Synthesis of Multicolor Core/Shell NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ Upconversion Nanocrystals

Abstract

The ability to synthesize high-quality hierarchical core/shell nanocrystals from an efficient host lattice is important to realize efficacious photon upconversion for applications ranging from bioimaging to solar cells. Here, we describe a strategy to fabricate multicolor core @ shell α-NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ (Ln = Er, Ho, Tm) upconversion nanocrystals (UCNCs) based on the newly established host lattice of sodium lutetium fluoride (NaLuF₄). We exploited the liquid-solid-solution method to synthesize the NaLuF₄ core of pure cubic phase and the thermal decomposition approach to expitaxially grow the calcium fluoride (CaF₂) shell onto the core UCNCs, yielding cubic core/shell nanocrystals with a size of 15.6 ± 1.2 nm (the core ~9 ± 0.9 nm, the shell ~3.3 ± 0.3 nm). We showed that those core/shell UCNCs could emit activator-defined multicolor emissions up to about 772 times more efficient than the core nanocrystals due to effective suppression of surface-related quenching effects. Our results provide a new paradigm on heterogeneous core/shell structure for enhanced multicolor upconversion photoluminescence from colloidal nanocrystals.

1. Introduction

Upconversion nanocrystals (UCNCs) are able to convert two or more long wavelength photons into short wavelength emissions through the use of real energy levels of trivalent lanthanide ions embedded in an inorganic host lattice [1]. Owing to the high physicochemical stability and intrinsic low phonon energy, fluoride-based UCNCs are able to minimize energy losses at the intermediate states of the incorporated lanthanide ions, thus generally exhibiting efficient upconversion (UC) luminescence efficiency [2]. Moreover, fluoride UCNCs also have superior features, such as low toxicity, non-blinking, non-photobleaching, absence of autofluorescence, and tissue-penetrable near-infrared (NIR) light excitation [3,4]. These superb attributes promise their applications in biological imaging [5,6,7,8], bio-detection [9,10], and three-dimensional display [11,12,13]. This motivation fuels a range of works to synthesize UCNCs with controlled size and morphology, as well as to prepare epitaxial core/shell UCNCs with enhanced efficiency and multifunction for theranostic applications, such as, NaYF₄:Yb/Er@NaYF₄ [14], (NaLuF₄:Gd³⁺/Yb³⁺/Er³⁺)@NaLuF₄:Yb³⁺ [15], and NaYbF₄:Er@NaGdF₄ core/shell nanostructures [16].

Sodium lutetium fluoride (NaLuF₄) has recently emerged as a new type of efficient host lattice for photon upconversion, similar to the well-established host material of sodium yttrium fluoride (NaYF₄). NaLuF₄-based UCNCs have been demonstrated to exhibit bright upconversion luminescence (UCL) [17,18,19,20,21,22,23,24] and show efficient five- and four-photon ultraviolet emissions under continuous wave excitation at 980 nm [25,26,27]. Despite recent success in synthesizing NaLuF₄ nanoplates or nanorods with size over ~30 nm using thermal decomposition method [28,29,30,31] or hydrothermal method [32,33,34,35,36,37,38], the ability to prepare uniform single crystal phase sub-10 nm NaLuF₄ UCNCs remains elusive. Moreover, doping of a high concentration of inert Gd³⁺ ions (≥20%) was typically required to prepare small-sized monodisperse NaLuF₄ UCNCs with single crystal phase previously [17,39], thus delivering traits actually from an entity of Lu-Gd alloyed host. On the other hand, small-sized UCNCs are important for single molecule imaging [40] and in vivo bioimaging with reduced toxicity, considering the renal clearance [41]. However, they always come at the sacrifice of UCL efficiency due to the increased size-induced surface-related surface quenching effect [42]. A core/shell geometric structure is therefore needed to eliminate or suppress this detrimental effect by spatial isolation of the core nanoparticle from the surrounding quenching sites. A straightforward approach is to grow a homogenous core/shell structure where the host of the shell is identical to the core [37,43]. However, the possible leaking of rare earth ions from the host lattice could possibly lead to diseases such as nephrogenic systemic fibrosis [44,45]. Compared with lanthanide fluorides, calcium fluoride (CaF₂) has unique advantages owing to its superior biocompatibility and high optical transparency [46,47,48,49,50,51]. It has recently been demonstrated that CaF₂ also has low lattice mismatch with NaReF₄ nanocrystals, and can efficiently prevent rare-earth ions from leaking [46,50]. This implies that the growth of a CaF₂ shell not only renders UCL enhancement, but also imparts biocompatibility with reduced leaking effect.

