Synthesis and characterization of amidoxime modified calix[8]arene for adsorption of U(VI) in low concentration uranium solutions

Abstract

A novel calixarene derivative, amidoxime group modified calix[8]arene (C8A-AO), was synthesized from the nitrile group modified calix[8]arene (C8A-CN), and then the nitrile group was changed into an amidoxime group by treating with hydroxylamine under alkaline solution. The products were characterized by nuclear magnetic resonance (NMR), Fourier transform infrared (FT-IR) spectroscopy, field-emitting scanning electron microscopy (FE-SEM), X-ray photoelectron spectroscopy (XPS), elemental analysis and thermogravimetric analysis (TGA). The prepared C8A-AO was applied to adsorb U(VI) in low concentration uranium solutions, less than 1 mg L⁻¹. The effects of pH, contact time, initial concentration, and temperature on the sorption of U(VI) on C8A-AO were investigated. The adsorption efficiency of U(VI) on C8A-AO reached more than 95% in neutral solution. The results showed the adsorbent C8A-AO had an excellent distribution coefficient (K_d) when the initial uranium concentration was less than 1 mg L⁻¹. Additionally, the adsorption results revealed that the kinetics of U(VI) adsorption followed a pseudo-second-order model and the adsorption isotherm fitted well with the Langmuir model. The results of thermodynamic parameters showed that the adsorption process of U(VI) was spontaneous, feasible and endothermic. Furthermore, the excellent selective adsorption capability of U(VI) in simulated seawater suggested that C8A-AO was a potential adsorbent for recovery of U(VI) from seawater.

---

1. Introduction

Nuclear energy, which is a low-carbon energy source, has been widely developed to provide base-load power. Uranium resources in nature mainly exist in two forms: dissolved in seawater and deposited in terrestrial ores.¹ With the rapid development of the nuclear industry, uranium resources on land could lead to its depletion in the near future. In seawater, although the average concentration of uranium is small, approximately 3.3 ppb,² the total amount reaches 4.5 billion tons, about 1000 times that of terrestrial ores.³ Moreover, removal of key radioactive nuclide uranium from wastewater in nuclear industry is necessary for reducing the loss of the uranium and protecting the environment during the mining and treating processes. There is a strong motivation to develop an efficient and economical seawater uranium extraction technology for the nuclear industry.

A number of technologies have been employed for uranium removal from either seawater or nuclear wastewater including biological processes,⁴ liquid–liquid extraction,⁵ chemical reduction,⁶ ion-exchange⁷ and absorption.⁸ Adsorption is a preferred method because of its easy operation and low cost. So adsorbents of rapid kinetics and high capacity have been consistently sought.⁹ So far, extensive efforts have been made towards adsorption uranium from seawater using various adsorbents such as inorganic nano-materials,¹⁰ montmorillonite–biomass complexes,¹¹ metal–organic frameworks (MOFs),¹² layered double hydroxide composites,¹³ synthetic and modified polymers.¹⁴ Among them, either inorganic substances or polymers based materials modified by amidoxime functional group have attracted increasing attentions due to the unique uranium (e.g., UO₂²⁺) adsorption properties.^15,16 In these cases, nitrile groups (–CN) were grafted onto carbon/silica supports to give amidoxime functional groups in the first step. Amidoxime, as an amphoteric functional group, is made up of a nucleophilic NH₂ and hydroxyl functionality. Its lone pairs of electrons in amino nitrogen and oxime oxygen can be donated to UO₂²⁺ center to form a stable five-membered chelate so that adsorption materials modified by amidoxime groups have potential sorption ability for uranium and safety of the environment.¹⁷

Recently, calixarenes have attracted extensive concerns in host–guest chemistry as a versatile class of macrocyclic compounds because of their simple preparations, easy modifications, low toxicity, and unique extraction properties of metal ions. A series of calixarene derivatives with different molecular structural parameters have been designed and synthesized as extractants by changing the size of the cavity and substituting the upper/lower groups into the different functional groups for different target cations.¹⁸ Calixarene derivatives, especially calix[4]arene derivatives, as materials for extracting metal ions are popular areas that have been well developed. Although much more attentions have still been devoted to the excellent extracting properties of calix[4]arene derivatives, some researchers have begun to develop the conformationally labile larger calixarenes. After suitably functionalized, the larger and conformationally flexible structures of calixarene derivatives can bind with larger cations and form strong inclusion complexes easily.¹⁹ On the other hand, in nuclear industry calixarene derivatives have mostly turned out to be a quite versatile class of extractants for uptaking of radionuclides from High Level Liquid Waste (HLLW).^20,21 Nevertheless, as one of the most important and useful compounds, only a few researches have been reported so far regarding the use of calixarene derivatives as uranium extractant.^22,23 For this reason it is necessary to investigate further their extraction ability towards uranium.

In this study, a new amidoxime modified calix[8]arene(C8A-AO) extractant was synthesized as potential uranium extraction from seawater for the first time. The conformationally labile larger calix[8]arene was utilized as the basic substrate. Then the nitrile groups were grafted onto the lower rim via an in situ chemical reaction to obtain the nitrile modified calix[8]arene(C8A-CN). After the conversion of the nitrile into amidoxime, amidoxime modified calix[8]arene(C8A-AO) were obtained. An NMR, FT-IR, XPS and other methods were used to characterize the C8A-AO. In order to check their extraction abilities with the uranyl ions, their adsorption capacities in different concentrations were studied systematically, and the effects of initial pH, contact time, temperature and related parameters were investigated to estimate the capacity in uranium adsorption. Moreover, based on the experimental findings, amidoxime modified calix[8]arene is expected to have potential application for uranium uptake in nuclear industry.

