Enhanced anti-stocks luminescence in LaNbO₄:Ln³⁺ (Ln³⁺ = Yb³⁺, Er³⁺/Ho³⁺/Tm³⁺) with abundant color

Abstract

Lanthanide ion doped up-conversion luminescent (UCL) materials, which have the ability to convert near-infrared longer wavelength excitation light into shorter wavelength visible light, have recently drawn much attention. Now, a series of Ln³⁺ (Yb³⁺, Er³⁺/Ho³⁺/Tm³⁺) ions doped LaNbO₄ (LNO) materials were prepared through a two-step route, including a facile sol–gel process and subsequent combustion. In our work, thermogravimetric and differential thermal analysis (TG-DTA), X-ray diffraction (XRD), field emission scanning electron microscopy (FE-SEM), up-conversion luminescence (UCL), energy level diagram, luminescent lifetimes and corresponding CIE were utilized to characterize the properties. A 980 nm laser was employed to excite Yb³⁺ ions to achieve the ²F_7/2 → ²F_5/2 transition, and the LNO:Yb³⁺,Er³⁺/Ho³⁺/Tm³⁺ samples exhibited intense green, yellowish-green and blue emission, respectively. Luminescent spectra at different powers demonstrated the two-photon absorption process; meanwhile, the corresponding energy transfer mechanisms from Yb³⁺ to Ln³⁺ and the emissions of Ln³⁺ under 980 nm laser excitation were described via an energy level diagram. Moreover, multicolor UCL emissions, the optimal intensities and corresponding mechanism were systematically explored via a suite of concentration spectra of different Ln³⁺ ions. At last, the decay lifetimes based on the concentration variation of Yb³⁺ ion demonstrated that the lifetimes are heavily influenced by Yb³⁺ ion.

---

1. Introduction

Up-conversion luminescence (UCL), a process that can convert near-infrared excitation into visible emission, also known as anti-Stokes shift, has recently gained more and more attention.^1–8 The energy transfer process, in which two or more low-energy photons are converted into one high-energy photon, has great potential applications in laser materials, data storage, near-infrared quantum counters, solar energy, traditional light, display technologies, biological imaging and so on.^9–20 Multiple metastable levels of the lanthanides just perfectly meet the requirement of these applications. The lanthanides are filled with the 4f-shell which range from element lanthanum (La) to lutetium (Lu), and their trivalent ions (Ln³⁺) generally possess several excited 4f energy level except for La³⁺, Yb³⁺, Ce³⁺, and Lu³⁺. Because the unique 4f–4f transitions of Ln³⁺ are shielded by the 5s² and 5p⁶ sub-shells, weak electron–phonon are coupling and consequently sharp and narrow f–f transition bands are come into being. Besides, the f–f transitions are Laporte forbidden which can lead to low transition probabilities and substantially long-lived excited states.²¹ Therefore, rare-earth doped up-conversion materials are extremely available for varied application.

UCL can be achieved by singly doped and double doped materials. Ln³⁺ ions such as Er³⁺, Ho³⁺ and Tm³⁺, which have abundant ladder-like arranged energy levels, are usually employed as activators in the UCL materials. However, the capacity absorbing cross-sections of Ln³⁺ is extremely low, and in turn the UCL intensity of singly doped materials is overall low. Additionally, high doping levels can result in harmful cross-relaxation and quenching excitation energy. Therefore, it is essential to reduce the concentration of Ln³⁺ and avoid the quenching effect. As a consequence, a sensitizer which has an abundant absorption cross-section in the near-infrared range is usually co-doped to enhance the UCL intensity. Trivalent Yb possesses an extremely simple energy level structure, including only one ground state level and one excited state level. In addition, the excited state level of Yb³⁺ is higher than the metastable energy levels of the representative up-converting Ln³⁺ (Er³⁺, Ho³⁺, and Tm³⁺) and the ²F_7/2 → ²F_5/2 transition is perfectly resonant with them because the energy state pair of Ln³⁺ is closed to the energy gap of Yb³⁺ (10⁴ cm⁻¹). With the assistance of Yb³⁺, a relatively low (less than 0.02) content of the up-converting activators can realize the UCL emission. In the experiments, the optimal concentrations of Er³⁺/Ho³⁺/Tm³⁺ are 0.02, 0.005 and 0.003, respectively. In our work, the Yb³⁺ ions with excellent optical characteristics are employed as a sensitizer in high-efficiency UCL materials.

Beyond that, it is extremely vital to find a suitable host compound with outstanding optical properties. Firstly, ideal host materials should be inorganic compounds with trivalent rare-earth ions, which ascribe to the similar ionic size and chemical properties of the whole rare-earth ions. Secondly, ideal host materials also require low lattice phonon energy, which can reduce non-radiative loss to increase the radiative emission. Although heavy halides and fluorides exhibit low phonon energy (less than 300 cm⁻¹ and 350 cm⁻¹), the applications are still limited due to their hygroscopic and instability. In comparison, oxides exhibit higher thermal and chemical stability, so appropriate oxides with relatively low phonon energy have more promising application. The third influencing factor is the crystal structure, which can be imputed to the difference of the crystal-fields around activators in various symmetrical matrices. Compared with high symmetry hosts, low symmetry hosts have more probability to achieve f–f transition for up-converting Ln³⁺, and this is because the electronic coupling between 4f energy levels and electronic configuration. Furthermore, the reduction of cation size in the host can lead to the enhancement of the crystal-field around the doped ions and subsequently increase UCL intensity.

Up to now, a series of niobates such as YNbO₄, LaNbO₄, GdNbO₄, CaNb₂O₆ have showed good mechanical and chemical stability, thus they have attracted widespread attention. Lin group have demonstrated that the niobates are a commendable host material in the down-conversion emission.^22–24 La³⁺ possesses the biggest radius which can be replaced by the other rare-earth ions and the symmetry of LaNbO₄ is pretty low, which are beneficial to enhance the optical properties, thus we select LaNbO₄ as our host material.

