulse laser as the excitation source (410–2400 nm, 10 Hz, pulse width of 5 ns, vibrant 355II, OPOTEK). The up-conversion of the quantum yields was measured using a Quanta Master-400 with a 980 nm DPSS laser. DCL spectra were recorded on a fluorescence spectrophotometer (JEOL, F-4500). All the measurements were performed at room temperature.

Results and discussion

Structure and morphology analysis

The phase purity of the as-prepared pure and doped YNbTiO₆ phosphors was measured using XRD. Fig. 1 shows that the different samples agree well with the euxenite-type YTiNbO₆ (Joint Committee on Powder Diffraction Standards JCPDS file card no. 01-83-1318) without redundant peaks or phases. Fig. 1(a) shows that as the annealing temperature rises from 800 to 1100 °C, the diffraction peaks successively become shaper and stronger, which suggests that the crystallinity becomes better and the size of the sample gets bigger. Fig. 1(b) presents the pure Er³⁺ doped and Er³⁺, Gd³⁺ co-doped samples that match well with euxenite-type YNbTiO₆. It confirms that the doped ions successfully enter into the host materials and no distinct Er³⁺ or Gd³⁺ related phases such as Er₂O₃, Gd₂O₃ or Y₂O₃ appear. This can be explained by the Y³⁺ which can certainly be replaced by the appropriate Er³⁺ and Gd³⁺, since they have similar physical properties and co-ordination structures.²¹ Furthermore, the inset, Fig. 1(c), shows that the diffraction peak (311) is slightly shifted to a higher degree when Er³⁺ or Er³⁺, Gd³⁺ is doped into the host material since the radius of the doped ions Er³⁺ (1.19 Å, CN = 8) and Gd³⁺ (1.14 Å, CN = 8) are bigger than Y³⁺ (1.02 Å, CN = 8).²⁰ The larger-sized Er³⁺, Gd³⁺ occupied the smaller-sized Y³⁺ sites, leading to a lattice distortion effect and the increasing of the lattice parameters and volume accordingly.²² From the analysis of the XRD patterns, it can be concluded that the doped ions have been successfully doped into the YTiNbO₆ host lattices.

Fig. 1 (a) XRD patterns of the as-synthesized YNbTiO₆: 4% Er, 0.5% Gd with different temperatures. (b) Pure YNbTiO₆, 4% Er single doped and 4% Er, 0.5% Gd co-doped phosphors. (c) Comparison of the as-synthesized samples in the 2θ range of 29.0 to 31.0 degrees.

Typical SEM pictures of the pure and doped phosphors prepared at 1100 °C are presented in Fig. 2. Higher annealing temperature leads to the agglomeration of the sample and irregular shapes. Fig. 2(a) displays the pure YTiNbO₆ shaped blocks or plates with a size of 500–800 nm. The morphology of YNbTiO₆: 4% Er is shown in Fig. 2(b). It has similar shapes as pure YTiNbO₆ and sizes of about 450–600 nm, which is smaller than pure YTiNbO₆. Fig. 2(c) exhibits the image of YNbTiO₆: 4% Er, 0.5% Gd. Its morphology becomes more uniform and agglomeration weaker and the size is about 400–550 nm. The decreasing crystallite size can be attributed to an enhanced nucleation versus crystal growth upon Gd³⁺ substitution.²³ Fig. S1† shows the SEM micrographs of the as-synthesized phosphors prepared with different concentrations of Gd³⁺. Compared with the undoped Gd³⁺ sample, the samples have a nanorod morphology. On increasing the Gd³⁺ concentration from 0.5% to 3%, the length of the nanorods became longer while the radius of the nanorod decreased. Furthermore, the EDS spectrum of the co-doped sample is presented in Fig. 2(d) where the ratio of M (Y, Er, Gd)/Ti/Nb/O is 1.04 : 0.97 : 1.01 : 5.97, near to 1 : 1 : 1 : 6, indicating that the sample is probably composed of YNbTiO₆. The inset shows that the ratio of Y/Er/Gd is 0.947 : 0.047 : 0.005, close to 0.945 : 0.04 : 0.005, which indicates that corresponding Er³⁺ and Gd³⁺ ions were successfully combined into the host lattice.