2. Experiment

2.1. Materials and reagents

Bromoacetonitrile (BrCH₂CN) and hydroxylamine hydrochloride (NH₂OH·HCl) were obtained from J&K Scientific Co., Ltd. The alkali used was sodium hydroxide (NaOH), and hydrochloric (HCl) and nitric (HNO₃) acids were used. The alkali, acid and all reagents were purchased from Beijing chemical works, Beijing, China. Stock solution of uranium was supplied by Analytical Laboratory, Beijing Research Institute of Uranium Geology, China. Other metal nitrates were purchased from Aladdin Co. Ltd., Beijing, China. All chemicals purchased from commercial sources were of analytical grade and used without further purification. Deionized water from a milli-Q plus water purification system (Millipore) was used in all experiments.

2.2. Characterization methods

NMR data were collected using a Bruker Biospin GmbH instrument with 700 MHz at 323 K. Fourier transform infrared (FT-IR) spectra were performed by using Perkin Elmer FT-IR spectrophotometer in the range of 400–4000 cm⁻¹ using the KBr pressed pellet technique. Morphological measurements were carried out using FEI Quanta-250 field-emitting scanning electron microscopy (FE-SEM) equipment. X-ray photoelectron spectroscopy (XPS) was recorded on a PHI Quantera system using monochromatic Al Kα (1486.6 eV) X-rays. Elemental analysis for C, H, O, and N in C8A-AO was performed on EuroVector EA3000 analyzer. About thermal analysis of the samples, thermogravimetric analyses (TGA) were measured using a TA SDTQ600 unit in nitrogen at a heating rate of 10 K min⁻¹. The concentration of uranyl ions were analyzed by the trace uranium analyzer (WGJ-III, China). The high resolution inductively coupled plasma mass spectrometry (ICP-MS, Element II) were used for measuring concentrations of the other metal ions.

2.3. Synthesis of C8A-AO

C8A-AO was synthesized by a three-step method. The synthesis routine of C8A-AO is shown in Fig. 1. In the first two steps, C8A and C8A-CN were synthesized respectively according to literatures.^24,25 In the last step, synthesis of C8A-AO was performed as follow: in a round-bottom flask C8A-CN (1.0 g, 0.64 mmol) was dissolved in 60 mL of tetrahydrofuran (THF) with magnetic stirring at ambient temperature. Meanwhile, NH₂OH·HCl (1.72 g, 24.8 mmol) was dispersed in a beaker of about 20 mL deionized water. The aqueous solution was adjusted to pH 8.0 by adding NaOH (1.0 g, 25.0 mmol) in order to obtain about 5 to 6 molar eq. NH₂OH solution. And then the hydroxylamine solution was added to the flask. The flask was fitted with reflux condenser, and the mixture was heated to reflux overnight with continuous stirring. After cooling, the reaction mixture was transferred to a separatory funnel. The THF solvent in the organic layer was removed and the residue was recrystallized by DMF/CH₃OH. Then the products were resolved completely in DMSO keeping the volume of DMSO as little as possible. After adding much water, the white product generated and kept stirring with a constant speed for about 1 h at the same time. The resulting product was filtered off, washed alternatively by plenty of water and ethanol several times. Finally, the obtained white product was dried in a vacuum at 60 °C for 24 h (yield: 0.94 g, 81%). ¹H NMR (700 MHz, DMSO-d₆, TMS): δ = 9.18 (1H, s, N–OH), 6.85 [2H, S, Ar–H], 5.40 [2H, S, Ar–CH₂–Ar], 4.13 (2H, s, O–CH₂), 4.04 (2H, s, NH₂), 1.00 (9H, s, C(CH₃)₃); ¹³C NMR (700 MHz, DMSO-d₆, TMS): δ = 152.30, 149.74, 145.30, 132.51, 125.20, 71.13, 33.54, 30.87, 29.16; IR (KBr) ν/cm⁻¹: 2964 (C–H), 1660 (C=N), 1479 (N–H), 1194 (C–N), 927 (N–O); calcd for C₁₀₄H₁₄₄O₁₆N₁₆: C, 66.64; H, 7.74; N, 11.96. Found: C, 66.49; H, 7.72; N, 11.92; FAB-MS: 1872.1 [M⁺].

Fig. 1 Scheme for the synthesis route of C8A-AO.

2.4. Uranium adsorption experiments

The adsorption was conducted in conical flasks for batch experiments to study the sorption behavior of C8A-AO power towards uranium in aqueous solution. The effects of pH (from 3 to 9), initial uranyl ions concentrations (50 μg L⁻¹ to 50 mg L⁻¹), temperature (288–308 K) and time (1 min to 48 h) were investigated in detail. Uranium solutions were adjusted to a certain degree of concentration using Milli-Q water and the pH of the solutions were adjusted to the desired values by adding negligible amount concentrated HCl or NaOH. In the adsorption experiment, the as-prepared amidoxime-functionalized C8A-AO powder was ultrasonically dispersed into the prepared solutions. The flasks were shaken in an oscillator with a frequency of 200 rpm at a certain temperature to reach the adsorption equilibrium. The suspension liquid was separated by filtration through a 0.22 μm-pore-size filter.

The concentration of the uranyl ions was measured by the trace uranium analyzer. All results of the tests were the average of triplicate determinations and the relative errors were less than 5%. The adsorption results including the percentage removal of uranyl ions (% R), the amount of uranyl ions equilibrium adsorption on the solid phase (Q_e) and the distribution coefficient (K_d) were calculated on the basis of the following equations.²⁶

where C₀ and C_e represent the initial and equilibrium concentrations of uranyl ions, respectively (mg L⁻¹). V is the volume of the solution used for adsorption (mL) and m stands for the weight of adsorbent (g).

