Fast and Efﬁcient Removal of Uranium onto a Magnetic Hydroxyapatite Composite: Mechanism and Process Evaluation

: The exploration and rational design of easily separable and highly efﬁcient sorbents with satisfactory capability of extracting radioactive uranium (U)-containing compound(s) are of paramount signiﬁcance. In this study, a novel magnetic hydroxyapatite (HAP) composite (HAP@ CoFe 2 O 4 ), which was coupled with cobalt ferrite (CoFe 2 O 4 ), was rationally designed for uranium(VI) removal through a facile hydrothermal process. The U(VI) ions were rapidly removed using HAP@ CoFe 2 O 4 within a short time (i.e., 10 min), and a maximum U(VI) removal efﬁciency of 93.7% was achieved. The maximum adsorption capacity (Q max ) of the HAP@CoFe 2 O 4 was 338 mg/g, which demonstrated the potential of as-prepared HAP@CoFe 2 O 4 in the puriﬁcation of U(VI) ions from nuclear efﬂuents. Autunite [Ca(UO 2 ) 2 (PO 4 ) 2 (H 2 O) 6 ] was the main crystalline phase to retain uranium, wherein U(VI) was effectively extracted and immobilized in terms of a relatively stable mineral. Furthermore, the reacted HAP@CoFe 2 O 4 can be magnetically recycled. The results of this study reveal that the suggested process using HAP@CoFe 2 O 4 is a promising approach for the removal and immobilization of U(VI) released from nuclear efﬂuents. and H.P.; Resources, M.S. and D.C.; Funding acquisition, M.S., N.C. and D.C.; Formal analysis, T.O. and J.T.; Writing—Review & Editing, T.O., M.S. and J.T.; Supervision, M.S. and D.C. authors read manuscript.


Introduction
Uranium (U), a long-persisting and highly hazardous radionuclide, is generally released from uranium mining, nuclear weapon testing, and nuclear accidents. Uranium released from untreated wastewater poses serious threats to aquatic life and human health [1][2][3][4]. Because of the high mobility of uranium(VI) ions under oxidizing conditions and the consequent environmental risks, the retardation and immobilization of UO 2 2+ , which is the most stable uranium specie, have attracted considerable attention [5]. Adsorption is the most facile and promising approach to the treatment of metal ions because of its low cost, high efficiency, and ease of operation [6][7][8][9][10].
Many cost-effective and ecofriendly adsorbents with components ubiquitous in the environment have been fabricated to remove U(VI) or other heavy metal ions, including iron and manganese oxides or hydroxides, organic matter, silicates, phosphate minerals, clay minerals and their modified forms [11][12][13][14][15][16][17]. Hydroxyapatite [Ca 10 (PO 4 ) 6 (OH) 2 , HAP] is a promising absorbent material because it can effectively immobilize considerable amounts of U(VI) due to its unique physical, chemical, mechanical, and biological properties [18,19]. Because UO 2 2+ exhibits a strong affinity for PO 4 3− , and their product uranyl phosphate generally has low solubility under most conditions, the phosphate group in the HAP structure dominates the transport and transformation behavior of U(VI) in nature [5,[20][21][22]. Autunite, meta-autunite, torbernite, meta-torbernite, and uranyl selenite are the principal uranyl minerals, and phosphorus can be used to promote the mineralization and retention of U(VI) to form stable secondary uranium minerals to mitigate U pollution in the environment [23]. Numerous studies have been performed on the morphology control, synthesis,

