1. Introduction

Rare earth ions have closely spaced energy levels and unique intra 4f transitions, which afford them the ability to absorb one or more low-energy near-infrared (NIR) photons and subsequently convert them to high-energy emissions [1]. This anti-Stokes emission process is well known as upconversion (UC). In Yb³⁺ sensitized rare earth ions doped materials, efficient NIR-to-ultraviolet (UV) UC emissions have widely been investigated due to the need for developing short-wavelength solid-state lasers and photodynamic therapies in biomedicine [2–4]. Therefore studies of new approaches to obtain efficient UV UC luminescence are very necessary and valuable.

Recently, by using Yb³⁺ and Tm³⁺ (Er³⁺) as double sensitizers and 980 nm NIR diode laser as pump light, Eu³⁺ doped fluoride nanocrystals have been demonstrated to exhibit the unusual UC spectrum from visible to UV [5, 6]. Analysis suggested upper levels of Eu³⁺ ions could be populated efficiently through internal ET between the optically active ions, while they could not be populated even in downconversion (DC) schemes of Eu³⁺ ions under vacuum UV excitations and at low temperatures. On the other hand, Eu³⁺ ions are excellent red emitters and famous structural probes under UV light excitation. Thus, it is attractive to explore the impact of ET routes on the spectra of Eu³⁺ ions. Particularly, Chen et al. have reported that Yb³⁺/Ho³⁺ doped NaYF₄ is also an ideal model for studying NIR-laser-induced UV UC radiation of Ho³⁺ ions [7]. It would be of great interest to extend the study and determine whether Ho³⁺ ions can sensitize other ions such as the Eu³⁺ ions and to uncover their fundamental optical UC mechanisms.

In this letter, we employed Ho³⁺ as the bridging ions between Yb³⁺ and Eu³⁺ ions to go on investigating the unusual radiative transitions of Eu³⁺ ions in NaYF₄ MCs under NIR 980 nm excitation. From a fundamental point of view, more detailed information about the ET mechanism and interactions between the optically active ions can be obtained by changing the bridging ions from Tm³⁺, Er³⁺ to Ho³⁺, or changing the excitation light sources. Hexagonal NaYF₄ was selected for embedding active ions, since it had been known to be an efficient host lattice for UC emissions [8–11]. Hexagonal micropillars of Yb³⁺/Ho³⁺/Eu³⁺ tri-doped β-NaYF₄ were synthesized via the ethylenediaminetetraacetic acid (EDTA)-assisted hydrothermal method. Mechanisms for UV-visible UC emissions of Eu³⁺ by two-step ET processes of Yb³⁺ → Ho³⁺ → Eu³⁺ have been proposed and demonstrated. The impact of energy populating routes on the spectra of Eu³⁺ in NaYF₄:Yb³⁺/Ho³⁺/Eu³⁺ MCs was also investigated based on experimental data and analysis.

2. Experimental

Analytical grade Y(NO₃)₃ • 6H₂O, Ho(NO₃)₃ • 6H₂O, Yb(NO₃)₃ • 6H₂O, Eu(NO₃)₃ • 6H₂O, NaF, ethanol, and EDTA were obtained from Beijing Chemical Reagents, China. Deionized water was used throughout. All other chemical reagents were of analytical reagent grade.

In a typical synthesis, 1 mL of 0.5 M Ln(NO₃)₃ aqueous solution and 0.5 mmol of EDTA were dispensed into 20 mL of deionized water and magnetically stirred for 1 h, forming a chelated Ln-EDTA complex. Then 16 mL of 0.5 M NaF aqueous solution was added to the solution. After vigorous stirring for 1 h, the mixture was transferred into a 50-mL Teflon-lined stainless steel autoclave. The autoclave was sealed and maintained in an oven at 160 °C for 18 h, and then cooled down slowly to room temperature. Subsequently, the suspension was centrifuged at 8000 rpm for 10 min. The resultant product was then washed thoroughly and dried in vacuum at 80 °C.