3. Results and discussion

3.1. Characterization

3.1.1 FT-IR. To identify the changes of functional groups during the synthesis process and show the banding configurations of the main elements, the compounds C8A, C8A-CN and C8A-AO were characterized by FT-IR. And the spectra are shown in Fig. 2(a)–(c). The characteristic absorption peak at 3242 cm⁻¹ for the C8A in Fig. 2a can be assigned to the stretching vibrations of O–H of the hydroxyl group. And the peaks at 2950–2800 cm⁻¹ are due to the C–H bands within the CH₃ and aromatic ring. After modifying the lower rim of C8A using the nitrile group, the spectrum of C8A-CN in Fig. 2b shows many obvious differences with that of C8A. The characteristic absorption band of the hydroxyl group at 3242 cm⁻¹ disappeared completely. And a new peak associated with –CN stretching vibrations was observed at 2248 cm⁻¹. Fig. 2c shows the characteristic absorption peaks of C8A-AO. Compared with the spectrum of C8A-CN in Fig. 2b, after amidoxime reaction the characteristic absorption band of the nitrile group at 2248 cm⁻¹ has disappeared as shown in Fig. 2c. However, new peaks at 3481, 3376, 1668 and 928 cm⁻¹ are assigned to the N–H, O–H, C=N and N–O stretching vibrations of the amidoxime group, respectively.²⁷ These new peaks indicate that the amidoxime group modified calix[8]arene(C8A-AO) was formed effectively via hydroxylamine treatment.

Fig. 2 FT-IR spectra of (a) C8A, (b) C8A-CN and (c) C8A-AO.

3.1.2 XPS. XPS was applied to confirm that the amidoxime group was attached to the lower rim of calix[8]arene after NH₂OH treatment. And detailed information about the surface chemical composition and bonding environment of the as-prepared C8A-AO were ascertained by the survey spectrum and high resolution spectrum of XPS. As is shown in the XPS spectrum of C8A-AO (Fig. 3a), it can be observed that three characteristic peaks have bonding energies of 283 eV, 398 eV and 531 eV, which are attributed to C 1s, N 1s, and O 1s, respectively.²⁸

Fig. 3 The typical XPS survey spectrum (a), high resolution XPS spectra of C 1s spectra (b), N 1s spectra (c) and O 1s spectra (d) of C8A-AO.

Fig. 3b shows a detailed spectrum of the C 1s signal for C8A-AO and the peak can be deconvoluted into five parts.²⁷ The main C 1s peak is dominated by the two peaks which are elemental carbon at 284.2 eV and 284.7 eV corresponding to sp³ hybridized carbon atoms (C–C or C–H) and sp² hybridized carbon atoms (C=C), individually.²⁹ The two peaks at 285.3 and 286.2 eV are assigned to the nitrogen-bound species C=N and C–N, respectively.^28,30 Another peak at 286.8 eV was also observed, which is attributed to the oxygen-bound species (C–O).²⁷ In N 1s spectra (Fig. 3c), a strong peak of 399.5 eV is assigned to C(NH₂)=N–OH,³¹ which provided strong evidence that the amidoxime group has successfully been anchored onto the calix[8]arene. The O 1s spectra shown in Fig. 3d can be fairly deconvoluted into two major components centered at 532.2 and 533.0 eV, which are attributed to oxygen atoms in hydroxyl groups and C–O, respectively.³⁰ The above results clearly confirmed that C8A-AO had been successfully prepared.

3.1.3 SEM. SEM images of the as-prepared sample C8A-AO from different preparation steps are given in Fig. 4. The images in Fig. 4a and b clearly show that the sample, after recrystallization from DMF/CH₃OH solution, exhibits large-sized lamellar structure. And the thickness of the layer was about 100 to 300 nm (Fig. 4b). Fig. 4c and d show the morphology of the sample C8A-AO, which were treated and obtained from DMSO/H₂O. From Fig. 4c, the image clearly shows that the form of the sample has changed into flocks, which appear to be composed of small particles with different degrees of adhesion. As shown in Fig. 4d (corresponding to the marked region of Fig. 4c), it is observed that the clarity of the C8A-AO particles are more aggregated with each other. And the size of the particles is in the range of about 50 to 100 nm.

Fig. 4 SEM images of C8A-AO.

3.1.4 Thermal analysis. Thermal gravimetric analysis (TGA) characterization was carried out to investigate the thermal stability of the as-prepared composites. The TGA curves for C8A, C8A-CN and C8A-AO are shown in Fig. 5. In the figure, the TGA curve of C8A in the range of room temperature to 380 °C shows the weight loss only about 6.4 wt%, which could be attributed to physically absorbed water. However, it decreases rapidly from 380 to 550 °C, owing to pyrolysis of the calix[8]arene framework and dissociating the covalently bound. Compared with C8A, the TGA curve of C8A-CN exhibit an obvious weight loss process at 300–340 °C, this may be attributed to the loss of the nitrile groups at the lower rim.

Fig. 5 The thermal analysis of C8A, C8A-CN and C8A-AO.

The TGA curve of C8A-AO, which is different from C8A and C8A-CN, has three main processes. In the first stage, the curve decreases slightly from room temperature to about 155 °C, over which the weight loss is about 5.1 wt%. This could be mainly related to physically absorbed water possibly with residual solvent.³² In the second stage, the curve decreases slightly at 160–340 °C and the weight loss (about 19.4 wt%) could be assigned to the amidoxime groups, which is close to the theoretical value of the amidoxime group ratio in the C8A-AO sample. In the process of decomposition of main functional group, the curve of C8A-AO is more slightly than C8A-CN because the amidoxime group (–C(NH₂)=N–OH) is composed of oxime group and amine group, and it is more complex than the nitrile group (–CN) in chemical structure. The last stage is above 340 °C, at which the weight loss is about 43.0 wt% and then the curve keeps stable. Interestingly, all the curves of samples almost kept stable after the same temperature at about 550 °C, which could be attributed to finish the process of pyrolysis of the calix[8]arene framework and dissociate the covalently bound. The TGA results confirmed the amount of main groups in the sample and indicated that the prepared materials have good thermal stability.