As far as we know, as of now, the UCL properties of niobate have rarely been reported. In our work, LNO and Ln³⁺ doped LNO have been fabricated via a facile sol–gel and combustion method. In general, the adulteration of various rare-earth ions is the easiest way to achieve multicolor in the UCL emission. Under the irradiation of 980 nm laser, four-color emissions from UCL materials NaYbF₄:Tm/Ho/Er/Yb were first demonstrated by Nann et al.²⁵ In our work, an array of up-conversion luminescent properties, including the up-conversion mechanism, power-dependent spectra, concentration-dependent spectra, lifetimes and CIE chromaticity coordinates. Additionally, we find the optimal concentration to achieve the strongest UCL intensities. It indicates that the samples have great prospect for applications such as 3D display and solid-state lasers.

2. Experimental

2.1. Synthetic procedures

A sequence of nano/micro-materials with chemical composition of La_{1−(y+e/h/t)}NbO₄:yYb³⁺,eEr³⁺/hHo³⁺/tTm³⁺ (y = 0.02–0.25, e = 0.005–0.04, h = 0.0025–0.02, t = 0.001–0.006), where La atoms occupations were considered to be occupied by other rare earth atoms, were synthesized by a facile sol–gel combustion process. Stoichiometric amounts of high-purity chemical compounds, including lanthanum oxide (La₂O₃), ytterbium oxide (Yb₂O₃), erbium oxide (Er₂O₃), holmium oxide (Ho₂O₃), thulium oxide (Tm₂O₃), niobium oxide (Nb₂O₅), citric acid (C₆H₈O₇·H₂O) and ammonium nitrate (NH₄NO₃), were used as the raw materials. The synthetic method is a two-step route, involving the sol–gel progress and the high-temperature combustion progress. First, La(NO₃)₃, Yb(NO₃)₃, Ho(NO₃)₃ and Er(NO₃)₃ aqueous solutions were prepared by dissolving the calculated amount of La₂O₃, Yb₂O₃, Er₂O₃, Ho₂O₃ and Tm₂O₃ into diluted nitric acid at a certain mole ratio, respectively. The residual nitric acid was removed via heating and evaporation. Second, excess hydrofluoric acid (HF, 40%) was employed to dissolve niobium oxide (Nb₂O₅, 99.9%) at 90 °C and the pH was regulated to 9.0 by adding aqueous ammonia solution, then a white precipitate of Nb₂O₅·nH₂O was obtained. Subsequently, the precipitate was washed with deionized water repeatedly by centrifugation to guarantee that the F⁻ ion was entirely removed. Third, citric acid and ammonium nitrate were used to dissolve the precipitate with a mole ratio of 3 : 5 : 1. In this process, citric acid and ammonium nitrate play the roles of complexant and fuel, respectively. Fourth, the RE nitrates were dissolved in the above solution under continuous stirring at a constant temperature of 85 °C to form the homogeneous sol. The transparent sol became a white gel with high viscosity by evaporating the water. After drying the gel for 12 h at 110 °C, a nut-brown precursor was formed. The precursor was directly annealed in air for 3 h at 1000 °C in a furnace and cooled down to room temperature naturally to obtain the final product.

2.2. Characterization

Thermal analysis of the precursor was performed on a thermogravimetric-differential thermal analyser (TG-DTA) (Perkin Elmer Corporation, Diamond TG-DTA) from 30 °C to 1100 °C with a constant heating rate of 20 °C min⁻¹. The X-ray diffraction (XRD) measurements were carried out on a Germany Bruker Axs D8-Focus diffractometer with graphite-monochromatized Cu Kα radiation (λ = 0.15418 nm) at a scanning rate of 0.04 s per step over the 2θ range from 10° to 80°. The morphology and size of the products were obtained using a scanning electron microscope (SEM, Hitachi, S-4800). The UCL spectra, decay times and were recorded on fluorescence spectrometer (Edinburgh FLS 980) equipped with a tuneable 980 nm semiconductor (1 W laser) as the excitation source.

3. Results and discussion

3.1. Structural, phase and morphological analysis

The TG-DTA curves of pure LNO precursor xerogel powder dried for 12 h at 110 °C is shown in Fig. 1. Throughout the temperature range from 30 °C to 1100 °C, there are three main weight loss stages in the TG curve. The first stage is from room temperature to 235 °C. In this stage, the evaporation of water, the combustion of organic phases (such as citric acid) and the thermal decomposition of NO₃⁻ collectively result in weight loss (at approximately 40.83%), an endothermic peak (at approximately 57 °C) and a strong exothermic peak (at approximately 212 °C). At the second stage, throughout which the temperature ranges from 235 °C to 508 °C, the weight loss is about 37.65%, and a high-intensitive exothermic peak occurs at 486 °C, which can be ascribed to the further combustion of the remaining organic materials. The last stage of weight loss occurring between 508 °C and 645 °C is about 2.52%, corresponding to a faint exothermic weak at 632 °C, which is mainly because of the process of the phase transformation from tetragonal to monoclinic and higher crystallization. The tetragonal phase can not be existent under high temperature.²⁶ No obvious weight loss is observed above 632 °C, which indicates that all organic compounds in the precursor had been decomposed completely below 632 °C, and the crystallized LNO inorganic phase is formed.

Fig. 1 TG-DTA curves of pure LNO precursor xerogel powder.

