I. INTRODUCTION
The rapid development of nuclear industry in China in recent years arises increased concerns on management of radioactive wastes, which contain uranium isotopes in various concentrations and oxidation states [1]. Uranium is of high toxicity [2, 3] and has long half-life isotopes, and its separation from radioactive waste is of importance in reducing potential harm to human health and environment.
While solvent extraction, ion exchange and extraction chromatography are mainly used to separate uranium from different samples, magnetically assisted chemical separation (MACS) process, which combines magnetic separation with the high selectivity afforded by solvent extractants, was developed for separating heavy metals elements [4, 5]. The advantages of MACS include its high separation efficiency, cost-effectiveness and reduced secondary waste streams [6]. The MACS method has been in rapid development. In Argonne Laboratory, for example, researchers used polymeric magnetic particles coated with organic extractants (CMPO, TBP, TOPO, Cyanex 272, or Cyanex 301) to separate actinides [7], uranium [8], transuranic element [5, 9] and the harmful metallic elements [10] in certain conditions. Also, magnetic particles containing crystalline silicotitanates (CST) were used to separate cesium from milk [11]; HEDPA solution [12] and polymer magnetic particles coated with Cyanex 923, DMDBTDMA, and TODGA were used to separate lanthanide and actinide elements [13-15]; and silanized magnetic particles with CMPO/TBP coating were for europium separation [16].
In this work, preparation of polymer-magnetic particles coated with poly (styrene-divinylbenzene) (Pst-DVB) and an adsorbed layer of TBP (tributyl phosphate) were prepared and characterized. Their capability of removing uranium from aqueous solutions was evaluated.
II. EXPERIMENTAL
A. Materials
Uranium standard (GBW(E) 070173) solution was obtained from National Research Center for Certified Reference Materials, Beijing, China. TBP in purity of >99.0% was produced by Fluka. All chemicals, including styrene (Tianjin Fuchen Chemical Reagents Factory, China), divinyl benzene (J&K Chemica, China), ferric chloride hexahydrate (FeCl36H2O) (Tianjin Shuangchuan Chemicals, China), ferrous chloride tetrahydrate (FeCl24H2O), ammonia water (25%), benzoyl peroxide(BPO) (Shantou Xilong Chemical Factory, China), ethanol, oleic acid, lauryl sodium sulfate (Beijing Chemical Works, China), were of analytical grade and used without any further purification.
B. Preparation of magnetic fluid
Magnetic fluid was prepared by conventional co-precipitation method with some modifications [17]. FeCl36H2O (23.6 g) and FeCl24H2O (8.6 g) were dissolved in 500 mL deionized water under nitrogen gas and vigorous stirring at 80 ℃. NH3 H2O (25 mL, 25 wt.%) was added to the solution. Then, oleic acid (10 mL) was added drop wise into the suspension within 20 min. After several minutes, the magnetic precipitates (Fe3O4 magnetic nanoparticles) were isolated from the solvent by magnetic decantation. The precipitates were washed with deionized water for several times to remove the excess oleic acid. The magnetic precipitate can be re-dispersed in water to form a water liquid based magnetic fluid.
C. Preparation of magnetic Pst-DVB particles
Styrene (150 mL) was mixed with 10 mL DVB, 20 g magnetic fluid and 1 g BPO to form an organic phase. SDS (0.2 g) was dissolved in 500 mL deionized water. The two mixtures were mixed together to form a suspension, which was transferred to a 500 mL beaker equipped with four vertical stainless steel paddles, a condenser, a nitrogen inlet, and a stirrer. The mixture was heated to 60 ℃ in 1 hour and maintained for 2 hours. The reaction continued for 2 hours at 70 ℃, stirring at 400 rpm. The magnetic microspheres were separated by magnetic decantation and washed with deionized water and acetone several times.
D. TBP coating
The magnetic particles were washed repeatedly with deionized water and 0.1 M NaOH solution, and dried overnight in a fume hood prior to use. TBP (0.25 mL) was diluted in 2 mL ethanol, and mixed with 0.1 g magnetic particles. The mixture was placed in a sonicating bath for 10 min, and heated in a water bath at 100 ℃, stirred constantly, until the ethanol was evaporated. The particles were baked in an oven at 120 ℃ for 17 hours to obtain the TBP-coated magnetic Pst-DVB particles [18].
E. Characterization
Size and morphology of the uncoated Fe3O4 particles and the magnetic Pst-DVB particles were examined by transmission electron microscopy (TEM, JEM 2100). The magnetic Pst-DVB particles before and after TBP coating were examined by scanning electronic microscopy (SEM, JSM-6700F). The magnetization curves of the samples were measured with a vibrating sample magnetometer (VSM, PPMS-9). The magnetite content of the magnetic polymer microspheres was determined by thermogravimetry (Netzsch, STA449). FTIR spectra were taken using a Fourier transform infrared spectrometer (Jasco FT/IR-660 plus).
F. Extraction process
The extraction efficiency was evaluated using the distribution coefficient, Kd. Portions of uranium standard solution (100 mg L-1) and appropriate volumes of ultrapure water were mixed in 5 mL test tubes. Then, 30 mg of TBP-coated magnetic Pst-DVB particles were added under mixing and the total volume of suspension was made up to 3.0 mL. The suspensions were mixed at room temperature. Subsequently, TBP-coated magnetic Pst-DVB particles were separated from the suspension using a permanent NdFeB magnet and uranium concentration in the clear supernatant was measured using inductively coupled plasma mass spectrometry (ICP-MS, Element2). The uranium concentration was determined from the calibration curve, which was prepared by determining a series of diluted uranium standard solution with 2% (v/v) HNO3. The Kd, was calculated by
where Ci and Ce are respectively the initial and equilibrium uranium concentrations in the aqueous phase, V is the total volume of the suspension (3 mL), and m is the weight (30 mg) of the added TBP-coated magnetic Pst-DVB particles.