3.2. Adsorption properties

3.2.1 Effect of pH on U(VI) adsorption by C8A-AO. The pH value had significant influence on the adsorption property of amidoxime functional group modified adsorbent for recovery of U(VI) from aqueous solution. Fig. 6 exhibits the effect of the uranium adsorption by C8A-AO at different pH values ranging from 3 to 9, with an initial uranyl ions concentration of about 0.5 mg L⁻¹ at 298 ± 0.5 K, an adsorbent dosage of 10 mg/50 mL and a contact time of 48 h. From Fig. 6, it can be seen clearly that the removal efficiency of U(VI) from solution increase gradually from about 25% to 96% at the pH range 3 to 8, and then the adsorption efficiency decreased approximately 5% in pH range 8 to 9. At low pH, the lower adsorption efficiency can be attributed to the strong acidic environment. The availability of more H⁺ had stronger competition than the uranyl ions for banding active sites on the adsorbent.³³ At the same time the amidoxime functional groups were protonated resulting in more positive charges on the adsorbent surface, which was unfavorable to enrich the uranyl ions.³⁴ While with the initial pH values increased from 3 to 7, the amount of H⁺ decreased gradually so that there was no other cations could compete with the uranyl ions in the adsorption process. Meanwhile, the amidoxime group became deprotonated with the pH value increased. Hence, the uranyl ions could be attached on the surface easily due to the formation of metal complexes through chelating or metal exchanges.³⁵ When the pH value went on increase(from 8 to 9), the adsorbed amount of uranium decreased due to the hydrolysis of U(VI) resulting in the formation of noncomplexible species UO₂(OH)₃⁻ and UO₃(OH)₇⁻, which increased the repulsion with the anions and negative charges on the surface of the adsorbent.³⁶ Therefore, to eliminate the effect of the hydrolysis in the high pH value, the subsequent adsorption experiments were carried out at pH 7.

Fig. 6 Effect of pH on U(VI) adsorption by C8A-AO (C₀ = 0.5 mg L⁻¹, t = 48 h, T = 298 ± 0.5 K, m/V = 10 mg/50 mL).

3.2.2 Effect of contact time on U(VI) adsorption by C8A-AO. The adsorption kinetics, as a significant part of the adsorption study, is related to the contact time and adsorption amount. To study the adsorption kinetics, the amounts of uranyl ions adsorbed by C8A-AO were measured in contact time from 1 min to 48 h. As presented in Fig. 7, the adsorption amounts increased rapidly from 1 min to 6 h resulting from the higher concentration uranyl ions solution at primary adsorption process. It was worth noting that, about 70% of the uranium was adsorbed in the initial 30 min from right axis of Fig. 7. While after 6 h, the adsorption amounts increased slowly contributed to the increase of diffusive resistance and the lack of adsorption sites on the internal and external surfaces of the material in the process of chelating uranyl ions with the amidoxime functional groups. After 24 h, the adsorption curves became very gentle, meaning that the adsorption process had reached equilibrium state. And more than 90% of uranium in aqueous solution had been removed [Fig. 7, right axis, where C_t (mg L⁻¹) represents the concentrations of uranyl ions at contact time (t)]. Based on the adsorption kinetics data, 24 h was selected to be the contact time in the adsorption process to ensure reaching the equilibrium on U(VI) adsorption for the following experiments.

Fig. 7 Adsorption kinetics of uranium on C8A-AO (C₀ = 0.549 mg L⁻¹, T = 298 ± 0.5 K, m/V = 10 mg/50 mL).

The experiment data analyzed by appropriate kinetic models could offer useful information to clarify the adsorption mechanism of metal ions on adsorbent. The adsorption kinetics of U(VI) adsorbed on C8A-AO was simulated by the pseudo-first-order, pseudo-second-order, and intraparticle diffusion models and all of them can be respectively represented as follows.^37–39

Pseudo-first-order model

Pseudo-second-order model

Intraparticle diffusion model

Q_t = k_intt^1/2 + θ (6)

where k₁ (h⁻¹), k₂ [g (mg h)⁻¹] and k_int [mg (g h^1/2)⁻¹] represent the pseudo-first-order rate constant, the pseudo-second-order rate constant and a rate constant of intraparticle diffusion, respectively; Q_e (mg g⁻¹) and Q_t (mg g⁻¹) are the sorption amounts of U(VI) at equilibrium time (h) and at contact time t (h), respectively; θ (mg g⁻¹) is a constant of the intraparticle diffusion model. In addition, the linear regression coefficients (R²) of these lines in plots indicate the applicability of different models for fitting the adsorption process.⁴⁰ From the linear plots in Fig. 8, these kinetic parameters of the pseudo-first-order (Fig. 8a) and pseudo-second-order (Fig. 8b) models can be obtained easily. And the calculated parameters of these three kinds of kinetic models by simulating the data for the adsorption uranyl ions were obtained and shown in Table 1. Obviously, the correlation coefficient (R²) of the pseudo-second-order model is higher than that of the pseudo-first-order model, indicating that it is better to describe the adsorption process by utilizing the pseudo-second-order model. Moreover, the calculated Q_e value of the pseudo-second-order model is 2.560 mg g⁻¹, which is similar to the experiment value, Q_e. Besides, the pseudo-second-order model is based on the assumption that the chemisorption between the metal ions and adsorbent is the rate-limiting step of the adsorption.⁴¹ Therefore, it can be inferred the chemical interaction between uranyl ions and C8A-AO is the mainly adsorption manner in the adsorption process.

Fig. 8 Fitting results of the pseudo-first-order and pseudo-second-order models.

