Facile Synthesis of Hollow MgO Spheres and Their Fluoride Adsorption Properties

In the present work, the hollow MgO spheres were synthesized through a facile wet-chemistry method. X-ray diffraction, scanning electron microscopy, transmission electron microscopy, and energy dispersive spectrum are employed to characterize the as-prepared sample. 0ose results indicate that the average size of the as-prepared MgO hollow spheres is about 2μm. Also, the MgO hollow spheres have the polycrystalline and porous structure, which would provide large specific surface area and plenty active sites for fluoride adsorption. 0e adsorption properties of the MgO hollow spheres towards fluoride are investigated. 0e fluoride adsorption kinetics of theMgOhollow spheres fits well the pseudo-second-ordermodel. Also, the kinetic data revealed that the fluoride adsorption was rapid, more than 83% of fluoride could be removed within 40min.0e fluoride adsorption capacity of the hollowMgO spheres is larger than 182.4mg/g at a pH of 7.0.0e as-prepared products maintain excellent adsorption performance in the pH range of 3–11. In addition, the adsorptionmechanism also has been discussed. From the FTIR results, a hydroxyl and carbonate coexchangemechanism is proposed. It is believed that the hollow MgO spheres are a potential candidate for fluoride removal.


Introduction
Excessive fluorine in drinking water is one of the world problems [1,2]. Only in the concentration range of 0.5-1.0 mg/L, fluoride-contained water is good for human health [3]. However, when the concentration of fluorine goes beyond this range, long-term intake of the fluoride-excessive water would cause dental and skeletal fluorosis. us, the World Health Organization (WHO) classified fluoride as one of the contaminants in ground water [4]. Over 200 million people drink fluoride-excessive water every day [5]. Many methods have been developed to the remove the excessive fluoride from drinking water, such as precipitation [6], ion exchange [7,8], nanofiltration [3,9], reverse osmosis [10,11], electrodialysis [12,13], and adsorption [14,15]. Due to the low cost and simplicity of operation, adsorption seems to be the attractive one among those technologies [16,17]. Various absorbents have been investigated for the fluoride removal [18][19][20][21][22]. For example, Zhang et al. applied bauxite in fluoride removal, the adsorption capacity of which is 1.775 mg/g [19]. Ku and Chiou studied the defluoride property of the alumina and found that the maximum fluoride removal capacity is 16.3 mg/g when pH � 5 [23]. However, the low adsorption capacity and narrow pH application range largely restrict the applications of those absorbents [24,25].
us, developing novel fluoride adsorbents with high adsorption capacity, low price, and wide pH application range is very important.
Magnesium oxide (MgO), as a low-cost absorbent, is attractive in fluoride removal due to its strong affinity, high adsorption capacity, nontoxic nature, and limited solubility in water [22,26,27]. It is reported that the MgO absorbents can efficiently remove fluoride in wide pH range. Liu found that the fluoride adsorption capacity of pure MgO was 4.5 mg/g, and the adsorption performance remained stable in the pH range of 2-10 [28]. Maliyekkal also found that the MgO/Al 2 O 3 composite can efficiently remove fluoride in the pH range of 3-10 [29]. However, the adsorption capacity of those absorbents is still limited. Many efforts have been taken to achieve high-performance MgO absorbents [30]. It is well accepted that the morphology can greatly affect the corresponding adsorption performance [31]. With the development of science and technology, nanostructures always exhibit better properties [32,33]. Among various nanostructures, hollow and porous nanostructures always possess higher surface area, subsequently bringing more surface defects. ose defects can work as active sites and accordingly increase the adsorption property [34]. us, hollow and porous structures are favorable for absorbents. Many pioneer works reported that the hollow-structured MgO absorbents would exhibit better adsorption properties. Kuang et al. synthesized spherical-like MgO nanostructures by calcinate magnesium acetate, which exhibited high adsorption performance for metal ions [35]. Song prepared the hierarchical flower-like MgO hollow spheres extremely through a solvothermal method, which are the excellent arsenic absorbents [36]. However, those methods always require high temperature, high pressure, and complex operation. It is still a big challenge to obtain the MgO hollow absorbents through a simple method for high-performance fluoride removal. In the present work, the MgO hollow spheres were synthesized through a facile wet-chemical method. e operation temperature is as lower as 100°C, and high temperature, high pressure, and autoclave are no longer needed. e as-prepared MgO hollow spheres are with the uniform sizes of 2 μm in average. Also, the MgO hollow spheres have the polycrystalline and porous structure, which would provide large specific surface area and plenty active sites for fluoride adsorption. e fluoride adsorption kinetics of the MgO hollow spheres fits well the pseudo-second-order model. Also, the kinetic data reveal that the fluoride adsorption was rapid, more than 83% of fluoride could be removed within 40 min. e MgO hollow spheres can efficiently remove fluoride at the condition of acid, neutral, and even weak alkaline. e adsorption capacity is over 182.4 mg/g at neutral condition. e fluoride adsorption mechanism of the absorbents is also investigated. It is found that, besides the hydroxyls, the surface carbonates formed during the synthesis process also can be exchanged with fluoride.