In this work, we describe our effort toward the controlled synthesis of single crystal phase sub-10 nm α-NaLuF₄:Yb³⁺/Ln³⁺ (Ln = Er, Ho, or Tm) nanoparticles using a liquid-solid-solution method without the involvement of doping with a high concentration of Gd³⁺, and then utilize them as the core to epitaxially grow a high quality α-NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ (Ln = Er, Ho, or Tm) core/shell UCNC via a thermal decomposition protocol. We found that the growth of a ~3 nm thin CaF₂ shell layer was able to enhance the multicolour UCL of the core nanocrystals by up to ~772-fold.

2. Results and Discussion

2.1. Synthesis of α-NaLuF₄:Yb³⁺/Ln³⁺ (Ln = Er, Tm, Ho) or α-NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ Core/Shell UCNCs

The crystal structure of NaLuF₄ has two forms of the cubic (α-) and the hexagonal (β-) phase. To prepare high-quality α-NaLuF₄@CaF₂ core/shell NCs, we firstly controlled the synthesis of cubic (α) NaLuF₄ core nanoparticles by varying the reaction temperature and the molar ratio of F⁻/Ln³⁺ (Ln = Lu + Yb + Er/Ho/Tm) precursor. Figure 1 shows X-ray diffraction (XRD) patterns of NaLuF₄:Yb³⁺/Er³⁺ (Tm³⁺, Ho³⁺) NCs hydrothermally prepared at various temperatures with the molar ratio of F⁻/Ln³⁺ (Ln = Lu + Yb + Er/Ho/Tm) fixed at 4:1. Figure 1 reveals that the phase transition process (β → α or α → β) occurs at a reaction temperature of T = 100 or 120 °C. The sample obtained at low temperature (T = 100 or 120 °C) shows nearly pure α-phase (JCPDS No. 27-0725). The diffraction peaks of β-NaLuF₄ appear at temperatures between 140 and 160 °C (See Figure 1c,d), while at 180 °C, only pure β-phase NaLuF₄ exists. It could be concluded that low temperature favors the formation of pure α-NaLuF₄ core NCs.

Figure 1.

The X-ray diffraction patterns of NaLuF₄:Yb³⁺/Ln³⁺ nanocrystals synthesized at different hydrothermal temperature: (a) 100 °C; (b) 120 °C; (c) 140 °C; (d) 160 °C; (e) 180 °C. All samples were prepared at F⁻/Ln³⁺ = 4:1.

Next, we investigated the role of the molar ratio of F⁻/Ln³⁺ (Ln = Lu + Yb + Er/Ho/Tm) precursor on the crystal phase of our product when setting the synthesis temperature at 140 °C. As can be seen in Figure 2, when the molar ratio of F⁻/Ln³⁺ is fixed at 4:1, the sample shows a mixture of α-phase (JCPDS No. 27-0725) and β-phase (JCPDS No. 27-0726). As the F⁻/Ln³⁺ ratio is reduced to 3.5:1, the intensities of peaks of β-NaLuF₄ decreases (Figure 2b). A pure α-phase NaLuF₄ NCs can be produced at the ratio of F⁻/Re³⁺ below 3:1, and the diffraction peaks of α-NaLuF₄ sample at F⁻/Re³⁺ = 3:1 show stronger intensity than at F⁻/Re³⁺ = 2.5:1 and 2:1. The average crystallite size of the nanocrystals was calculated according to Scherrer’s equation [51],

$$D=K\mathsf{\lambda }/\mathsf{\beta }\cos \mathsf{\theta }$$ (1)

where K = 0.89, D represents the crystallite size (in nanometers), λ is the wavelength of the Cu Kα radiation, β is the corrected half-width of the diffraction peak, and θ is Bragg’s angle of the diffraction peak. According to Equation (1) and the half width of the main diffraction peak at 28° in Figure 2, the average size was calculated to be about 10 nm for F⁻/Re³⁺ = 3:1, in good agreement with the TEM result (Figure 3b). As a consequence, we selected α-NaLuF₄ nanoparticles prepared at F⁻/Re³⁺ = 3:1 and T = 140 °C to epitaxially grow the α-NaLuF₄:Yb/Ln@CaF₂ core/shell structure.

