Micro-nano structure nanofibrous p-sulfonatocalix[8]arene complex membranes for highly efficient and selective adsorption of lanthanum(III) ions in aqueous solution

Abstract

In this study, micro-nano structured p-sulfonatocalix[8]arene (calix8) complex membranes prepared by electrostatic adsorbing anionic calix8 onto the cationic nanofibrous mats with micro-nano structure were utilized as an affinity membrane for the selective adsorption of lanthanum(III) ions, where the cationic nanofibrous mats were fabricated by wet-electrospinning technique from polyacrylonitrile (PAN) solution with the aid of pore-forming agent poly(vinyl pyrrolidone) (PVP) and followed by the amination with diethylene triamine (DETA). The as-prepared nanofibrous calix8 complex membranes were subject to selective adsorption of La(III) ions in aqueous solution and showed very high adsorption capacity and selectivity for La³⁺ from other metal ions such as Fe³⁺, Al³⁺, Cu²⁺, Ca²⁺, Mg²⁺ and K⁺. The resultant membranes adsorbed with La(III) ions could be desorbed and regenerated successfully without significantly affecting their adsorption capacity. The adsorption data at equilibrium were well fitted to Langmuir isotherm equation with a maximum adsorption capacity of 155.1 mg g⁻¹ for La(III) ions. Furthermore, the possible adsorption mechanism of La(III) ions onto the calix8 membrane was discussed based on the FTIR and XPS data. This study demonstrated a facile route for highly efficient and selective separation of lanthanide ions from aqueous solutions.

---

1 Introduction

Over the past decades, due to the unique properties in optics, electricity, and magnetism, the rare earth elements have received increasing demands in many hi-tech industries such as nuclear, optical materials, magnetic materials, laser materials, superconducting materials, batteries, etc.^1,2 However, the effluents containing rare earth ions discharged from industrial factories such as mining, metallurgy, manufacture, and so on, will contaminate the environment and bring great harm to human's health. Moreover, the limited resource availability decreased dramatically owing to the uncontrolled mining since 1980's in China. Therefore, the importance of separation and recovery of rare earth elements from the discharged effluents and out-of-date materials becomes increasingly conspicuous.³ Recently, there are many reports on the adsorption materials used in the recovery and separation of rare earth elements.^4–6 Wu et al. investigated the competitive adsorption between La(III) and alkali ions, alkali earth ions and some transition metal ions using 2-ethylhexyl phosphonic acid mono-2-ethylhexyl ester-grafted magnetic silica nanocomposites as adsorbent.⁵ Zhu et al.⁷ reported that the saponified sec-octylphenoxy acetic acid coated magnetic nanosorbent showed higher affinity toward La(III) ions over alkali metal ions, alkali earth metal ions, transition metal ions, and other rare earth ions in aqueous solution.

Calixarenes are considered as the third generation host supermolecules following crown ethers and cyclodextrins owing to their basket shape suitable for the complexation of small molecules and ions.⁸ Among the various calixarene derivatives, the water-soluble p-sulfonatocalix[n]arenes (with n being 4, 6, 8 named as calix4, calix6, and calix8, respectively) attracted much attention due to the fact that they possess the three-dimensional, flexible, and π-rich cavities, and also provide additional anchoring points from sulfonated groups resulting in the versatile inclusion/complexation properties for certain guest molecules.⁹ The complexation of lanthanide metal ions with calix4, calix6 and calix8, has been reported respectively by many research groups,^10–13 especially for the water-soluble compounds with different cavity size possessing special binding ability to the particular molecules or ions. For example, the calix8 (the chemical structure as shown in Fig. 1) can form weak 1 : 1 stoichiometry complexes with some divalent metal ions except for Cu²⁺, and form relatively stable 1 : 2 stoichiometry complex with UO₂²⁺ determined by the cavity of calix8, revealing its selective binding behaviors.¹⁴ However, calix8 is water-soluble, it's inconvenient to be used as adsorbent in aqueous solution unless it was immobilized on an insoluble substrate such as nanoparticles or porous membranes.

Fig. 1 Chemical structure of calix8.

