Understanding the effect of Mn²⁺ on Yb³⁺/Er³⁺ upconversion and obtaining a maximum upconversion fluorescence enhancement in inert-core/active-shell/inert-shell structures

Abstract

In this study, the energy transfer (ET) mechanism among Yb³⁺/Er³⁺/Mn²⁺ has been revealed by download conversion (DC) and upconversion (UC) transients and steady-state fluorescence spectra. Moreover, the novel structure, inert-core/active-shell/inert-shell cubic NaYF₄@NaYF₄:Er³⁺/Yb³⁺/Mn²⁺@NaYF₄, has been synthesized via a convenient hydrothermal method. With the excitation of a 980 nm laser diode, bright single-red upconversion (UC) fluorescence is observed, which shows the longest UC lifetime and highest efficiency compared with the other structures of NaYF₄:Er³⁺/Yb³⁺/Mn²⁺, NaYF₄:Er³⁺/Yb³⁺/Mn²⁺@NaYF₄, and NaYF₄@NaYF₄:Er³⁺/Yb³⁺/Mn²⁺. The inert-core/active-shell/inert-shell nanocubes are homogeneous with an ultra-small (sub-20 nm) size, which is suitable for distribution and elimination in biological tissue. Our results indicate that the Mn²⁺ doping of δ-doped structures is an effective method to significantly improve UC single-red light, which shows great potential application in vivo bioimaging.

---

1. Introduction

Recently, lanthanide-doped NaYF₄ nanocrystals have attracted a great deal of attention due to their unique UC luminescence properties and broad potential applications. In comparison with organic fluorophores and semiconducting nanocrystals, upconversion nanoparticles (UCNPs) offer high photochemical stability, sharp emission bandwidths, and large anti-Stokes shifts that separate discrete emission peaks from infrared excitation.^1–4 Moreover, NIR sources are usually inexpensive and easy to use. In particularly, UCNPs with a small diameter and intense red emission should be comparable in size to the biomolecules they label and simultaneously not cause tissue absorption and autofluorescence problems by their emission.⁵ Therefore, luminescence imaging on the basis of single-red UCNPs is becoming a popular tool for biomedical assays and diagnosis.

Since they are used for bioimaging, the brightness of UCNPs must be taken into account. In general, the brighter the UCNPs, the higher the signal-to-noise ratio. Despite the significant progress in the use of UCNPs, their low emission efficiency still limits their practical application in vivo imaging.^6,7 This restriction is usually solved by employing many strategies, such as constructing composite NPs with Au and Ag nanoparticles,^8,9 co-doping with other ions,^10–12 and the use of core/shell^13–16 or Janus structures.¹⁷ However, most of these methods are either difficult to implement or only suitable for individual lanthanide ions, and much less realized due to the small size of hydrophilic biological cells.

Among the UC materials, α-phase NaYF₄ is reported as one of the most efficient hosts for greatly enhancing near-infrared (NIR) to single-red band up-conversion when tri-doped with Yb³⁺/Er³⁺/Mn²⁺ ions.¹⁸ Similar effects have been reported in several other KMnF₃:Yb³⁺/Er³⁺, NaLuF₄:Yb³⁺/Er³⁺/Mn²⁺, NaMnF₃:Yb³⁺/Er³⁺, and MnF₂:Yb³⁺/Er³⁺ materials. The explicit UC energy transfer mechanism among Yb³⁺/Er³⁺/Mn²⁺ is not very clear by sufficient experimental evidence. In this study, we attempt to reveal this mechanism. Moreover, in order to improve the UC efficiency in the Yb³⁺/Er³⁺/Mn²⁺ system, several types of core/shell structures of UCNPs are synthesized for investigation. Experimental results indicate that the NaYF₄@NaYF₄:Er³⁺/Yb³⁺/Mn²⁺@NaYF₄ structure is the most efficient method to improve UC single-red luminescence.

2. Experimental

Chemicals and sample preparation

The chemicals used in this study were sodium fluoride (99.99%, Aladdin), sodium hydroxide (AR, Sinopharm Chemical), oleic acid (OA) (AR, Sinopharm Chemical), manganese chloride (AR, Aladdin) and the rare earth oxides yttrium oxide, ytterbium oxide and erbium oxide, all at the same concentration (99.99%, Aladdin). Before the synthesis, 2.5 mmol rare earth oxide was weighed and reacted with hydrochloric acid to generate a rare earth chloride solution in the concentration of 0.5 mmol mL⁻¹. In addition, 20 mmol of NaF powder was dissolved in deionized water to prepare a 30 mL solution.