Figure 2.

The X-ray diffraction patterns of NaLuF₄:Yb³⁺/Er³⁺ nanocrystals synthesized with molar ratio of F⁻/Re³⁺ = (a) 4:1; (b) 3.5:1; (c) 3:1; (d) 2.5:1; (e) 2:1. (T = 140 °C).

Figure 3.

(a) Schematic illustration of the epitaxial growth of CaF₂ shell on α-NaLuF₄:Yb³⁺/Ln³⁺ core nanoparticles (NPs); (b) TEM image of α-NaLuF₄:Yb³⁺/Ln³⁺ core nanocrystals prepared at F⁻/Re³⁺ = 3:1 and T = 140 °C. TEM images of (c) NaLuF₄:Yb³⁺/Er³⁺@CaF₂ core/shell upconversion nanocrystals (UCNCs); (d) α-NaLuF₄:Yb³⁺/Ho³⁺@CaF₂ core/shell UCNCs; (e) α-NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ core/shell UCNCs. Particles in (c–e) were synthesized by the thermal decomposition method.

The principle for the epitaxial growth of the core/shell structure is illustrated in Figure 3a, which involves an injection of the (CF₃COO)₂Ca solution into the growing solution containing pre-synthesized α-NaLuF₄:Yb/Ln core nanocrystals prepared at F⁻/Re³⁺ = 3:1 and T = 140 °C. The morphologies and sizes of the α-NaLuF₄:Yb³⁺/Ln³⁺ core and the resulting α-NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ core/shell nanoparticles were examined by transmission electron microscopy (TEM), and the results are shown in Figure 3b. As one can see, the α-NaLuF₄:Yb³⁺/Ln³⁺ core has sphere-like morphology with a size of 9 ± 0.9 nm, in good agreement with the XRD result in Figure 2c.

After growing the CaF₂ shell, the core/shell NCs showed a cubic morphology with a size of 15.6 ± 1.2 nm (Figure 3c–e). The size difference indicated that a CaF₂ shell with a thickness of about 3.3 ± 0.3 nm was grown on the surface of α-NaLuF₄:Yb³⁺/Ln³⁺ core. Moreover, the formation of core/shell structure can be seen in the TEM images, shown in Figure 3c–e.

The XRD peaks of the α-NaLuF₄:Yb³⁺/Ln³⁺ core, the α-NaLuF₄:Yb³⁺/Er³⁺@CaF₂ core/shell, the NaLuF₄:Yb³⁺/Ho³⁺@CaF₂ core/shell, and the NaLuF₄:Yb³⁺/Tm³⁺@CaF₂ core/shell UCNCs are presented in Figure 4. As one can see, the core and core/shell NCs have identical peak positions, agreeing well with the standard JCPDF27-0725 sample of cubic NaLuF₄ and JCPDF 02-1302 sample of cubic CaF₂. The well-defined peaks are indicative of the high crystallinity of both the core and the core/shell NCs. The narrower XRD peaks of the core/shell NCs compared to that of core NPs indicate the larger size of core/shell NCs, in accordance with the TEM results in Figure 3.

Figure 4.

The X-ray diffraction patterns of the (a) α-NaLuF₄:Yb³⁺/Ln³⁺ NCs; (b) α-NaLuF₄:Yb³⁺/Er³⁺@CaF₂ core/shell UCNCs; (c) α-NaLuF₄:Yb³⁺/Ho³⁺@CaF₂ core/shell UCNCs; and (d) the α-NaLuF₄:Yb³⁺/Tm³⁺@CaF₂ core/shell UCNCs, in reference to the standard diffraction patterns of the α-phase NaLuF₄ (JCPDS 27-0725) and cubic phase CaF₂ (JCPDS 02-1302).

