1 Introduction

Engineering a material into a core–shell hierarchical heterostructure has attracted extensive attention owing to the capability of fine-tuning physicochemical properties for various applications, including catalysis [1], photocatalysis [2], luminescence [3], photothermal conversion [4], lithium ion batteries [5], and ion separation [6], etc [7,8,9]. It is generally recognized that the formation of a shell on a core enables the protection of the core as an active center, improving the stability of core materials, especially active nanomaterials, e.g., a nanocatalyst [10]. Indeed, a large number of studies have demonstrated that direct synthesis of nanomaterials without a protection effect from a shell leads to the aggregation of nanomaterials and consequently performance degradation, e.g., in photocatalysis [11] and luminescence [12], of metal oxides involving calcination treatment.

Although the core–shell heterostructure engineering has been widely explored and recognized as an effective technology for synthesis of high-performance hierarchical structures, the core–shell-core hierarchical heterostructures with double cores are rarely investigated. For the first time, this paper presents the synthesis of a core–shell-core hierarchical heterostructure for luminescence applications [13,14,15,16,17,18,19,20,21,22,23,24]. The phosphor was effectively engineered into a well-constructed core–shell-core heterostructure Y2O3:Eu3+/MCM-41/YVO4:Eu3+, exhibiting enhanced red emission, along with tunable, broadened-band response to excitation. The broadening of band response of the phosphor could facilitate the fluorescent labeling applications involving organisms vulnerable to UV radiation at short wavelengths, e.g., smaller than 365 nm [25, 26]. Therefore, our phosphor shows great potential to replace conventional fluorescent organic dye and quantum dots (QDs), which thereby circumvents the insurmountable problem of the conventional bioimaging techniques, including the photobleaching and quenching of organic dyes and the toxicity of QDs [27]. Additionally, there will be many other potential applications for our prepared luminescent material, such as cathode ray tubes, plasma display panels, projection television screens, light-emitting diodes, and field emission displays [28,29,30,31,32,33,34,35,36,37,38,39].

The present synthesis of such a phosphor takes advantage of the fascinating properties (such as homogeneous mesopore distribution and satisfactory thermal stability) of a mesoporous molecule sieve working as a bridging material, in this case MCM-41. It possesses abundant hydroxyl groups on its surface, which offers an excellent platform for further modification and also facilitates the host–guest assembly chemistry [40,41,42]. Furthermore, its mesoporous structure endues tunable optical properties since the energy transfer of the emission activator [43] (namely Eu3+) of our phosphor can be varied with the extension direction along the pore walls and channels, providing a way for unraveling the photoluminescence mechanism. The hosts Y2O3 and YVO4 were selected because of suitable crystallographic sites for Eu3+ ion doping in the crystals [44]. This study aims to explore a novel core–shell-core hierarchical heterostructure Y2O3:Eu3+/MCM-41/YVO4:Eu3+, along with its photoluminescence properties. An understanding of the energy transfer and phosphorescence mechanism is investigated as well.

2 Experimental

2.1 Materials

Tetraethoxysilane (TEOS, Si(OC2H5)4), tetrabutyl ammonium bromide (TBAB, C16H36BrN) and cetyl trimethyl ammonium bromide (CTMAB, C16H33(CH3)3NBr), of AR grade, were supplied from Guanghua Chemical Factory (Shantou, China), Bodi Chemical Co., Ltd. (Tianjin, China), and Fuchen Chemical Reagents Factory (Tianjin, China), respectively. Y2O3 and Eu2O3 (high-purity grade, 99.999%) were obtained from Zhujiang Refinery Factory (Guangdong, China). All other chemicals were AR grade and used without further purification.

