A facile approach to synthesize SiO2 · Re2O3 (Re = Y, Eu, La, Sm, Tb, Pr) hollow sphere and its application in drug release

Multifunctional SiO2 · Re2O3 (Re = Y, Eu, La, Sm, Tb, Pr) hollow spheres (HSs) have been fabricated using an acidic Re3+ ion solution. Under ultraviolet radiation, functional HSs emit different colors of light according to the different rare-earth ions embedded into the shell of SiO2 hollow spheres. The as-prepared hollow capsules were characterized by X-ray diffraction, transmission electron microscopy, high-resolution transmission electron microscopy, Brunauer-Emmett-Teller method, scanning electron microscopy, and energy-dispersive spectrometry. Drug loading and release experiments have been carried out using SiO2 · Eu2O3 HSs that acted as drug carriers. The results demonstrate that the multifunctional HSs exhibit a high storage capacity and the ability of retaining drug stability and activity, which indicates that the as-synthesized fluorescent hollow capsules are a potential candidate as drug delivery materials.

Herein, we report a one-step self-driven process to synthesize multifunctional HSSs under an acidic condition with rare-earth ion assistance. Compared with Wang's report, the synthetic approach of HSSs is simpler. Being synthesized with the assistance of rare-earth ions, the as-prepared HSSs can emit bright fluorescence under ultraviolet radiation, which is convenient to be detected in real time if it is used in biological applications. Typical drug loading and release experiments are carried out using our prepared multifunctional HSSs, SiO 2 · Eu 2 O 3 HSs.

Synthesis of monodisperse silica spheres
Silica spheres with a diameter of 200 to 500 nm were prepared by the hydrolysis of TEOS in the mixture of ethanol, water, and ammonium using the Stöber process [37][38][39].

Synthesis of SiO 2 · Re 2 O 3 hollow spheres
In a typical synthesis, silica spheres (0.06 g) were added to 20 mL Re(NO 3 ) 3 (0.06 mol/L) and stirred for 30 min. The pH of the solution is 4.5 (adjusted with dilute nitric acid). The mixture was transferred into a Teflon-lined stainless autoclave (capacity 25 mL) and heated at 250°C for 12 h. After the products naturally cooled down to room temperature, they were washed with deionized water and separated by centrifugation (4,000 rpm) for three times and then dried at 60°C for 4 h in the air.

Drug storage and release
The steps of drug storage and release are as follows: 1. SiO 2 · Eu 2 O 3 HSs (1 g) were added into a 50-mL hexane solution containing 40 mg/mL ibuprofen (IBU). The mixture was sealed and stirred for 24 h. Then the sample was separated by centrifugation and dried at 60°C in the air. The filter was characterized by UV-visible (UV-vis; 264 nm) spectroscopy. 2. The dry SiO 2 · Eu 2 O 3 loaded with IBU (0.1 g) was immersed into 50 mL of simulated body fluid (SBF; pH = 7.4) at 37°C and stirred at the rate of 100 rpm. Three milliliters from the top of the solution was used for release measurement at different intervals, and then 3 mL of fresh SBF is added into the solution to keep the volume unchanged.

Characterization and instruments
The characterization and instruments used are detailed as follows: 1. The samples were characterized by X-ray diffraction (XRD) with a Philips X'Pert Super diffractometer (Amsterdam, The Netherlands) with graphitemonochromatized Cu Kα radiation (λ = 1.54178 Å) in the 2θ range of 1.5°to 10°and 10°to 80°.
The samples used for characterization were ultrasonically dispersed in absolute ethanol for 30 min before the TEM and HRTEM tests.