Fig. 2 SEM micrographs of the as-synthesized phosphors prepared at 1100 °C: (a) YNbTiO₆, (b) YNbTiO₆: 4% Er, (c) YNbTiO₆: 4% Er, 0.5% Gd, and (d) the EDS spectrum of YNbTiO₆: 4% Er, 0.5% Gd sample; the inset shows the atom percent of all elements.

The structure of euxenite-type YTiNbO₆ is shown in Fig. 3. Nb⁵⁺ and Ti⁴⁺ ions are located at the same crystallographic position. The double-layer distorted Ti(Nb)O₆ octahedra join together by sharing edges or corners. The YO₈ polyhedra share edges with four others to form a single layer. Fig. 3(b) and (c) show the polyhedra of Ti(Nb)O₆ and YO₈. For the site of the cations in YTiNbO₆, the Y³⁺ ions occupy a C_s^xy(4) site and the Ti/Nb ions occupy a C₁(8) site of general symmetry.²⁴ The oxygen atoms occupy three different sites of C₁(8) as O₁, O₂ and O₃. O₁ and O₂ sites are tetrahedrally coordinated with two Ti(Nb) and two Y³⁺ cations, while O₃ is linked with four Ti(Nb) cations.²⁵

Fig. 3 (a) Three-dimensional space crystal structures of YNbTiO₆, (b) distorted Ti(Nb)O₆ octahedron, and (c) YO₈ irregular polyhedron.

The TG-DTA curves of the Er, Gd co-doped YNbTiO₆ xerogel powders dried at 1100 °C for 24 h are shown in Fig. 4. As the temperature increased to 293 °C, the TG curve displays the first weight loss which is about 73.64%, accompanied by weak endothermic peaks around 161 °C and exothermic peaks at 290 °C in the DTA curve. The weight loss is due to the evaporation of water and the combustion of organic components such as citric acid or the remaining organic components from tetra-n-butyl titanate and decomposition. The endothermic peaks are due to the evaporation of water and the exothermic peaks are due to the cross-linking effect and the combustion of the organic components. In the range of 293–563 °C, the second weight loss stage is due to the further combustion of the citrate and the organic remains, accompanied by a broad exothermic peak in the range of 499–560 °C. The third stage of weight loss comes at the temperatures between 563 and 1200 °C, accompanied by one weak exothermic peak at 714 °C which can be ascribed to the process of the orthorhombic phase formation, crystallization and transition. When the temperature reaches 1200 °C, the sample underwent no more transformations and there was nearly no weight loss in the TG curve, which proves that the sample has formed a relatively stable state.

Fig. 4 TG-DTA curves of the Er, Gd co-doped YNbTiO₆ precursor xerogel powder.

The Raman spectrum of the pure YTiNbO₆ is shown in Fig. 5 and the main peaks are summarized in Table 1. The sample did not contain rare earth dopants in order to prevent obscuring of the Raman signal by photoluminescence. The vibration of the Ti(Nb)–O bonds produces two Raman active ν₁A_1g and ν₂E_g modes; the ν₁A_1g mode corresponds to the peak at 857.72 cm⁻¹ and the ν₂E_g mode leads to the weak asymmetric shaped Raman band at 630.63 cm⁻¹.²⁶ For wavenumbers between 200 and 400 cm⁻¹, the bands are due to the symmetric bending vibrations of O–Ti(Nb)–O, corresponding to ν₃F_1u, ν₅F_2g, and ν₆F_2u modes.²⁷ The last Raman band located at 155.43 cm⁻¹ would be the euxenite-type lattice vibration, mainly associated with the RE ions.²⁴ Furthermore, the statistics confirm that the highest phonon energy of YTiNbO₆ is about 860 cm⁻¹, which is relatively lower than some inorganic oxide compounds such as phosphate, silicate and aluminate, etc.²⁸ The low phonon energy is better for phonon-assisted energy transfer and may suggest that YTiNbO₆ is a suitable host lattice for effective UCL with appropriate RE ion doping.

