Introduction

Thorium (Th) is widely distributed in the environment and is present at low levels in water, soil, rocks, plants and animals. It is an important element not only in industrial applications but also in energy and environmental issues. Because of the extensive usage of Th for various industrial purposes and also because of its toxicity and radioactivity, pre-concentration or separation of Th in environmental and biological samples is significant [1, 2]. Numerous methods have been developed to pre-concentrate or separate Th from environmental and biological sources, including co-precipitation [3], liquid–liquid extraction [4, 5], solid-phase extraction (SPE) [6–8], extraction chromatography [9, 10], ion exchange [11], and membrane separation [12].

The SPE method uses ion surface imprinted polymers (IIP), which can avoid the grinding and sieving that can occur in other methods. IIP, the best used method for pre-concentration or separation of trace metals, also has high selectivity, good mass transfer, and fast binding kinetics [13–15]. And the IIP was used as the adsorbent of SPE and applied to the extraction of thorium ion by more and more researchers. Sibel and his fellows [16] prepared the Th(IV)-imprinted beads with N-methacryloyl-(L)-glutamic acid (MAGA) as the complexing monomer by suspension polymerization. Qun He et al. [17] synthesized Th(IV)-imprinted polymers (IIPs) by surface imprinting technology with a new functional monomer N-(o-carboxyphenyl)maleamic acid (CPMA) on the surface of silica gel. The Th(IV)-imprinted chitosan-phthalate particles, which were synthesized by the complexing monomer and epichlorohydrin as crosslinking agent, were reported by Birlik et al. [15]. We also reported the preparation and characterization of the Th(IV)-imprinted material based on the surface of silica gel with methacrylic acid (MAA) as a functional monomer [18].

Magnetic materials, which support surface imprinted polymers, have been widely used in the pre-concentration and separation process because they are simple to use [19–21], but not found in reports of the Th(IV)-imprinted polymer.

During the ion imprinting process, a functional ligand determines selectivity and affinity for the target ion. Furthermore, it is trend for IIP to choose ion coordinate with ligand. Salophen schiff bases strongly coordinate to metals ions [22–24], and were found to possess high affinity to form complexes with Th(IV) [25, 26].

Thus, to further our previous research [27, 28], the aim of this study was to synthesize a new magnetic surface Th(IV)-imprinted sorbent, based on a novel complex of N,N’-bis(3-allylsalicylidene)o-phenylenediamine (BASPDA) with Th(IV) as the functional monomer and template, and using a modified magnetic Fe3O4 particle as its support. In this manner, we have developed a new method to pre-concentrate and separate trace Th(IV). We describe, in this article, the interaction of the sorbent with the Th ion and its corresponding mechanism of adsorption. As the paragraph said, the magnetic.

Materials and methods

Materials

Azobisisobutyronitrile (AIBN, 98 %), 3-Aminopropyltriethoxysilane (APTES), maleic anhydride, salicylaldehyde, o-phenylenediamine, allylchloride, 2-propanol, Tetraethyl orthosilicate (TEOS), Arsenazo III, Chlorophosphonazo(CPA)-III, iron(II) chloride tetrahydrate (FeCl2·4H2O), iron(III) chloride hexahydrate (FeCl3·6H2O) were obtained from Aladdin (Shanghai, China). Ethylene glycol dimethacrylate (EGDMA) was purchased from EHSY (Shanghai, China), and Th(NO3)4·4H2O (SP) was purchased from Hengrui New Materials Co. Ltd. (Chengdu, Sichuan).

Unless otherwise stated, reagents of analytical purity were used for all experiments and double-distilled water (DDW) was used throughout. Standard labware and glassware were washed with 10 % HNO3 and rinsed with DDW. Arsenazo III solution was prepared by dissolving an appropriate amount of reagent in DDW. The pH adjustments were made with 0.1 mol L−1 HCl or 0.1 mol L−1 NH4OH. Standard stock solutions (1 mg mL−1) of Th(IV), U(VI), Ce(III), La(III) and Nd(III) were prepared by dissolving appropriate amounts of nitrate salts in DDW containing 1.0 mol L−1 HNO3 and were further diluted daily prior to use.