Synthesis and Characterizations of HAP@CoFe 2 O 4
Nanosized HAP that was used as a precursor was prepared using a chemical precipitation method described in our previous study [40]. The HAP@CoFe 2 O 4 composite was hydrothermally fabricated (Text S1 in Supplementary Materials). The initial mole ratio of HAP/CoFe 2 O 4 is 0.5/1. First, 30 mL of ethylene glycol and 20 mL of ultrapure water were added to the beaker, and subsequently, 0.5 mmol of HAP precursors were added to the beaker and sonicated for 15 min to ensure uniform dispersion. Then, 2 mmol of ferric chloride and 1 mmol of cobalt chloride were added to the beaker and stirred until they were dissolved. Aqua ammonia was titrated to increase the pH of the solution to 10.0 with magnetic stirring at room temperature for 1 h. The mixed solution was transferred into a 100 mL Teflon-lined, stainless-steel autoclave and stored at 180 • C for 24 h to facilitate reactions in an oven. Then, the autoclave was cooled to room temperature naturally, and the precipitates were obtained. The precipitates were centrifuged and washed with Processes 2021, 9,1927 3 of 13 alcohol and ultrapure water 3-4 times, and finally, the obtained solid was placed in a constant-temperature oven and dried at 60 • C for 4 h.
The prepared adsorbent samples' phases and crystal structures were determined through an X-ray diffraction (XRD) instrument with Cu K α radiation (λ = 1.5418 Å) operated at a 2000 W power and 10 • /min scanning rate in the 2θ range from 10 • to 90 • . Additionally, the specific surface area, magnetic properties, and functional groups of the as-prepared samples were analyzed using the Brunauer-Emmett-Teller specific surface analyzer (ASAP 2020, Micromeritics, Norcross, GA, USA), magnetometer (VSM, Dexing, Model: 250), and Bruker Tensor 27 FT-IR spectrometer (Bruker Tensor27, Germany), respectively. The surface morphology and elemental compositions of the HAP@CoFe 2 O 4 samples were characterized by a transmission electron microscope (JEOLJSM-2100F, Japan) and X-ray photoelectron spectroscopy (Thermo Fisher Scientific ESCALAB250Xi spectrometer, Waltham, MA, USA). The point of zero charge (pzc) of the HAP@CoFe 2 O 4 was measured by a Zeta potentiometer (NanoBrook Zeta PALS Potential Analyzer, Brookhaven Instruments, Holtsville, NY, USA).

U(VI) Removal Experiments
A series of batch sorption experiments were performed to investigate the U(VI) removal performance of HAP@CoFe 2 O 4 . The solution volume was 50 mL. The influence of pH was investigated in the range of 2.0-6.0. The dosage of the sorbent varied from 0.1 to 0.3 g/L. A dosage of 0.2 g/L of HAP@CoFe 2 O 4 exhibited excellent U(VI) removal ability and was thus selected as the optimal dosage for the experiments, while the initial U(VI) concentration and contact time varied. The residual U in the solution after filtration was determined using a uranium microanalyzer. The amount of U(VI) ions adsorbed by the adsorbent at time t was estimated using Equation (1): In Equation (1), Q t , C 0 , C t , V, and m denote the adsorption capacity of HAP@CoFe 2 O 4 for U(VI) at time t, initial U(VI) concentration (mg/L), U(VI) concentration measured at time t, volume of the reaction system, and the adsorbent's mass, respectively.