Fig. 5 Raman spectrum of pure YTiNbO₆ phosphor.

Table 1 Prominent Raman peaks and characteristic modes of YTiNbO₆

Wavenumber (cm^−1)	Stretching vibration	Assignments
857.72	Ti–O, Nb–O	ν1A1g
636.63	Y–O, T–O, Nb–O	ν2Eg
399.28	O–Ti–O, O–Nb–O	ν3F1u
350.51	O–Ti–O, O–Nb–O	ν5F2g
278.98	O–Ti–O, O–Nb–O	ν6F2u
226.96	O–Ti–O, O–Nb–O	ν6F2u
155.43	Lattice vibration	Lattice modes

---

UC and DC emission properties

It is well known that the doping concentration and co-doped ions have a great impact on UC and DC luminescence properties. In order to reveal the concentration dependency of UCL and DCL in YTiNbO₆, emission spectra with different Er³⁺ doping are shown in Fig. 6(a) and (b). Fig. 6(a) shows the DCL excited at 378 nm has two main emission peaks centred at 522 and 552 nm and Fig. 6(b) displays the UCL excited at 980 nm that has three main emission peaks located at 522, 552 and 668 nm. All the emission peaks can be explained using the energy transitions of Er³⁺ displayed in Fig. 6(d). The green emissions at 522 and 552 nm are due to the ²H_11/2 → ⁴I_15/2 and ⁴S_3/2 → ⁴I_15/2 transitions, while for the red emission at 668 nm, the ground state electrons were excited to the excited state ⁴F_7/2 through ETU and ESA processes and then transferred to the ⁴F_9/2 state by multi-phonon relaxation, whereafter the excited electrons jumped to the ground state with the transition ⁴F_9/2 → ⁴I_15/2. Compared with the DCL, the UCL has strong red emission which can be explained that under 980 nm excitation, the lower lever ⁴F_9/2 can be more effectually populated than under 378 nm excitation.²⁹ Furthermore, the highest intensity of UCL is achieved when doped with 4% Er³⁺, while for DCL the best concentration is 2% Er³⁺.

Fig. 6 (a) DC emission spectra under 378 nm excitation with doping x% mol Er³⁺ (x = 0, 0.5, 1, 2, 3, 4, 5). (b) UC emission spectra under 980 nm excitation with different concentrations of Er³⁺. (c) The intensity change of the DC and UC emission spectra at 522 nm with increasing Er³⁺ concentration. (d) Simplified energy level diagram and the UC and DC energy transfer mechanisms in Er³⁺ doped YTiNbO₆; ground state absorption (GSA), excited state absorption (ESA), energy transfer up-conversion (ETU).

To study the different influence of the Er³⁺ doping concentration on UCL and DCL, the green emission intensity at 522 nm for UC and DC is calculated from the emission spectra in Fig. 6(a) and (b), as shown in Fig. 6(c). It can be seen that with 378 nm excitation, the intensity rapidly rises as the doping concentration increases from 0.5% to 2% and then slowly decreases as the doping concentration varies from 2% to 5%, while under 980 nm excitation, the intensity gradually goes up as the concentration increases from 0.5% to 4% and is followed by a decrease as the doping concentration increases from 4% to 5%. They both increase at first and reach their own maximum values, and then decrease with rising Er³⁺ concentration. The influence of Er³⁺ concentration on the UCL and DCL is different mainly due to the different excitation routes and quenching mechanisms.²²