Apparatus

A Perkin-Elmer Lambda 45 UV–Visible spectrometer (USA) was used for determination of the metal ion concentration. IR Spectra (4,000–400 cm−1) were recorded on IR Prestige-21 (Shimadzu, Japan) using KBr pellets. 1H NMR spectra were recorded with a Bruker Avance III 500 MHz spectrometer with CDCl3 as the solvent. Powder XRD patterns of the particles were collected on a Rigaku D/max-RA powder diffraction meter (Rigaku, Japan), and the single-crystal XRD was collected on a Bruker APXII CCD diffractometer using graphite monochromated Mo Kα radiation (λ = 0.71073 Å). A pHs-l0C digital pH meter (Pengshun Scientific Instruments Research, Shanghai, China) was used for the pH adjustments.

Synthesis of N,N’-bis(3-allylsalicylidene)o-phenylenediamine (BASPDA)

The mechanism of BASPDA synthesis is presented in Scheme 1. The mixture of KOH (80 mmol, 4.48 g) and salicylaldehyde (80 mmol, 9.76 g) was stirred at room temperature for 30 min. Then, 6.88 g allylchloride was added to this solution, and the mixture was refluxed for 17 h. The mixture was then allowed to cool gradually to room temperature and filtered. When the solvent was removed, an oily liquid was obtained; the oily liquid was then mixed with ether and water (at a ratio of 1:1), and the organic phase was washed sequentially with 10 % NaOH solution, water and brine, and then dried with anhydrous Na2SO4. Finally, the crude 2-(vinyloxy)benzaldehyde was obtained by removing the solvent [29]. Then, the 2-(vinyloxy)benzaldehyde was incubated for 24 h at 190 °C under a nitrogen atmosphere. After cooling the mixture slightly, the light-yellow 2-hydroxy-3-vinylbenzaldehyde was obtained by distilling the mixture under a vacuum (bp, 142–145 °C, 18 mmHg) [30].

Scheme 1
scheme 1

A schematic representation of the mechanism of synthesis of the functional monomer and preparation of the Th(IV)-imprinted polymers

Full size image

Sequentially, 50 mL ethanol, followed by the synthesized 2-hydroxy-3-vinylbenzaldehyde (3.24 g) and then o-phenylenediamine (1.09 g) were added to a three-neck flask, and the mixture was refluxed for 2 h. After cooling the mixture to 0 °C, yellow solids were obtained by filtering the mixture. When the yellow compound was re-crystallized using absolute ethanol, yellow crystals were obtained; these yellow crystals were BASPDA and they were analyzed by IR and 1H NMR. IR analysis results (cm−1) are as follows: 3,452 (O–H), 2,902 (C–H), 1,637 (C=N), 1,573 (C=C), 906 (=C–H), 746 (C–H). The 1H NMR(CDCl3) δ analysis results are as follows: 13.09 (s, 1H,–OH), 8.58 (s, 2H,–CH=N), 6.86–7.32 (m, 10H, Ar–H), 6.04 (2H, m,–CH=), 5.00 (4H, d, =CH2), 3.41 (4H, d, –CH2–).

Preparation of Th(BASPDA)2 complexes

The Th(BASPDA)2 complexes were formed through a chemical reaction as previously described [25]. Briefly, BASPDA (1.58 g), Th(NO3)4·4H2O (1.10 g) and absolute ethanol (30 mL) were refluxed for 1 h, and the Th(BASPDA)2 complexes were obtained by filtering the mixture. When the yellow compound was re-crystallized using DMF and anhydrous ethanol (1:1), 2.08 g of yellow crystals were obtained. These yellow crystals were analyzed using IR, and the results (cm−1) are as follows: 420 (Th–O), 530 (Th–N), 1,637 (C=N), 1,581 (C=C).

Preparation of vinyl modified Fe3O4

Preparation of Fe3O4@SiO2

Fe3O4 was prepared as previously described [31], and SiO2 was used to prepare Fe3O4@SiO2 as was also previously described [32, 33]. Briefly, 5.00 g Fe3O4 was dissolved in 250 mL 2-propyl alcohol and 20 mL DDW by sonication for 15 min, followed by the sequential addition of 20 mL ammonia solution and 33.3 mL TEOS. The resulting product was collected by an external magnetic field after the reaction had continued for 12 h at room temperature with continuous stirring, followed by six rinses with purified water, and thorough drying.