2.2 Synthesis of a (Y,Eu)(OH)3 hydrosol as a precursor of the Y2O3:Eu3+ core

According to a homogeneous co-precipitation method, the precursor (Y,Eu)(OH)3 was firstly synthesized using urea as a precipitation agent. A mixed solution (pH 4.3) of urea (2 mol/L), Y(NO3)3 (0.04 mol/L) and Eu(NO3)3 (0.002 mol/L) were agitated at 80 °C. When the solution turned out to be turbid, the reaction was stopped after a further 1 h reaction. The product was then collected by filtration, followed by through washing with deionized (DI) water and by re-dispersing, leading to the (Y,Eu)(OH)3-based hydrosol. The preparation of the Y3+(Eu3+)-based nitrates was described as follows: a given amount of Y2O3 and Eu2O3 was mixed to concentrated nitric acid in an excess quantity. The excessed nitric acid was volatilized by heating treatment. After cooling to room temperature, DI water was mixed to generate the final Y(NO3)3 (0.04 mol/L) and Eu(NO3)3 (0.002 mol/L) mixed nitrate solution.

2.3 Synthesis of core–shell heterostructure Y2O3:Eu3+/MCM-41

To prevent the dissolution of the core Y2O3:Eu3+ during imbedding YVO4:Eu3+ as the second core material, the surface of the nanoparticles in the (Y,Eu)(OH)3-based hydrosol was coated with a thin layer of SiO2, before further encapsulation with MCM-41, as schematically illustrated in Fig. 1. The precursor of Y2O3:Eu3+ (namely (Y,Eu)(OH)3, 1.0 g) was treated in 20 mL of ethanol by ultrasonication for 15 min, and then 3 mL of ammonia (25 wt%) was added as a catalyst. Under continuous stirring, 2.8 mL of TEOS was dropwise added, and a 1.5 h reaction at room temperature was conducted. After that, the product was collected by filtration, thoroughly washed with DI water three times, and finally dried at 90 °C for 12 h. The obtained dried product was then re-dispersed into 60.0 mL of DI water using 9.0 g of ammonia (25 wt%) as the catalyst. Separately, 0.6125 g of TBAB was dissolved into 10.0 mL of DI water. Similarly, CTMAB (2.7698 g) was mixed into 50.0 mL of DI water. All the three solutions were subsequently mixed together under stirring for 0.5 h, which was then agitated intensely, along with dropwise adding 10.48 g of TEOS. The time used for the addition of TEOS lasted for 1.5 h. Upon completion of the TEOS addition, the reaction further proceeded for 3 h at room temperature. The molar ratio of the reactants, namely CTMAB/TEOS/ammonia/TBAB, was calculated to be 0.152/1/2.8/141.2/0.038. The product was filtered, washed with DI water, and then dried at 363 °C for 12 h. The dried product in the powder form was employed as the precursor of Y2O3:Eu3+/MCM-41, which was placed in a muffle furnace for calcination treatment. The programming temperature was set as follows: a temperature ramp of 1.0 °C/min from room temperature to a final temperature of 823.0 °C, which was held for 9 h to produce the final Y2O3:Eu3+/MCM-41.

Fig. 1
figure 1

Schematic illustration of the synthesis of the present core–shell (Y2O3:Eu3+/MCM-41) and core–shell-core (Y2O3:Eu3+/MCM-41/YVO4:Eu3+) hierarchical heterostructures. The effective synthesis mainly includes the following three steps: (i) formation of protective SiO2 shell on (Y,Eu)(OH)3, (ii, iii) generation of the core–shell (ii) and core–shell-core (iii) phosphors

2.4 Synthesis of core–shell-core heterostructure Y2O3:Eu3+/MCM-41/YVO4:Eu3+

A given amount of NH4VO3 was dissolved into ammonia with a mass ratio of 1:1 by heating treatment, leading to the generation of a clear solution. A mixed solutions of Y(NO3)3 and Eu(NO3)3 was homogenized and then added into the solution of NH4VO3 (pH ~ 4), and agitated for 0.5 h to form a hydrosol in orange color. The as-synthesized Y2O3:Eu3+/MCM-41 was then added to the hydrosol under magnetic stirring for 2 h. The product was collected by filtration, drying treatment at 90 °C for 12 h, and calcination at 550 °C, resulting in the core–shell-core heterostructured phosphor Y2O3:Eu3+/MCM-41/YVO4:Eu3+.