The composition, crystal structure, and phase purity of the as-prepared samples LNO host under various temperatures and times, LNO:0.10Yb³⁺,0.02Er³⁺, LNO:0.15Yb³⁺,0.005Ho³⁺ and LNO:0.15Yb³⁺,0.003Tm³⁺ are respectively determined from the XRD patterns, which are clearly shown in Fig. 2. All of the diffraction peaks could be well matched with the XRD pattern of monoclinic LaNO₄ phase corresponding to standard reference JCPDS card no. 81-1973. As shown in Fig. 3, the crystallization belongs to monoclinic system, and space group is I112/b (15), and cell parameters are a = 5.5634 Å, b = 5.2030 Å, c = 11.5227 Å, V = 332.69(3) ų, and Z = 4. Significantly, as the annealing temperature increases from 500 °C to 1000 °C and reaction time grows from 0.5 h to 3 h, the diffraction peaks become obviously stronger and sharper, implying that the high annealing temperature and longer reaction time contribute to the high crystallinity of the samples. Based on the higher crystallinity of the heat treated samples, we can choose more appropriate condition to prepare the samples. In general, the samples along with high crystallinity show higher emission intensity. Further, no traces of impurity phases are observed, demonstrating little change of this crystal structure accompanied by introducing such Yb³⁺, Er³⁺, Ho³⁺ and Tm³⁺ ions into the host. This is because such Ln³⁺ ions have similar radius and the La³⁺ ions is just located at the central position surrounded by eight oxygen atoms and the LaNbO₄ phase has a low symmetry crystal structure. Only one kind of La cationic position with a Wyckoff position 4e could be taken in for activators due to the analogical ionic radius of La³⁺ [r = 1.061 Å, CN = 8], Yb³⁺ (r = 0.858 Å, CN = 8), Er³⁺ (r = 0.881 Å, CN = 8), Ho³⁺ (r = 0.894 Å, CN = 8) and Tm³⁺ (r = 0.87 Å, CN = 8). Furthermore, in the inset of Fig. 2, the diffraction peaks (−121) (121) are slightly shifted to a higher degree and the crystallinity lower down accompanied with Yb³⁺, Er³⁺/Ho³⁺/Tm³⁺ doped into the host material.

Fig. 2 XRD patterns of the as-prepared (a) LaNbO₄ host with different temperature, time and the standard pattern (JCPDS 81-1973) of LaNbO₄. (b) LaNbO₄:0.10Yb³⁺,0.02Er³⁺, LNO:0.15Yb³⁺,0.005Ho³⁺, LNO:0.15Yb³⁺,0.003Tm³⁺ samples.

Fig. 3 Crystal structure of the monoclinic LaNbO₄ phase.

The SEM images of the as-prepared LNO, LNO:0.10Yb³⁺,0.02Er³⁺, LNO:0.15Yb³⁺,0.005Ho³⁺, LNO:0.15Yb³⁺,0.003Tm³⁺ samples are shown in Fig. 4. It can be clearly found that the morphology and particle size of the as-prepared samples have been slightly influenced by the dopant, which is corresponding to the XRD patterns.

Fig. 4 SEM images of the as-prepared (a) LNO, (b) LNO:0.10Yb³⁺,0.02Er³⁺, (c) LNO:0.15Yb³⁺,0.005Ho³⁺, (d) LNO:0.15Yb³⁺,0.003Tm³⁺.

The agglomeration of the samples is observed especially in un-doped LNO sample, which is attributed to the high-temperature combustion process. The morphology and size of the samples are not comparatively homogeneous based on the same reason. The diameter of all the samples ranges from 200 nm to 500 nm except that the morphology become more uniform along with Yb³⁺, Er³⁺, Ho³⁺, Tm³⁺ co-doped in LNO. The TEM images which shown in Fig. S1† are also demonstrated the phenomenon.

3.2. Upconversion photoluminescence properties and mechanism

In order to obtain multicolor in the UCL, Er³⁺, Ho³⁺, Tm³⁺ together with Yb³⁺ are doped in LNO, respectively. The UCL spectra of LNO:Yb³⁺,Er³⁺/Ho³⁺/Tm³⁺ under 980 nm excitation and their corresponding CIE (Commission Internationale de I'Eclairage 1931 chromaticity) coordinates positions are shown in Fig. 5. Sharp characteristic emission peaks are all obvious in the UCL spectra. Fig. 5a exhibit the UCL spectrum of the as-prepared LNO:0.10 Yb³⁺,0.02 Er³⁺ and there are mainly two typical emission bands. The shorter-wavelength (green) emission in the range of 510 to 580 nm, corresponding to the transition from the excited state ²H_11/2 and ⁴S_3/2 to the ground state ⁴I_15/2. While the longer-wavelength (red) emission ranging from 635 to 690 nm, which belongs to ⁴F_9/2 → ⁴I_15/2 transition of Er³⁺. Because the intensity of the green emission is obviously stronger than the red emission, the final color is green and CIE chromaticity diagram for it is (0.26790, 0.71130) as shown in Fig. 5d(E) under the excitation of 980 nm laser.

Fig. 5 The up-conversion emission spectra of (a) LNO:0.10Yb³⁺,0.02Er³⁺, (b) LNO:0.15Yb³⁺,0.005Ho³⁺, (c) LNO:0.15Yb³⁺,0.003Tm³⁺, (d) the CIE chromaticity coordinates of the as-prepared samples under the 980 laser excitation.

Fig. 5b shows the UCL spectrum of the as-prepared LNO:0.15Yb³⁺,0.005Ho³⁺, radiative relaxation processes of ⁵F₄/⁵S₂ → ⁵I₈, ⁵F₅ → ⁵I₈ and ⁵S₂ → ⁵I₇ levels corresponding to green (about 540 nm), red (about 660 nm) and NIR (about 753 nm) emissions, respectively. In addition, the CIE chromaticity coordinates of LNO:0.15Yb³⁺,0.003Ho³⁺ is (0.32549, 0.66416) shown in Fig. 5d(H) under the excitation of 980 nm laser.