Experimental
2.1. Synthesis. All reagents were purchased from Sinopharm Chemical Reagent Co., Ltd. (China) with analytical grade. In the synthesis process, 1 g MgSO 4 ·7H 2 O and 3 g urea were added into the mixed solution of 20 mL deionized water and vigorously stirred for 10 min. en, the abovementioned solution was transferred into a 100 mL conical flask, sealed, and heated at 100°C for 720 min. After reaction, the white precipitate was collected by centrifugation. e resultant precipitate was washed and dried at 60°C. After annealing at 500°C for 4 h, the white MgO hollow spheres can be obtained.

Characterization.
e morphologies of the samples were examined by field-emission scanning electron microscopy (FE-SEM; JEOL JSM-7500F) and high-resolution transmission electron microscopy (HRTEM; FEI Talos F200X). e crystal structure of the samples was determined by X-ray diffraction (XRD; Bruker D8) with Cu Kα radiation. e infrared spectrum of the samples was tested by Fourier-transform infrared spectrometer (FTIR; ermo Scientific Nicolet 8700).

Adsorption Experiments.
e fluoride stock solution with the concentration of 1000 ppm was prepared by dissolving NaF in deionized water and stored under dark conditions at 4°C. e fluoride standard solutions with different concentrations were prepared by rational dilution of the stock solution. e isotherm adsorption tests were carried out in a centrifuge tubes containing 10 mg absorbent and 10 mL fluoride solution. en, the centrifuge tubes were shaked in a shaker at 25°C for 12 h. e pH value was not further adjusted. en, the upper layer solution was carefully pipetted and filtered through a 0.45 μm filter for collection. e fluoride-ion-selective electrode PF-202-CF was applied to measure the residual fluoride concentration of the solution. e residual MgO hollow spheres were centrifuged, washed, and dried in an oven at 60°C.
In the kinetics test, the initial fluoride concentration was 10 mg/L, the dose of adsorbents was 1.0 g/L, and the volume of the fluoride solution was 100 mL. e samples were shaken at 150 rpm in the shaker at 25°C. At certain time, 6 mL of supernatant solutions was pipette and centrifuged for the remaining fluoride concentration determination.
In the experiment of pH effect on fluoride adsorption, the pH value was adjusted by HCl and NaOH solution. In the experiment of the coexisting anions' effects on fluoride adsorption, the concentration of coexisting anions was set as 300 mg/g.

Results and Discussion
3.1. Characterization. Firstly, the morphology of the MgO hollow spheres was characterized by SEM. Figure 1(a) presents the SEM image of the MgO hollow spheres. It can be clearly seen that the MgO hollow spheres are with the uniform sizes of 2 μm in average. Figure 1(b) shows the highresolution SEM images of the MgO hollow spheres. Obviously, the as-prepared MgO hollow spheres have rough external surface, which is probably due to the chaos accumulation of small nanoparticles. In the direction of the white arrow, the broken spheres can be clearly observed, and the hollow nature of MgO spheres is verified. During the experimental process, the precursors were firstly synthesized, and then, the MgO hollow spheres can be obtained after the heat treatment. Figure 1(c) presents the XRD patterns of the precursors and the final MgO hollow spheres. All peaks of the precursors can be indexed to Mg 5 (CO 3 ) 4 (OH) 2 •(H 2 O) 4 hydromagnesium (JCPDS 70-1177) [5]. After annealing at 600°C for 2 h in the air, the XRD pattern has totally changed, which can be indexed to (111), (200), and (220) planes of the hexagonal structure MgO (JCPDS card no. 89-4248). Also, no peak for precursor hydromagnesium can be identified, indicating that the precursor is fully decomposed to MgO.