Characterization of SiO 2 · Eu 2 O 3 HSs
Newly prepared silica spheres were used to fabricate HSSs. The monodispersed SiO 2 spheres with an average diameter of 230 nm ( Figure 1A) were fabricated using the Stöber method [37][38][39] and acted as the template. The hollow SiO 2 · Eu 2 O 3 HSs were uniform, as shown in the HRTEM image in Figure 1B, whose size was nearly unchanged. XRD curves in Figure 1C demonstrate that both the SiO 2 sphere and SiO 2 · Eu 2 O 3 hollow sphere are amorphous (compared with ICSD #174). The absence of diffraction peaks for Eu 2 O 3 was owing to the few content of Eu 2 O 3 in the sample. Figure 2 shows the HRTEM image and energy-dispersive spectrometer (EDS) analysis of SiO 2 · Eu 2 O 3 HSs. A large number of holes with different sizes on the surface of SiO 2 · Eu 2 O 3 HSs could be observed in Figure 2A, which belonged to a range of mesoporous structures according to the diameter of holes. The SiO 2 · Eu 2 O 3 HSs with numerous mesoporous structures indicated that they are potential drug carriers for application in medicine, e.g., targeting therapy. The results of the EDS analysis showed that the content of O, Si, and Eu was 72.43%, 25.15%, and 2.22%, respectively. The microcontent of Ge (0.19%) was due to the impurity coming from the reagent of Eu 2 O 3 . The SiO 2 HSs were amorphous according to their XRD pattern, so the lattice fringe that appeared on the HRTEM image ( Figure 2B ions that formed a mesoporous structure with different orientations. The measured emission spectrum curve of SiO 2 · Eu 2 O 3 HSs is shown in Figure 3. The strong red emission peak further suggested that Eu 3+ existed in the surface of the SiO 2 hollow sphere. The emission spectrum of SiO 2 · Eu 2 O 3 HSs consisted of peaks mainly located in the wavelength range from 570 to 700 nm. These peaks corresponded to transitions from the excited state 5 D 0 to the ground state 7 F J (J = 0, 1, 2, 3, 4) of the 4f 6 configuration of Eu 3+ , as marked in Figure 3. Luminescence originating from transitions between 4f levels is predominant due to electric dipole or magnetic dipole interactions [40][41][42][43][44]. As can be seen in Figure 3, the strong red emission peak at 612 nm originating from the electric dipole transition 5 D 0 to 7 F 2 was the dominant band in the measured spectrum. The emission peak at around 590 nm was attributed to the 5 D 0 to 7 F 1 transition. The peaks located at 648 and 695 nm corresponded to 5 D 0 to 7 F 3 and 5 D 0 to 7 F 4 transitions, respectively. Figure 4 shows the Brunauer-Emmett-Teller (BET) N 2 adsorption-desorption isotherms and the pore size distribution of the as-prepared SiO 2 · Eu 2 O 3 HSs. The BET specific surface area and the total pore volume of the SiO 2 · Eu 2 O 3 HSs were measured to be 308.6 m 2 /g and 0.307 cm 3 /g, respectively. The pore diameter distribution was relatively wide according to the data of the adsorption branch of the isotherm. The as-prepared SiO 2 · Eu 2 O 3 HSs with a mesoporous structure may possess good performance in drug delivery efficiency, catalytic activity, and so on.
Influencing factors of the synthetic process of SiO 2 · Re 2 O 3 (Re = Y, Eu, La, Sm, Tb, Pr) hollow structures The experiments showed that the pH value of the solution, reaction temperature and time, and different rareearth ions and concentrations played an outstanding role in the synthesis of SiO 2 · Re 2 O 3 hollow structures, which are discussed in detail as follows.

Effect of the pH value of the solution
The pH value of the solution was adjusted with dilute nitric acid. The studied pH range was from 7 to 3 under the following reaction conditions: Re 3+ = 0.06 mol/L and T = 250°C. Hollow-structure particles could be obtained under the range of 4 ≤ pH < 5.5, and the optimum pH value was 4.5. No hollow structure products appeared when 6 ≤ pH ≤ 8. No HSSs appeared when 2 < pH < 3. Normally, a few HSSs could have emerged in the product at the conditions of 3 < pH < 4.0 or 5 < pH < 6 if the   reaction time was more than 10 h. The detailed results are shown in Additional file 1: Table S1 and Figure S3. It is known that SiO 2 is an amphoteric oxide which can dissolve into an acidic or basic solvent. The experiments showed that a weak acid solution was in favor of hollow structure formation. The experiments also revealed that the experimental results were unaffected when using other acids such as HCl or H 2 SO 4 to adjust the pH value of the solution.

Effect of reaction temperature
The temperature of the hydrothermal reaction affected greatly not only the reaction (going or not) but also the reaction rate (slow or fast). Additional file 1: Figure S1 shows the TEM images of the as-prepared samples at different reaction temperatures. No hollow-structure products appeared if the temperature T < 230°C in our experiments. The morphology and size of nanocrystals became difficult to control when the temperature was up to 260°C or higher because the higher the temperature was, the faster the reaction rate was. When T = 255°C, the quality of the obtained SiO 2 · Re 2 O 3 HSs was always poor. The experiments verify that the moderate temperature was 250°C.