Table 1 Adsorption kinetics fitting results for U(VI) on C8A-AO by pseudo-first-order, pseudo-second-order, and intraparticle diffusion models

Pseudo-first-order model
Qe (mg g^−1)	k1 (h^−1)	R^2
0.441	0.3244	0.8984
Pseudo-second-order model
Qe (mg g^−1)	k2 [g (mg h)^−1]	R^2
2.560	1.0572	0.9997
Intraparticle diffusion model
θ (mg g^−1)	kint [mg (g h^1/2)^−1]	R^2
2.122	0.0856	0.9995

---

The experimental kinetic data were also analyzed by the intraparticle diffusion model. About this model, if the plot is a straight line, which passes through the origin, it indicates the adsorption process is only controlled by intraparticle diffusion. While, when the result exhibit multilinear plots by simulating, it implies that this process is influenced by two or more different steps⁴² As can be seen from Fig. 9, the adsorption process is consisted of three different sloped portions: the first steep-sloped line from 0 to 6 h denotes external mass transfer that includes the uranyl ions remove from the aqueous solution to the solid phase; then the second gentle-sloped line from 6 to 24 h is attributed to the intraparticle diffusion and the kinetic process of uranyl ions reacting with the functional groups on the adsorbent; and the third fitting line represents the equilibrium adsorption process. Based on the results shown above, it can be concluded that the plot is not the line passing through the origin. It suggested that the intraparticle diffusion is not the only rate-limiting step.

Fig. 9 Plots of U(VI) adsorption by C8A-AO according to the intraparticle diffusion model.

3.2.3 Effect of initial uranium concentration on U(VI) adsorption by C8A-AO. To evaluate the adsorption performance of C8A-AO in aqueous solutions with different initial uranium concentrations, batch adsorption studies were performed for initial uranium concentration ranging from about 50 μg L⁻¹ to 3 mg L⁻¹ at pH 7, for the C8A-AO dosage of 0.01 g/50 mL, a temperature of 298 ± 0.5 K, and a contact time of 24 h. As indicated in Table 2, the amount of uranium adsorbed on C8A-AO increased with increasing initial uranium concentration. This is attributed to the higher uranium concentration, the more chance for the adsorbent contracted with the uranyl ions, and thus the larger adsorption amount. And the U removal gets to high values of above 92% in the range of initial uranium concentrations at about 50–1000 μg L⁻¹ separately. Although the adsorption amounts are larger in the initial uranium concentrations more than 3 mg L⁻¹, their U removal can not reach 50%. The distribution coefficient (K_d) as one of the main parameters to evaluate the adsorption performance of adsorbent is shown in Table 1. Generally, an adsorbent with a larger K_d value (more than 10⁴ mL g⁻¹) is considered to be an excellent adsorbent material.^13,43 In the case of C8A-AO, the K_d values can reach high values of 5.85 × 10⁴ to 2.50 × 10⁵ at uranium concentration range of about 50–1000 μg L⁻¹. Both the U removal values and the distribution coefficient values clearly indicated that the C8A-AO adsorbent has an excellent adsorption performance at the low initial uranium concentration of aqueous solution.

Table 2 Effect of initial uranium concentration on U(VI) adsorption by C8A-AO^a

C0 (ppm)	Ce (ppm)	Qe (mg g^−1)	Removal (%)	Kd (mL g^−1)
a m: 0.01 g, V: 50 mL, V/m = 5000 mL g^−1, T = 298 ± 0.5 K, contact time: ∼24 h. C0, initial uranium concentration. Ce, final uranium concentration after 24 h.
0.051	0.001	0.250	98.04	2.50 × 10^5
0.115	0.004	0.555	96.52	1.39 × 10^5
0.417	0.019	1.990	95.44	1.05 × 10^5
0.726	0.043	3.415	94.08	7.94 × 10^4
1.029	0.081	4.740	92.13	5.85 × 10^4
3.079	1.776	6.515	42.32	3.67 × 10^3
5.611	3.330	11.41	40.67	3.43 × 10^3
25.40	17.10	41.50	32.68	2.43 × 10^3
38.80	28.50	51.50	26.55	1.81 × 10^3
53.60	41.30	61.50	22.95	1.49 × 10^3

---

To evaluate the adsorption capacity of C8A-AO adsorbent for uranyl ions, the adsorption data at higher initial uranium concentrations were treated with Langmuir and Freundlich isotherm models, respectively. The Langmuir isotherm model is always used on the base of an assumption that adsorption occurs on a homogenous surface by monolayer sorption with a finite number of homogeneous sites.⁴⁴ While the Freundlich isotherm model assumes that uptake of metal ions occurs on heterogeneous surfaces and active sites.⁴⁵ Both Langmuir equation (eqn (7)) and Freundlich equation (eqn (8)) can be described separately with the form as:

where C_e is the final equilibrium uranium concentration (mg L⁻¹), Q_e is the equilibrium adsorption capacity (mg g⁻¹), Q_m is the maximum adsorption capacity (mg g⁻¹), b and K_F are the Langmuir constant (L mg⁻¹) and Freundlich constant (mg g⁻¹)(L mg⁻¹)^1/n, respectively. And n is an empirical parameter related to the adsorption intensity.

The resulted data, analyzed by the Langmuir and Freundlich isotherm models respectively, are shown in Fig. 10a and b. And from the parameters and correlation coefficients (R²) of these models shown in Table 3, it could be found that the correlation coefficient of the Langmuir model (R² = 0.9959) is higher than the Freundlich model (R² = 0.9877), fitting better with Langmuir model, which indicates that the surfaces of the adsorbents are homogenous and monolayer adsorption in this process. Besides, the Q_m calculated by Langmuir model is 98.425 mg g⁻¹, which is used to compare to other similar uranium-selective sorbents reported so far. And all of them were summarized in Table 4. According to this table, the C8A-AO adsorbent has certain adsorption property with uranium. It also shows a higher adsorption capacity for uranyl ions than other similar adsorbents.

Fig. 10 Fitting results for the Langmuir (a) and Freundlich (b) isotherm models.