2.5 Characterization

The X-ray diffraction (XRD) patterns of the synthesized samples in the powder form were inspected on a Dandong-aolong Y-2000 diffractometer using Cu Kα radiation at λ = 0.15418 nm with a single-crystal silicon powder as an external standard. The structure and morphology of the samples were examined using a high resolution transmission electron microscope (HR-TEM, JEOL JEM 2010). The excitation and emission spectra were recorded on a Hitachi F-2500 spectrofluorimeter equipped with a 150 W xenon lamp as the excitation source. All the measurements were carried out at room temperature. Low-temperature N2 adsorption–desorption isotherms were recorded on a Micromertics ASAP 2020 M system at liquid nitrogen temperature. Before the measurement, the sample was degassed in vacuum at 90 °C for 1 h, and then vacuum-dried at 350 °C for 8 h (less than 5 × 10−3 mmHg of vacuum degree).

3 Results and discussion

Figure 1 presents the schematic illustration of the synthesis process of core–shell and core–shell-core heterostructures Y2O3:Eu3+/MCM-41 and Y2O3:Eu3+/MCM-41/YVO4:Eu3+, respectively. For the synthesized heterostructures, their XRD patterns are shown in Figs. 2 and 3. For both core–shell (see Fig. 2a) and core–shell-core (see Fig. 2b) phosphors, the diffraction angles (2θ) appear at 28.9°, 33.7°, 48.3° and 57.2°, corresponding to (222), (400), (440), and (622) crystallographic planes of Y2O3 (in a body-centered cubic structure) respectively (JCPDS No. 25-1011) [45]. An amorphous broad peak at 22° can be an indication that the pore channels of the mesoporous molecule sieve are made from amorphous SiO2 [46, 47]. As for core–shell-core heterostructure Y2O3:Eu3+/MCM-41/YVO4:Eu3+, XRD peaks emerge at 2θ ≈ 24.6° and 49.5°. This characteristic of diffraction bands can be indexed to (200) and (312) crystallographic planes of YVO4 (with a quartet zircon structure) respectively (JCPDS No.17–0341) [48, 49], as displayed in Fig. 2b. It is also found that the characteristic XRD bands derived from Y2O3 remain almost unchanged after imbedding the second core YVO4:Eu3+ into Y2O3:Eu3+/MCM-41.

Fig. 2
figure 2

XRD patterns of Y2O3:Eu3+/MCM-41 (a) and Y2O3:Eu3+/MCM-41/YVO4:Eu3+ (b). Filled square YVO4, filled circle Y2O3

Fig. 3
figure 3

SAXRD patterns of Y2O3:Eu3+/MCM-41 (dashed line) and Y2O3:Eu3+/MCM-41/YVO4:Eu3+ (solid line)

Small angle X-ray diffraction was also employed to further characterize the core–shell-core heterostructure (see Fig. 3). Apart from two small (110) and (200) diffraction peaks of MCM-41, a strong (100) diffraction band can be observed around 2.35° [50]. According to the Brag’s law, i.e., d 100  = λ/2sinθ, the spacing d 100 can be calculated as 3.76 nm, which is consistent with the literature [51]. Such a strong (100) plane diffraction band is characteristic of mesoporous materials with a hexagonal close-packed structure, which also reveals that the formation of periodically aligned liquid crystal and texture takes places through self-assembly interactions between inorganic silicate and surfactant ions. This result verifies that the formed inorganic skeleton of the shell possesses the hexagonal mesoporous structure typical of MCM-41 [50]. On the other hand, the incorporation of YVO4:Eu3+ into Y2O3:Eu3+/MCM-41 largely broadens the (100) diffraction peak (Fig. 3), suggesting a significant loss of the hexagonal mesostructure order due to the blocking of the mesopores as a result of deposition of the phosphor YVO4:Eu3+ [50, 52].

To further investigate the structural features of the single core–shell and double core–shell heterostructures, TEM images together with the selected-area electron diffraction (SAED) patterns are presented in Fig. 4. Both the core–shell (Fig. 4a) and core–shell-core (Fig. 4d) phosphors exhibit a sphere-like core–shell structure, with ordered hexagonal arrays of mesoporous channels. Such ordered hexagonal arrays can be more clearly observed as for the core–shell phosphor Y2O3:Eu3+/MCM-41 (Fig. 4c) relative to that for the core–shell-core phosphor Y2O3:Eu3+/MCM-41/YVO4:Eu3+ (Fig. 4f); this is because the deposition of the YVO4:Eu3+ enables the hexagonal mesoporous arrays to be covered to some extent, in good consistence with the small small-angle XRD patterns.