Effect of Re 3+ ion and its concentration
It was reported that Na 2 SO 4 and NaCl were advantageous to HSS formation [52] and the work matter was Na + cation, which was in line with our experimental data. Hereby, we investigated the synthesis of HSSs under different rareearth ions and bivalent cations.
In order to get uniform hollow structures, the optimal concentration of the rare-earth ions was usually kept in the range of 0.04 to 0.08 mol/L. The experimental data and TEM images are depicted in Additional file 1: Table S1 and Figure S2. The concentration less than 0.03 mol/L resulted in poor quality in production, and the concentration greater than 0.08 mol/L always led to products with not all having a hollow structure. The experiments showed that the lower or higher concentration of Re 3+ ions was not good for HSS formation and 0.06 mol/L was the optimal concentration.
Although the SiO 2 · Re 2 O 3 HSs were obtained based on the rare-earth ion assistance strategy, their quality was quite different under assistance of different kinds of rare-earth ions. By keeping other reaction conditions unchanged such as the pH value of the solution, reaction time, and reaction temperature, the influence of different Re 3+ ions (Re = Y, Eu, La, Sm, Tb, Pr) on the structure of the as-prepared products was investigated (see Additional file 1: Table S2 and Figure S4). Additional file 1: Figure S4 clearly shows that the influence sequence of Re 3+ was as follows: Eu 3 + ≈ Sm 3 + > Y 3 + > Pr 3 + ≈ La 3 + > Tb 3 + . Nearly all of the as-prepared samples were hollow spheres with good quality under the effect of Eu 3+ and Sm 3+ existence, and the experiments showed good reproducibility and satisfactory results. With Y 3+ , Pr 3+ , and La 3+ ions included, all of the products always formed a mixture of HSSs and core/shell structure. Furthermore, all of the samples can be formed into a hollow sphere if the reaction time is prolonged, but the yield of HSSs was lower. Only a small amount of HSs could be obtained with Tb 3+ existence. The experiments indicated that changing the reaction time did not work.
We compared the synthesis approach under the same reaction conditions using Mg 2+ , Ca 2+ , Mn 2+ , Zn 2+ , and Cu 2+ instead of Re 3+ ions. The results showed that SiO 2 · HSs could barely be obtained at the above situations, which indicated that rare-earth ion was an indispensable factor in hollow structure formation.
Experimental data showed that the rare-earth ions were advantageous to HSS formation; however, further study is needed to understand why the effect of different Re 3+ ions on the formation of HSSs has a different role.

Effect of reaction time
The reaction time will determine the deepening of the reaction after fixing other reaction conditions. Figure 5 shows the structures of the as-prepared particles with a variety of reaction time. As can be seen, rattle-type particles appeared after 6 h of reaction, and then the core of particles gradually disappeared and finally became HSs at about 8 h, meanwhile many tiny particles accompanied with the HSs. After 10 h, the shapes of HSs were clearer, though many tiny particles were around them. The tiny particles came from the dissolved SiO 2 , which disappeared with reaction time extension. The highquality HSs with clear edges were obtained when the reaction lasted for 12 h; simultaneously, the tiny particles disappeared too. It was noticed that the shell of hollow spheres was getting thinner and thinner when the reaction time was over 12 h. As can be seen, a handful of HSs had cracked after 14 h. The experiments indicated that the reaction time would significantly vary the influence on the shell thickness of SiO 2 · Re 2 O 3 HSs. Therefore, the shell thickness of HSs can be controlled by adjusting the reaction time.
From the above, our synthesis procedure of HSSs is very simple and effective compared with those previously reported.