Fig. 5c displays the UCL spectrum of the as-prepared LNO:0.15Yb³⁺,0.003Tm³⁺, which exhibit one extremely strong blue emission, one weak green emission and two very weak red emission ascribed to ¹G₄ → ³H₆ (located at 475 nm), ¹D₂ → ³H₅ (between 515 nm to 560 nm) and ¹G₄ → ³F₄, ³F₃ → ³H₆ (ranging from 625 nm to 720 nm) respectively. The final color is relative pure blue and CIE chromaticity diagram is (0.13167, 0.13330) as shown in Fig. 5d(T) under the excitation of 980 nm laser.

As is well-known to all, the UCL is an anti-Stokes luminescent process, in which two or more low-energy photons are consecutively absorbed and followed by emission of a higher-energy photon in quick succession.²⁷ To explore the populating mechanism, the up-conversion luminescent properties are compared at different powers, which are shown in Fig. 6a–c. According to the spectra, the transition intensity of the as-prepared samples becomes stronger corresponding to the increasing pump powers.

Fig. 6 UCL spectra of the as-prepared (a) LNO:0.10Yb³⁺,0.02Er³⁺, (b) LNO:0.15Yb³⁺,0.005Ho³⁺, (c) LNO:0.15Yb³⁺,0.003Tm³⁺ under the excitation of 980 nm laser at the different power and the corresponding power dependence of UCL in (d) LNO:0.10Yb³⁺,0.02Er³⁺, (e) LNO:0.15Yb³⁺,0.005Ho³⁺, (f) LNO:0.15Yb³⁺,0.003Tm³⁺.

For any unsaturated UCL process, the visible emission intensity (I_em) and the pump power P of infrared excitation resource is expressed as

I_em ∝ Pⁿ

where n is the positive integer of NIR photons absorbed to convert a higher-energy, which can be required from the slope in Fig. 6d–f.^28–30 By fitting the data points, the slopes of log(I_em) versus log(P) for the ²H_11/2 → ⁴I_15/2, ⁴S_3/2 → ⁴I_15/2, ⁴F_9/2 → ⁴I_15/2 transitions of LNO:0.10Yb³⁺,0.02Er³⁺ are calculated to be 1.865, 1.861 and 1.969, respectively. (shown in Fig. 6d.) The strait lines illustrate that two photons are absorbed in the UCL process. For LNO:0.15Yb³⁺,0.005Ho³⁺, the slopes at ⁵F₄/⁵S₂ → ⁵I₈, ⁵F₅ → ⁵I₈ and ⁵S₂ → ⁵I₇ transitions are 2.041, 2.016 and 2.195, respectively, indicating a two-photon absorption process. (shown in Fig. 6e.) The value of n for LNO:0.15Yb³⁺/0.003Tm³⁺ are 2.331, 2.352 and 1.677, which are corresponding to ¹G₄ → ³H₆, ¹G₄ → ³F₄ and ³F₃ → ³H₆, respectively, which demonstrate the transition from Yb³⁺ to Tm³⁺ involving two NIR photons. (shown in Fig. 6f.) However, as the photons enter the crystals, the loss of energy and nonluminous relaxation cause the emitted energy to be less than the absorbed energy, and thus the slope is less than 2. But the slopes of Ho³⁺ and Tm³⁺ are nearly more than 2 and this phenomenon illustrates existence of multiphoton absorption in LNO:Yb³⁺,Ho³⁺ and LNO:Yb³⁺,Tm³⁺ samples.

In general, three luminescence mechanism have been proved in the UCL process, and they are ground/excited state absorption (GSA/ESA), energy transfer up-conversion (ETU) and photon avalanche (PA).³¹ In the Fig. 6d–f, each slope of all the samples is not changed, which demonstrates PA mechanism is not involved in the UCL process. To explore the UCL populating mechanism, the typical energy level diagrams and the possible mechanism for LNO:Yb³⁺/Er³⁺, LNO:Yb³⁺/Ho³⁺, LNO:Yb³⁺/Tm³⁺ are exhibited in Fig. 7a–c. The solid, dotted and curly arrows represent the emission, energy transfer, and multi-phonon relaxation process, respectively. In the Yb³⁺ and Ln³⁺ co-doped LNO samples, Yb³⁺ ions absorb energy from the 980 nm laser excitation, while Er³⁺, Ho³⁺ and Tm³⁺ generate visible emission. First of all, the ground state ²F_7/2 of two Yb³⁺ are promoted to the ²F_5/2 by absorbing the excitation of 980 nm laser. The transition of ²F_7/2 → ²F_5/2 is usually resonant with the f–f transition of Er³⁺, Ho³⁺ and Tm³⁺, prompting the efficient energy transform from Yb³⁺ to these ions. For LNO:Yb³⁺/Er³⁺, the ground state ⁴I_15/2 of Er³⁺ is excited to the state ⁴I_11/2. Then under the effect of ET mechanism of Yb³⁺ → Er³⁺, the excited electrons on the level ⁴I_11/2 are partly excited to the energy level ⁴F_7/2. Afterwards, the electrons relax from ⁴F_7/2 to ²H_11/2, ⁴S_3/2, generating green ²H_11/2, ⁴S_3/2 → ⁴I_15/2 transition. In another case, if the excited electrons on the energy level ⁴I_11/2 are relaxed to the energy level ⁴I_13/2 accompanied by another timely ET from Yb³⁺ to the energy level ⁴I_13/2 of Er³⁺, the energy from ⁴I_13/2 will be transferred to ⁴F_9/2, and subsequently return to the ground state via radiative transition, producing red ⁴F_9/2 → ⁴I_15/2 transition.^32–35 For LNO:Yb³⁺/Ho³⁺, the mechanism is somewhat different from LNO:Yb³⁺/Er³⁺. The ground state absorption (GSA) is the phonon assisted process because the excitation energy between donor and acceptor could not match well (about 1580 cm⁻¹) and hence there is no resonant transition. Firstly, the energy of Ho³⁺ ions is excited from the ground state ⁵I₈ to the excited state ⁵I₆ via phonon-assisted ET. Following this process, the second photon absorbed from energy level ²F_5/2 of Yb³⁺ or the second pump photons make Ho³⁺ excite to the state ⁵F₄, ⁵S₂ and then decay to the ground state or the state ⁵I₇, in which process green ⁵F₄, ⁵S₂ → ⁵I₈ and NIR ⁵F₄, ⁵S₂ → ⁵I₇ are produced. If direct multiphotons relax to the energy level from ⁵F₄, ⁵S₂ to ⁵F₅, a strong red emission ⁵F₅ → ⁵I₈ can be generated.^36,37 Non-radiative multiphoton relaxation rate also determines the efficiency of the UCL through impacting the population of the intermediate and the emitting levels. The non-radiative multiphoton relaxation rate constant is expressed as below:

k_nr ∝ exp(−βΔE/hω_max)

where k_nr represents the multiphoton relaxation rate constant, β is the empirical constant, ΔE is the energy gap, and hω_max is the highest energy vibrational mode of the host lattice. From the formula, k_nr decreases exponentially accompanied by the rise of energy gap. Compared with Yb³⁺ and Er³⁺, Ho³⁺ is more likely to realize non-radiative multiphoton relaxation, which probably reducing the efficiency of Ho³⁺.³⁸ For LNO:Yb³⁺/Tm³⁺, Tm³⁺ ions are excited from ground state ³H₆ level to ¹G₄ level by three energy transfer processes and the mechanism is expressed as follows: firstly, ³H₆ (Tm) + ²F_5/2 (Yb) → ³H₅ (Tm) + ²F_7/2 (Yb), and then the energy level ³H₅ relaxes to ³F₄. Secondly, ³F₄ (Tm) + ²F_5/2 (Yb) → ³F₂ (Tm) + ²F_7/2 (Yb), and next go back to the energy level ³H₄ or ³F₃ by relaxation progress corresponding to the red emission ³F₃ → ³H₆. Lastly, ³H₄ (Tm) + ²F_5/2 (Yb) → ¹G₄ (Tm) + ²F_7/2 (Yb). The photons transfer from the energy level ¹G₄ to the ground state or ³F₄ level, corresponding to blue color ¹G₄ → ³H₆ or red color ¹G₄ → ³F₄. Another excited energy from Yb³⁺ excites the photons from ¹G₄ to ¹D₂ and decay to ³H₅ level resulting in relative weak green emission ¹D₂ → ³H₅.^39–41

Fig. 7 The energy level diagrams of LNO:Yb³⁺/Er³⁺, LNO:Yb³⁺/Ho³⁺, LNO:Yb³⁺/Tm³⁺ samples up-conversion mechanisms under 980 nm laser excitation.

The concentration of sensitizer and activator is a significant factor of the UCL properties. The concentration affects the relative amount of Ln³⁺ in the samples and the average distance between adjacent dopant ions, leading to a significant impact on the optical properties of the samples. Fig. 8 exhibit various UCL spectra under 980 nm excitation for LNO:yYb³⁺,eEr³⁺/hHo³⁺/tTm³⁺ (y = 0.02–0.25, e = 0.005–0.04, h = 0.0025–0.02, t = 0.001–0.006) with different concentrations of up-conversion Ln³⁺ ions, respectively. All the as-prepared samples have the similar spectral profiles apart from the UCL intensity, indicating the crystal field is nearly consistent. Overall, as the concentration of Yb³⁺, Er³⁺/Ho³⁺/Tm³⁺ ions increases, the emission intensities of them in the LNO:yYb³⁺,eEr³⁺/hHo³⁺/tTm³⁺ systems are clearly rising to the maximum at y = 0.1, e = 0.02; y = 0.15, h = 0.005; y = 0.15, t = 0.003 and descending subsequently. The spectra of Er³⁺, Ho³⁺ and Tm³⁺ with different concentration are shown in Fig. 8a–c. When the concentrations of Er³⁺, Ho³⁺ and Tm³⁺ are lower than the maximum, the intensities of UCL keep rising corresponding to the increasing concentration. Energy transition is induced by the shorter distance between the activators, so the probability of energy transfers from sensitizer to activator increases.

Fig. 8 Dependence of UCL emission spectra of (a) LNO:0.05Yb³⁺,eEr³⁺, (b) LNO:0.05Yb³⁺,hHo³⁺, (c) LNO:0.05Yb³⁺,tTm³⁺, (d) LNO:yYb³⁺,0.02Er³⁺, (e) LNO:yYb³⁺,0.005Ho³⁺, (f) LNO:yYb³⁺,0.003Tm³⁺.