Electrostatic attraction is an effective method for the preparation of functional membranes from oppositely charged molecules, providing opportunities to achieve ideal controllable surface, which can be employed in numerous applications with great benefits such as low cost, high efficiency, and ease to be fabricated.¹⁵ Especially, it has been widely used in the field of gas separation, alcohol–water separation, metal ions separation, water softening and seawater desalination.^16,17 The rich negatively charged p-sulfonatocalix[n]arenes (n = 4, 6, 8) have been alternating layer-by-layer self-assembled on porous polymer substrate via electrostatic interaction with polyvinylamine (PVA) to obtain an ultrathin membrane used in the selective ion transport from mixed aqueous solution containing lanthanide metal ions, alkali metal ions, alkali earth metal ions, and transition metal ions.^8,17,18

Nanofibers or nanostructure materials fabricated by electrospinning have been widely used as affinity nanofibrous membranes,^6,19 owing to their superior properties such as fine diameters, large specific surface area, and high porosity, small interfibrous pore size and stability in liquid media. In the present work, a novel nanofibrous calix8 complex membrane was prepared via electrostatic attraction to immobilize anionic calix8 on the surface of cationic aminated polyacrylonitrile (APAN) nanofibers and utilized as an affinity membrane for the selective adsorption of lanthanum(III) ions. It is clear that large surface area and numerous adsorption sites of the matrix are essential for adsorption affinity membranes. In this regard, hierarchically structured APAN nanofiber mat should be considered to prepare nanofibrous calix8–APAN complex membrane with large surface area and more efficient adsorption sites. We believe that large surface area and abundant calixarene group will ensure that the nanofibrous calix8 complex membranes possess high adsorption capacity and selectivity toward lanthanum(III) ions.

2 Experimental

2.1 Chemicals and materials

PAN (, T_g = 98 °C, ρ = 1.184 g cm⁻³) was kindly provided by Shanghai Jinshan Petroleum Chemical Co., Ltd.; p-tert-butylphenol (99.5%), LaCl₃·6H₂O (99.9%), DyCl₃·6H₂O (99.0%), PrCl₃·6H₂O (99.9%), SmCl₃·6H₂O (99.0%), ErCl₃·6H₂O (99.9%), NdCl₃·6H₂O (99.9%), EuCl₃·6H₂O (99.9%), TbCl₃·6H₂O (99.9%) and CeCl₃·7H₂O (99.9%) were purchased from Sigma-Aldrich China; PVP (K-30, ), N,N-dimethylformamide (DMF), DETA, Phenol, toluene, xylene, chloroform, methyl alcohol, NaOH, Na₂CO₃, CuSO₄·5H₂O, MgCl₂, AlCl₃, Na₂SO₄, FeCl₃·6H₂O, KCl, CaCl₂, HCl (aqueous solution, 36.5%), paraformaldehyde, and concentrated sulphuric acid (98%) were purchased from China State Medicinal Group Chemical Reagent Co., Ltd, as analytical grade. All the chemicals were used directly without further purification.

2.2 Synthesis of calix8

The functional compound of calix8 (p-sulfonatocalix[8]arene) was synthesized according to the method reported by Gutsche et al.^20,21 and Shinkai et al.,²² m.p. 380 °C; FTIR (ATR): ν_max/cm⁻¹ 3467 (OH), 1046 and 1159 (SO₃); ¹H-NMR: δ_H (400 MHz, D₂O, Me₄Si): 4.6–4.9 (16H, s, ArCH₂Ar), 7.2 (16H, s, ArH), 10.0 (8H, s, ArOH). The results were in accordance with the reports in literature.²²