Fig. 4
figure 4

Microstructural analysis of the phosphors Y2O3:Eu3+/MCM-41 and Y2O3:Eu3+/MCM-41/YVO4:Eu3+by means of TEM and SAED. TEM (a, c) and SAED (b) analyses of Y2O3:Eu3+/MCM-41, TEM (d, f) and SAED (e) analyses of Y2O3:Eu3+/MCM-41. c and f are the TEM images at higher magnifications in comparison with (a) and (d), respectively

The crystalline phase (indexed to Y2O3:Eu3+) can be well evidenced by the SAED pattern of Y2O3:Eu3+/MCM-41 shown in Fig. 4b. Ordered diffraction spots can be unambiguously seen, indicating the high lattice ordering of Y2O3:Eu3+ crystalline particles that exist in the single-crystalline nature [53, 54]. On the other hand, the SAED pattern of Y2O3:Eu3+/MCM-41/YVO4:Eu3+features the concentric rings with dispersed spots (Fig. 4e), which is an indication of the polycrystalline nature of YVO4:Eu3+ nanoparticles strictly confined in the mesoporous channels of the MCM-41 shell [54]. The thickness of the bridging shell MCM-41 is approximately 20 nm. Such a nano-thin shell implies the imbedded YVO4:Eu3+ particles at a nanoscale even an atomic scale (Fig. 4d). Despite the deposition of YVO4:Eu3+ onto the core–shell phosphor Y2O3:Eu3+/MCM-41, the ordered mesoporous structure of the core–shell-core heterostructure Y2O3:Eu3+/MCM-41/YVO4:Eu3+ can still be noted, without observable large YVO4:Eu3+ aggregates (Fig. 4d), revealing uniform deposition of YVO4:Eu3+ nanoparticles into the mesopore channels of the MCM-41 shell. Furthermore, from the TEM images at a higher magnification (Fig. 4f), the lattice fringes of the crystalline phase (YVO4) are distinguishable; the distance between the adjacent lattice fringes can be calculated as approximately 0.35 nm, corresponding well to the interplanar distance of YVO4 (200) crystallographic planes [50]. This result is consistent with the values obtained from JCPDS 17–0341 and with our XRD analysis.

The surface area and pore size distribution of our prepared core–shell and core–shell-core heterostructures are further probed by N2 adsorption–desorption isotherms (Fig. 5). Both the isotherms feature Langmuir IV adsorption, with H1 hysteresis loop; this is a typical adsorption–desorption curve of mesoporous material, indicating an ordered mesoporous composition of the shell material [8, 55, 56]. At a low partial pressure range (P/P0 < 0.3), the adsorption amount of N2 exhibits a linear increase with increasing the partial pressure, which can be due to the monolayer adsorption of N2 on the mesopore walls. As the partial pressure further raises beyond P/P0 of 0.3, the adsorption amount is apparently increased, as a result of the capillary condensation as generated by the N2 adsorbed in the mesoporous skeleton. With ongoing increase of the partial pressure, the curve becomes slowly upward, indicating the saturation of the N2 adsorption in the capillary. When the P/P0 is close to 1, a steep raise of the N2 adsorption can be noted, implying the filling of large pores; this is associated with the capillary condensation of interparticle pores and channels. The main information read from the N2 adsorption–desorption isotherms is also summarized in Table 1. After embedding YVO4:Eu3+ into the core–shell heterostructure Y2O3:Eu3+/MCM-41, the surface area, and pore size and volume values decrease, revealing the well-dispersed YVO4:Eu3+ nanoparticles in the ordered pore channels of the mesopore SiO2 and consequent filling of a part of mesopore channels.