Under the effect of cross-relaxation, the intensities decline occurs, when the concentrations surpass the optimal one. In Fig. 8d–f, as Yb³⁺ absorbing energy from 980 nm laser excitation, more and more energy can be absorbed and transferred to Ln³⁺ ions, leading to the enhancement of UCL intensities.⁴² However, once the concentration of Yb³⁺ ions exceeds the maximum, concentration quenching occurs immediately. There are two explanations to account for the phenomenon: on the one hand, the pairing or aggregation of Yb³⁺ ions could induce a quenching effect. On the other hand, excessive Yb³⁺ ions act as the energy migrating bridge to defects, which can waste the excited energy through the non-radiative energy transition from Yb³⁺ ions to the defects.⁴³ The large absorption cross-section of Yb³⁺, which minimizes the cross-relaxation energy loss, makes the activator content is relatively low(less than 2 mol%). In addition, from the spectra of Fig. 8a and d, the green emission of quenching concentration of LNO:yYb³⁺,eEr³⁺ is not equal to the red emission, which the maximum concentrations are respectively 0.1 and 0.15. The enhancement of Yb³⁺ concentration leads to a relative increase in the red emission intensity of Er³⁺. A similar phenomenon has been found by other teams, and the factor on green emission and red emission has been studied. The reason for the variation of relative intensity of red (⁴F_9/2 → ⁴I_15/2) and green emission (²H_11/2/⁴S_3/2 → ⁴I_15/2) is possibly the cross-relaxation process of ²H_11/2/⁴S_3/2 (Er) + ²F_7/2 (Yb) → ⁴I_13/2 (Er) + ²F_5/2 (Yb), resulting in the population increasing on the ⁴F_9/2 level. Compared with ⁴F_9/2 → ⁴I_15/2 transitions, the ²H_11/2/⁴S_3/2 → ⁴I_15/2 are restrained by the cross-relaxation process. In consequence, the phenomenon can be attributed to back-energy transition from Er³⁺ to Yb³⁺.^44–46

The corresponding CIE coordinates positions as well as the chromaticity coordinates for the as-prepared LNO:yYb³⁺,eEr³⁺/hHo³⁺/tTm³⁺ samples under 980 nm laser excitation are shown in Fig. 9 and Table 1, respectively. It can be found that the emitted light of LNO:yYb³⁺,Er³⁺ samples shift from shorter wavelength to longer wavelength accompanied with increasing value of y, which exactly demonstrates the back-energy transition from Er³⁺ to Yb³⁺. In addition, the chromaticity coordinates generally move regularly. When the ratio of Yb³⁺/Ho³⁺ increases, the CIE positions shift from light yellow to green. It may be caused by the concentration of Yb³⁺. For LNO:yYb³⁺/tTm³⁺, T1(LNO:0.02Yb³⁺,0.003Tm³⁺), T2(LNO:0.05Yb³⁺,0.003Tm³⁺) are respectively situated at (0.21049, 0.36642), (0.16614, 0.25068) corresponding to cyan and light blue and T3–T14 are assembled around dark blue area. Thus we can achieve different color through controlling the relative concentrations of sensitizer and activators.

Fig. 9 CIE chromaticity diagram for LNO:yYb³⁺,eEr³⁺(E1–E10), LNO:yYb³⁺,hHo³⁺(H1–H10), LNO:yYb³⁺,tTm³⁺(T1–T14) (y = 0.02–0.25, e = 0.005–0.04, h = 0.0025–0.02, t = 0.001–0.006).

Table 1 Variations of CIE chromaticity coordinate for LNO:yYb³⁺,eEr³⁺, LNO:yYb³⁺,hHo³⁺, LNO:yYb³⁺,tTm³⁺ (y = 0.02–0.25, e = 0.005–0.04, h = 0.0025–0.02, t = 0.001–0.006)

No. of points	LNO:yYb^3+,0.02Er^3+		CIE(x, y)
E1	y = 0.02	e = 0.02	(0.26514, 0.71330)
E2	y = 0.05	e = 0.02	(0.26473, 0.71428)
E3	y = 0.10	e = 0.02	(0.26790, 0.71130)
E4	y = 0.15	e = 0.02	(0.26973, 0.70855)
E5	y = 0.20	e = 0.02	(0.28336, 0.69640)

---

No. of points	LNO:0.05Yb^3+,eEr^3+		CIE(x, y)
E6	y = 0.05	e = 0.005	(0.26614, 0.71319)
E7	y = 0.05	e = 0.01	(0.27386, 0.70567)
E8	y = 0.05	e = 0.02	(0.28428, 0.68682)
E9	y = 0.05	e = 0.03	(0.29076, 0.69105)
E10	y = 0.05	e = 0.04	(0.29964, 0.68363)

---

No. of points	LNO:yYb^3+,0.05Ho^3+		CIE(x, y)
H1	y = 0.02	h = 0.05	(0.39424, 0.59712)
H2	y = 0.05	h = 0.05	(0.37109, 0.61964)
H3	y = 0.10	h = 0.05	(0.33925, 0.65076)
H4	y = 0.15	h = 0.05	(0.32549, 0.66416)
H5	y = 0.20	h = 0.05	(0.32951, 0.66023)

---

No. of points	LNO:0.05Yb^3+,hHo^3+		CIE(x, y)
H6	y = 0.05	h = 0.0025	(0.36330, 0.62725)
H7	y = 0.05	h = 0.005	(0.36480, 0.62580)
H8	y = 0.05	h = 0.0075	(0.40780, 0.58411)
H9	y = 0.05	h = 0.01	(0.36171, 0.62881)
H10	y = 0.05	h = 0.02	(0.37292, 0.61810)

---

No. of points	LNO:yYb^3+,0.003Tm^3+		CIE(x, y)
T1	y = 0.02	t = 0.003	(0.21049, 0.36642)
T2	y = 0.05	t = 0.003	(0.16614, 0.25068)
T3	y = 0.10	t = 0.003	(0.16215, 0.14648)
T4	y = 0.15	t = 0.003	(0.13167, 0.13330)
T5	y = 0.20	t = 0.003	(0.13410, 0.13978)
T6	y = 0.25	t = 0.003	(0.14384, 0.14477)
T7	y = 0.30	t = 0.003	(0.13575, 0.14112)
T8	y = 0.35	t = 0.003	(0.13235, 0.13685)

---

No. of points	LNO:0.05Yb^3+,tTm^3+		CIE(x, y)
T9	y = 0.05	t = 0.001	(0.15857, 0.16977)
T10	y = 0.05	t = 0.002	(0.17098, 0.14516)
T11	y = 0.05	t = 0.003	(0.14840, 0.13937)
T12	y = 0.05	t = 0.004	(0.16342, 0.15238)
T13	y = 0.05	t = 0.005	(0.16899, 0.13647)
T14	y = 0.05	t = 0.006	(0.17106, 0.14151)