Adsorption Properties.
e fluoride adsorption performances of the MgO hollow spheres were systematically studied. Due to the hollow structure and porous surface, the as-prepared samples are expected with good adsorption properties. Figure 3(a) presents the fluoride adsorption kinetics of the MgO hollow spheres. It can be seen that the adsorption happens rapidly in the first 40 min, where over 83% of the fluoride ions can be adsorbed. en, the adsorption capacity increases gradually until it reaches the equilibrium, where more than 95% of the fluoride ions can be removed by the MgO hollow spheres. To further quantify the changes of fluoride adsorption with time on the MgO hollow spheres, the Lagergren pseudo-first-order Model (1) [37] and Ho's pseudo-second-order Model (2) [38] were used to simulate the kinetics.
ln q e − q t � ln q e − k 1 t, where q e and q t are the amounts of adsorbed fluoride at equilibrium and at any time t; k 1 and k 2 are the equilibrium rate constants for pseudo-first-order and pseudo-secondorder sorption. e adsorption kinetic experimental data fitted into the pseudo-second-order model are presented in Figure 3(b). Also, Table 1 shows the summarized results fitted by the pseudo-first-order and pseudo-second-order models. e higher correlation coefficient value of the pseudo-secondorder model (R 2 � 0.998) indicates that the fluoride adsorption kinetics of the MgO hollow spheres fit well with the pseudo-second-order kinetics model. When the initial fluoride concentration is 10 mg/L, the equilibrium adsorption capacities (qe, cal) can be calculated as 9.54 mg/g, which is in agreement with experimental equilibrium adsorption capacities.  the MgO hollow spheres increases. e adsorption capacity of the spheres is 119.3 mg/g when the equilibrium concentration is 80.7 mg/g. As the equilibrium concentration increases, the adsorption capacity of the MgO hollow spheres is larger than 182.4 mg/g. Langmuir and Freundlich models are applied for data fitting, and the results are presented in Figures 4(b) and 4(c). Table 2 summarizes the isotherm parameters computed from the Langmuir and Freundlich models. e R 2 value of the Langmuir and Freundlich fitting model is 0.962 and 0.996, respectively. e higher value of R 2 suggests that the fluoride adsorption isotherm of the MgO hollow spheres follows the Freundlich models. e results are similar with the previous reported MgO fluoride adsorbent [26]. A adsorption capacity comparison of different adsorbents is listed in Table 3. It can be seen that the MgO hollow spheres have a much higher adsorption capacity than other adsorbents. e excellent fluoride adsorption performance can be attributed to the hollow structure and porous surface. e hollow structure would bring relatively large surface area and more

Equations
Pseudo-first-order kinetic model Pseudo-second-order kinetic model ln(q e − q t ) � ln q e − k 1 t (t/q t ) � (1/k 2 q 2 e ) + (t/q e ) C 0 (mg/L) qe, exp (mg/g) Note: k 1 is the adsorption rate constant for the pseudo-first-order reaction (1/min). k 2 is the rate constant for the pseudo-second-order reaction (g/(mg min)). q e and q t are the amounts of solute sorbed at equilibrium and at any time t (mg/g), respectively. qe, exp is the adsorption capacity evaluated from batch experiment, while qe, cal is the adsorption capacity calculated on the basis of the pseudo-first-order and pseudo-second-order equations. Advances in Condensed Matter Physics active sites for fluoride adsorption. Commonly, the adsorbents work when its surfaces contact with the fluoride solution, indicating that only the outer surface of normal sphere works during the adsorption process. However, the MgO hollow spheres have two surfaces, an inner and outer surface. ey both can work during the fluoride adsorption process, which largely improve the utilization of the adsorbents. Furthermore, the porous structure can provide abundant channels for fluoride ion diffusion, which would efficiently improve the adsorption performance of the absorbent.
It is well accepted that the pH value of the fluoride solution can largely affect the performance of the adsorbents.
us, the fluoride adsorption property of the MgO hollow spheres were studied under different pH values. It can be seen from Figure 4 that, within 2-10, the pH value has little influence on the fluoride removal efficiency of the MgO hollow spheres. However, when the pH value is beyond 11, the fluoride removal efficiency of the absorbent dramatically decreases. e quick reduction of fluoride removal efficiency of the MgO hollow spheres in the alkaline condition could be attributed to competition of hydroxyl ions with fluoride for adsorption sites [46]. In practical application, various anions exist in the actual groundwater; thus, it is useful to investigate the interference of coexisting ions on the fluoride adsorption of MgO hollow spheres. As presented in Figure 6, these anions at different Table 2: Langmuir and Freundlich adsorption isotherm parameters for fluoride on the hollow MgO spheres.

Equations
Langmuir model Freundlich model q e � q m k L C e /1 + k L C e q e � k F C e Values 200 0.030 0.962 162.9 2.73 0.996 Note: C e is the equilibrium concentration of fluoride (mg/L); q e is the amount of fluoride adsorbed on per weight of adsorbent after equilibrium (mg/g); q m represents the maximum adsorption capacity of fluoride on per weight of adsorbent (mg/g); k L is the Langmuir constant related to the energy of adsorption (L/mg); q m and k L were calculated from the slope and intercept of the linear plots of C e /q e vs C e . e Freundlich constant k is correlated to the relative adsorption capacity of the adsorbent (mg/g), and 1/n is the adsorption intensity.