Preparation of amino-functionalized Fe3O4@SiO2 (Fe3O4@SiO2–NH2)

As previously described [34], 4.00 g Fe3O4@SiO2 and 100 mL of toluene were added to a 250 mL three-necked flask and then ultrasonically dispersed for 15 min. APTES (8 mL) was then added into the flask, and the mixture was refluxed at 110 °C with continuous stirring for 12 h under a flow of nitrogen. After the mixture was allowed to cool gradually to room temperature, the resulting amino-functionalized Fe3O4@SiO2 was collected by filtering the mixture, washing it several times with ethanol and acetone, and then drying it.

Preparation of vinyl-functionalized Fe3O4@SiO2–NH2

As previously described [24], a solution of maleic anhydride (2.50 g) in DMF (25 mL) was added to the three neck flask containing Fe3O4@SiO2–NH2 (4.00 g) and DMF (25 mL). After the mixture was stirred for 24 h at room temperature, the resulting products were collected by filtration, washed sequentially with DMF, anhydrous ethanol and anhydrous ether, and dried under a vacuum at room temperature. Vinyl-functionalized magnetic particles were obtained.

Synthesis of magnetic surface Th(IV)-imprinted (Fe3O4@SiO2-IIPs) and non-imprinted polymers (Fe3O4@SiO2-NIPs)

The mechanism of synthesis for the magnetic surface Th(IV)-imprinted polymers (Fe3O4@SiO2-IIPs) is presented in Scheme 1. Th(BASPDA)2 complexes (1.02 g) dissolved in DMF (25 mL), vinyl-functionalized magnetic particles (2.00 g), AIBN (0.15 g) and EGDMA (6 mL) were added to a three neck flask. Then, the oxygen was removed by sonication for 15 min and nitrogen was bubbled through the mixture for 25 min. The mixture was stirred and heated under nitrogen at 60 °C for 24 h. After the mixture was cooled to room temperature, the products were collected by filtration, washed with methanol/water/DMF (1:1:1) and treated with 3 mol L−1 HCl for 24 h to remove Th(IV) from the polymers. The final products were rinsed with DDW several times until all the acid was removed, and the exiguous granules were removed by rinsing with acetone, then drying under a vacuum at 70 °C for 24 h. In this manner, the magnetic surface Th(IV)-imprinted polymers were obtained. Non-imprinted polymers (Fe3O4@SiO2-NIPs) were also prepared using the same method, except that the Th(NO3)4·4H2O was omitted.

Adsorption experiments

A series of metal ion standard solutions (Th(IV), Th(IV)/U(VI), Th(IV)/La(III), Th(IV)/Ce(III), or Th(IV)/Nd(III)) were transferred into a 25 mL color comparison tube, and the pH values were adjusted to the desired value with 0.1 mol L−1 HCl or 0.1 mol L−1 NH3·H2O. After the volume was adjusted to 25 mL with DDW, Fe3O4@SiO2-IIPs or Fe3O4@SiO2-NIPs polymers (30 mg) were added, and the mixture was shaken vigorously for 30 min. Then the mixture was centrifuged (3,000 rpm for 10 min), and the concentrations of Th ions in the solutions were determined by spectrophotometry, using Arsenazo-III (0.05 %, w/v) as a chromogenic agent in a 3 mol L−1 HNO3 medium at a wavelength of 660 nm. The concentrations of U(IV) (at 658 nm) and Ce(III)–La(III) (at 652 nm) were determined in buffer solutions of chloroacetic acid-sodium chloroacetate of pH 2.5 and 2.8, respectively, while the concentration of Nd(III) was determined at a wavelength of 670 nm using Chlorophosphonazo(CPA)-III as a indicator agent, and the concentrations were compared to a reagent blank that was used as a reference for all samples. All the experiments were carried out in triplicate. The amount of adsorption, the removal efficiency, the distribution coefficient K d (mL g−1), the selectivity coefficient k, and the relative selectivity coefficient k’ were calculated by Eqs. 1–5, respectively.