2.3 Preparation of nanofibrous calix8 complex membrane

The procedure of preparation of nanofibrous calix8–APAN complex membranes was divided into three steps. Firstly, PAN nanofibrous mats were prepared by electrospinning technique with two different manners. One is to prepare nonporous PAN nanofibers by typical dry-electrospinning method (without the collecting bath): PAN was dissolved in DMF by mild stirring at 50 °C for 24 h to make 10 wt% homogeneous transparent solution, and the typical electrospinning parameters were as follows: the applied electric voltage was 25 kV, the distance between the spinneret and the grounded drum was 15 cm, and the solution feed rate was 1.0 mL h⁻¹. The other is to prepare porous PAN nanofibers by wet-electrospinning method (with hot water as the collecting bath),²³ as shown schematically in Fig. 2. Porous PAN nanofibers were prepared with in situ pore-forming method with the aid of pore-forming agent PVP. Here, the 20 wt% homogeneous polymer blend solution for electrospinning was obtained by dissolving polymer powders of PAN and PVP with PAN/PVP weight ratio of 1 : 2.5 in DMF at 50 °C for 24 h. The typical parameters for wet-electrospinning experiments were as follows: the applied electric voltage was 30 kV, the distance between the spinneret and collector was 15 cm, and the solution feed rate was 0.5 mL h⁻¹. The surface velocity of the drum for nanofibers collection was about 3 m min⁻¹. The leaching bath was hot water with the temperature of 85 °C for the removal of PVP during the electrospinning process. It should be noted that the samples of electrospun nonporous and porous PAN nanofibers were denoted as PAN-N and PAN-P nanofibers, respectively.

Fig. 2 Schematic diagram of the wet-electrospinning setup for the in situ pore-formation.

Secondly, the as-prepared two kinds of PAN nanofibrous mats were aminated with DETA.²⁴ In brief, the amination of PAN nanofibrous mat was carried out in a 250 mL three-necked round bottom flask equipped with thermometer, a nitrogen gas inlet, a reflux condenser and magnetic stirrer. A 100 mL deionized water, 1.0 g PAN nanofibrous mats, 31.52 g DETA and 1.0 g Na₂CO₃ were added into the flask and heated to 90 °C with stirring under nitrogen atmosphere for 2 h. Afterwards, the flask was cooled down to ambient temperature, and then the mats were taken out from the liquid and washed with deionized water until neutral. The obtained APAN nanofibrous mats were dried in vacuum at 50 °C for 24 h, and stored in a desiccator.

Subsequently, nanofibrous calix8 complex membranes were fabricated by anionic calix8 adsorbed onto the cationic APAN nanofibrous mats. The APAN mat was cut into the size of 5 cm × 5 cm (about 100 mg), and then dipped into 200 mL calix8 aqueous solution at certain concentration. In consideration of the possible pH value for the following adsorption of metal ions on calix8–APAN complex membrane, the initial pH of calix8 aqueous solution will be set at 5.0 to prepare the calix8–APAN complex membrane. The membranes were kept in the solution mounted on a horizontal shaking at 25 °C and 75 rpm for 24 h to ensure the sufficient adsorption of calix8 onto APAN nanofibers to obtain nanofibrous calix8–APAN complex membrane. Afterwards, the membrane was taken out and dipped into deionized water for 2 h and rinsed with deionized water twice, and then the calix8–APAN membrane was dried in vacuum at 50 °C till constant weight and stored in the desiccator for further use.

The amount of calix8 adsorbed onto the APAN nanofibrous membrane was calculated according to the following equation:

where q is the mass of calix8 adsorbed on APAN nanofibrous membrane (g g⁻¹), m₁ is the quantity of the membrane after adsorption (g), m₀ is the mass of the APAN nanofibrous membrane used in the adsorption experiments (g).

2.4 Adsorption experiments

2.4.1 Adsorption of La³⁺ in aqueous solution. The adsorption of La³⁺ onto calix8–APAN complex membrane in aqueous solution was carried out in a series of 250 mL flasks containing 100 mL aqueous solution of LaCl₃ with different initial concentration, wherein a piece of membrane was dipped into the solution. The flasks were equilibrated in a thermostatic water-bath shaker operated at 25 °C and 75 rpm for 24 h. After adsorption experiments, the calix8–APAN membranes were taken out and rinsed with deionized water twice, and then dried in vacuum at 60 °C till constant weight for later analysis. The concentration of La³⁺ before and after adsorption was determined. The amount of La³⁺ absorbed on calix8–APAN membranes at adsorption equilibrium, q_e (mg g⁻¹) was calculated according to the following equation:⁶

The removal percentage (R_p) can be calculated with the following equation:⁶

where C₀ and C_e are the initial and equilibrium concentration of La³⁺ ions (mg L⁻¹), respectively. V is the volume of the solution (L) and m is the weight of the calix8–APAN membrane used (g).