Table 3 Langmuir and Freundlich model fitting parameters for uranium adsorption on C8A-AO

Temperature (K)	Langmuir parameters			Freundlich parameters
	Qm (mg g^−1)	B (L mg^−1)	R^2	KF (mg^1−1/n L^1/n g^−1)	n	R^2
298	98.425	0.0401	0.9959	4.6030	1.3814	0.9877

---

Table 4 Comparison of sorption capacity of U(VI) on various uranium sorbents

Sorbents	Capacity (mg g^−1)	Ref.
Amidoximation of polyacrylonitrile (PAN) fibers	2.3	1
Amidoximated macroporous beads	3.5	46
AO-functionalized mesoporous carbon	4.6	15
AO-mesoporous imprinted polymer	19.1	47
Amidoxime modified bentonite	33.3	48
Amidoximated copolymer(PAMSA)	39.5	49
AO-poly(acrylonitrilec-methacrylic acid)	51.5	50
Fe3O4@SiO2-amidoxime	105.0	36
Mesoporous carbon–O–PO(OH)2	112.1	51
Amidoxime modified calix[8]arene(C8A-AO)	98.4	This work

---

3.2.4 Effect of temperature on U(VI) adsorption by C8A-AO. To investigate the influence of temperature on U(VI) adsorption by C8A-AO, the adsorption experiments at different temperatures of 293, 298, 303 and 308 K were performed at pH = 7, with an initial uranyl ions concentration of about 0.5 mg L⁻¹, an adsorbent dosage of 10 mg/50 mL and a contact time of 24 h. The adsorption results, at different temperatures, are listed in Table 5. As indicated in the table, the adsorbent kept high values of the removal ratio and distribution coefficient at different temperatures in the adsorption process. Uranium adsorption on C8A-AO was increased with increasing temperature at equilibrium time. And the amount of U(VI) adsorbed by C8A-AO increased from 2.505 to 2.640 mg g⁻¹. The increment of adsorption capacity can be attributed to the increased diffusion rate of the metal ions and decreased viscosity of the solution with the temperature increased.⁴⁰

Table 5 Effect of temperature on U(VI) adsorption by C8A-AO^a

T (K)	C0 (ppm)	Ce (ppm)	Qe (mg g^−1)	Removal (%)	Kd (mL g^−1)
a m: 0.01 g, V: 50 mL, V/m = 5000 mL g^−1, contact time: ∼24 h. C0, initial uranium concentration. Ce, final uranium concentration after 24 h.
293	0.549	0.048	2.505	91.26	5.22 × 10^4
298	0.549	0.039	2.550	92.90	6.54 × 10^4
303	0.549	0.028	2.605	94.90	9.30 × 10^4
308	0.549	0.021	2.640	96.17	1.26 × 10^5

---

The thermodynamic parameters, involving the standard Gibbs free energy change (ΔG°), standard enthalpy change (ΔH°) and standard entropy change (ΔS°), can be calculated from the temperature dependence of adsorption results. The Gibbs free energy change (ΔG°) was calculated from the following equation (eqn (9)) and according to Fig. 11 and the following equation (eqn (10)), standard enthalpy change (ΔH°) and standard entropy change (ΔS°) were calculated from the slope and intercept, of a plot of ln K_d versus 1/T, respectively.⁵²

ΔG° = −RT ln K_d (9)

where K_d is the distribution coefficient and R is the ideal gas constant [8.314 J (mol K)⁻¹].

Fig. 11 Linear plot of ln K_d versus 1000/T for of U(VI) adsorption by C8A-AO.

The obtained values of the thermodynamic parameters (ΔG°, ΔH° and ΔS°) were listed in Table 6. All the Gibbs free energy changes at 293, 298, 303 and 308 K are all negative, indicating adsorption of U(VI) on C8A-AO at different temperatures are all spontaneous processes. And the values of ΔG° become even more negative with the increasing of the temperature, implying that higher temperature can help to promote the adsorption process efficiently. The positive value of standard enthalpy change (ΔH° = 45.27 kJ mol⁻¹) indicates that the adsorption of U(VI) on C8A-AO is an endothermic process in nature, which agrees with the fact that the adsorption process is favorable at higher temperature. In addition, the positive value of standard entropy change (ΔS° = 187.09 J K⁻¹ mol⁻¹) revealed the increased randomness at the adsorbent–solution interface during the process of uptaking U(VI) on C8A-AO.

Table 6 Thermodynamic parameters for the adsorption of U(VI) by C8A-AO

T (K)	ΔG° (kJ mol^−1)	ΔH° (kJ mol^−1)	ΔS° (J K^−1 mol^−1)
293	−9.62	45.27	187.09
298	−10.36
303	−11.41
308	−12.39

---

3.2.5 Selective adsorption of C8A-AO towards uranium. The selective adsorption of U(VI) on the as-prepared adsorbent C8A-AO was investigated via batch adsorption experiment in simulated seawater containing Na, Mg, Ca, Sr, etc.⁵³ The adsorption experiment were performed in the simulated seawater at pH = 8, with an adsorbent dosage of 10 mg/50 mL and a contact time of 24 h at a temperature of 298 ± 0.5 K. The initial concentrations and equilibrium concentrations of these main metal ions in seawater are shown in Table 7. The distribution coefficient (K_d) and the selective adsorption coefficient (K_U/M) were calculated from the eqn (2) and the eqn (11), respectively.where K_d(U) and K_d(M) represent the distribution coefficient of uranium and the distribution coefficient of other metal ions, respectively.