$$ Q = \frac{{C_{0} - C_{\text{e}} }}{W} \times V $$
(1)
$$ E = \frac{{C_{0} - C_{\text{e}} }}{{C_{0} }} \times 100\,\% $$
(2)
$$ K_{d} = \left( {\frac{{C_{0} - C_{\text{e}} }}{{C_{\text{e}} }}} \right)\frac{V}{W} $$
(3)
$$ k = \frac{{K_{{{\text{d}}({\text{Th}})}} }}{{K_{{{\text{d}}({\text{M}})}} }} $$
(4)
$$ K^{\prime} = \frac{{k_{{({\text{IIP}})}} }}{{k_{{({\text{NIP}})}} }} $$
(5)

where Q represents the amount of Th(IV) adsorbed on the polymer (mg g−1), E represents the percent (%) removal efficiency, C 0 and C e represent the initial and equilibrium concentrations, respectively, of the given metal ions in solution (μg mL−1), W is the amount of polymer (mg), and V is the volume of metal ion solutions (mL). K d(Th) and K d(M) represent the distribution coefficients of Th(IV) and M ions, respectively. k (IIP) and k (NIP) represent the selectivity coefficient of Fe3O4@SiO2-IIPs and Fe3O4@SiO2-NIPs, respectively.

Desorption procedure

Magnetic Th(IV)-imprinted polymers (50 mg) that had adsorbed Th(IV) ions were packed into a glass column (120 mm × 5.5 mm i.d.). The ends of the column were plugged with a small portion of glass wool to prevent loss of the sorbents during sample loading. Then, the stoichiometric HCl solutions were passed through the column by a peristaltic pump at a flow rate of 1.0 mL min−1 until no Th(IV) ion was detected in the eluent.

Results and discussion

Characteristics of the adsorbent

The FT-IR spectra

FT-IR was used to characterize Fe3O4@SiO2, Fe3O4@SiO2–NH2, vinyl-functionalized Fe3O4@SiO2–NH2, Fe3O4@SiO2-IIPs containing Th(IV), and Fe3O4@SiO2-NIPs (Fig. 1). For all six particles absorption peaks at 586 cm−1 were observed, corresponding to the Fe–O vibration from the magnetite phase. As presented in Fig. 1a, the peaks at 401, 814, 957 and 1,084 cm−1 were attributed to the stretching of SiO2 [27, 34]. The presence of adsorbed water is reflected by the –OH vibration at 1,630 and 3,441 cm−1, which suggests that the SiO2 has been successfully immobilized on Fe3O4 surface [35]. Successful amino-functionalization of Fe3O4@SiO2 was also verified by the absorption peak at 1,409 cm−1 and 3,439 cm−1 (Fig. 1b), attributed to the stretching and bending vibrations of amino groups. Moreover, there were absorption peaks from methylene groups of Fe3O4@SiO2–NH2 at 2,951 and 2,981 cm−1. The peaks at 1,720 and 1,544 cm−1 represent the stretching vibration of C=O and C=C, respectively (Fig. 1c).

Fig. 1
figure 1

FT-IR spectra: (a) Fe3O4@SiO2; (b) Fe3O4@SiO2–NH2; (c) vinyl-functionalized Fe3O4@SiO2–NH2; (d) Th(IV) Fe3O4@SiO2-IIP containing Th(IV); (e) Fe3O4@SiO2-IIPs with Th(IV) removed; (f) Fe3O4@SiO2–NIPs

Full size image

The spectra presented in Fig. 1d–f have similar characteristics. Comparing the spectrum in Fig. 1c with those in Fig. 1d–f, the vibration intensity of the methylene groups on Fe3O4@SiO2-IIPs and Fe3O4@SiO2-NIPs are greater than that on Fe3O4@SiO2–NH2. In addition, the spectrum in Fig. 1d–f had the following unique peaks: 1,500–1,600 cm−1 and 600–900 cm−1 (benzene ring); 1,635–1,641 cm−1 (C=N) (Fig. 1d–f); 420 cm−1 (Th–O stretching vibrations, Fig. 1d [36]; and 530 cm−1 (Th–N stretching vibrations, Fig. 1d) [37]. There were absorption peaks of the Th–O and Th–N groups of Th(IV) attached to Fe3O4@SiO2-IIPs at 420 cm−1 and 530 cm−1, respectively, but these peaks could not be seen in the IR spectra of Fe3O4@SiO2-IIPs and Fe3O4@SiO2-NIPs. These results suggested that the Th(IV) ion-imprinted magnetic polymer with BASPDA as a functional monomer was formed, and that the Th(IV) ion had been removed completely.