Characterization
The X-ray diffraction (XRD) patterns of the HAP@CoFe 2 O 4 are illustrated in Figure 1a Figure 1b indicated that the HAP@CoFe 2 O 4 exhibited a specific surface area of 12.161 m 2 /g, which suggested that the as-obtained HAP@CoFe 2 O 4 could be beneficial for interface reactions because of the sufficient active sites. The N 2 adsorption-desorption isotherm of HAP@CoFe 2 O 4 was a type IV curve and belonged to the H3 type hysteresis loop, which indicated that HAP@CoFe 2 O 4 is a flat slit structure. The average pore size of the product calculated according to the Barrett-Joyner-Halenda (BJH) method was 15.292 nm.
A broad peak near 3448.3 cm −1 , which can be attributed to the stretching vibration of the -OH groups, was observed in the Fourier-transform infrared (FT-IR) spectra of the HAP@CoFe 2 O 4 ( Figure 1c). This spectrum, which exhibited bands at 2360.99 cm −1 and between 1476.14 and 1383.85 cm −1 could be explained by the adhesion of CO 2 from the atmosphere during the synthesis of the HAP in highly alkaline conditions [41][42][43]. The peak at 1638.06 cm −1 was related to the -COO group, possibly because of the citric acid added during synthesis. The peaks at approximately 572.89 and 602.94 cm −1 represented asymmetric bending vibration, and the peak at 1044.82 cm −1 represented symmetric stretching vibration; these were attributed to the tetrahedron PO 4 3− groups of HAP. For FT-IR, Co/Fe-O bonds are typical in the 580-598 banding [44][45][46]. In this case, the peak of the Co/Fe-O bond may have overlapped with the typical vibration of the PO 4 3− group of HAP. The charge on the surface of the material affects its adsorption performance. As displayed in Figure 1d, the pHpzc of the HAP@CoFe 2 O 4 was approximately 2.8, even in the solution with 15 ppm U(VI). A broad peak near 3448.3 cm −1 , which can be attributed to the stretching vibra the -OH groups, was observed in the Fourier-transform infrared (FT-IR) spectra HAP@CoFe2O4 (Figure 1c). This spectrum, which exhibited bands at 2360.99 cm −1 a tween 1476.14 and 1383.85 cm −1 could be explained by the adhesion of CO2 from mosphere during the synthesis of the HAP in highly alkaline conditions [41][42][43]. Th at 1638.06 cm −1 was related to the -COO group, possibly because of the citric acid during synthesis. The peaks at approximately 572.89 and 602.94 cm −1 represented Because the morphology including size and shape of nanomaterials considerably affected the physical and chemical properties of the product, the morphology of the HAP@CoFe 2 O 4 was investigated through transmission electron microscopy (TEM). As presented in Figure 1e, HAP@CoFe 2 O 4 morphology showed a typical bundle-like structure with numerous CoFe 2 O 4 nanoparticles. CoFe 2 O 4 was evenly dispersed and tightly adhered to the surface of the HAP. The results from hysteresis loop analysis showed that the as-prepared HAP@CoFe 2 O 4 possessed excellent magnetic properties, suggesting that HAP@CoFe 2 O 4 could be easily separated using an extra magnet within a short time after the reaction (Figure 1f). Compared with other separation technology (e.g., filtering and centrifugal separation) applying an external magnetic field and using magnetic material(s) allows the adsorbent to be more quickly recovered from the reaction system and thus saving time and cost. These results revealed that the HAP@CoFe 2 O 4 composites were successfully synthesized and can be applied in the treatment process of eliminating U(VI) from the aqueous solution.

U(VI) Removal by Magnetic HAP@CoFe 2 O 4
Dosage experiments were conducted for confirming the optimal dosage of adsorbents. As shown in Figure 2a, with an increase in the dosage, the adsorption sites increased in the solution, which were more conducive to the binding of UO 2 2+ , leading to the improvement of removal efficiency. The HAP@CoFe 2 O 4 maintained the high efficiency adsorption of U(VI). The HAP considerably dominated the adsorption process, whereas the adsorption contribution of the CoFe 2 O 4 to U(VI) was insignificant.