The luminescence intensity is mainly affected by the electron transition, non-radiative relaxation and energy transfer processes for UCL and DCL. For the DCL excited at 378 nm in Fig. 6(c), when the Er³⁺ concentration is lower than 2%, the luminescence intensity is nearly linearly proportional to the Er³⁺ concentration. This is because the ²H_11/2 level is populated via a cascade multi-phonon nonradiative transition from ⁴G_11/2 while the cross relaxation (CR) between Er³⁺ ions can be disregarded. As the concentration of Er³⁺ surpasses 2%, the intensity begins to decline because the CR interactions between Er³⁺ ions begin to take effect and concentration quenching appears. The influence of the CR factor is inversely proportional to R^Q (R is the average distance between Er³⁺ ions, Q is the interaction between electric-dipoles and magnetic dipoles).²⁹ R^Q will rapidly decline with increasing Er³⁺ concentration, and the CR effect will be enhanced. The CR in the energy levels can be described as ²H_11/2 + ⁴I_15/2 → ⁴I_13/2 + ⁴I_9/2 (CR1), ²H_11/2 + ⁴I_13/2 → ⁴I_11/2 + ⁴F_9/2 (CR2) and ⁴S_3/2 + ⁴I_15/2 → ⁴I_9/2 + ⁴I_13/2 (CR3).³⁰ Therefore, these CR processes reduce the carrier number in the ²H_11/2 and ⁴S_3/2 levels and lead to fluorescence quenching of the DCL.

For the 522 nm emission band excited with the 980 nm laser, when the Er³⁺ concentration is lower than 4%, the luminescence intensity slowly increases, which is different to the DCL. Furthermore, the quenching concentration of Er³⁺ in UC is higher than DC and the UCL keeps increasing while the DCL has reached its quenching concentration. This can be explained from the UC energy transfer in Fig. 6(d). When the Er³⁺ concentration is lower, the ²H_11/2 level is mainly populated by the excited state absorption (ESA) process. As the concentration increases, the energy transfer between Er³⁺ will be developed. Energy transfer up-conversion (ETU) due to the strong multipolar ion–ion interactions is induced by the short distance between Er³⁺ ions. The phonon-assisted energy transfer also promotes Er³⁺ ions at the ⁴I_11/2 level to consecutively transfer to the ⁴F_7/2 level via the energy transfer of Er³⁺ ions in the excited ⁴I_11/2 level to the ground level ⁴I_15/2.³¹ The ETU process between Er³⁺ can be described as ⁴I_11/2 + ⁴I_11/2 → ⁴F_7/2 + ⁴I_15/2. Therefore, the ETU covers a wide range of Er³⁺ concentrations and the UCL has a higher fluorescence quenching concentration than the DCL. As the concentration of Er³⁺ surpasses 4%, the intensity begins to rapidly decline due to energy migration and concentration quenching appears.³² Furthermore, the UCL excited with the 785 nm laser shown in Fig. S2† revealed only weak green emission at 522 nm and 552 nm and no red emission. This was because the Er³⁺ had a little absorbance at 785 nm and after multi-phonon relaxation there were hardly any electrons in the excited state of ⁴F_9/2.