2.2. Characterization of α-NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ (Ln = Er, Tm, or Ho) Core/Shell UCNCs

We compared the UCPL from the corresponding core NaLuF₄:20%Yb³⁺/2%Ln³⁺ and the core/shell NaLuF₄:20%Yb³⁺/2%Ln³⁺@CaF₂ UCNCs (Ln = Er, Ho, and Tm) (Figure 5). As shown in Figure 5a, two UC bands with maxima at 525/540 nm and 660 nm were observed from both the NaLuF₄:20%Yb³⁺/2%Er³⁺ core and the NaLuF₄:20%Yb³⁺/2%Er³⁺@CaF₂ core/shell UCNCs, corresponding to the ²H_11/2/⁴S_3/2 → ⁴I_15/2 and the ⁴F_9/2 → ⁴I_15/2 transitions of Er³⁺ ions, respectively. Moreover, the intensity of UCPL from core/shell UCNCs is dramatically higher than that of core nanoparticles. The intensity for UCPL band at 540 nm from the NaLuF₄:20%Yb³⁺/2%Er³⁺@CaF₂ core/shell NCs were found to be about 656 times higher than that from the α-NaLuF₄:Yb³⁺/Er³⁺ core nanocrystals. Moreover, according to Figure 5b,c, when compared with the α-NaLuF₄:Yb³⁺/Ho³⁺ and α-NaLuF₄:Yb³⁺/Tm³⁺ core nanocrystals, the intensity of UCPL from the corresponding core/shell UCNCs showed a 772- and 75-fold enhancement, respectively. Taken together, we conclude that the core/shell structure resulted in a significant UCPL enhancement of the core nanocrystals. Since the CaF₂ is an inactive layer, such an impressive enhancement undoubtedly arises from the suppression of surface quenching effects on the surface of core nanoparticles. In fact, the ultrasmall size (9 nm) of the core UCNCs is able to expose most of the doped sensitizer (Yb³⁺) and the doped activator (Er³⁺/Ho³⁺/Tm³⁺) to surface quenching effects (surface defects, surface strains, ligand and solvent molecules with groups possessing high vibration energy) [42] due to the extremely high “surface-to-volume” ratio. The epitaxial growth of the thin CaF₂ shell not only eliminates the quenching defects on the surface of the core, but also shields all the sensitizer and activator ions in the core from the environmental quenching factors.

Figure 5.

The upconversion luminescence spectra under excitation at 976 nm using a fiber-coupled laser diode: (a) the α-NaLuF₄:Yb³⁺/Er³⁺ core and α-NaLuF₄:Yb³⁺/Er³⁺@CaF₂ core/shell NPs; (b) the α-NaLuF₄:Yb³⁺/Ho³⁺ core and α-NaLuF₄:Yb³⁺/Ho³⁺@CaF₂ core/shell NPs; (c) the α-NaLuF₄:Yb³⁺/ Tm³⁺ core and α-NaLuF₄:Yb³⁺/Tm³⁺@CaF₂ core/shell NPs. The concentration of Ln³⁺ in all samples was kept identical at about 0.5 mmol nanoparticles (i.e., nanoparticles formed by 0.5 mmol lanthanide precursors) per 10 mL cyclohexane.