Synthesis of undoped α-phase NaYF₄. For this initial step, 0.5 mmol YCl₃ solution was added to a stirring precursor solution of 5 mL deionized water, 0.3 g NaOH, 5 mL OA and 15 mL ethanol at a controlled rate, and the mixture was stirred at 50 °C for 15 minutes. Then, 3 mL (2 mmol) of sodium fluoride (NaF) aqueous solution was slowly added to the mixed solution to form an emulsion. After vigorous magnetic stirring for 0.5 h, the colloidal solution was transferred to a 50 mL Teflon-lined autoclave and heated at 180 °C for 0.5 h. The final product was collected by centrifugation and washed with ethanol/cyclohexane (v/v = 1 : 1) several times to remove possible remnants.

Synthesis of NaYF₄@NaYF₄:Er³⁺/Yb³⁺/Mn²⁺. 94 mg (0.5 mmol) of the as-synthesized NaYF₄ core was added to the shell precursor solution containing 0.5 mmol MCl₃ (M = 30% Mn²⁺, 20% Yb³⁺, 2% Er³⁺ and 48% Y³⁺). Then, 3 mL (2 mmol) sodium fluoride (NaF) aqueous solution was added, and the mixture was stirred at room temperature for 0.5 h. Finally, the mixed reactants were transferred to a 50 mL Teflon-lined autoclave and heated at 180 °C for 0.5 h. The method described previously was used to collect the final product.

Synthesis of NaYF₄@NaYF₄:Er³⁺/Yb³⁺/Mn²⁺@NaYF₄. To form the second shell, 146 mg NaYF₄@NaYF₄:Er³⁺/Yb³⁺/Mn²⁺ (0.5 mmol) nanocrystals was used as the nucleation seeds for the growth of the inert capping layer. The experimental procedure used is similar to the synthetic procedure for NaYF₄@NaYF₄:Er³⁺/Yb³⁺/Mn²⁺ except for the reaction time, which is reduced to 0.5 h and the replacement of the 0.5 mmol MCl₃ (M = 30% Mn²⁺, 20% Yb³⁺, 2% Er³⁺ and 48% Y³⁺) shell precursor solution by a 0.5 mmol YCl₃ solution.

Sample characterization

Phase identification of the core, core–shell and core–shell–shell crystal structure was done via X-ray diffraction (XRD) using a Philips X'Pert Pro X-ray diffractometer with Cu Kα radiation. Nanostructure and morphology were characterized via transmission electron microscopy (TEM) (JEOL JEM-2100, JP). Energy spectra were obtained via field emission scanning electron microscopy (FESEM) (HITACHI S4700, JP). Absorption spectra were measured on a Cary Series UV-Vis-NIR spectrophotometer (Agilent Technologies, USA). UC luminescence spectra, luminescence decay curves and time-resolved luminescence spectra were obtained on an FLS980E spectrometer (Edinburgh Instruments Ltd, UK).

3. Results and discussion

Energy transfer among Yb³⁺, Er³⁺, and Mn²⁺ ions

In order to investigate the energy transfer mechanism among Er³⁺/Yb³⁺/Mn²⁺, five NaYF₄ samples were synthesized, each doped with 20 mol% Yb³⁺, 2 mol% Er³⁺ and 0–40 mol% Mn²⁺ ions. Then, the crystal structure and phase purity of the five samples were examined via XRD, as shown in Fig. 1. The XRD pattern of NaYF₄:Yb³⁺/Er³⁺ NPs can be assigned to a pure cubic NaYF₄ crystal (JCPDS 06-0342), and no obvious extra diffraction peaks are detected, even when the Mn²⁺ ion concentration increases to 30 mol%. However, at 40 mol% Mn²⁺ doping, the appearance of some weak diffraction peaks, marked by asterisks, indicates the emergence of an impurity phase, which is from the NaMnF₃ crystals (JCPDS 18-1224). Moreover, as guided by the dash line, the diffraction peak (111) shifts slightly to the higher-angle side with an increase in the concentration of the Mn²⁺ dopant, due to the substitution of Y³⁺ ions by the smaller Mn²⁺ ions in the host lattice, which results in a decrease in the unit-cell volume. Hence, in order to ensure a pure cubic phase, the highest Mn²⁺ doping concentration of 30 mol% must be controlled under the current synthetic conditions.