2.4.2 The selective adsorption toward La³⁺ onto calix8–APAN nanofibrous affinity membrane. The selectivity of calix8–APAN nanofibrous membrane was executed in a 250 mL flask containing 100 mL aqueous solution containing miscellaneous metal ions with the same concentration of each (150 mg L⁻¹), and a piece of membrane about 50 mg was dipped into the solution. The flask was equilibrated in a thermostatic water-bath shaker operated at 25 °C and 75 rpm for 24 h. After adsorption experiments, the calix8–APAN membranes were taken out and rinsed with deionized water twice, and then dried in vacuum at 60 °C till constant weight for later analysis. The concentrations of mixed metal ions in solution before and after adsorption were determined. The amount of metal ions absorbed on calix8–APAN membranes at adsorption equilibrium (q_e, mg g⁻¹) was calculated according to the eqn (2), and the selective coefficient of S_La/M could be calculated with the following equation:¹⁸where q_e,La, q_e,M is the amount of La³⁺ ions and metal ions adsorbed onto calix8–APAN membrane at equilibrium, respectively (mmol g⁻¹).

2.5 Desorption and regeneration experiments

In our previous work, the 0.01 M H₃PO₄ aqueous solution was used as the efficient eluant for La³⁺ desorption from polydopamine complex nanofibrous membrane due to the stronger coordination of La³⁺ with H₃PO₄.⁶ Thus, the 0.01 M H₃PO₄ aqueous solution was chosen as eluant in desorption and regeneration experiments after La³⁺ equilibrated onto calix8–APAN complex membrane. The calix8–APAN membranes were put into the desorption solution for 24 h, and then the membranes were washed with plenty of deionized water three times and reused in the next cycle of adsorption experiment. The adsorption–desorption experiments were carried out for 3 cycles. The desorption efficiency (D) and regeneration percentage (R) were determined from the following equations:⁶where C₀ is the concentration of La³⁺ in desorption solution (mg L⁻¹), V is the volume of the desorption solution (L), q_d is the amount of La³⁺ adsorbed on calix8–APAN membrane before desorption experiment (mg g⁻¹), q_eq is the maximum amount of La³⁺ adsorbed onto calix8–APAN membrane at equilibrium (mg g⁻¹), and m_d is the amount of the calix8–APAN membrane used in the desorption experiments (g).

2.6 Measurements and characterization

The morphology of APAN nanofibrous membranes were examined by field emission scanning electron microscopy (FESEM) (SU8000, Hitachi, Japan). The FTIR spectroscopy (Nicolet-6700, Thermometer, USA) in attenuated total reflectance (ATR) mode was used to analyze the synthesized calix8 and calix8–APAN complex nanofibrous membranes before and after La³⁺ uptake with a resolution of 4 cm⁻¹ and in the range of 4000–650 cm⁻¹. ¹H-NMR spectra were recorded on a Bruker Avance 400 MHz NMR Spectrometer. The Brunauer–Emmett–Teller (BET) surface area was obtained by a surface area analyzer (ASAP 2010, Micromeritics Instruments, USA) using N₂ adsorption and desorption isotherms. The concentrations of La³⁺ and foreign metal cations in aqueous solution were investigated by inductively coupled plasma-atomic emission spectroscopy (ICP-AES, Prodigy, USA). The pH of aqueous solution was obtained with pH meter (PB-10, Sartorius, Germany). The X-ray photoelectron spectroscopy (XPS) measurements were executed to identify the interaction of La³⁺ with the surface chemical groups on calix8–APAN self-assembly membranes by using a Kratos Axis Ultra^DLD spectrometer (Kratos Analytical-A Shimadzu Group Company, Japan) with monochromatic Al Kα radiation as the excitation and an X-ray power of 75 W. Survey scans were taken from 0 to 1200 eV binding energies, using both electrostatic and magnetic time of 100 ms. Elemental peaks were fitted using Casa XPS software (Casa Software Ltd, Teignmouth, Devon, UK). Component peak shapes were fitted using a Gaussian–Lorentzian model.