Formation mechanism of SiO 2 · Re 2 O 3 HSs
In our experiments, SiO 2 · Re 2 O 3 HSs were synthesized in an acidic solution. It was reported that colloid SiO 2 would carry a negative charge when pH > 3 [45]. The following equilibriums existed in the intermediate between the liquid and solid interfaces [48]: When Re 3+ ions are added into the solution, an electrostatic force is produced between the surface of silica and Re 3+ , Re 3+ ions are absorbed onto the surface of SiO 2 spheres at first, and then insoluble compounds SiO 2 · Re 2 O 3 are formed. Meanwhile, SiO 2 cores keep dissolving in the acidic solution, as shown in Figure 5 (6 h). At the initial stage, most of the Re 3+ ions are absorbed onto the surface of SiO 2 spheres, and numerous insoluble tiny particles that come from the residual Re 3+ ions meet with the negative ions in the solution, as shown in Figure 5 (8 and 10 h). As the reaction continues, the tiny particles are swallowed by the SiO 2 · Re 2 O 3 lamella due to Ostwald ripening, and clear SiO 2 · Re 2 O 3 HSs are obtained at last, as shown in Figure 5 (12 h). Figure 6 is the sketch map of SiO 2 · Re 2 O 3 HS formation. In the first step, Re 3+ would be adsorbed on the surface of the silica colloid, and then the silica colloid turns into a rattle-type particle and finally turns into a HS through Ostwald ripening after being treated in hydrothermal conditions. The experiments showed that the diameter of SiO 2 · Re 2 O 3 HSs was almost the same as that of the template, which indicated that the size of SiO 2 · Re 2 O 3 HSs was determined by the SiO 2 solid spheres. Therefore, we can control the size of SiO 2 · Re 2 O 3 HSs by controlling the diameter of SiO 2 solid spheres.

Drug delivery and release
Considering that HSs have numerous mesoporous structures on the surface, they can act as drug loading capsules. IBU, a typical anti-inflammatory drug, is a good example used for drug loading experiments [49,53]. Herein, IBU was used to study the drug delivery and release behavior of nanostructured HSs.
The SiO 2 · Re 2 O 3 HSs were 1 g after loading IBU (see the 'Methods' section), and the IBU storage in nanostructured SiO 2 · Re 2 O 3 HSs reached 287.8 mg/g, which means that the as-prepared SiO 2 · Re 2 O 3 HSs have a high loading capacity.
The rate of drug release determines the drug effect. Slow and sustained drug release ensures a long drug effect. First of all, a phosphate buffer solution (PBS) of IBU (0.1 μg/mL) was prepared to find out the maximum absorption wavelength using a UV-visible spectrophotometer. The experiments indicated that the maximum absorption wavelength of IBU was 264 nm. According to the Lambert-Beer law, A = kC, where A is the absorbency, k is a constant, and C is the concentration of IBU in PBS. The insert of Figure 7A is the working curve of IBU in PBS, which was obtained by the measured absorbency of different PBS concentrations. The relationship between the concentration of IBU in PBS and absorbency was as follows: þ 0:00109 mg=ml ð Þ; its related coefficientR 2 ¼ 0:9948: The released IBU concentration in SBF could be calculated using the following equation: The total release rate of IBU can be calculated by the following equation: where R is the total release rate, C i is the IBU concentration in SBF at time i, i is the time of the IBU medium solution taking out from the SBF, and m represents the total mass of the IBU in the HSs.  Figure 7A shows the release behavior of the IBU-loaded SiO 2 · Eu 2 O 3 HSs in SBF. Compared with the pure IBU disk release in SBF, the release rate of the IBU-loaded SiO 2 · Eu 2 O 3 HSs lasted long. The drug release rate was very fast within 12 h, which showed a nearly linear relationship between drug release rate and release time at the first 12 h. Then the drug release process was very slow and finished after 60 h. The results indicated that the sustained release behavior of the drug carrier was in favor of a durative drug effect.
In order to investigate the properties of the loaded drug, the UV-vis absorption spectra of the IBU hexane solution before and after IBU loading in SiO 2 · Eu 2 O 3 HSs were measured. The results are shown in Figure 7B. Curves a, b, and c were the IBU hexane solution before drug loading, the SBF solution after the release of IBUloaded SiO 2 · Eu 2 O 3 HSs for 4 h, and the SBF solution after the release of IBU-loaded SiO 2 · Eu 2 O 3 HSs for 70 h, respectively. It was noticed that the shape of the absorption curves was essentially the same, which demonstrated that the property of IBU was not changed in the loading and release processes.
We noticed that the samples still emitted fluorescence after the experiments of drug delivery and release, which indicated that the leftover via the loading and release processes can be tracked and detected.

Conclusions
We have reported an approach of the synthesis of functional SiO 2 · Re 2 O 3 HSs using silica spheres, rare-earth ions, and an acidic environment. The size of synthesized hollow capsules can be modulated by controlling the diameter of the silica template. The facile and economical synthesis protocol is valuable and convenient for wide use. Acting as drug-loaded capsules, the SiO 2 · Re 2 O 3 HSs demonstrated much excellent properties of high payloads, retained drug activity and stability, and slow drug release rate. Furthermore, real-time detection may be carried out during drug delivery and release with SiO 2 · Re 2 O 3 HSs by measuring their fluorescence.