To illustrate the UC mechanisms of the NaLuF₄:20%Yb³⁺/2%Ln³⁺@CaF₂ (Ln³⁺ = Er³⁺, Ho³⁺, Tm³⁺) NCs, possible UC processes are schematically given in the energy level diagrams of Yb³⁺, Er³⁺, Ho³⁺, and Tm³⁺ ions in Figure 6. The observed green UC bands (²H_11/2 → ⁴I_15/2, 525 nm; ⁴S_3/2 → ⁴I_15/2, 540 nm) and red UC band (⁴F_9/2 → ⁴I_15/2, 660 nm) from Er³⁺ ions in NaLuF₄:20%Yb³⁺/2%Er³⁺@CaF₂ UCNCs may take place via the following process: Yb³⁺ ion absorbs one laser photon and gets excited from the ground ²F_7/2 state to the exclusive excited ²F_5/2 state. The Yb³⁺ ions in the excited state transfer their absorbed energy to neighboring Er³⁺ ions and excite them from the ground ⁴I_15/2 state to the ⁴I_11/2 state, then to the ⁴F_7/2 state. Multiphonon assisted relaxations from the ⁴F_7/2 state can decay nonradiatively to the lower ²H_11/2 and ⁴S_3/2 levels, emitting the 525 and 540 nm UCL, respectively. The red emission 660 nm originates from the ⁴F_9/2 → ⁴I_15/2 transition, and the ⁴F_9/2 state can be populated either from nonradiative relaxations from the ⁴S_3/2 level or the energy transfer from Yb³⁺ ions to the Er³⁺ ion at the ⁴I_13/2 state. In addition, the processes for green (⁵F₄ → ⁵I₈, 537 nm) and red (⁵F₅ → ⁵I₈, 645 nm) UCL of Ho³⁺ ions involved two different centers—the sensitizer (Yb³⁺), the activator (Ho³⁺)—along with two successive transfers from Yb³⁺ ions to Ho³⁺ ions. In the first transfer, the Yb³⁺ ion absorbs the excitation photons through the ground state absorption and transfers its absorbed energy to the neighboring Ho³⁺ ion to populate to its intermediate (⁵I₆ level). The energy difference between the two levels was abridged by the vibration energy of the host lattice. The second transfer is to promote the population from the intermediate ⁵I₆ level to the emitting energy levels (⁵S₂) by energy transfer (ET) from another excited Yb³⁺ ion [52]. Once the ⁵S₂ level is populated, the excited electron can release its energy by emitting green emissions. The red emission at 645 nm can be produced by radiative decay to the ground ⁵F₅ state. The blue UCL of Tm³⁺ occurs via a three-step ET from Yb³⁺ to Tm³⁺. First, the Tm³⁺ ion in the ground state ³H₆ is excited to the state ³H₅ via an ET from a neighboring excited Yb³⁺ ion. Subsequent nonradiative relaxation of ³H₅/³F₄ populates the ³F₄ level. In the second-step excitation, the Tm³⁺ ion in the ³F₄ state is excited to the ³F_2,3 state via another ET from a neighboring excited Yb³⁺ ion. The populated ³F₂ level may nonradiatively relax to the ³F₃ level. When the Tm³⁺ ion at the ³F₃ level decays to the ground state, a weak red emission (³F₃ → ³H₆) is produced. Additionally, the near-infrared UCL at 802 nm arises from the ³H₄ → ³H₆ transition, where the ³H₄ state is populated by the efficient nonradiative relaxation from the ³F_2,3 state. A third energy transfer from Yb³⁺ excites Tm³⁺ at the ³F₃ level to the ¹G₄ level, from which the blue emission (¹G₄ → ³H₆) occurs by radiative decay to the ground state. The fourth energy transfer from Yb³⁺ promotes the Tm³⁺ at the ¹G₄ level to the ¹D₂ level, from which a 360 nm ultraviolet UCL (¹D₂ → ³H₆) is generated.

Figure 6.

The energy level diagrams of Yb³⁺, Er³⁺, Tm³⁺, and Ho³⁺ ions, showing the proposed upconversion mechanisms in the α-NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ (Ln = Er, Ho, or Tm) core/shell UCNCs.

3. Materials and Methods

3.1. Materials

All Ln₂O₃ (99.9%, Ln = Lu, Yb, Er, Tm, Ho) were obtained from Jianfeng Rare-Earth Limited Company, Conghua, China. The basic chemical reagents, such as sodium hydroxide, oleic acid (OA), absolute ethyl alcohol, trifluoroacetic acid (TFA), calcium oxide, sodium fluoride and octadecene (ODE) were purchased from Sinopharm Chemical Reagent Co., Ltd., Beijing, China. All chemicals were of analytical grade and were used as received without further purification.