Fig. 1 XRD patterns of NaYF₄:Yb³⁺/Er³⁺ (20/2 mol%) with various concentrations of Mn²⁺ added to the initial reaction solution (0, 10, 20, 30 and 40 mol%). The dash line is shown to guide the eye.

Fig. 2 shows the absorption spectra of the as-synthesized Yb³⁺/Er³⁺/Mn²⁺ tri-doped NaYF₄ nanoparticles. As shown, absorption peaks at 363 nm, 380 nm, 448 nm, 488 nm, 520 nm, 650 nm and 800 nm are detected, which should be assigned to the ⁴I_15/2 to ²G_7/2, ⁴G_11/2, ⁴F_5/2, ⁴F_7/2, ²H_11/2, ⁴S_3/2, ⁴F_9/2 and ⁴I_9/2 transitions of Er³⁺. In addition to the absorption of Er³⁺, two broad absorptions in the visible range (400–600 nm) and NIR wavelength range (900–1100 nm) are detected as well, which may be ascribed to ⁶A₁ → ⁴T₁ of Mn²⁺ and ²F_7/2 → ²F_5/2 of Yb³⁺, respectively. This result indicates that the Yb³⁺, Er³⁺ and Mn²⁺ ions are successfully doped into the host lattices; moreover, manganese in the nanocrystals are all in pure valence state of Mn²⁺.

Fig. 2 Absorption spectrum of the as-synthesized NaYF₄:Yb³⁺/Er³⁺/Mn²⁺ (20/2/30 mol%) nanocrystals.

The UC spectra of NaYF₄:Yb³⁺/Er³⁺ co-doped with various amounts of Mn²⁺ were detected under the excitation of a Continuous Wave (CW) 980 nm NIR diode laser. As shown in Fig. 3, the luminescence intensity of the green emission (520–570 nm) decreases; moreover, the red emission increases monotonously with an increase in the dopant concentration. In addition, a “single-band” red UC emission is obviously seen in the 30 mol% and 40 mol% doped NPs. The green emission bands centered at 525/540 nm are attributed to the electronic transitions ²H_11/2 → ⁴I_15/2 and ⁴S_3/2 → ⁴I_15/2 of the Er³⁺ ions, respectively. The red UC luminescence band centered at 655 nm and 670 nm are assigned to the Er³⁺:⁴F_9/2 → ⁴I_15/2 transitions.

Fig. 3 Upconversion luminescence spectra of NaYF₄:Yb³⁺/Er³⁺ (20/2 mol%) nanoparticles with different concentrations of Mn²⁺ added to the initial reaction solution (0, 10, 20, 30 and 40 mol%).

Fig. 4(a) further shows the UC luminescence decay curves of the ⁴S_3/2 and ⁴F_9/2 levels of Er³⁺, which correspond to various amounts of Mn²⁺. It is observed that the presence of Mn²⁺ facilitates the relaxation of the ⁴S_3/2 level and reduces the decay time, and meanwhile increases the lifetime of the red region emission of Er³⁺. It is well known that the lifetime of the excited state is inversely proportional to the radiative transition rate, the non-radiative transition rate and the energy transfer rate.¹⁹ The presence of Mn²⁺ in NaYF₄:Yb³⁺/Er³⁺ may create a non-radiative energy transfer channel from the levels of ²H_11/2 and ⁴S_3/2 of Er³⁺ to Mn²⁺:⁴T₁ owing to the resonances between the Mn²⁺ absorption bands and ²H_11/2/⁴S_3/2 levels of Er³⁺, which leads to an increase in the overall transition rate and thus a reduction in the fluorescence lifetime. For UC materials, a longer lifetime usually means a higher energy transfer rate from other levels to the emission level and a more efficient UC luminescence.²⁰ The increase in the lifetime of Er³⁺:⁴F_9/2 may be attributed to the energy transfer process from Mn²⁺ to Er³⁺ ions (Fig. 4(b)).

Fig. 4 UC luminescence decay curves of green (550 nm) (a) and red (650 nm) (b) emission of NaYF₄:Yb³⁺/Er³⁺ with different Mn²⁺ concentrations added to the initial reaction solution (0, 10, 20 and 30 mol%).