3 Results and discussion

3.1 Calix8–APAN complex nanofibrous affinity membrane

In general, the as-prepared PAN nanofibers by conventional typical electrospinning technique were nonporous as shown in Fig. 3A. In order to endow PAN nanofibers with larger surface area, a hierarchically structured surface should be constructed. In this work, wet-electrospinning technique was performed to prepare porous PAN nanofibrous membrane with the aid of pore-forming agent PVP from PAN/PVP blend solution with hot water bath for in situ leaching of PVP. The leaching of PVP and pore formation was occurred simultaneously during the electrospinning process, the typical FESEM image of the as-prepared PAN-P nanofibrous membrane was shown in Fig. 3B. The numerous nano-pores and nano-protuberances were observed on the surface of the fibers, indicating that the nanofibers were porous throughout. The throughout porous structure resulted in a high amination degree (up to 88.0%) of PAN-P nanofibers with DETA providing sufficient amine and imine groups for aminated PAN-P (APAN-P) nanofiber membranes, and the resultant APAN-P nanofibers showed a unique micro-nano structure (as shown in Fig. 3C) which ensured a large specific surface area. Actually, the specific surface area of the APAN-P nanofiber membrane was 29.1 m² g⁻¹ which was almost 4 times of APAN-N (7.3 m² g⁻¹) attributing to the micro-nano structure on the APAN-P nanofibers. Correspondingly, a relative lower amination degree (45.2%) was achieved for the APAN nanofibers prepared from normal PAN-N nanofibers.

Fig. 3 Typical FESEM images of nanofibrous membranes: (A) PAN-N, (B) PAN-P, (C) APAN-P, (D) calix8–APAN-P complex.

Obviously, the unique morphology combined with abundant amine and imine groups resulted in micro-nano structured APAN nanofibrous membrane with high positive charge density on fiber surface, which will enable the APAN nanofiber to attract more negatively charged calix8 by electrostatic adsorption and also endow the calix8–APAN-P complex membrane with its similar micro-nano rough structure (as shown in Fig. 3D). Fig. 4 showed the adsorption capacity of calix8 onto the two kinds of APAN nanofibrous membranes with the change of initial solution concentration of calix8. As can be seen, the amount of calix8 adsorbed onto APAN nanofibrous membranes increased with the increase of initial solution concentration and then tended to level off, indicating that the adsorbate calix8 reached saturation on the APAN nanofibrous membranes finally. The maximum capacity of calix8 on APAN-P nanofibrous membrane was 2.93 g g⁻¹ (Fig. 4A), which was about 4.6 times of that (0.63 g g⁻¹, Fig. 4B) from APAN-N nanofibrous membrane at equilibrium. It should be concluded that, the achieved APAN-P nanofibrous mat showed unique micro/nano structures with abundant amine and imine groups, which was beneficial to the electrostatic adsorption of anionic calix8 and resulted in the calix8–APAN-P complex membrane with high specific surface area and high calix8 loading.

Fig. 4 Effect of the initial concentration on the adsorption of calix8 onto the aminated PAN nanofibrous membranes: (A) APAN-P and (B) APAN-N (initial pH, 5.0; membrane dosage, 100 mg L⁻¹).

3.2 Adsorption of La(III) ions on nanofibrous calix8 complex membrane

Calix8 and protonated APAN are negatively and positively charged polyelectrolytes respectively, and both of them are sensitive to pH as well as the species of La(III) ions. The effect of initial pH on the adsorption capacity of La³⁺ onto calix8–APAN nanofibrous membrane was shown in Fig. 5. The amount of La³⁺ adsorbed onto calix8–APAN-P membrane increased with increasing pH from 2.0 to 5.0, and then decreased at pH > 5.0. This can be explained that the calix8–APAN membrane has stronger binding ability to H⁺ and the membrane showed strong electrostatic repulsion to La³⁺ ions at pH lower than 3.0, resulting in the low amount of La³⁺ adsorbed onto the adsorbent, and the electrostatic repulsion will decline with the increase of pH resulting in the increased uptake of La³⁺.²⁵ Moreover, La³⁺ hydroxide adducts will be formed when pH was higher than 5.0, resulting in diminished activity of La(III) ions.²⁶ The protonated amine group will decrease with the increase of pH resulting in the decrease of adsorbed calix8 when pH was higher than 5.0. Therefore, the optimal pH of 5.0 was selected for the following adsorption experiments.