Adsorption Mechanism.
To reveal the adsorption mechanism, zeta potential analysis was firstly employed. Figure 7 presents the zeta potential of the MgO hollow spheres at different pH values. When the pH � 2, the zeta potential of the sample is 13.8 eV. Under acidic conditions, as the pH value increases, the zeta potential of the MgO hollow spheres slightly decreases. However, when the pH value continuously increases to alkaline, the zeta potential of the MgO hollow spheres drops sharply. It can be seen that the zero point of potential of the MgO hollow spheres is close to pH 10. e higher zero point of the MgO hollow spheres is benefit to the fluoride adsorption via electrostatic attraction, subsequently resulting in the good fluoride adsorption performance in neutral and even weak alkaline conditions.
Obviously, only zeta potential is inadequate to illustrate the adsorption mechanism of the MgO hollow spheres; thus, the FTIR analysis was also employed. Figure 8 indicates the FTIR spectra of the MgO hollow spheres before and after fluoride adsorption. ere are a broad-band peak located at 3466 cm −1 and a relatively small peak located at 1641 cm −1 existing in both the samples, which correspond to the stretching vibration modes of OH bands and the bending vibration of the H-O-H band of the adsorbed water, respectively [43]. After fluoride adsorption, the FTIR peak at 3695 cm −1 become strong, suggesting the formation of Mg(OH) 2 on the surface of MgO hollow spheres during the adsorption process [47]. Furthermore, before fluoride adsorption, the peak at 429.7 cm −1 can be obviously observed, which corresponds to the Mg-O stretching vibration.
However, after adsorption, the peak at 429.7 cm −1 vanishes, and a new peak at 492.7 cm −1 appears. e new peak can be assigned to the Mg-F stretching vibration, indicating that MgF forms during the adsorption process. For the MgO hollow spheres, the peak at 1453 cm −1 can be clearly observed, which corresponds to the asymmetric stretching vibration of carbonates [48]. MgO, as basic oxide, easily reacts with the CO 2 molecule in the air, forming a magnesium carbonate species on its surface [26]. It is noteworthy that, after adsorption, the intensity of the peak at 1453 cm −1 largely decreases, indicating the great reduction of the surface carbonates during the fluoride adsorption process. e hydroxyl exchange adsorption mechanism is well accepted in fluoride adsorption. In our case, we also found that the surface carbonates also can be exchanged with F − during the adsorption process.
As presented in Figure 9, the schematic fluoride adsorption mechanism is proposed on the MgO hollow spheres. It is well accepted that the MgO can react with   H 2 O, forming MgOH on the surface. By exchanging with the surface hydroxyls, fluoride ions can be adsorbed on the surface. In acidic condition, the large amount of H + ions are conducive to the adsorption equilibrium; thus, the MgO hollow spheres exhibit higher adsorption property in acidic condition. When the pH value increase, the redundant hydroxyl ions are not in favor for the fluoride removal, which will result in the decrease of the removal capacity. Furthermore, due to the preparation environment, there is a considerable quantity of carbonates adsorbed on the surface of MgO hollow spheres. During the adsorption process, those surface carbonates also can exchange with the fluoride ions in the water, which is benefit to the adsorption property of the absorbents. e coexchange of the hydroxyls and surface carbonates would largely increase the adsorption performance of the MgO hollow spheres in neutral and alkaline conditions. In this way, the MgO hollow spheres the exhibit superior performance in fluoride adsorption over a wide pH range.

Conclusions
In conclusion, the MgO hollow spheres have been synthesized via a facile wet-chemistry method. e hollow and porous structure endorses the MgO hollow spheres the good fluoride adsorption property. e removal efficiency of the MgO hollow spheres to fluoride is very fast, and the result can be well fitted into the pseudo-secondorder rate kinetic model. e adsorption capacity of the MgO hollow spheres is larger than 182.4 mg/g at pH 7.0. e MgO hollow spheres maintain excellent fluoride adsorption performance over a wide pH range. From the FTIR results, a hydroxyl and carbonate coexchange mechanism is proposed. It is believed that the as-prepared MgO hollow spheres can be a potential candidate for fluoride removal.
Data Availability e raw data required to reproduce these findings cannot be shared at this time as the data also form part of an ongoing study.

Conflicts of Interest
e authors declare that they have no conflicts of interest.