In order to verify in depth the existence of the energy transfer process from Mn²⁺ to Er³⁺ ions, we further synthesized some pure α-NaYF₄ nanocrystals with three types of doping: Mn²⁺ 30 mol%, Er³⁺ 2 mol% and Mn²⁺/Er³⁺ (30/2 mol%). 400 nm photons directly excite the high level of Mn²⁺ to ensure the strongest intensity of the orange emission of Mn²⁺:⁴T₁ → ⁶A₁ due to the fact that 400 nm is the excitation peak of NaYF₄:Mn²⁺, which is shown in the insert of Fig. 5. By comparing the down-conversion (DC) luminescence spectrum of the three samples in Fig. 5, an obvious red emission band centered at 655 nm and 670 nm appears for the NaYF₄:Mn²⁺/Er³⁺ nanoparticles, whereas it does not appear for both the NaYF₄:Mn²⁺ and NaYF₄:Er³⁺ samples. Moreover, this result indicates that the energy transfer process from Mn²⁺ to Er³⁺ ions really exists, and the proposed DC mechanism between Mn²⁺ and Er³⁺ under the excitation of 400 nm is shown in Fig. 6.

Fig. 5 Luminescence spectra of NaYF₄ doped with different elements and excited by a xenon lamp at 400 nm. Inset is the excited spectrum of NaYF₄:Mn²⁺ monitored at 560 nm.

Fig. 6 Energy level diagrams of the Er³⁺ and Mn²⁺ dopant ions and down-conversion mechanisms by 400 nm photon excitation.

In order to more clearly and intuitively observe the variation in the UC luminescence process, time-resolved emission spectra for NaYF₄:Yb³⁺/Er³⁺ (20/2 mol%) and NaYF₄:Yb³⁺/Er³⁺/Mn²⁺ (20/2/30 mol%) were obtained under the excitation of a pulse 980 nm NIR diode laser. The spectra are shown in Fig. 7, in which the measuring time ranges from 0 μs to 500 μs, and the spectral wavelength are in the range of 500–700 nm. Obviously, in the 30 mol% Mn²⁺ doped NPs, the weak green emission still exists at 350 μs, which shows the “single-red-band” luminescence in Fig. 3. Furthermore, a faster decrease in the intensity of the green emission can be clearly seen in 30 the mol% Mn²⁺ doped nanocrystals compared with the Mn²⁺-absent nanocrystals. Moreover, the decrease rate of the intensity of the red light emission is slower in the Mn²⁺ doped samples. Moreover, the intensity of the red band emission in the range of 620 nm to 670 nm increases apparently in the 30 mol% Mn²⁺ doped nanoparticles at each time in the range of 0 μs to 500 μs compared with the Mn²⁺-absent nanocrystals. The results of the time-resolved emission spectra intuitively show that a longer lifetime of the red region emission of Er³⁺ will be obtained by the introduction of Mn²⁺ into NaYF₄:Yb³⁺/Er³⁺ NPs.

Fig. 7 Time-resolved luminescence spectra of NaYF₄:Yb³⁺/Er³⁺ (20/2 mol%) (a) and NaYF₄:Yb³⁺/Er³⁺/Mn²⁺ (20/2/30 mol%) (b).

According to the above mentioned analysis, the UC mechanism of Yb³⁺/Er³⁺/Mn²⁺ is proposed and shown in Fig. 7. Without Mn²⁺ doping, three UC fluorescence bands at 525 nm, 540 nm and 655 nm are detected under the excitation of 980 nm, due to the energy transfer from Yb³⁺ to Er³⁺. When Mn²⁺ ions are introduced into NaYF₄:Yb³⁺/Er³⁺, a new energy transfer process between Mn²⁺ and Er³⁺ arises, which results in a decrease in the radiative transition rate of Er³⁺:⁴S_3/2 and Er³⁺:²H_11/2 levels to the ground state, whereas the population density of Mn²⁺:⁴T₁ increases due to the resonance energy transfer. Subsequently, back-energy transfer from ⁴T₁ of Mn²⁺ to the ⁴F_9/2 level of Er³⁺ leads to an enhancement in the red emission. Since the Er³⁺:⁴F_9/2 lifetime is shorter than Mn²⁺:⁴T₁, there is no orange luminescence that corresponds to the Mn²⁺:⁴T₁ → ⁶A₁ transition (Fig. 8).