Adsorption Isotherms and Kinetics
To investigate the adsorption behavior, Langmuir (Equation S2) and Freundlich (Equation S3) isotherm models were used to simulate the adsorption process of the HAP@CoFe2O4 for U(VI). The correlation coefficients (R2) of the HAP@CoFe2O4 were 0.9390 and 0.8663 for the Langmuir and Freundlich isotherm models, respectively ( Figure  3a and Table 1). This result indicated that the adsorption process of U(VI) onto Because pH is a critical factor affecting the interaction between HAP@CoFe 2 O 4 particles and U(VI), experiments were performed to investigate the effect of the solution pH on U(VI) adsorption by changing the pH value (2.0-6.0) of the solution (Figure 2b). When the pH value was less than 3.0, U(VI) was mainly present in the form of uranyl ions (UO 2 2+ ) in the solution [47]. Moreover, HAP@CoFe 2 O 4 is not stable under the conditions with a pH below 3.0 because the HAP might dissolve. A higher pH reduced protonation in the solution, and the U(VI) changed. At a low pH, the high H + concentration of the solution competed intensely with the UO 2 2+ , which limited the UO 2 2+ adsorption by the HAP@ CoFe 2 O 4 . At pH = 3.0, the adsorption capacity considerably increased and reached the maximum. As the pH increased, the adsorption of U(VI) by HAP@CoFe 2 O 4 weakened marginally. When the pH was 3.0 or higher, the charge of as-fabricated HAP@CoFe 2 O 4 composite decreased and became negative [19]. Meanwhile, under high pH conditions, the UO 2 2+ began to hydrolyze and form new species, such as (UO 2 ) 2 (OH) 2 2+ , (UO 2 ) 3 (OH) 5+ , (UO 2 ) 4 (OH) 7+ , UO 2 (OH) + , and (UO 2 ) 3 (OH) 7− . The affinity of these species is generally lower than that of the UO 2 2+ [48]. Therefore, the HAP@CoFe 2 O 4 composite does not easily absorb these species. Nevertheless, the HAP@CoFe 2 O 4 particles still maintained high adsorption of U(VI), which indicated that the composite material could efficiently adsorb U(VI) in a wide pH range, making it capable of removing uranium from contaminated wastewater under various conditions.
The reaction system comprising 0.2 g/L of sorbent and approximately 15 mg/L U(VI) exhibited the highest removal efficiency (Figure 2c). Figure 2d indicated that the adsorption efficiency of the HAP@CoFe 2 O 4 was as high as 93% in 10 min. Initially, sufficient active surface sites were present, which resulted in a high adsorption efficiency and intensive interaction between the HAP@CoFe 2 O 4 and U(VI). On prolonging the contact time, U(VI) ions occupied most of the active sites of the HAP@CoFe 2 O 4 , and the adsorption reached equilibrium at 30 min. Even when the time was increased, stable adsorption efficiency was maintained.

Adsorption Isotherms and Kinetics
To investigate the adsorption behavior, Langmuir (Equation (S2)) and Freundlich (Equation (S3)) isotherm models were used to simulate the adsorption process of the HAP@CoFe 2 O 4 for U(VI). The correlation coefficients (R 2 ) of the HAP@CoFe 2 O 4 were 0.9390 and 0.8663 for the Langmuir and Freundlich isotherm models, respectively ( Figure 3a and Table 1). This result indicated that the adsorption process of U(VI) onto HAP@CoFe 2 O 4 could be better interpreted using the Langmuir isotherm model than the Freundlich isotherm. On the basis of the Langmuir isotherm model, the Q max of the HAP@CoFe 2 O 4 for U(VI) was 338 mg/g, which was almost consistent with the experimental results. The two lower correlation coefficients may be due to the combined effects of physical adsorption and surface precipitation. In the process of the HAP@CoFe 2 O 4 absorbing uranium, the dominant adsorption mechanism is related to the initial uranium concentration. Surface adsorption may contribute considerably to the total uranium adsorption at low uranium concentrations [49]. The fitting constants for pseudo-first-order (Equation (S4)) and pseudo-second-order (Equation (S5)) kinetic models are displayed in Figure 3b and Table 2. The pseudo-secondorder kinetic model exhibited greater R 2 (0.9922) than that of the pseudo-first-order kinetic model (0.9790), which indicated that the kinetic process of the HAP@CoFe 2 O 4 for U(VI) Processes 2021, 9,1927 7 of 13 was well described by the pseudo-second-order kinetic model. Furthermore, the adsorption capacity (67.61 mg/g) calculated using the pseudo-second-order kinetic model was consistent with the experimental result (68.85 mg/g), which indicated that the adsorption process of U(VI) by the HAP@CoFe 2 O 4 was controlled by chemical adsorption. The fitting constants for pseudo-first-order (Equation (S4)) and pseudo-second-order (Equation (S5)) kinetic models are displayed in Figure 3b and Table 2. The pseudo-secondorder kinetic model exhibited greater R 2 (0.9922) than that of the pseudo-first-order kinetic model (0.9790), which indicated that the kinetic process of the HAP@CoFe2O4 for U(VI) was well described by the pseudo-second-order kinetic model. Furthermore, the adsorption capacity (67.61 mg/g) calculated using the pseudo-second-order kinetic model was consistent with the experimental result (68.85 mg/g), which indicated that the adsorption process of U(VI) by the HAP@CoFe2O4 was controlled by chemical adsorption.