3.2. Hydrothermal Synthesis of α-NaLuF₄:Yb³⁺/Ln³⁺ UCNCs

We synthesized α-NaLuF₄:Yb³⁺/Ln³⁺ (Ln = Er, Tm, or Ho) nanocrystals by a hydro-thermal method adapted from the literature [53]. Typically, 0.6 g of NaOH, 3 mL of water, 14 mL of oleic acid (OA) (90 wt. %), and 10 mL (120 mmol) of ethanol were well-mixed at room temperature to yield a white viscous solution. Then, 1 mmol Ln(NO₃)₃ (Ln = Lu, Yb, Er, Tm, Ho, and Lu³⁺:Yb³⁺:Er³⁺/Tm³⁺/Ho³⁺ = 78%:20%:2%/2%/2%) was added into the above solution and kept vigorous stirring. After aging for 30 min, 4 mL (4 mmol) of NaF (F⁻/Ln³⁺ = 4:1) solution was added under vigorous stirring for 30 min. Subsequently, the mixture was transferred to a 50-mL Teflon-lined autoclave and heated at 140–180 °C for 24 h. After washing with ethanol, the final products were dispersed in cyclohexane.

3.3. Thermal Decomposition Synthesis of α-NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ Core/Shell UCNCs

The core/shell nanoparticles were synthesized using the thermal decomposition method. Typically, 0.5 mmol CaO was first added to a 50 mL flask containing 5 mL deionized water and 5 mL trifluoroacetic acid (TFA). The solution was heated at 90 °C until the solution became transparent, and was then dried at this temperature with nitrogen purge to yield the shell precursor (CF₃COO)₂Ca. After obtaining the (CF₃COO)₂Ca powders, 10 mL of OA, 10 mL of ODE, and the pre-prepared α-NaLuF₄:Yb³⁺/Ln³⁺ (0.5 mmol) in cyclohexane were added. The solution was then vacuum-degassed at 120 °C for 30 min to remove water, oxygen, and cyclohexane. Subsequently, the solution was heated to 300 °C at a rate of 15 K·min⁻¹ under nitrogen protection. After maintaining at 300 °C for 30 min, the reaction was stopped and cooled down to room temperature. After washing with ethanol, the products were dispersed in cyclohexane for further use.

3.4. Thermal Decomposition Synthesis of α-NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ Core/Shell UCNCs

The size and morphology of the α-NaLuF₄:Yb³⁺/Ln³⁺ and α-NaLuF₄:Yb³⁺/Ln³⁺@CaF₂ core/shell nanocrystals were characterized by transmission electron microscopy (TEM) using a JEOL JEM-2010 microscope (JEOL Ltd., Tokyo, Japan) at an acceleration voltage of 200 kV. The powder X-ray diffraction (XRD) patterns were recorded by a Siemens D500 diffractometer (Bruker Beijing Scientific Technology Co. Ltd, Beijing, China) using Cu Kα radiation (λ = 0.15418 nm). The 2θ angle of the XRD spectra was recorded at a scanning rate of 5°/min. The UCPL spectra were obtained using a Zolix monochoromator (Beijing Zolix Instruments CO., Ltd., Beijing, China) under excitation at 976 nm using a fiber-coupled laser diode (BWT Beijing Ltd., Beijing, China).

4. Conclusions

In summary, cubic phase α-NaLuF₄:Yb/Ln cores can be precisely controlled through a simple variation of reaction temperature and the added amount of NaF in a hydrothermal method. Moreover, a seed-mediated growth protocol with selected parameters favorable for shell growth yields the α-NaLuF₄:Yb/Ln@CaF₂ (Ln = Er, Ho, Tm) core/shell structure NPs having a core size of ~9 nm and shell thickness of ~3.3 nm. Moreover, we found that the growth of the inert thin shell of CaF₂ onto the α-NaLuF₄:Yb/Ln (Ln = Er, Ho, Tm) core could enhance its multicolor UCL by up to 772-fold, being attributed to effective suppression of surface-related quenching effects via spatial isolation of the core from the surrounding environment. Small-sized α-NaLuF₄:Yb/Ln@CaF₂ (Ln = Er, Ho, Tm) UCNCs developed here have implication for uses in a range of biophotonic applications, such as bioimaging.

Author Contributions

Hui Li and Guanying Chen conceived and designed the experiments; Hui Li performed the experiments and drafted the manuscript; Chunhui Yang provided valuable suggestions; Guanying Chen and Shuwei Hao advised the work and edited the manuscript.

Conflicts of Interest

The authors declare no conflict of interest.