3. Results and discussion

3.1 Structure and morphology of NaYF₄:Yb³⁺/Ho³⁺/Eu³⁺ microcrystals

The crystal structure was analyzed by a Rigaku RU-200b X-ray powder diffractometer (XRD) using a nickel-filtered Cu-Ka radiation (λ = 1.4518Å). The size and morphology were investigated by scanning electron microscope (SEM, KYKY 1000B). A power-adjustable laser diode (980 nm, 0 to 2W) with a lens making the beam parallel was employed as the UC pump source and a 100 W xenon lamp equipped in the Hitachi F-4500 fluorescence spectrophotometer was used as the DC pump source. The UC and DC luminescence spectra were recorded with a Hitachi F-4500 fluorescence spectrophotometer (1.0 nm for spectral resolution (FWHM) of the spectrophotometer and 400 V for PMT voltage) at room temperature. A 10-ns Raman laser running at 953.6 nm was used as the pulsed excitation source for temporal investigations. Low temperature emission spectra were obtained at 10 K with the samples mounted in a helium exchange gas chamber of a closed cycle refrigeration system. Figure 1 shows the XRD pattern of the MCs. All the diffraction peaks can be indexed to the pure hexagonal NaYF₄ (JCPDS 16-0334). No other impurity peaks were detected. The corresponding SEM image (Fig. 1) shows that the NaYF₄:Yb³⁺/Ho³⁺/Eu³⁺ MCs are hexagonal pillars.

 

Fig. 1 SEM and XRD patterns of sample NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺, 3%Eu³⁺ MCs.

Download Full Size | PDF

3.2 Upconversion emissions of Ho³⁺ and Eu³⁺ ions

For discernible distinction between the emissions of Ho³⁺ and Eu³⁺ ions, the enlarged UC spectra of NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺, 3%Eu³⁺ (a, b) and NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺ (c, d) MCs in the range of 230-660 nm are shown in Fig. 2 and Fig. 3 . Most emissions of Ho³⁺ ions correspond well to what observed in Yb³⁺/Ho³⁺ codoped crystals [7]. Figure 2 shows UV UC luminescence of the samples under 980 nm excitation from a diode laser. As can be seen in Fig. 2c, the shortest UV UC emission of Ho³⁺ ions centered at 241 nm was observed, corresponding to ³F₄/⁵D₄ → ⁵I₈ emission. Besides them, characteristic peaks of Eu³⁺ ions can be found explicitly. The wide range UC luminescence of Eu³⁺ ions includes: ⁵D₀ → ⁷F_J, ⁵D₁ → ⁷F₃ (red emissions, J = 1, 2), ⁵D₂ → ⁷F_J (blue-green emissions, J = 0, 3), ⁵D₃ → ⁷F_J (violet emissions, J = 2, 3) and ⁵L₆ → ⁷F₀, ⁵H_3-7 → ⁷F_J (ultraviolet emissions, J = 0, 1, 2, 3). Although, compared with our previous results based on YF₃:Yb/Tm/Eu samples, Ho³⁺ is not as effective as Tm³⁺ serving as a bridge to transfer energy from Yb³⁺ to Eu³⁺ under 980 nm excitation [6]. To the best of our knowledge, such efficient UC luminescence of Eu³⁺ has not been reported through Ho³⁺ serving as a bridge to transfer energy from Yb³⁺ to Eu³⁺ under 980 nm excitation.

 

Fig. 2 UV UC spectra of NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺, 3%Eu³⁺ (a, b) and NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺ (c, d) MCs under 980 nm excitation. Excitation power density is 320 W/cm².

Download Full Size | PDF

 

Fig. 3 UC spectra of NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺, 3%Eu³⁺ (a, b) and NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺ (c, d) MCs under 980 nm excitation. Excitation power density is 320 W/cm².