U(VI) Removal Mechanisms
XRD, FT-IR, and XPS were performed to understand the adsorption mechanism of U(VI) by the HAP@CoFe2O4 composite. The XRD patterns (Figure 4a) revealed that a uranium-containing compound was formed. After adsorption, numerous diffraction peaks were observed at 2θ = 10. . These results were in good agreement with the standard pattern of autunite, which can be explained by the dissolution of the HAP and then precipitation of the autunite on the surface of the HAP [50]. The presence of autunite indicated a successful fixation of uranium on the adsorbent. The XRD pattern also illustrated the presence of CoFe2O4 in the adsorbed material, which proved that the adsorbent was still magnetic after the reaction.
FT-IR analysis (Figure 4b) revealed that both the adsorbents before and after the reaction exhibited a distinct single and wide peak at approximately 3445 cm −1 , which was attributed to the -OH groups. This result indicated that the existence of -OH groups in the HAP@CoFe2O4 is a typical feature of HAP [51]. The peak at 1616 cm −1 is characteristic of -COO, which could be attributed to the absorption of CO2 in air during the test [40,41]. The peaks at 1044.82 and 602.94 cm −1 , (before adsorption) and 1007.83 and 600.90 cm −1 (after adsorption) were attributable to the PO4 3-group in the adsorbent [40,52], and the  6 , PDF no. 72-2117]. These results were in good agreement with the standard pattern of autunite, which could be explained by the dissolution of the HAP and then precipitation of the autunite on the surface of the HAP [50]. The presence of autunite indicated a successful fixation of uranium on the adsorbent. The XRD pattern also illustrated the presence of CoFe 2 O 4 in the adsorbed material, which proved that the adsorbent was still magnetic after the reaction.
FT-IR analysis (Figure 4b) revealed that both the adsorbents before and after the reaction exhibited a distinct single and wide peak at approximately 3445 cm −1 , which was attributed to the -OH groups. This result indicated that the existence of -OH groups in the HAP@CoFe 2 O 4 is a typical feature of HAP [51]. The peak at 1616 cm −1 is characteristic of -COO, which could be attributed to the absorption of CO 2 in air during the test [40,41]. The peaks at 1044.82 and 602.94 cm −1 , (before adsorption) and 1007.83 and 600.90 cm −1 (after adsorption) were attributable to the PO 4 3− group in the adsorbent [40,52], and the position and intensity of the characteristic peaks in the adsorbent changed considerably before and after the reaction. The results revealed that the PO 4 3− in the HAP@CoFe 2 O 4 is involved in the reaction. After the adsorption reaction, a new peak appeared at 546.327 cm −1 , which corresponded to Co/Fe-O bonding [53,54]. A new Processes 2021, 9, 1927 8 of 13 absorption peak of the HAP@CoFe 2 O 4 after the reaction appeared at 911 cm −1 , which was characteristic of UO 2 2+ , providing strong evidence of uranium loading. The essence of removing U(VI) by using the HAP@CoFe 2 O 4 is the chemical reaction between UO 2 2+ and the HAP. The UO 2 2+ was adsorbed and eventually incorporated into the Ca 2+ -PO 4 3− -OH according to Equation (2): Processes 2021, 9, x FOR PEER REVIEW 8 of 13 before and after the reaction. The results revealed that the PO4 3-in the HAP@CoFe2O4 is involved in the reaction. After the adsorption reaction, a new peak appeared at 546.327 cm −1 , which corresponded to Co/Fe-O bonding [53,54]. A new absorption peak of the HAP@CoFe2O4 after the reaction appeared at 911 cm −1 , which was characteristic of UO2 2+ , providing strong evidence of uranium loading. The essence of removing U(VI) by using the HAP@CoFe2O4 is the chemical reaction between UO2 2+ and the HAP. The UO2 2+ was adsorbed and eventually incorporated into the Ca 2+ -PO4 3--OH according to Equation (2): (2) The XPS survey spectra (Figure 5a) revealed the peaks of C 1s, O 1s, Ca 2p, P 2p, Fe 2p, and Co 2p, which indicated that the CoFe2O4 was successfully loaded on the HAP. A new strong double peak for the antisymmetric vibration of [O=UVI=O] 2+ appeared in the XPS spectra of the HAP@CoFe2O4 reacted with U(VI), which was consistent with the EDS elemental mapping results ( Figure S1), suggesting that uranium was successfully adsorbed on the surface of the material. The high resolution of the U 4f spectrum (Figure 5b) can be well fitted at binding energies of 383.3 (U 4f7/2) and 394.0 eV (U 4f5/2), indicating that the valence of uranium did not change during the adsorption process [55,56]. The Co spectrum ( Figure S2a) revealed two peaks at binding energies 782.58 and 786.84 eV, which were attributed to Co 2p3/2 and its shake-up satellites, respectively, whereas peaks at higher binding energies (~797.81 and 804.44 eV) corresponded to Co 2p1/2 and their shakeup satellites, respectively [57]. The Fe 2p spectrum ( Figure S2b) displayed into two peaks at binding energies of 712.13 and 726.02 eV, which respectively correspond to Fe 2p3/2 and Fe 2p1/2, evidencing the existence of Fe 3+ [58][59][60]. As displayed in Figure 5d, after reaction with U(VI), the P 2p peaks shifted slightly towards higher binding energy, which indicated that the presence of phosphorus considerably affected the adsorption of uranium [19]. The spectra of O 1s (Figure 5c) can be divided into four peaks, namely, anion oxide (530.64 eV, Fe2O4 2− ), phosphate group (531.53 eV, PO4 3− ), hydroxyl groups (532.50 eV, −OH), and adsorbed groups (H2O, 533.57 eV), respectively. The peak area ratio of -OH varied considerably (from 50.33% to 40.26%) before and after U(VI) adsorption, which evidenced that the surface hydroxyl groups (-OH) played a key role in the adsorption of U(VI) through the sharing of electrons to form U-O bonds [61,62].