Fig. 5 Effect of pH values on adsorption of La³⁺ on calix8–APAN complex nanofibrous membranes prepared from APAN-P nanofiber membrane.

The effect of initial solution concentration of La³⁺ was shown in Fig. 6. The results indicated that the adsorption capacity of La³⁺ on both calix8–APAN complex membranes all increased with the initial concentration increasing until the equilibrium between the adsorbents and La³⁺ solution was reached. When the chelating sites on the affinity membrane are occupied sufficiently, the complexes between the chelating sites and the adsorbate will be saturated and the adsorption capacity will reach a maximum level. At the initial La³⁺ concentration of 150 mg L⁻¹, the maximum adsorption capacity for La(III) ions onto the calix8–APAN-P membrane was high to 149.5 mg g⁻¹, which was about 4.4 times of that (34.2 mg g⁻¹) for the calix8–APAN-N membrane. It was attributed to the amount ratio (4.6) of the calix8 adsorbed on APAN-P and APAN-N membranes. In brief, the two advantages of unique micro-nano structure and high amine loading made the APAN-P nanofibrous membrane excellent adsorption capability for anionic calix8, and sequentially resulted in calix8–APAN-P complex nanofiber membrane with micro-nano structure and high calix8 loading for high efficient adsorption of La³⁺ ions. Therefore, it should be noted that the calix8–APAN complex membranes mentioned in the later test were from APAN-P nanofibrous membranes.

Fig. 6 Effect of La³⁺ initial concentration on the adsorption capacity onto calix8–APAN complex nanofibrous membranes prepared from: (A) APAN-P; (B) APAN-N nanofiber membranes (membrane dosage, 100 mg L⁻¹; pH, 5.0).

For a given initial concentration of adsorbate solution, the membrane dosage can affect the adsorption capacity of adsorbent. The experiments were carried out at initial concentration of 150 mg L⁻¹ with pH of 5.0, and the effect of the membrane dosage on the adsorption of La³⁺ ions was investigated by varying the amount of calix8–APAN membranes added into the solution. The effect of calix8–APAN membrane dosage on the removal percentage of La³⁺ ions was shown in Fig. 7. It can be seen that the removal percentage of the La³⁺ ions increased rapidly with the increase of membranes dosage up to 0.95 g L⁻¹ due to the increase of the available sorption sites on the surface of calix8–APAN membrane,¹⁹ and then remained constant. At the adsorption equilibrium, the maximum percentage removal of La³⁺ ions reached 99.93%.

Fig. 7 The effect of calix8–APAN membrane dosage on the removal percentage of La³⁺ ions (initial concentration, 150 mg L⁻¹; pH, 5.0).

The Langmuir and Freundlich isotherm models are commonly used to analyze the equilibrium data for describing the interaction between adsorbate and adsorbent. Langmuir isotherm model expresses the monolayer adsorption occurring on a homogeneous surface with negligible interaction between adsorbates. Therefore, adsorption equilibrium is achieved once the monolayer is completely saturated. Freundlich isotherm model describes the adsorption on an energetically heterogeneous surface with non-uniform energies of active sites.^6,19 The two widely used isotherm models can be expressed as the following equations in mathematically:^6,19

where q_e is the amount of adsorbate adsorbed on the membrane at adsorption equilibrium (mg g⁻¹), C_e is the equilibrium adsorbate concentration in solution (mg L⁻¹), q₀ expresses the maximum uptake of the adsorbate (mg g⁻¹), b is the Langmuir constant (L mg⁻¹) related to the binding energy of adsorption (affinity), K_F is the Freundlich constant depicting adsorption capacity related to bond strength (mg g⁻¹) and the slope 1/n is a measure of the adsorption intensity or surface heterogeneity, respectively.