Download Full Size | PDF

3.3. UC mechanisms for Eu³⁺ and Ho³⁺ions

Figure 4 describes schematically energy level diagrams of Eu³⁺, Ho³⁺, and Yb³⁺ ions and possible UC processes under 980 nm excitation [12]. Yb³⁺ ions continuously absorb 980 nm photons and then transfer the energy to populate the ⁵I₆, ⁵F₂/⁵F₄ and ⁵F₂/³F₂/⁵G₂ levels of Ho³⁺ in turn. Different from Ho³⁺ ions, the population of the levels of Eu³⁺ ions can only originate through the ETs from excited Ho³⁺ ions. This can be confirmed from our experimental result: we have not observed any emissions of Eu³⁺ by UC in NaYF₄: 20%Yb³⁺, 3%Eu³⁺ MCs. Multiple ET processes may exist since the trivalent Ho³⁺ ion has abundant energy levels and their small energy mismatches between levels of Eu³⁺ and Ho³⁺ ions. For briefness, four possible and representative ET processes should be considered in our paper: ET1 ⁵S₂/⁵F₄ → ⁵I₈ (Ho³⁺): ⁷F₀ → ⁵D₀ (Eu³⁺); ET2 ⁵G₄ → ⁵I₈ (Ho³⁺): ⁷F₀ → ⁵D₃ (Eu³⁺); ET3 ⁵F₂/³F₂/⁵G₂ → ⁵I₈ (Ho³⁺): ⁷F₀ → ⁵D₄ (Eu³⁺); ET4 ⁵G₄/⁵D₄/³G₄ → ⁵I₈ (Ho³⁺): ⁷F₀ → ⁵I_J/³P₀/⁵F_J (Eu³⁺). Simultaneously, the low energy levels ⁵D_3,2,1,0 of Eu³⁺ ions could be populated through a series of nonradiative relaxations from the neighboring higher excited levels.

 

Fig. 4 Energy level diagrams of Eu³⁺, Yb³⁺ and Ho³⁺ ions and possible UC emission processes.

Download Full Size | PDF

To understand the UC processes well, we investigated the excitation power dependence of UC luminescence intensities. For an unsaturated UC process, the integrated UC luminescence intensity I_f is proportional to Pⁿ [13], where P is the pumping laser power, and n is the number of laser photons required in populating the upper emitting state. Figure 5 shows the typical pump-power dependence of UC luminescence of NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺, 3%Eu³⁺. The values of photon number n are 1.81 for the ⁵D₀ → ⁷F₀ transition of Eu³⁺, 2.15 for the ⁵D₂ → ⁷F₀ transition of Eu³⁺, 2.75 for the ⁵L₆ → ⁷F₀ transition of Eu³⁺, 2.98 for the ⁵G₄ → ⁵I₈ transition of Ho³⁺, 3.38 for the ³H₄/⁵G₄/¹D₄ → ⁵I₈ transition of Ho³⁺, and 3.67 for the ⁵H_3-7 → ⁷F₀ transition of Eu³⁺, indicating that these transitions are of two-, three- and four-photon UC processes, respectively. Power dependence analyses illustrate that these levels of Eu³⁺ have the same multi-photon UC characters with the corresponding levels of Ho³⁺ ions and confirm that they are populated by the ETs from the corresponding levels of Ho³⁺ ions.

 

Fig. 5 Excitation power dependence of UC luminescence in NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺, 3%Eu³⁺ MCs under 980 nm excitation.

Download Full Size | PDF

The energy transfer from excited Ho³⁺ to Eu³⁺ ions can be further proved by the dynamical analysis on Ho³⁺ excited states. The decay curves for the representative emissions from the ⁵F₂/³F₂/⁵G₂, ⁵G₄, ⁵S₂/⁵F₄ levels of Ho³⁺ ions in NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺ and NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺, 3%Eu³⁺ MCs were recorded under 953.6 nm pulsed Raman laser, as shown in Fig. 6 . Each of the decay curves can be fitted well into a single-exponential function as I = I₀exp(−t/τ) (I₀ is the initial emission intensity, τ is the lifetime of the level). The best-fitted results were listed in Table 1 . As can be seen clearly from Table 1, with the addition of Eu³⁺ ions, all the lifetimes of the ⁵F₂/³F₂/⁵G₂, ⁵G₄, ⁵S₂/⁵F₄ levels decrease greatly which explicitly demonstrate the energy transfer from Ho³⁺ to Eu³⁺ ions.