Preliminary Evaluation of the U(VI) Removal Process
The suggested schematic of the removal of the U(VI) from wastewater using HAP@CoFe2O4 is presented in Figure 6, 93.7%, where the U(VI) was removed within 10 min. As a magnetic separable material, the advantage of using HAP@CoFe2O4 is that the adsorption/separation is a simple and efficient process, U(VI) can be effectively removed from wastewater and the U(VI) loaded can be easily separated from water using magnetic field, which is beneficial for the minimization of secondary waste. The process using HAP@CoFe2O4 as an adsorbent for the removal of U(VI) has economic feasibility. The use The XPS survey spectra (Figure 5a) revealed the peaks of C 1s, O 1s, Ca 2p, P 2p, Fe 2p, and Co 2p, which indicated that the CoFe 2 O 4 was successfully loaded on the HAP. A new strong double peak for the antisymmetric vibration of [O=UVI=O] 2+ appeared in the XPS spectra of the HAP@CoFe 2 O 4 reacted with U(VI), which was consistent with the EDS elemental mapping results ( Figure S1), suggesting that uranium was successfully adsorbed on the surface of the material. The high resolution of the U 4f spectrum (Figure 5b) can be well fitted at binding energies of 383.3 (U 4f 7/2 ) and 394.0 eV (U 4f 5/2 ), indicating that the valence of uranium did not change during the adsorption process [55,56]. The Co spectrum ( Figure S2a) revealed two peaks at binding energies 782.58 and 786.84 eV, which were attributed to Co 2p 3/2 and its shake-up satellites, respectively, whereas peaks at higher binding energies (~797.81 and 804.44 eV) corresponded to Co 2p 1/2 and their shake-up satellites, respectively [57]. The Fe 2p spectrum ( Figure S2b) displayed into two peaks at binding energies of 712.13 and 726.02 eV, which respectively correspond to Fe 2p 3/2 and Fe 2p 1/2 , evidencing the existence of Fe 3+ [58][59][60]. As displayed in Figure 5d, after reaction with U(VI), the P 2p peaks shifted slightly towards higher binding energy, which indicated that the presence of phosphorus considerably affected the adsorption of uranium [19]. The spectra of O 1s (Figure 5c) can be divided into four peaks, namely, anion oxide (530.64 eV, Fe 2 O 4 2− ), phosphate group (531.53 eV, PO 4 3− ), hydroxyl groups (532.50 eV, −OH), and adsorbed groups (H 2 O, 533.57 eV), respectively. The peak area ratio of -OH varied considerably (from 50.33% to 40.26%) before and after U(VI) adsorption, which evidenced that the surface hydroxyl groups (-OH) played a key role in the adsorption of U(VI) through the sharing of electrons to form U-O bonds [61,62].