Powder and single-crystal XRD analysis

The crystalline structures of the synthesized magnetic particles were studied with XRD. There were six characteristic peaks observed for Fe3O4 (2θ = 30.3°, 35.7°, 43.6°, 53.9°, 57.4°, and 63.0°) (Fig. 2). These six peaks correspond to the six crystal faces of Fe3O4 and are (220), (311), (400), (422), (511) and (440), respectively [38]. In Fig. 2b–f, the same set of characteristic peaks as for the Fe3O4@SiO2, Fe3O4@SiO2–NH2, vinyl-functionalized Fe3O4@SiO2, Fe3O4@SiO2-NIP, and Fe3O4@SiO2-IIP were observed. In addition, the six peaks of the Fe3O4 (Fig. 2d–f) were observed for the synthesized vinyl-functionalized Fe3O4@SiO2, Fe3O4@SiO2- NIP and Fe3O4@SiO2-IIP, but the intensity of the peaks was slightly decreased in these latter 3 compounds. Thus, it was concluded that the modified magnetic particles and the imprinted polymer were in a spinel structure, and that the crystalline structure of the magnetite was essentially maintained; it was also concluded that SiO2, NH2 and BASPDA were successfully immobilized on the surface of Fe3O4.

Fig. 2
figure 2

XRD patterns of (a) Fe3O4, (b) Fe3O4@SiO2, (c) Fe3O4@SiO2–NH2, (d) vinyl-functionalized Fe3O4@SiO2–NH2, (e) Fe3O4@SiO2-NIPs and (f) Fe3O4@SiO2-IIPs

Full size image

The crystal structure of the compound Th(BASPDA)2 was determined, and X-ray intensity data were recorded on a Bruker APXII CCD diffraction meter using graphite monochromated Mo Kα radiation (λ = 0.71073 Å). A total of 47,678 reflections were measured, of which 15,646 were unique (Rint = 0.032) in the range of 1.14° < θ < 25.00° (h, −17 to 19; k, −20 to 20; l, −22 to 20), and 12434 observed reflections with I > 2σ(I) were used in the refinement on F 2. The structure was solved by direct methods with the SHELXS-97 program. All of the non-H atoms were refined anisotropically by a full-matrix least-squares analysis to give the final R = 0.0314 and wR = 0.0914 (w = 1/[σ2 (F 2o ) + (0.0477P)2 + 3.7353P], where P = (F 2o  + 2F 2c )/3 with (Δ/σ)max = 0.004 and S = 1.043 by using the SHELXL program. The hydrogen atoms were located using a difference Fourier map and refined isotropically. The details of the crystal structure were published elsewhere, and the cell structure of the crystal is shown in Fig. 3.

Fig. 3
figure 3

The shell structure of Th(BASPDA)2 crystal

Full size image
Fig. 4
figure 4

The effect of pH on adsorption of Th(IV) on the Th(IV)-imprinted polymers; 10 mL Th(IV) 1.0 μg mL−1; temperature 25 °C

Full size image

Affect of pH

The effect of pH was investigated in the pH range of 2.0 to 6.0 using the batch procedure with 10 mL of Th(IV) (1.0 μg mL−1) (Fig. 4). The adsorption percentage of the Th(IV) increased as the pH increased from pH 2.0 to 4.5, and there was a further increase between pH 4.5 and pH 6.0. At pH ≤ 2.5, the adsorption percentage was very low. From the structure of the compound Th(BASPDA)2, it was found that the interactions of Th(IV) with Fe3O4@SiO2-IIPs were the coordination of the Th(IV) with the N and O atoms on the BASPDA groups in the polymers. There were also competitive reactions between the N and O in the sorbent and the metal ions and hydrogen ions in the solution. At pH ≤ 2.5, most of N and O of the BASPDA was protonated; by increasing the pH, the ratio of the free BASPDA group to the protonated BASPDA group increases, resulting in a greater chance of interaction between BASPDA and Th(IV); at pH ≥ 7, the Th(IV) ion is easily hydrolyzed and precipitated. At pH = 4.5, the adsorption rate of Th(IV) was 99.55 %, and thus pH 4.5 was picked as the optimum pH level for subsequent experiments.