 

Fig. 6 Time-dependent emission profile of ⁵F₂/³F₂/⁵G₂, ⁵G₄, ⁵S₂/⁵F₄ levels of Ho³⁺ in NaYF₄: Yb/Ho/Eu (|) and NaYF₄: Yb/Ho (■) samples under 953.6 nm pulsed Raman excitation. The decay parts can be fitted into a single-exponential function as I = I₀exp(-t/τ) (I₀ is the initial emission intensity, τ is the lifetime of the emission center).

Download Full Size | PDF

Table 1. Average Lifetimes τ of Emission Levels of Ho³⁺ in NaYF₄: Yb/Ho/Eu and NaYF₄: Yb/Ho Samples Calculated from Time-dependent Emission Profiles.

View Table

3.4. The spectral difference and analysis of Eu³⁺ ions under 980 nm and 394 nm excitation

Figure 7 presents the absorption spectra of the NaYF₄: Yb/Ho/Eu around 980 nm and 394 nm. The absorption of ⁵L₆ level of Eu³⁺ at 394 nm and the ²F_5/2 level of Yb³⁺ at 980 nm can be observed, clearly. Consequently, the different excitation lights associate with different energy populating processes of Eu³⁺ in NaYF₄: 20%Yb³⁺, 1.5%Ho³⁺, 3%Eu³⁺ MCs. For comparison, Fig. 8 shows the luminescence integral intensity ratios of (⁵D₁ → ⁷F₃)/(⁵D₀ → ⁷F₂) and (⁵D₂ → ⁷F₃)/(⁵D₀ → ⁷F₂) at the different doping concentration of Eu³⁺ under 394 nm (Fig. 8a) and 980 nm (Fig. 8b) excitation, respectively. It can be found clearly that the ratios exhibit gradual decrease with increasing the doping concentration of Eu³⁺. The decreases can be associated with the quenching processes of ⁵D₁ and ⁵D₂ emissions. According to previous research, the hexagonal NaYF₄ matrix has phonon energies (~520 cm⁻¹). The energy gaps of ⁵D₂ → ⁵D₁ and ⁵D₁ → ⁵D₀ are estimated to be 1771 and 2464 cm⁻¹, respectively. Correspondingly, it needs 3 and 5 phonons to bridge the gaps of ⁵D₂ → ⁵D₁ and ⁵D₁ → ⁵D₀ in Eu³⁺ doped NaYF₄ MC. The emissions from the levels ⁵D₁ and ⁵D₂ of Eu³⁺ may be quenched through thermally enhanced multiphonon relaxation which can be related to the temperature through , where is the nonradiative relaxation rate at 0 K [14,15]. On the other hand, the concentration quenching occurs at higher doping concentrations of Eu³⁺ [16]. Besides, the ⁵D₁ emissions can be quenched by the cross relaxation between two Eu³⁺ ions: ⁵D₁(Eu₁) + ⁷F₀(Eu₂) → ⁵D₀(Eu₁) + ⁷F₃(Eu₂). This can be confirmed from the fact in Fig. 8b that the ratio of the (⁵D₁ → ⁷F₃)/(⁵D₀ → ⁷F₂) decreases faster than ratio of (⁵D₂ → ⁷F₃)/(⁵D₀ → ⁷F₂) with increasing the doping concentration of Eu³⁺ ions under 980 nm excitation.

 

Fig. 7 Absorption spectra of NaYF₄: Yb/Ho/Eu samples around 980 nm and 394 nm.