Fig. 5
figure 5

N2 absorption–desorption isotherms of the samples Y2O3:Eu3+/MCM-41 and Y2O3:Eu3+/MCM-41/YVO4:Eu3+ (insets present the corresponding pore diameter distribution curves)

Table 1 Structural information of the samples Y2O3:Eu3+/MCM-41 and Y2O3:Eu3+/MCM-41/YVO4:Eu3+ as summarized from N2 adsorption–desorption isotherms

The luminescence properties of our prepared core–shell and core–shell-core heterostructures are finally investigated. Figure 6 presents the UV excitation spectra of the heterostructures Y2O3:Eu3+/MCM-41 and Y2O3:Eu3+/MCM-41/YVO4:Eu3+, as monitoired at 612 (Fig. 6a) and 620 (Fig. 6b) nm. For the sample Y2O3:Eu3+/MCM-41 with a core–shell structure, the absoprtion band at 256 nm can be ascribed to the charge transfer band (CBT) of Eu–O (Fig. 6a). By contrast, the sample Y2O3:Eu3+/MCM-41/YVO4:Eu3+ features the CBT-reated broad absorption in the range of 200–350 nm, centered at 254 nm. It also presents a typical VO4 3−-based CBT-related absorption at 307 nm, which is the main absorption characteristics of the spectrum. In the range of 350–450 nm, sharp absorption peaks can be noted, as generated by the f–f electron forbidden transition of Eu3+. These Eu3+-related peaks at 394, 363, 382, and 416 nm, can be attributed to 7F05L6, 7F05D4, 7F05G2, and 7F05D3 electron transitions, respectively. Concerning the sample Y2O3:Eu3+/MCM-41/YVO4:Eu3+, the excitation spectra shown in Fig. 6b display only a strong broad excitation band at 307 nm when monitored at 620 nm, suggesting that the excitation of Eu3+ mainly comes from the energy transfer from VO4 3− to Eu3+ that is embedded within the pore channels of the shell. This also reveals that VO4 3− can transfer energy to Eu3+ ions more efficiently in the mesopore channels.

Fig. 6
figure 6

Excitation spectra of Y2O3:Eu3+/MCM-41 (solid line) and Y2O3:Eu3+/MCM-41/YVO4:Eu3+ (dotted line) monitored at 612 nm (a) and 620 nm (b), respectively

Figure 7 shows the emission spectra of the samples Y2O3:Eu3+/MCM-41 and Y2O3:Eu3+/MCM-41/YVO4:Eu3+ at different excitation wavelengths of 255 and 303 nm. Upon excitation at 255 nm (Fig. 7a), the emission peak at 612 nm exhibits the highest intensity for the sample Y2O3:Eu3+/MCM-41, corresponding to the typical 5D07F2 electron transition of Eu3+. As for the core–shell-core sample Y2O3:Eu3+/MCM-41/YVO4:Eu3+, the biggest emission peaks are located at 616 and 620 nm (along with a shoulder peak at approximately 612 nm), corresponding to the main emission from YVO4:Eu3+. These fluorescence features can be primarily generated by the strong dependency of the emission-related ultrasensitive electron transition 5D07F2 on the surroundings and local symmetry of Eu3+ ions. Upon excitation at 303 nm, the emission intensity is much higher for Y2O3:Eu3+/MCM-41/YVO4:Eu3+ relative to Y2O3:Eu3+/MCM-41, as contributed by the emission from the phosphor YVO4:Eu3+. These results thus verify that embedding YVO4:Eu3+ into Y2O3:Eu3+/MCM-41 (Fig. 1) enhances the red emission performance under flexible excitation conditions. A schematic diagram is also given to show the luminescence routes upon excitation into both CTBs, i.e., Eu–O and VO4 3+, as well as the emission process of Eu3+ (Fig. 8).