Fig. 8 Energy level diagrams of the Er³⁺, Yb³⁺ and Mn²⁺ dopant ions and upconversion mechanisms by 980 nm photon excitation.

Further luminescence enhancement in inert-core/active-shell/inert-shell nanocubes

The red emission is located in the optical transmission window of biological tissues, which shows minimum absorption and maximum penetration depth in biological tissue. Hence, it is meaningful to do some deeper work for further enhance the UC red luminescence intensity. Here, we try to synthesize δ-doped cubic NaYF₄@NaYF₄:Yb³⁺/Er³⁺/Mn²⁺@NaYF₄ (C/S_d/S) nanocrystals with the doping ratio of Yb³⁺/Er³⁺/Mn²⁺ of 20/2/30. For comparison, three other types of structures of NaYF₄:Yb³⁺/Er³⁺/Mn²⁺ (C_d), NaYF₄@NaYF₄:Yb³⁺/Er³⁺/Mn²⁺ (C/S_d) and NaYF₄:Yb³⁺/Er³⁺/Mn²⁺@NaYF₄ (C_d/S) are also prepared with the same doping proportion of Yb³⁺/Er³⁺/Mn²⁺ via similar synthetic methods. All the samples were characterized via XRD first, the results of which are exhibited in Fig. 9. It is observed that all the diffraction peaks of the samples correspond to the cubic NaYF₄ crystal (JCPDS 06-0342), which indicate that all the crystals keep the cubic phase. According to the Scherrer formula, the average size of the as-prepared C_d, C_d/S, C/S_d and C/S_d/S nano-particles are 9.3 nm, 13.8 nm, 14.2 nm and 19.4 nm, respectively.

Fig. 9 XRD patterns of NaYF₄:Yb/Er/Mn (20/2/30 mol%) (C_d), NaYF₄:Yb/Er/Mn (20/2/30 mol%) @NaYF₄ (C_d/S), NaYF₄@NaYF₄:Yb/Er/Mn (20/2/30 mol%) (C/S_d), and NaYF₄@NaYF₄:Yb/Er/Mn (20/2/30 mol%) @NaYF₄ (C/S_d/S).

Fig. 10 shows the TEM images and energy dispersive X-ray analysis (EDXA) of the different stages for the synthesis of the C/S_d/S samples. As shown, the average size and homogeneity of the nanocubes increase increasing shell layers. The average diameters of C, C/S_d and C/S_d/S are about 8 nm, 11 nm, and 18 nm, respectively, which are in accordance with the XRD analysis results. These are deemed as ideal sizes for use as biological labels. Moreover, with an increase in the reaction process, the nanoparticles show a clearer cubic shape, which indicates better crystallinity. Furthermore, it further indicates the success of the core–shell and core–shell–shell wrapping. The results of the EDXA analysis show auxiliary evidence for the successful preparation of all the C, C/S_d and C/S_d/S structures. By analyzing the EDXA data obtained from the inert NaYF₄ core, we find that the elements in this NPs are only C, O, Na, Y and F, and no Yb³⁺/Er³⁺/Mn²⁺ ions are detected in the sample. The carbon and oxygen elements may be ascribed to the substrate, and as a result, it is certain that the core NPs have been successfully synthesized. The detection of Yb³⁺, Er³⁺ and Mn²⁺ atoms in the C/S_d structure NPs confirms that the obtained sample is a core–shell structure, with an NaYF₄:Yb³⁺/Er³⁺/Mn²⁺ shell on the surface. By comparison with the C/S_d structure, it is noted that the ratio of Y³⁺/(Yb³⁺/Er³⁺/Mn²⁺) increases as an inert NaYF₄ shell coating on the surface of C/S_d. This indicates that the doping is predominantly confined to the center layer and the C/S_d/S structure UCNPs have been synthesized successfully.

Fig. 10 TEM images, EDXA patterns and schematic of the as-synthesized NaYF4 core (C), NaYF4@NaYF4:Yb³⁺/Er³⁺/Mn²⁺ (C/S_d) and NaYF₄@NaYF₄:Yb³⁺/Er³⁺/Mn²⁺@NaYF₄ (C/S_d/S) samples. The left column is in the same scale bar of 50 nm, and the middle column is in the same scale bar of 10 nm.