Download Full Size | PDF

 

Fig. 8 Luminescence intensity ratio of (⁵D₁ → ⁷F₃)/(⁵D₀ → ⁷F₂) and (⁵D₂ → ⁷F₃)/(⁵D₀ → ⁷F₂) of Eu³⁺ as a function of the doping concentration of Eu³⁺ ions under 394 nm excitation (T = 300K) (a) and 980 nm excitation (T = 300K and 10K) (b), respectively. Excitation power densities are 20 W/cm² under 394 nm excitation and 320 W/cm² under 980 nm excitation.

Download Full Size | PDF

Comparing Fig. 8a and Fig. 8b, we can observed obviously that the luminescence integral intensity ratios of (⁵D₁ → ⁷F₃)/(⁵D₀ → ⁷F₂) and (⁵D₂ → ⁷F₃)/(⁵D₀ → ⁷F₂) under 980 nm excitation are much larger than those under 394 nm excitation at the same doping concentration of Eu³⁺. Such results confirm again that the ETs from Yb³⁺ to Ho³⁺ and then from Ho³⁺ to Eu³⁺ play more efficient roles in populating the high-energy states of Eu³⁺ ions under 980 nm excitation than pumping them directly under 394 nm excitation [6].

Besides, we can found clearly in Fig. 8 that the intensity ratios of the (⁵D₁ → ⁷F₃)/(⁵D₀ → ⁷F₂) and (⁵D₂ → ⁷F₃)/(⁵D₀ → ⁷F₂) decrease dramatically with increasing the doping concentration of Eu³⁺ ions under 980 nm excitation comparing with those under 394 nm excitation at 300 K. The fast decrease of the ratios under 980 nm excitation can be attributed to the thermal effect of 980 nm infrared light, which increase the local temperature of samples. Cross relaxation, concentration quenching, and multiphonon relaxation rates increased sharply at higher temperatures under 980 nm excitation [17]. The ⁵D₁ and ⁵D₂ emissions were quenched more and more with increasing the doping concentration of Eu³⁺ ions. Accordingly, the intensity ratios of the (⁵D₁ → ⁷F₃)/(⁵D₀ → ⁷F₂) and (⁵D₂ → ⁷F₃)/(⁵D₀ → ⁷F₂) decreased dramatically under 980 nm excitation, which is demonstrated in Fig. 8b. This can be further confirmed by the experimental data in Fig. 8b that when the samples were cooled from room temperature to 10 K, the intensity ratios of the (⁵D₁ → ⁷F₃)/(⁵D₀ → ⁷F₂) and (⁵D₂ → ⁷F₃)/(⁵D₀ → ⁷F₂) decrease gently with increasing the doping concentration of Eu³⁺ ions under 980 nm excitation, which are analogous to those under 394 nm excitation at a room temperature.

4. Conclusions

In summary, Yb³⁺/Ho³+/Eu³⁺ tri-doped NaYF₄ MCs were fabricated through a simple hydrothermal process. It was found that not only unusual UV emissions of Ho³⁺ ions, but also unusual UV emissions of Eu³⁺ ions could be observed in this tri-doped system. Power-dependence analysis and dynamical analysis on Ho³⁺ excited states confirm that Ho³⁺ is an important dopant that served as a “bridging ion” in the efficient UC excitation processes of Eu³⁺. Besides, the spectra of NaYF₄: Yb³⁺/Ho³⁺/Eu³⁺ MCs under 980 nm and 394 nm excitation were compared. The luminescence integral intensity ratios of (⁵D₁ → ⁷F₃)/(⁵D₀ → ⁷F₂) and (⁵D₂ → ⁷F₃)/(⁵D₀ → ⁷F₂) under 980 nm excitation are much larger than those under 394 nm excitation at the same doping concentration of Eu³⁺. These results indicate that a UC scheme is a better way to populate the high-energy levels of Eu³⁺ than a DC scheme.