Adsorption kinetics

The adsorption rate of the imprinted polymer for Th(IV) ions was determined to be between 1 and 40 min. In the experiment, 30 mg of the polymer was added to 25 mL of 10 μg mL−1 Th(IV) aqueous solution after adjusting the pH to 4.5; the equilibration time was approximately 20 min (Fig. 5).

Fig. 5
figure 5

The adsorption rate of Th(IV) on Th(IV)-imprinted adsorbent. pH 4.5; 25 mL Th(IV) 10 μg mL−1; temperature 25 °C

Full size image

In order to investigate the mechanism of adsorption kinetics, two different kinetic models were tested to interpret the data obtained from the batch experiments [39].

The pseudo-first-order equation is:

$$ \ln (q_{\text{e}} - q_{\text{t}} ) = \ln q_{\text{e}} - k_{1} t $$
(6)

The pseudo-second-order equation is:

$$ \frac{t}{{q_{\text{t}} }} = \frac{1}{{k_{2} q_{\text{e}}^{2} }} + \left( {\frac{1}{{q_{\text{e}} }}} \right)t $$
(7)

where qe (mg g−1) and qt (mg g−1) are the adsorption amount at equilibrium and the time t (min), respectively. The qe values were calculated from the intercept of the ln(q e −q t) versus t plot. The slope of t/q t versus t is defined as the theoretical qe(cal) value of pseudo-first-order and pseudo-second-order model, respectively. k 1 (min−1) and k 2 (mg g−1 min−1) are pseudo-first-order and pseudo-second-order rate constants of adsorption, respectively.

The pseudo-first-order kinetic model indicates that the rate of adsorption site occupation is proportional to the number of unoccupied sites; the pseudo-second-order kinetic model assumes that the chemical reaction mechanisms [40] and also the adsorption rate are controlled by chemical adsorption through sharing or exchange of electrons between the adsorbate and adsorbent [41]. The correlation coefficients (R 2) of the pseudo-first-order kinetic model and pseudo-second-order kinetic model for adsorption of Th(IV) ions onto the Fe3O4@SiO2-IIPs were 0.8462, 0.9994, respectively (Fig. 6). Also, the q e value estimated from pseudo-second-order kinetic model was closer to the experimental value than that of pseudo-first-order kinetic model (Table 1). These results demonstrated that the pseudo-second-order adsorption mechanism was the best-fit model to describe the adsorption behavior of Th(IV) onto the synthesized imprinted particles, and that this adsorption mechanism was a chemical coordination process.

Fig. 6
figure 6

a pseudo-first-order kinetic model, b pseudo-second-order kinetic model

Full size image
Table 1 Kinetic parameters of the pseudo-order rate equation for Th(IV) adsorption onto Th(IV)-IIP
Full size table

Adsorption isotherm

The adsorption capacity of Fe3O4@SiO2-IIPs and Fe3O4@SiO2-NIPs for Th(IV) was investigated by batch experiments in the concentration range of 5 μg mL−1 to 90 μg mL−1. The amount of Th(IV) ions adsorbed per unit mass of the Fe3O4@SiO2-IIPs and Fe3O4@SiO2-NIPs increases with the increasing initial concentration of Th(IV), until a saturation value was achieved (Fig. 7). The maximum adsorption capacities of the imprinted and non-imprinted particles were 42.54 and 14.10 mg g−1, respectively. Compared with Fe3O4@SiO2-NIPs, the adsorption capacity of Th(IV) on Fe3O4@SiO2-IIPs was almost three times that of Fe3O4@SiO2-NIPs, which indicates that the cavities created after removal of Th(IV) in Fe3O4@SiO2-IIP are complementary to the imprint ion in size, shape, coordination geometries and chemical functionality, whereas in Fe3O4@SiO2-NIPs there is a random distribution of ligand functionalities in the polymeric network.