The adsorption isotherm experiments were carried out at 298 K and the data were fitted with the two equations as shown in Fig. 8. The obtained constants were listed in Table 1. It was found that the experiment data was fitted to Langmuir equation well and the value of the correlation coefficient R² calculated from Langmuir equation was 0.9959 significantly higher than that of Freundlich equation (0.9563), suggesting that the adsorption of La³⁺ onto calix8–APAN self-assembly nanofibrous membrane at equilibrium was better fitted to Langmuir equation. Moreover, the maximum uptake at equilibrium calculated from the Langmuir equation was 155.1 mg g⁻¹, this value was very comparable with those (normally no more than 100 mg g⁻¹ for La(III) adsorption) from other adsorbents except from activated carbon (175.4 mg g⁻¹) and bamboo charcoal (215 mg g⁻¹).^5,6,26 Although the active carbon and bamboo charcoal showed relatively good capacity for lanthanum adsorption, the regeneration of these materials was not satisfied. Recently, Wu et al. reported that organophosphonic acid functionalized magnetic silica nanocomposites were used for the adsorption of La(III) with the maximum adsorption capacity of 55.9 mg g⁻¹.⁵

Fig. 8 Adsorption isotherms of La³⁺ ions onto calix8–APAN nanofibrous membranes according to Langmuir equation (A) and Freundlich equation (B) (pH, 5.0; contact time, 24 h).

Table 1 Langmuir and Freundlich constants for the adsorption of La³⁺ ions onto calix8–APAN nanofibrous membranes

Langmuir constants			Freundlich constants
q0 (mg g^−1)	b (L mg^−1)	R^2	KF (mg g^−1)	n	R^2
155.12	1.04	0.9949	63.53	3.77	0.9563

---

Fig. 9A showed the effect of contact time on the adsorption of La³⁺ ions onto calix8–APAN self-assembly nanofibrous membrane. It was found that the adsorption amount of La³⁺ ions increased with contact time and achieved equilibrium at 2 h. Within the first 30 min, the adsorption amount reached 114.9 mg g⁻¹, which was 75.8% of the adsorption capacity at equilibrium. Afterwards, the adsorption rate decreased slightly from 30 min to 2 h. This can be explained that many sorption sites are available in the initial stage on the surface of affinity membrane for La³⁺ binding, and then the adsorption rate dropped off with the decrease of available sorption sites. The adsorption achieved equilibrium when the available sorption sites and the amount remained invariant. Thus, for adsorption experiments, 24 h of contact time is enough to establish the equilibrium.

Fig. 9 Effect of contact time (A) adsorption kinetics based on the pseudo-first-order model (B) and based on the pseudo-second-order model (C) for the adsorption of La³⁺ ions onto calix8–APAN membranes (initial concentration, 150 mg L⁻¹; membrane dosage, 0.95 g L⁻¹; pH, 5.0).

Based on the results of contact time study, the adsorption kinetic data was analyzed within 2 h with the two common models: pseudo-first-order model and pseudo-second-order model. The two models can be expressed as eqn (9) and (10), respectively.¹⁹

ln(q_e − q_t) = ln q_e − k₁t (9)

where q_t is the amount of La³⁺ adsorbed on calix8–APAN membrane at an arbitrary time t (mg g⁻¹), q_e is the adsorption capacity at equilibrium (mg g⁻¹), k₁ is the pseudo-first-model rate constant (min⁻¹), k₂ is the pseudo-second-model rate constant (g mg⁻¹ min⁻¹). The analyses of the data with the two models were shown in Fig. 9B and C, and all the obtained kinetic parameters were listed in Table 2. The correlation coefficient (R² = 0.9990) of the pseudo-second-order model was higher than that of the pseudo-first-order model (R² = 0.8644), indicating that the pseudo-second-order model was more suitable for explaining the adsorption kinetics of La³⁺ ions onto calix8–APAN membrane, which also revealed that the chemisorption was the rate-determining step.¹⁹

Table 2 Adsorption kinetic parameters for the adsorption of La³⁺ ions on calix8–APAN nanofibrous membranes