Fig. 7
figure 7

Emission spectra of Y2O3:Eu3+/MCM-41 (solid line) and Y2O3:Eu3+/MCM-41/YVO4:Eu3+ (dotted line) with λex = 255 nm (a) and λex = 303 nm (b), respectively

Fig. 8
figure 8

Part of the energy level scheme of the sample Y2O3:Eu3+@SiO2@YVO4:Eu3+

To probe the impact of temperature on the fluorescence of the prepared core–shell and core–shell-core phosphors, temperature-dependent fluorescence properties were measured, with the results shown in Figs. 9 and 10. Note that the core–shell phosphor Y2O3:Eu3+/MCM-41 (Fig. 9) exhibits a significantly high fluorescence stability in the temparature range of 300–500 K, i.e., 27–227 °C. Such a stability can stem from the SiO2 and MCM-41 shells that protect Y2O3:Eu3+ from thermal quenching (Fig. 1), in addition to the high stability of Y2O3:Eu3+ as reported elsewhere [57]. The highly efficient fluorescence signal and a good thermal stability imply that our core–shell phosphor Y2O3:Eu3+/MCM-41 is suited for successful practical applications [58]. In contrast, the fluorescence stability against temperature variation is lowered when YVO4:Eu3+ nanoparticles are incorporated into the Y2O3:Eu3+/MCM-41 (Fig. 10a, b). The lowered stability can be due to the YVO4:Eu3+ nanoparticles that are directly exposed to heat (see Fig. 1). Despite the fluorescence degradation by heat, the fluorescence intensity still keeps a high level even at 500 K, i.e., 227 °C (Fig. 10b), revealing that it is also promising for applications of our phosphor Y2O3:Eu3+/MCM-41/YVO4:Eu3+under high-temperature conditions. It can be found that there exist around two stages of the temperature-dependent fluorescence spectra, i.e., stage I and stage II in the range of below and over 100 °C, respectively. This finding is expected to contribute to future research on the intelligent system that can sense temperatures of lower or higher than 100 °C .

Fig. 9
figure 9

Study of the impact of temperature (from 300 to 500 K) on the emission spectra at λex = 303 nm for the core–shell phosphor Y2O3:Eu3+/MCM-41. Temperature-dependent fluorescence spectra (the inset is the magnification of the box as marked

Fig. 10
figure 10

Study of the impact of temperature (from 300 to 500 K) on the emission spectra at λex = 303 nm for the core–shell-core phosphor Y2O3:Eu3+/MCM-41/YVO4:Eu3+. a Temperature-dependent fluorescence spectra (the inset is the magnification of the box as marked, b plot of the characteristic fluorescence peak intensity as a function of temperature, clearly showing the impact of temperature on the fluorescence emission properties of the core–shell-core heterostructure Y2O3:Eu3+/MCM-41/YVO4:Eu3+

4 Conclusions

In this work, a novel core–shell-core heterostructure has been presented, which consists of the phosphor Y2O3:Eu3+ (as an inner core), a mesoporous amorphous SiO2 material (MCM-41) serving as a bridging shell, and the phosphor YVO4:Eu3+ (as an outer core). The mesoporous MCM-41 shell plays a pivotal role in forming the well-constructed core–shell-core hierarchical heterostructure. Highly-ordered mesoporous channels are evidenced in the MCM-41 shell, and the ultrasmall YVO4:Eu3+nanoparticles are effectively incorporated into the MCM-41 mesoporous channels. This heterostructure with double cores exhibits enhanced red emission and tunable, broadened-band response to excitation properties, which hold great potential for fluorescence labeling applications. Furthermore, our core–shell phosphor (with only an inner core) exhibits impressive fluorescence stability in the temperature range of 300–500 K, while the core–shell-core phosphor (with double cores) shows a lowered stability against temperature variation due to the outer core (namely YVO4:Eu3+ nanoparticles) that is directly explored to heat (Fig. 1). Interestingly, such temperature-dependent fluorescence changes may be useful for the future exploration of intelligent system that can detect temperatures of higher or lower than 100 °C. This work also opens up a new avenue for design and synthesis of core–shell-core hierarchical phosphors with many promising applications, even beyond fluorescence/luminescence. Additionally, our synthetic route can also be generalized to produce double core–shell materials for various applications. For instance, these double cores can either synergistically contribute to performance enhancement, or play completely different roles in generating multifunctional materials (such as a magnetic inner core and a catalytic outer core for producing magnetically separable catalysts).