Fig. 7
figure 7

The effect of Th(IV) initial concentration (C 0) on the adsorption quantity (Q) of Th(IV)-IIP and NIP adsorbent, pH 4.5; sample volume 25 mL; temperature 25 °C

Full size image

The equilibrium experimental data were analyzed using the Langmuir and the Freundlich isotherms.

The Langmuir isotherm is represented by the following expression [42]:

$$ \frac{{C_{\text{e}} }}{{Q_{\text{e}} }} = \frac{1}{{Q_{\max } b}} + \frac{{C_{\text{e}} }}{{Q_{ \max } }} $$
(8)

The Langmuir dimensionless separation factor RL is expressed as follows:

$$ R_{\text{L}} = \frac{1}{{1 + bC_{0} }} $$
(9)

The Freundlich isotherm equation is [43]:

$$ \ln Q_{\text{e}} = \ln K_{\text{F}} + \left( \frac{1}{n} \right)\ln C_{\text{e}} $$
(10)

Where Q e (mg g−1) is the amount of Th(IV) adsorbed at equilibrium, C e (μg mL−1) is the equilibrium concentration of Th(IV) in the solution, Q max is the maximum adsorption quantity; b (L mg−1) is the Langmuir constant; and K F and n are the Freundlich constants, which indicate the adsorption capacity and the adsorption intensity, respectively, in the Freundlich isotherm equation.

The Langmuir isotherm is a fixed monolayer adsorption onto a surface with a finite number of identical sites. The Freundlich isotherm is an empirical equation based on adsorption onto a heterogeneous surface. The Langmuir linear equation (C e/Q e = 0.06549 + 0.02155C e, R 2 = 0.9963) and Freundlich isotherm equation (lnQ e = 2.3728 + 0.4519lnC e, R 2 = 0.9561) for the adsorption of Th(IV) were obtained from experimental data (Figs. 8, 9). From the value of correlation coefficients (R 2), it was concluded that the Langmuir isotherm is a better fit to our data and best describes the adsorption isotherm in our experimental results. The values of R L (Table 2), which were all between 0 and 1, indicate that favorable adsorption conditions were present in this process. Furthermore, the calculated value of Q max (46.40 mg g−1) from the Langmuir equation was comparable to the value (42.54 mg g−1) obtained in our experiments, indicating that the process involves a monolayer adsorption and that Th(IV) was adsorbed onto the surface of the synthesized imprinted particles.

Fig. 8
figure 8

The Langmuir isotherm for the adsorption of Th(IV) on the Th(IV)-imprinted, polymers; pH 4.5; temperature 25 °C

Full size image
Fig. 9
figure 9

The Freundlich isotherm for the adsorption of Th(IV) on the Th(IV)-imprinted, polymers; pH 4.5; temperature 25 °C

Full size image
Table 2 RL values in the Langmuir isotherm of Th(IV)-IIP
Full size table

Uptake thermodynamics

In this research, the thermodynamic parameters for the adsorption of Th(IV) onto the Fe3O4@SiO2-IIP were studied in the temperature range of 298.15 – 318.15 K. The thermodynamic parameters can be determined using the following equations:

$$ \Updelta G^\circ = \Updelta H^\circ - T\Updelta S^\circ $$
(11)
$$ \Updelta G^\circ = - RT\ln K $$
(12)
$$ \ln K = \frac{\Updelta S^\circ }{R} - \frac{\Updelta H^\circ }{RT} $$
(13)
$$ K = \frac{{q_{\text{e}} }}{{C_{\text{e}} }} $$
(14)

where K is equilibrium constant, T (K) is the temperature, and R is the gas constant; and ΔS° (kJ mol−1 K−1), ΔH  (kJ mol−1) and ΔG° (kJ mol−1) are the standard entropy change, enthalpy and Gibbs free energy, respectively.