Table 7 Selective adsorption of metal ions in simulated seawater by C8A-AO

	U	Na	Mg	Ca	Sr	Rb	Fe	Zn
C0 (ppm)	0.471	2282	385.5	40.77	24.03	3.470	0.319	0.847
Ce (ppm)	0.202	2012	382.8	39.63	23.03	3.090	0.280	0.803
Kd (mL g^−1)	6658	671.0	35.53	143.8	217.1	614.9	696.4	274.0
KU/M	1	9.920	187.4	46.30	30.67	10.83	9.560	24.30

---

As shown in Table 7, the initial concentration of U(VI) is lower than that of many other competitive metal ions. Besides, the distribution coefficient (K_d) value of U(VI) expressing the degree of sorption by C8A-AO reached up to 6658 mL g⁻¹ in the adsorption process from simulated seawater, which is lower than the K_d value of U(VI) adsorbed in single uranium aqueous solution(more than 1 × 10⁴ mL g⁻¹). However, the K_d values of all the other main metal ions in simulated seawater are fairly low. So the K_U/M values of the other metal ions are higher than that of U(VI). These findings provided further support for the desirable selectivity of C8A-AO to uptake U(VI) from simulated seawater. This indicates it has high potential application in recovery of uranium from seawater with a very low concentration.

3.2.6 Adsorption mechanism of U(VI) by C8A-AO. The whole working principle of U(VI) adsorption on C8A-AO is summarized as follow: first, the larger calixarene has a greater conformational flexibility than calix[4]arene and its derivatives, which is attributed to its larger ring structure.^19,54 It was found that the effect of disperse of C8A-AO precipitated again from DMSO/water solution was improved during the adsorption process in the aqueous solution. This facilitated adsorption could be due to the rearrangement of the conformation that the hydrophobic alkyl groups turn to the inside of the molecular structure and the hydrophilic amidoxime groups turn to the outside. Second, calixarene, as one of the macrocyclic compounds with molecular recognition function, can be modified easily to obtain the desired structure for uptaking uranium. Their capacity as the cation receptors is highly dependent on the functional groups they carry and the cavity size of calixarene.⁵⁵ The calix[8]arene has bigger cavity, which can be more beneficial to deal with the metal ions with smaller diameter (Na, Mg, Ca, etc.) that can pass through it easily. So the influence of the smaller metal ions for enriching the uranyl ions in the adsorption process can be reduced. Third, after being modified by amidoxime groups, the calixarene has plenty of new functional groups. It is reasonable that the amidoxime groups provide so many active adsorption sites that they could chelate with U(VI) on the surface of the C8A-AO and help to improve the adsorption capacity of U(VI). In addition, there are abundant ether and cavities, which also could help to form complexes or inclusion of uranyl ions. The amount of U(VI) on the adsorbent is attributed to the chemical interaction that the amidoxime groups reacted with U(VI) to form complexes.

4. Conclusions

A novel method to synthesize amidoxime group modified calix[8]arene (C8A-AO) was reported for the first time by a three-steps processes. The material was found to have adsorption ability for U(VI) due to the strong chelation of amidoxime group to U(VI). The removal efficiency of U(VI) from solution reached maximum when the pH of the uranium solution was neutral. The adsorption efficiency of C8A-AO towards U(VI) from the aqueous solution with low concentration of less than 1.0 mg L⁻¹ elevated to more than 95% and had desirable values of distribution coefficient (K_d). And the results revealed the kinetics of U(VI) adsorption process were in good agreement with pseudo-second-order model, indicating chemical interaction was the rate-control step in the adsorption process. The adsorption isotherm was well fitted to the Langmuir model, implying the monolayer adsorption was dominated. According to the values of thermodynamic parameters, it could be found that the adsorption process was spontaneous, feasible and endothermic. The adsorption experiment in simulate seawater demonstrates that C8A-AO has excellent selective adsorption capacity. These adsorption results indicate that C8A-AO has a great potential for uptaking uranium from seawater.

Acknowledgements

This work was supported by the International Science & Technology Cooperation Program of China (2014DFR61080).