The UC luminescence spectra of C_d, C/S_d, C_d/S and C/S_d/S are shown and compared in Fig. 11. Obviously, the NPs with the C/S_d/S structure show the highest quantum yield of other all the C_d, C/S_d and C_d/S structures. From calculations, the red UC fluorescence exhibits an approximate double enhancement after coating with the inert NaYF₄ shell to the C/S_d structure. Due to the ultra-small C/S_d (about 11 nm) nanocrystals, the doped layer is exposed to the surface where strong quenching is expected, owing to surface recombination. However, in the C/S_d/S architecture, the outermost inert NaYF₄ shell serves as a passivation layer, which decouples the dopant ions from the surface defects, resulting in an increase in UC fluorescence. Eliminating the influence of the surface, it is interesting to see that the UC luminescence intensity of C/S_d is greater than that of C_d.

Fig. 11 Upconversion luminescence spectra of NaYF₄:Yb/Er/Mn (C_d), NaYF₄@NaYF₄:Yb/Er/Mn (C/S_d), NaYF₄:Yb/Er/Mn@NaYF₄ (C_d/S) and NaYF₄@NaYF₄:Yb/Er/Mn@NaYF₄ (C/S_d/S) nanoparticles under 980 nm laser diode excitation.

To further investigate the UC process in different structures, the UC luminescence decay curves of the red emission of C_d, C/S_d, C_d/S and C/S_d/S were detected and compared in Fig. 12. It is noted that the C/S_d/S samples have the longest lifetime of the samples. Furthermore, it can also be clearly seen that the lifetime of C/S_d is longer than that of C_d. All the results show that the samples that incorporate dopant ion S_d layers have a higher quantum yield than that of the dopant ions distributed uniformly throughout the particle, which is expressed as C_d.

Fig. 12 UC luminescence decay curves of red (660 nm) emission of NaYF₄:Yb/Er/Mn (C_d), NaYF₄@NaYF₄:Yb/Er/Mn (C/S_d), NaYF₄:Yb/Er/Mn@NaYF₄ (C_d/S) and NaYF₄@NaYF₄:Yb/Er/Mn@NaYF₄ (C/S_d/S) nanoparticles.

The UC luminescence enhancement of the inert-core/active-shell/inert-shell structure in our experiments is in accord with the work of Z. Li et al.²¹ According to their theoretical analysis results, the 2D transfer rate is greater than the 3D transfer rate owing to the fact that the inert-core/active-shell/inert-shell structure has a minimum defect density. When activators are confined in C_d, a three dimensional distribution (3D) appears, whereas in the case of doping in S_d, it shows a two dimensional (2D) distribution owing to the thinner layer thickness. A concrete analysis of the Yb³⁺, Er³⁺ and Mn²⁺ tri-doped UC energy transfer process in our UC mechanism is as follows: the Yb³⁺ ions act as donors, which provide energy to the activator Er³⁺ ions, and the Mn²⁺ ions play the role of tuning the spectra of Er³⁺. Because the UC luminescence intensity of the Er³⁺ ions relies mainly on the energy provided by Yb³⁺, the energy transfer processes of Yb³⁺ ions should be considered and studied. It is known that there are three principal energy transfer pathways for Yb³⁺, which are Yb³⁺–Yb³⁺, Yb³⁺–Er³⁺ and Yb³⁺–defect transfers. The defects provide paths for nonradiative relaxation and are always distributed randomly throughout the nanoparticles. As a result, the energy transfer processes between the Yb³⁺ ions and defects occur randomly in the C_d doping samples. However, due to the planar doped C/S_d/S structure, the energy transfer processes between Yb³⁺ ions and defects are strongly suppressed. In addition, the Yb³⁺–Yb³⁺ energy transfer is a resonant transfer; therefore, the efficiency of the energy migration among the Yb³⁺ ions gradually increases along with a decrease in the effect of the quenching of the defect. Then, the energy transfer rate between Yb³⁺ and Er³⁺ increases. Therefore, in Fig. 12, the lifetime of the ⁴F_9/2 levels of Er³⁺ increases obviously. The C_d/S structure only eliminates surface defects from the “exterior” surface of the doped core and cannot reduce inner (core) defects and suppress the energy transfer from Yb³⁺ to the inner defects in the nanocrystals. While keeping the same doping density as C_d, the C/S_d/S structure shows a shallow doping thickness; in addition, the two interfaces of the doping-layer are in contact with the conformal inert host material (NaYF₄), which reduces more surface defects than that of the C_d structure. Therefore, the C/S_d/S structure shows a superior performance than the traditional core–shell structures and further causes a greater enhancement of the UC single-red emission, which will increase the signal-to-noise ratio in biological imaging and labeling.