Trivalent gadolinium ions (Gd³⁺) have a 4f⁷ electronic configuration and have many excellent characteristics when co-doped in host materials.^33,34 Fig. 7(a) and (b) show the DCL and UCL curves for YTiNbO₆: 4% Er, x% Gd (x = 0, 0.5, 1, 2, 3) and the insets (1) and (2) are samples of the UC pictures. It is observed that incorporation of Gd ions in YTiNbO₆: 4% Er does not have any effect on the position of the emission band but does change the UCL and DCL intensity, while for the DCL shown in Fig. 7(a), it is found that the emission intensity of Er³⁺ prominently increases as the Gd³⁺ concentration increases to 1% for YTiNbO₆: 4% Er and then decreases. The co-doping of Gd³⁺ induces higher local distortion in the YTiNbO₆: 4% Er host lattice and increases odd-rank crystal field components which can significantly enhance the DCL intensity.³⁵ When the concentration of Gd³⁺ exceeds 1%, the intensity slowly declines, which may be due to surface defects and the concentration quenching effect. Fig. S3† shows the excitation spectra of YNbTiO₆: 4% Er doped with different concentrations of Gd³⁺ ions at the emission band at 552 nm. No other excitation peaks appeared except the Er³⁺ ions. The relative intensity of the excitation spectra was consistent with the DCL shown in Fig. 7(a). For the UCL shown in Fig. 7(b), the YTiNbO₆: 4% Er co-doped with 0.5% Gd results in a higher UC emission intensity than without Gd³⁺ doping, which can be ascribed to the fact that after Gd³⁺ modified the shape and morphology, the particles formed uniform nanorods which influenced the structural cell distortion, bond vibration frequencies and surface defects, and changing of the surrounding environment of Er³⁺ can impact the energy transfer and phonon relaxation. On further increasing the concentration of Gd³⁺, the UCL decreases evidently. This is mainly due to that the morphology of the nanorods largely increasing the surface quenching sites and the excess Gd³⁺ ions significantly increase the lattice distortion and defect concentration.³⁶

Fig. 7 (a) DC emission spectra of YTiNbO₆: 4% Er, x% Gd (x = 0.5, 1, 2, 3), and (b) UC emission spectra of YTiNbO₆ 4% Er, with the corresponding concentrations of Gd. The insets (1) and (2) show pictures of YTiNbO₆: 4% Er and YTiNbO₆: 4% Er, 0.5% Gd excited with 980 nm laser (600 mW).

For the UC mechanisms of YTiNbO₆: 4% Er, 0.5% Gd, the integer n in the relationship I_up ∝ Pⁿ is the number of NIR photons absorbed to generate a higher frequency photon.^37,38 The pump power dependency is shown in Fig. 8(a). The slopes of In(I_up) versus In(P) for the 552 nm and 668 nm emission bands are found to be 2.24 and 1.76, respectively, which illustrates that the transitions of ⁴S_3/2 → ⁴I_5/2 and ⁴F_9/2 → ⁴I_15/2 involve a two-photon absorption process. The slopes deviate from 2, which might be caused by multi-phonon relaxation to a lower-lying state, nonradiative transition or radiative decay to the ground state.³⁷ The quantum yield of YTiNbO₆: 4% Er, 0.5% Gd was 3.03 × 10⁻⁷ which was lower than the Er³⁺, Yb³⁺ co-doped fluorochemical.³⁹ This was because the phonon energy of YTiNbO₆ was higher than the fluorochemical and Er³⁺ acted as both sensitizer and activator. Compared with the fluorochemical, YTiNbO₆ was more chemically and thermally stable and could expand the up-conversion properties in other application fields.

Fig. 8 (a) Pump power dependence of the green and red UCL integrated intensities of YNbTiO₆: 4% Er, 0.5% Gd under 980 nm excitation. (b) UCL decay curves of Er³⁺ measured at 668 nm for YNbTiO₆: 4% Er and YNbTiO₆: 4% Er, 0.5% Gd. (c) UCL decay curves of Er³⁺ measured at 552 nm for YNbTiO₆: 4% Er and YNbTiO₆: 4% Er, 0.5% Gd.

In order to fully understand the influence of Gd³⁺ on the UCL of YNbTiO₆: 4% Er, it is necessary to compare the decay curves before and after doping Gd³⁺ ions. To quantify the decay constant, all the decay curves I(τ) were fitted with a double exponential equation:

where I(τ) is the luminescence intensity at the maximum of the emission band, a₁ and a₂ are the pre-exponential factors and τ₁ and τ₂ are the fitted decay times. The shorter time τ₁ may be related to the surface Er³⁺ of the phosphors, while the long decay time τ₂ could be attributed to the Er³⁺ located in the bulk phase of the phosphors.⁴⁰ Furthermore, the average decay time constant τ can be determined using the equation:

Fig. 8(b) and (c) show the recorded decay curves of the ⁴F_9/2 and ⁴S_3/2 states of YNbTiO₆: 4% Er before and after doping with 0.5% Gd. Both of the decay times increase after doping with 0.5% Gd compared with the undoped sample, which could be ascribed to the doping Gd³⁺ ions which lead to a lower symmetry of the surrounding environment of Er³⁺ and have an effect on the energy transfer of the UC process. What is more, the red emission decay time is found to be longer than the green emission, which is related to the red decay rate being lower than the green decay rate.⁴¹

Conclusions

Euxenite-type YTiNbO₆ doped with Er³⁺, Gd³⁺ has remarkable UCL and DCL properties. Single Er³⁺ doped YTiNbO₆, and the mechanisms of UC and DC excited at 980 nm and 378 nm were characterized. Both UCL and DCL were observed to have strong green emission at 522 and 552 nm while the UCL also had strong red emission at 668 nm. The concentration of Er³⁺ had different effects on the UCL and DCL and they had different quenching concentrations of Er³⁺ due to the different depopulation routes and energy transfer. When Gd³⁺ ions were co-doped in YTiNbO₆: 4% Er, the morphology was that of uniform nanorods, particle agglomeration got weaker and particle size was smaller. In addition, it has been found that co-doped Gd³⁺ does not change the UC and DC emission peak position but increases the intensity through changing the surrounding environment of the Er³⁺ ions, enhancing the phonon-assisted energy transfer and enhancing the symmetry breaking.

Acknowledgements

This work was supported by projects from the Chinese PLA Medical Science and Technique Foundation (CWS11J243) and the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry.