References

1. H. Zhao, X. Liu, Y. Ming, Z. Wang, B. Zhang, H. Ma, W. Min and J. Li, Ind. Eng. Chem. Res., 2015, 54, 3101–3106.
2. F. R. Alain, G. Lebuzit, J. Foos, A. Guy, M. Draye and M. Lemaire, Ind. Eng. Chem. Res., 2003, 42, 5900–5904.
3. H. J. Schenk, L. Astheimer, E. G. Witte and K. Schwochau, Sep. Sci. Technol., 1982, 17, 1293–1308.
4. M. Liu, F. Dong, X. Yan, W. Zeng, L. Hou and X. Pang, Bioresour. Technol., 2010, 101, 8573–8580.
5. K. Chandrasekaran, D. Karunasagar and J. Arunachalam, Anal. Methods, 2011, 3, 2140–2147.
6. B. Hua and B. L. Deng, Environ. Sci. Technol., 2008, 42, 8703–8708.
7. S. Akyil, M. A. A. Aslani and M. Eral, J. Radioanal. Nucl. Chem., 2003, 256, 45–51.
8. C. Gunathilake, J. Górka, S. Dai and M. Jaroniec, J. Mater. Chem. A, 2015, 3, 11650–11659.
9. L. Wen, Z. Xiao, T. Wang, D. Zhao and J. Ni, Chem. Eng. J., 2015, 286, 427–435.
10. L. Qian, M. Ma and D. Cheng, J. Radioanal. Nucl. Chem., 2015, 303, 161–170.
11. M. S. Olivelli, G. A. Curutchet and R. M. T. Sánchez, Ind. Eng. Chem. Res., 2013, 52, 2273–2279.
12. Z. Q. Bai, L. Y. Yuan, L. Zhu, Z. R. Liu, S. Q. Chu, L. R. Zheng, J. Zhang, Z. F. Chai and W. Q. Shi, J. Mater. Chem. A, 2015, 3, 525–534.
13. S. Ma, L. Huang, L. Ma, Y. Shim, S. M. Islam, P. Wang, L. D. Zhao, S. Wang, G. Sun and X. Yang, J. Am. Chem. Soc., 2015, 137, 3670–3677.
14. M. Monier and N. H. Elsayed, J. Colloid Interface Sci., 2014, 423, 113–122.
15. J. Górka, R. T. Mayes, L. Baggetto, G. M. Veithand and S. Dai, J. Mater. Chem. A, 2013, 1, 3016–3026.
16. D. Yuan, C. Long, X. Xin, L. Yuan, S. Liao and W. Yun, Chem. Eng. J., 2016, 285, 358–367.
17. Y. Zhao, J. Li, S. Zhang and X. Wang, RSC Adv., 2014, 4, 32710–32717.
18. M. Kyrš, K. Svoboda, P. Lhoták and J. Alexová, J. Radioanal. Nucl. Chem., 2003, 258, 497–509.
19. B. B. Adhikari, X. Zhao, S. Derakhshan and M. P. Schramm, Dalton Trans., 2014, 43, 15560–15565.
20. H. Huang, S. Ding, L. Ning, Y. Wu, D. Su and H. Song, Sep. Purif. Technol., 2014, 123, 235–240.
21. V. Kumar, J. Sharma, P. Achuthan and R. Hubli, J. Radioanal. Nucl. Chem., 2014, 299, 1547–1553.
22. W. S. Bliwa, Croat. Chem. Acta, 2002, 75, 131–153.
23. S. Shinkai, H. Kawaguchi and O. Manabe, J. Polym. Sci., Part C: Polym. Lett., 1988, 26, 391–396.
24. C. D. Gutsche, B. Dhawan, K. H. No and R. Muthukrishnan, J. Am. Chem. Soc., 1981, 103, 3782–3792.
25. S. Erdemir, O. Sahin, A. Uyanik and M. Yilmaz, J. Inclusion Phenom., 2009, 64, 273–282.
26. C. He, D. Shao, J. Li and X. Wang, Chem. Eng. J., 2014, 254, 623–634.
27. X. Yang, J. Li, J. Liu, Y. Tian, B. Li, K. Cao, S. Liu, M. Hou, S. Li and L. Ma, J. Mater. Chem. A, 2013, 2, 1550–1559.
28. S. Liu, J. Tian, L. Wang, Y. Zhang, X. Qin, Y. Luo, A. M. Asiri, A. O. Al-Youbi and X. Sun, Adv. Mater., 2012, 24, 2037–2041.
29. G. Wu, L. Li, J. H. Li and B. Q. Xu, Carbon, 2005, 43, 2579–2587.
30. J. H. Zhou, Z. J. Sui, J. Zhu, P. Li, D. Chen, Y. C. Dai and W. K. Yuan, Carbon, 2007, 45, 785–796.
31. X. Jin, Y. Cui, Y. Li, Y. Qi, L. Yang, G. Zhao and H. Hu, J. Hazard. Mater., 2011, 186, 1672–1680.
32. P. J. Lebed, J. D. Savoie, J. Florek, F. Bilodeau, D. Larivière and F. Kleitz, Chem. Mater., 2012, 24, 4166–4176.
33. S. Chen, J. Hong, H. Yang and J. Yang, J. Environ. Radioact., 2013, 126, 253–258.
34. V. C. Taty-Costodes, H. Fauduet, C. Porte and A. Delacroix, J. Hazard. Mater., 2003, 105, 121–142.
35. A. Masoumi, K. Hemmati and M. Ghaemy, RSC Adv., 2015, 5, 1735–1744.
36. Y. Zhao, J. Li, L. Zhao, S. Zhang, Y. Huang, X. Wu and X. Wang, Chem. Eng. J., 2013, 235, 275–283.
37. J. Hu, C. Chen, X. Zhu and X. Wang, J. Hazard. Mater., 2009, 162, 1542–1550.
38. Y. S. Ho and G. Mckay, Process Biochem., 1999, 34, 451–465.
39. G. Annadurai, R. S. Juang and D. J. Lee, J. Hazard. Mater., 2002, 92, 263–274.
40. Z. Li, D. Xiao, Y. Ge and S. Koehler, ACS Appl. Mater. Interfaces, 2015, 7, 15000–15009.
41. Y. Ge, D. Xiao, Z. Li and X. Cui, J. Mater. Chem., 2014, 2, 2136–2145.
42. İ. A. Şengil and M. Özacar, J. Hazard. Mater., 2008, 157, 277–285.
43. J. Lehto and A. Clearfield, J. Radioanal. Nucl. Chem., 1987, 118, 1–13.
44. I. Langmuir, J. Am. Chem. Soc., 1918, 143, 1361–1403.
45. H. M. F. Freundlich, J. Phys. Chem., 1906, 57, 385–470.
46. K. Singh, C. Shah, C. Dwivedi, M. Kumar and P. N. Bajaj, J. Appl. Polym. Sci., 2013, 127, 410–419.
47. D. James, G. Venkateswaran and T. P. Rao, Microporous Mesoporous Mater., 2009, 119, 165–170.
48. S. Şimşek and U. Ulusoy, J. Radioanal. Nucl. Chem., 2012, 292, 41–51.
49. O. Hazer and Ş. Kartal, Talanta, 2010, 82, 1974–1979.
50. N. Sahiner, H. Yu, G. Tan, J. He, V. T. John and D. A. Blake, ACS Appl. Mater. Interfaces, 2012, 4, 163–170.
51. M. Carboni, C. W. Abney, K. M. L. Taylorpashow, J. L. Viveroescoto and W. Lin, Ind. Eng. Chem. Res., 2013, 52, 15187–15197.
52. Y. Liu, J. Chem. Eng. Data, 2009, 54, 1981–1985.
53. T. Saito, S. Brown, S. Chatterjee, J. Kim, C. Tsouris, R. T. Mayes, L. J. Kuo, G. Gill, Y. Oyola and C. J. Janke, J. Mater. Chem. A, 2014, 2, 14674–14681.
54. N. Basílio, L. Garciario and M. Martínpastor, Langmuir, 2012, 28, 2404–2414.
55. B. B. Adhikari, M. Gurung, A. B. Chetry, H. Kawakita and K. Ohto, RSC Adv., 2013, 3, 25950–25959.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra23764a

---