The linear equation representing the thermodynamic parameters is lnK = 0.0382−10.588/T, with the slope represented by −ΔH°/R and the intercept value described by ΔS°/R (Fig. 10). The values for ΔH°, ΔS° and ΔG° are presented in Table 3. The positive ΔH° values suggest that the adsorption reaction was an endothermic reaction, and the positive ΔS° values indicate an increase in the randomness of the adsorption process in the experimental system (Table 3). The ΔG° values were obtained using Eq. 11 (Table 3), and the negative ΔG° values indicate the spontaneous nature of Th(IV) adsorption.

Fig. 10
figure 10

The effect of temperature on the adsorption of Th(IV) on the Th(IV)-imprinted polymers; Th(IV) 10 μg mL−1; pH 4.5

Full size image
Table 3 Thermodynamic parameters for the adsorption of Th(IV)-IIP
Full size table

Adsorption selectivity

The competitive adsorption of Th(IV)/U(VI), Th(IV)/La(III), Th(IV)/Ce(III) and Th(IV)/Nd(III) from these combinations of ions (25 mL) was also investigated using a batch process because these ions often coexist in minerals or products, or have a similar ionic radius or the same ionic charge. The distribution ratio (K d) values of Th(IV) in the Th(IV)-Fe3O4@SiO2-IIP adsorbent was greater than that of the other metals (Table 4). The selectivity coefficient (k) for Th(IV)/U(VI), Th(IV)/La(III), Th(IV)/Ce(III) and Th(IV)/Nd(III) was 71.1, 649.2, 595.8 and 96.6, respectively. Also, the k of Fe3O4@SiO2-IIP was 21 times to 93 times greater than that of Fe3O4@SiO2-NIPs; this difference is likely due to the random distribution of ligand functionalities in the polymeric network of Fe3O4@SiO2-NIPs, whereas this random distribution was not observed for Fe3O4@SiO2-IIP. For Fe3O4@SiO2-IIP, the cavities created after removal of the template were complementary to the size, shape and coordination geometries of the imprinted ion. The results described in this section show that the Th(IV)-imprinted adsorbent has a high selectivity for Th(IV) and that it Th(IV) is adsorbed even in the presence of La(III), Ce(III), Nd(III), and U(VI).

Table 4 Competitive adsorption of different ions by Th(IV)-IIP and NIP sorbent (initial concentrations of metal ions: 10 μg mL−1; sorbent 30 mg; temperature 25 °C)
Full size table

Desorption and regeneration

The uptake ratio onto the imprinted adsorbent is low below a pH of 2.5, as described above, which suggests that the adsorbed Th(IV) can be released into an acidic solution. Thus, the desorption of the adsorbed Th(IV) and regeneration of the polymer were tested in a desorption procedure conducted using different concentrations (0.1 to 2.5 mol L−1), volumes (2 to 8 mL) and flow rates (0.5 to 6 mL min−1) of HCl solutions. The results indicated that 1.0 mol L−1 HCl (5 mL) is sufficient for quantitative recovery of Th(IV) using a flow rate of 1.0 mL min−1. The successive adsorption–desorption processes indicated that the uptake capacity of the Fe3O4@SiO2-IIPs for Th(IV) was only reduced by approximately 5 % after nine repetitions under the optimal conditions. These results demonstrate that the Fe3O4@SiO2-IIP can be used recovered and reused.

The comparisons of Th(IV-imprinted material with the results from different references were given in Table 5. The results from the table 5 show that the prepared sorbent gives better Q max and repeated times.

Table 5 Comparisons of Th(IV-imprinted material with the results from different references
Full size table

Conclusions

This research has described a novel magnetic Th(IV)-imprinted polymer based on a new complex of the schiff base N,N’-bis(3-allyl salicylidene)o-phenylenediamine, with Th(IV) as the functional monomer. The template for this novel polymer was prepared successfully using surface imprinting technology. The imprinted polymers demonstrated acceptable characteristics, such as high affinity, selectivity and adsorption capacity, good reusability, and fast process kinetics for Th(IV). The adsorption behavior of Th(IV) was best described by a pseudo-second-order kinetic model, and the adsorption mechanism was a chemical coordination process. A Langmuir isotherm best described the experimental data and indicated a monolayer formation of Th(IV) ions at the surface of the sorbent. In addition, the thermodynamic studies indicated that the adsorption of Th(IV) was an endothermic and spontaneous process.