All the experiments were performed twice and mean values were used in data analysis. Control experiments without the TBP-coated magnetic Pst-DVB particles were carried out to determine the degree of uranium removal.
III. RESULTS AND DISCUSSION
A. Characterization
Structure of the synthesized Fe3O4 and Pst-DVB magnetic particles were examined using the transmission electron microscope (TEM). The Fe3O4 nanoparticles were of spherical structure, in relatively uniform size (Fig. 1(a)) of around 10 nm, while the Pst-DVB particles, containing the Fe3O4 nanoparticles (Fig. 1(b)), were sized at about 1 μm.
Morphologies of Pst-DVB particles before and after the TBP-coating were examined by scanning electron microscope (SEM) (Fig. 2). The SEM images indicated that the TBP-coating changed the surface morphology of Pst-DVB particles.
There was no hysteresis in the magnetization with both remanence and coercity being zero, showing that the Fe3O4 magnetic nanoparticles and TBP-coated magnetic Pst-DVB particles are superparamagnetic (Fig. 3). Magnetization of the Fe3O4 magnetic nanoparticles was saturated at 39.41 emu/g. After polymerization of the DVB and styrene, and the coating process of TBP, saturation magnetization of the TBP-coated magnetic Pst-DVB particles reduced to 4.32 emu/g. This is due to the embedment of Fe3O4 magnetic nanoparticles inside the magnetic polymer microspheres, the increased particle size, and the decreased unit mass content of magnetic particles.
Figure 4 shows TGA curve of the TBP-coated magnetic Pst-DVB particles. At 600 ℃, organic substances of the TBP-coated magnetic Pst-DVB particles decomposed, with just the magnetic Fe3O4 being left, the content of which was about 10% of the TBP-coated magnetic Pst-DVB particles.
From the FTIR spectra of the Pst-DVB particles before and after TBP-coating (Fig. 5), the characteristic peaks at 1600 cm-1, 1449 cm-1, 760 cm-1 and 700 cm-1 are of styrene. And a band at around 627 cm-1 and 585 cm-1 is assigned to Fe-O. In Spectrum (b), the band at 1028 cm-1 corresponding to P-O-C group indicates the presence of TBP. Also, the P=O group of TBP can be seen at 1264 cm-1.
B. Effect of HNO3concentrations on uranium uptake by TBP-coated magnetic Pst-DVB particles
Uptake of uranium in 0.05 mg L-1 concentration by the TBP-coated magnetic Pst-DVB particles was studied with nitric acid in concentrations of 0.1–0.7 mol/L. The distribution coefficient of uranium, Kd (mL/g), increased with the HNO3 concentration (Fig. 6), reaching a peak value of 109 mL/g at 5 mol/L HNO3 and decreasing at higher HNO3 concentrations. Therefore, the optimal initial nitric acid concentration of 5 mol/L was used for the TBP-coated magnetic Pst-DVB particles to remove uranium from water solutions.
C. Adsorption isotherms
Adsorption isotherm describes how a solute interacts with the adsorbent under equilibrium conditions, providing important information for understanding mechanism of the adsorption systems. Linear isotherm can be expressed as
where qeq is the amount of adsorbed solute per unit weight of adsorbent (mg/g), Kd is the distribution coefficient (mL/g) and Ceq (mg/L) is the residual liquid phase concentration after equilibrium. If the number of adsorption sites is large relative to the number of solute molecules, it is possible to use the Freundlich isotherm. The model is based on the assumption that ions are accumulated infinitely on the sorbent as described by
where Kf and n are the Freundlich constants and are indicators of adsorption capacity and intensity, respectively. If n=1, the Freundlich isotherm is identical to the linear isotherm. By taking the log of the terms, a straight line develops making it easier to obtain the slope and intercept of the line
The linear and Freundlich isotherms were determined for uranium concentration of 0.2–1 mg L-1 and are given graphically in Fig. 7. The isotherm model parameters were obtained by non-linear least square fitting of the experimental data. We had Kd=0.0963 (R2=0.9728) by using the linear model to fit the data of uranium adsorption on the TBP-coated magnetic Pst-DVB particles, while Kf = 0.1186 and 1/n=1.2391 (R2=0.9975) by using the Freundlich models. The Freundlich model fits the experimental data better, and the highest value of sorption capacity was 0.050 mg g-1, but it was not possible to determine the maximum sorption capacity because the concentration range of uranium was below saturation.
IV. CONCLUSION
The synthesized TBP-coated magnetic Pst-DVB particles show characteristics of the P-O-C and P=O groups and the changes in surface morphology. The distribution coefficient of the TBP-coated magnetic Pst-DVB particles increases with HNO3 concentration, reaching a peak value at 5 mol/L nitric acid. The adsorption equilibrium data are fitted by the linear and Freundlich models, and the results show that the Freundlich model fits the data better. In addition, the TBP-coated magnetic Pst-DVB particles enable simple removal from the suspensions using a permanent magnet. These results indicate that the TBP-coated magnetic Pst-DVB particles can be used to separate uranium from aqueous media.
Feasibility of Separation and Utilization of Caesium and Strontium from High Level Liquid Waste, Technical Reports Series No. 356
,