---

To elucidate the quenching effect, we studied lifetimes of LNO:yYb³⁺,0.02Er³⁺ (monitored at 552 nm and 669 nm for ⁴S_3/2 → ⁴I_15/2 and ⁴F_9/2 → ⁴I_15/2), LNO:yYb³⁺,0.005Ho³⁺ (monitored at 539 nm and 660 nm for ⁵F₄/⁵S₂ → ⁵I₈ and ⁵F₅ → ⁵I₈) and LNO:yYb³⁺,0.003Tm³⁺ (monitored at 477 nm and 649 nm for ¹G₄ → ³H₆ and ¹G₄ → ³F₄) under the excitation of a 980 nm laser as shown in Fig. 10. The decay curves of characteristic emissions are obviously changed with the concentration variation of Yb³⁺. Generally, double-exponential decay behavior of the activators is usually accompanied with the energy transition from sensitizer to activator. Thus all of the samples can be well fitted with the double-exponential formula as follows:

I(τ) = I₁ exp(−t/τ₁) + I₂ exp(−t/τ₂)

in the formula, I represents the fluorescent intensity, I₁, τ₁ and I₂, τ₂ are the shorter and longer lifetime constants, and t is the time. On the basis of above parameters, decay time can be calculated by the formula τ =(I₁τ₁² + I₂τ₂²)/(I₁τ₁ + I₂τ₂), the values of calculated lifetimes were concluded and exhibited in Fig. 10. In general, we can observe that the value of τ decreases monotonously along with the increasing concentration of Yb³⁺ ion.

Fig. 10 Dependence of UCL decay lifetimes of (a and b) LNO:yYb³⁺,0.02Er³⁺ monitored at 552 nm and 669 nm, (c and d) LNO:yYb³⁺,0.005Ho³⁺ monitored at 539 nm and 660 nm, (e and f) LNO:yYb³⁺,0.003Tm³⁺ monitored at 477 nm and 649 nm, respectively.

4. Conclusions

In summary, a series of Ln³⁺ (Yb³⁺, Er³⁺/Ho³⁺/Tm³⁺) ions doped LNO materials have been successfully prepared through a facile sol–gel process and subsequent combustion, and the up-conversion results are presented. The results of TG-DTA, XRD and SEM analyses demonstrate that the samples annealed at 1000 °C present as the monoclinic phase of LNO with no additional phases and exhibit homogeneous and spherical grains. Under 980 nm laser excitation, LNO:Yb³⁺/Er³⁺, LNO:Yb³⁺/Ho³⁺, LNO:Yb³⁺/Tm³⁺ samples exhibit intense up-conversion emissions. The power-dependent UCL spectra of LNO:Yb³⁺,Er³⁺/Ho³⁺/Tm³⁺ indicate a two-photon absorption is the main process. The corresponding mechanism, including the emission, energy transfer, multi-phonon relaxation process, has been discussed detailedly. UCL multicolor fine-tuning can be achieved via changing the concentration of the dopant ions. We also receive the optimal UCL intensity, which are separately LNO:0.10Yb³⁺/0.02Er³⁺, LNO:0.15Yb³⁺/0.005Ho³⁺, LNO:0.15Yb³⁺/0.003Tm³⁺, respectively.

The relative variation of red and green emission for LNO:yYb³⁺/eEr³⁺ has also given detailed explanation. And meanwhile, spectra, emitting colors as well as lifetimes have been certified the relation with the concentration of Yb³⁺ ion. All in all, it can be concluded that we have found an excellent matrix to achieve efficient UCL emission. The UCL properties have shown potential application in the field of 3D display and solid-state lasers and provided a new insight to design a novel UCL materials with high efficiency.

Acknowledgements

The authors would like to acknowledge the financial support from National Science Foundation of Shandong Province (ZR2016EMM20, ZR2016EMQ09), the Science and Technology Foundation of Shenzhen, Shenzhen Science and Technology Innovation Committee (JCYJ20160331173823401) and National Science Foundation of China (51372138).