4. Conclusions

In summary, we revealed the UC energy transfer mechanism of Yb³⁺/Mn²⁺/Er³⁺ tri-doped in uniform cubic NaYF₄ nanoparticles through steady and dynamic fluorescence analyses. Furthermore, in order to further increase the UC red emission, a novel C/S_d/S nanostructure, which incorporates Yb³⁺/Mn²⁺/Er³⁺ ions in the middle two-dimensional layer, has been designed and synthesized via a convenient and fast hydrothermal method. In addition, for comparison, the C_d, C/S_d and C_d/S structures have also been prepared in our work. The measurement results of UC luminescence spectra and UC luminescence decay curves show that the C/S_d/S structure has the highest quantum yield and longest lifetime compared with the C_d, C/S_d and C_d/S structures. This is due to the fact that the C/S_d/S nanoparticles have smallest amount of defects and the highest energy transfer rate. The results suggest an effective method to significantly improve the red-light emission of lanthanide-doped nanocrystals, which show great potential application in vivo bioimaging.

Acknowledgements

This study is supported by the Science and Technology Research Project of Henan Province (No. 142102210389), the National Science Foundation of China (Grant no. 21103043) and the Program for Innovative Research Team (in Science and Technology) in University of Henan Province (No. 13IRTSTHN017).

Notes and references

1. F. Wang, R. Deng, J. Wang, Q. Wang, Y. Han, H. Zhu, X. Chen and X. Liu, Nat. Mater., 2011, 10, 968.
2. F. Wang, Y. Han, C. S. Lim, Y. H. Lu, J. Wang, J. Xu, H. Y. Chen, C. Zhang, M. Hong and X. Liu, Nat. Lett., 2010, 463, 1061.
3. M. Haase and H. Schafer, Angew. Chem., Int. Ed., 2011, 50, 5808.
4. J. Shan and Y. Ju, Nanotechnology, 2009, 20, 11969.
5. D. Tian, D. Gao, B. Chong and X. Liu, Dalton Trans., 2015, 44, 4133.
6. L. Cheng, C. Wang and Z. Liu, Nanoscale, 2013, 5, 23.
7. G. Chen, J. Shen, T. Y. Ohulchanskyy, N. J. Patel, A. Kutikov, Z. Li, J. Song, R. K. Pandey, H. Agren, P. N. Prasad and G. Han, ACS Nano, 2012, 6, 8280.
8. X. Chen, W. Xu and L. Zhang, Adv. Funct. Mater., 2015, 25, 5462.
9. P. Yuan, Y. H. Lee and M. K. Gnanasammandhan, Nanoscale, 2012, 4, 5132.
10. J. Tang, L. Chen and J. Li, Nanoscale, 2015, 7, 14752.
11. Y. Li, X. Cao and G. Wang, Sci. Adv. Mater., 2014, 6, 1037.
12. C. Zha, X. Kong and X. Liu, Nanoscale, 2013, 5, 8084.
13. J. Shen, G. Chen and T. Y. Ohulchanskyy, Small, 2013, 9, 32139.
14. G. Chen, H. Ågren and T. Y. Ohulchanskyy, Chem. Soc. Rev., 2015, 44, 1680.
15. X. Chen, D. Peng and Q. Ju, Chem. Soc. Rev., 2015, 44, 1318.
16. X. Huang, Opt. Lett., 2015, 40, 3599.
17. Q. Ma, J. Wang and X. Dong, Chem. Eng. J., 2015, 260, 222.
18. G. Tian, Z. Gu and L. Zhou, Adv. Mater., 2012, 24, 1226.
19. X. Guo, W. Song and C. Chen, Phys. Chem. Chem. Phys., 2013, 15, 14681.
20. F. Shi, J. Wang and X. Zhai, CrystEngComm, 2011, 13, 3782.
21. Z. Li, W. Park, G. Zorzetto, J.-S. Lemaire and C. J. Summers, Chem. Mater., 2014, 26, 1770.

---