Notes and references

1. Y. Mao, T. Tran, X. Guo, J. Y. Huang, C. K. Shih, K. L. Wang and J. P. Chang, Adv. Funct. Mater., 2009, 19, 748.
2. G. Liu, Chem. Soc. Rev., 2015, 44, 1635.
3. N. C. Dyck, F. C. J. M. van Veggel and G. P. Demopoulos, ACS Appl. Mater. Interfaces, 2013, 5, 11661.
4. Y. Wang, W. Xu, S. Cui, S. Xu, Z. Yin, H. Song, P. Zhou, X. Liu, L. Xu and H. Cui, Nanoscale, 2015, 7, 1363.
5. J. Han, C. Zhang, F. Liu, B. Liu, M. Han, W. Zou, L. Yang and Z. Zhang, Analyst, 2014, 139, 3032.
6. A. Sedlmeier and H. H. Gorris, Chem. Soc. Rev., 2015, 44, 1526.
7. J. Zhao, Y. Zhu, J. Wu and F. Chen, J. Colloid Interface Sci., 2015, 440, 39.
8. F. Xin, S. Zhao, G. Jia, L. Huang, D. Deng, H. Wang and S. Xu, Mater. Chem. Phys., 2012, 137, 177.
9. X. Huang, S. Han, W. Huang and X. Liu, Chem. Soc. Rev., 2013, 42, 173.
10. R. Dey and V. Kumar Rai, Dalton Trans., 2014, 43, 111.
11. C. S. Lim, Infrared Phys. Technol., 2014, 67, 371.
12. R. Krishnan and J. Thirumalai, New J. Chem., 2014, 38, 3480.
13. O. Ehlert, R. Thomann, M. Darbandi and T. Nann, ACS Nano, 2008, 2, 120.
14. W. Feng, C. Han and F. Li, Adv. Mater., 2013, 25, 5287.
15. A. K. Singh, S. K. Singh and S. B. Rai, RSC Adv., 2014, 4, 27039.
16. X. Qi, R. Illingworth, H. G. Gallagher, T. P. J. Han and B. Henderson, J. Cryst. Growth, 1996, 160, 111.
17. Y. Shi, Y. Wang and Z. Yang, J. Alloys Compd., 2011, 509, 3128.
18. Q. Ma, M. Lu, P. Yang, A. Zhang and Y. Cao, Mater. Res. Bull., 2013, 48, 3677.
19. Q. Ma, M. Lu, P. Yang, A. Zhang and Y. Cao, Luminescence, 2014, 29, 386.
20. Q. Ma, Y. Zhou, A. Zhang, M. Lu, G. Zhou and C. Li, Solid State Sci., 2009, 11, 1124.
21. X. Zhang, G. Zhou, J. Zhou, H. Zhou, P. Kong, Z. Yu and J. Zhan, RSC Adv., 2014, 4, 13680.
22. Y. Wang, W. Xu, S. Cui, S. Xu, Z. Yin, H. Song, P. Zhou, X. Liu, L. Xuand and H. Cui, Nanoscale, 2015, 7, 1363.
23. X. Wu, J. G. Li, Q. Zhu, J. Li, R. Ma, T. Sasaki, X. Li, X. Sun and Y. Sakka, Dalton Trans., 2012, 41, 1854.
24. C. W. A. Paschoala, R. L. Moreira, C. Fantini, M. A. Pimenta, K. P. Surendranb and M. T. Sebastian, J. Eur. Ceram. Soc., 2003, 23, 2661.
25. B. P. Singh, A. K. Parchur, R. K. Singh, A. A. Ansari, P. Singh and S. B. Rai, Phys. Chem. Chem. Phys., 2013, 15, 3480.
26. S. Joseph, M. K. Suresh, J. K. Thomas, A. John and S. Solomon, Int. J. Appl. Ceram. Technol., 2010, 7, 129.
27. S. Solomon, D. B. Dhwaja, G. R. Remya, A. John and J. K. Thomas, J. Alloys Compd., 2010, 504, 151.
28. B. R. Diamente, M. Raudsepp and F. C. J. M. Veggel, Adv. Funct. Mater., 2007, 17, 363.
29. J. Li, J. Sun, J. Liu, X. Li, J. Zhang, Y. Tian, S. Fu, L. Chen, H. Zhong, H. Xia and B. Chen, Mater. Res. Bull., 2013, 48, 2159.
30. H. Lu, W. P. Gillin and I. Hernández, Phys. Chem. Chem. Phys., 2014, 16, 20957.
31. G. Chen, T. Y. Ohulchanskyy, A. Kachynski, H. Ågren and P. N. Prasad, ACS Nano, 2011, 5, 4981.
32. E. M. Chan, G. Han, J. D. Goldberg, D. J. Gargas, A. D. Ostrowski, P. J. Schuck, B. E. Cohen and D. J. Milliron, Nano Lett., 2012, 12, 3839.
33. J. Wang, Y. Cheng, Y. Huang, P. Cai, S. I. Kim and H. J. Seo, J. Mater. Chem. C, 2014, 2, 5559.
34. R. Krishnan and J. Thirumalai, New J. Chem., 2014, 38, 3480.
35. B. P. Singh, A. K. Parchur, R. S. Ningthoujam, A. A. Ansari, P. Singh and S. B. Rai, Dalton Trans., 2014, 43, 4779.
36. S. Zeng, J. Xiao, Q. Yang and J. Hao, J. Mater. Chem., 2012, 22, 9870.
37. J. Zhao, Y. Sun, X. Kong, L. Tian, Y. Wang, L. Tu, J. Zhao and H. Zhang, J. Phys. Chem. B, 2008, 112, 15666.
38. F. Vetrone, V. Mahalingam and J. A. Capobianco, Chem. Mater., 2009, 21, 1847.
39. X. Li, R. Wang, F. Zhang and D. Zhao, Nano Lett., 2014, 14, 3634.
40. D. T. Klier and M. U. Kumke, J. Phys. Chem. C, 2015, 119, 3363.
41. J. Zhao, Z. Lu, Y. Yin, C. McRae, J. A. Piper, J. M. Dawes, D. Jin and E. M. Goldys, Nanoscale, 2013, 5, 944.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra11603a

---