Preliminary Evaluation of the U(VI) Removal Process
The suggested schematic of the removal of the U(VI) from wastewater using HAP@CoFe 2 O 4 is presented in Figure 6, 93.7%, where the U(VI) was removed within 10 min. As a magnetic separable material, the advantage of using HAP@CoFe 2 O 4 is that the adsorption/separation is a simple and efficient process, U(VI) can be effectively removed from wastewater and the U(VI) loaded can be easily separated from water using magnetic field, which is beneficial for the minimization of secondary waste. The process using HAP@CoFe 2 O 4 as an adsorbent for the removal of U(VI) has economic feasibility. The use of HAP@CoFe 2 O 4 for the U(VI) removal/separation could result in a significant reduction of costs due to easier separation of the solid phase and energy savings. HAP@CoFe 2 O 4 can be considered to be an effective and environmental material for the elimination of U(VI) from mining wastewater.

Conclusions
This study developed a promising and easily separable composite material, HAP@CoFe2O4, for removing uranium ions from radioactive wastewater. A U(VI) removal efficiency of 93.7% can be achieved by the HAP@CoFe2O4 in 10 min, and the maxi-

Conclusions
This study developed a promising and easily separable composite material, HAP@CoFe 2 O 4 , for removing uranium ions from radioactive wastewater. A U(VI) removal efficiency of 93.7% can be achieved by the HAP@CoFe 2 O 4 in 10 min, and the maximum adsorption capacity was 338 mg/g. Therefore, HAP@CoFe 2 O 4 can be used as an emergency material to treat uraniumcontaining radioactive wastewater generated in nuclear accidents. The UO 2 2+ were transferred to the HAP@CoFe 2 O 4 by reacting with dissolved calcium and phosphate, thereby forming a more stable compound-autunite [Ca(UO 2 ) 2 (PO 4 ) 2 (H 2 O) 6 ]. The adsorption and incorporation are the main ways for U(VI) removal. Simulation of the interaction process between the HAP@CoFe 2 O 4 and UO 2 2+ by using adsorption isotherm models and adsorption kinetic models revealed that the adsorption process followed the Langmuir isotherm and pseudo-second-order dynamic model, which indicated that the adsorption of U(VI) by the HAP@CoFe 2 O 4 was mainly controlled by chemical adsorption. Moreover, the magnetic property of the as-prepared HAP@CoFe 2 O 4 was measured and found to be used for magnetic separation. Therefore, the developed process using HAP@CoFe 2 O 4 as an adsorbent is promising to mitigate uranium pollution generated in a nuclear accident, and a pilot scale test for process optimization is planned in the future.

Data Availability Statement:
The data presented in this study are available in this article.