Notes and references

1. M. Haase and H. Schäfer, Angew. Chem., Int. Ed., 2011, 50, 5808.
2. X. Teng, Y. Zhu, W. Wei, S. Wang, J. Huang, R. Naccache, W. Hu, A. Y. Tok, Y. Han, Q. Zhang, Q. Fan, W. Huang, J. A. Capobianco and L. Huang, J. Am. Chem. Soc., 2012, 134, 8340.
3. M. Challenor, P. Gong, D. Lorenser, M. Fitzgerald, S. Dunlop, D. D. Sampson and K. S. Iyer, ACS Appl. Mater. Interfaces, 2013, 5, 7875.
4. Q. C. Sun, H. Mundoor, J. C. Ribot, V. Singh, I. I. Smalyukh and P. Nagpal, Nano Lett., 2013, 14, 101.
5. J. H. Kim and J. H. Kim, ACS Photonics, 2015, 2, 633.
6. S. Guo, X. Xie, L. Huang and W. Huang, ACS Appl. Mater. Interfaces, 2015, 8, 847.
7. K. Yamamoto, M. Fujii, S. Sowa, K. Imakita and K. Aoki, J. Phys. Chem. C, 2015, 119, 1175.
8. X. Chen, W. Xu, H. Song, C. Chen, H. Xia, Y. Zhu, D. Zhou, S. Cui, Q. Dai and J. Zhang, ACS Appl. Mater. Interfaces, 2016, 8, 9071.
9. T. Liu, X. Bai, C. Miao, Q. Dai, W. Xu, Y. Yu, Q. Chen and H. Song, J. Phys. Chem. C, 2014, 118, 3258.
10. A. Stepuk, G. Casola, C. M. Schumacher, K. W. Krämer and W. J. Stark, Chem. Mater., 2014, 26, 2015.
11. X. Ma and X. Ni, J. Nanopart. Res., 2013, 15, 1.
12. J. Yu, Y. Yang, R. Fan, D. Liu, L. Wei, S. Chen, L. Li, B. Yang and W. Cao, Inorg. Chem., 2014, 53, 8045.
13. J. Jin, Y. J. Gu, C. W. Y. Man, J. Cheng, Z. Xu, Y. Zhang, H. Wang, V. H. Y. Lee, S. H. Cheng and W. T. Wong, ACS Nano, 2011, 5, 7838.
14. R. Wei, Z. Wei, L. Sun, J. Z. Zhang, J. Liu, X. Ge and L. Shi, ACS Appl. Mater. Interfaces, 2015, 8, 400.
15. V. Muhr, S. Wilhelm, T. Hirsch and O. S. Wolfbeis, Acc. Chem. Res., 2014, 47, 3481.
16. X. Yu, M. Li, M. Xie, L. Chen, Y. Li and Q. Wang, Nano Res., 2010, 3, 51.
17. S. M. Ametamey, M. Honer and P. A. Schubiger, Chem. Rev., 2008, 108, 1501.
18. T. V. Esipova, X. Ye, J. E. Collins, S. Sakadžić, E. T. Mandeville, C. B. Murray and S. A. Vinogradov, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 20826.
19. D. Patel, A. Kell, B. Simard, B. Xiang, H. Y. Lin and G. Tian, Biomaterials, 2011, 32, 1167.
20. X. Ge, L. Dong, L. Sun, Z. Song, R. Wei, L. Shi and H. Chen, Nanoscale, 2015, 7, 7206.
21. F. Auzel, Chem. Rev., 2004, 104, 139.
22. Y. Lu, X. Tang, L. Yan, K. Li and X. Liu, J. Phys. Chem. C, 2013, 117, 21972.
23. K. Li, Y. Zhang, X. Li, M. Shang, H. Lian and J. Lin, Phys. Chem. Chem. Phys., 2015, 17, 4283.
24. K. Li, X. Liu, Y. Zhang, X. Li, H. Lian and J. Lin, Inorg. Chem., 2014, 54, 323.
25. O. Ehlert, R. Thomann, M. Darbandi and T. Nann, ACS Nano, 2008, 2, 120.
26. G. Blasse and L. H. Brixner, Chem. Phys. Lett., 1990, 173, 409.
27. F. Auzel, Chem. Rev., 2004, 104, 139.
28. J. Zhao, Y. Sun, X. Kong, L. Tian, Y. Wang, L. Tu, J. Zhao and H. Zhang, J. Phys. Chem. B, 2008, 112, 15666.
29. F. Vetrone, V. Mahalingam and J. A. Capobianco, Chem. Mater., 2009, 21, 1847.
30. R. Balakrishnaiah, D. W. Kim, S. S. Yi, K. Jang, H. S. Lee and J. H. Jeong, Opt. Mater., 2009, 31, 959.
31. F. Wang and X. Liu, Chem. Soc. Rev., 2009, 38, 976.
32. J. Tang, L. Chen, J. Li, Z. Wang, J. Zhang, L. Zhang, Y. Luo and X. Wang, Nanoscale, 2015, 7, 14752.
33. Y. Tian, Y. Tian, P. Huang, L. Wang, Q. Shi and C. Cui, Chem. Eng. J., 2016, 297, 26.
34. F. Zhang, G. Li, W. Zhang and Y. L. Yan, Inorg. Chem., 2015, 54, 7325.
35. W. Yao, Q. Tian, J. Liu, Z. Wu, S. Cui, J. Ding, Z. Dai and W. Wu, J. Mater. Chem. C, 2016, 4, 6327.
36. P. Anurag, K. Vinod, R. E. Kroon and H. C. Swart, J. Alloys Compd., 2016, 672, 190.
37. A. Dwivedi, A. K. Singh and S. B. Rai, Dalton Trans., 2014, 43, 15906.
38. J. M. F. Van Dijk and M. F. H. Schuurmans, J. Chem. Phys., 1983, 78, 5317.
39. M. Rathaiah, P. Haritha, A. D. Lozano-Gorrín, P. Babu, C. K. Jayasankar, U. R. Rodríguez-Mendoza, V. Lavín and V. Venkatramu, Phys. Chem. Chem. Phys., 2016, 18, 14720.
40. H. Guo, N. Dong, M. Yin, W. Zhang, L. Lou and S. Xia, J. Phys. Chem. B, 2004, 108, 19205.
41. M. Li, X. Liu, L. Liu, B. Ma, B. Li, X. Zhao, W. Tong and X. Wang, Inorg. Chem., 2016, 3, 1082.
42. D. Xu, C. Liu, J. Yan, S. Yang and Y. Zhang, J. Phys. Chem. C, 2015, 119, 6852.
43. J. Wang, R. Deng, M. A. MacDonald, B. Chen, J. Yuan, F. Wang, D. Chi, T. S. A. Hor, P. Zhang, G. Liu, Y. Han and X. Liu, Nat. Mater., 2014, 13, 157.
44. G. Chen, G. Somesfalean, Y. Liu, Z. Zhang, Q. Sun and F. Wang, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 75, 195204.
45. D. Xu, C. Liu, J. Yan, S. Yang and Y. Zhang, J. Phys. Chem. C, 2015, 119, 6852.
46. A. Li, D. Xu, Y. Zhang, H. Lin, S. Yang, Z. Chen and Y. Shao, J. Am. Ceram. Soc., 2016, 99, 1657.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra28592a

---