Pseudo-first-order model			Pseudo-second-order model
qe (mg g^−1)	k1 (min^−1)	R^2	qe (mg g^−1)	k2 (g mg^−1 min^−1)	R^2
189.6	0.0509	0.8644	147.1	4.62 × 10^−4	0.9990

---

3.3 Selective adsorption toward La³⁺ onto calix8–APAN nanofibrous affinity membrane

In general, waste water contains different metal ions depending on the source, and these metal ions may affect the adsorption of the affinity membrane due to the competition for the sorption sites between different ions. To investigate the selectivity of the calix8–APAN nanofibrous membrane toward La³⁺, the adsorption experiment was conducted in aqueous solution containing miscellaneous ions of La³⁺, Fe³⁺, Al³⁺, Cu²⁺, Ca²⁺, Mg²⁺, K⁺. At the same time, the adsorption of various metal ions in its single ions solutions onto the calix8–APAN membrane was also tested respectively. The amount of every metal ions adsorbed onto the calix8–APAN membrane was evaluated and the results were summarized in Table 3. As can be seen, the calix8–APAN nanofibrous membrane showed excellent adsorption selectivity toward La³⁺, the coexistence of most selected metal ions didn't significantly interfere with the adsorption of La³⁺ binding to calix8–APAN complex membrane. The amount of adsorbed La³⁺ in the presence of interference ions was 1.03 mmol g⁻¹, which is slightly lower than that (1.09 mmol g⁻¹) in single ions solution attributing to the competitive adsorption from the miscellaneous metal ions, such as Cu²⁺, Fe³⁺, which could lead to a little change in adsorption equilibrium. The lowest and highest selective adsorption coefficient was 129 of S_La/Cu and 1144 of S_La/K, respectively, revealing that the calix8–APAN membrane showed high selectivity toward La³⁺ in the model waste solution. These results should be related to the ion radius and chelating bond orbital of different metal ions, the La³⁺ ion can coordinate with the 8 phenoxide atoms of calix8, resulting in the embedment of La³⁺ into the cavity of calix8 and formation of stable complex.

Table 3 Adsorption capacity of different metal ions in wastewater solution on calix8–APAN nanofibrous membranes and their selectivity coefficients

	K^+	Mg^2+	Ca^2+	Cu^2+	Fe^3+	Al^3+	La^3+
qe,single (mmol g^−1)	0.09	0.11	0.10	0.24	0.18	0.06	1.09
qe,mix (mmol g^−1)	0.9 × 10^−3	3.0 × 10^−3	3.6 × 10^−3	8.0 × 10^−3	7.0 × 10^−3	7.0 × 10^−3	1.03
SLa/M	1144	343	286	129	147	147	—

---

The selective adsorption of calix8–APAN for La³⁺ and other lanthanide ions was also investigated in the aqueous solution containing La³⁺ and other lanthanide ions such as Ce³⁺, Nd³⁺, Er³⁺, Eu³⁺, Pr³⁺, Dy³⁺, Tb³⁺, Sm³⁺, etc. with same concentration of 150 mg L⁻¹ of each. The results were summarized in Table 4. It can be seen that the maximum uptakes of these rare earth metal ions were all slightly lower than that of La³⁺ and the adsorption selective coefficients were in the range of 1.3–2.2, revealing that the calix8–APAN membrane showed slightly higher affinity toward La³⁺ than other lanthanide ions. This could be attributed to the ion radius and bonding orbital of lanthanide ions in a close range.

Table 4 Adsorption capacity of different lanthanide metal ions in wastewater solution on calix8–APAN nanofibrous membranes and their selectivity coefficients

	Ce^3+	Pr^3+	Nd^3+	Sm^3+	Eu^3+	Tb^3+	Dy^3+	Er^3+	La^3+
qe,single (mmol g^−1)	1.06	1.08	1.04	1.07	1.06	1.04	1.06	1.06	1.09
qe,mix (mmol g^−1)	0.15	0.13	0.10	0.10	0.10	0.095	0.091	0.090	0.20
SLa/M	1.3	1.5	2.0	2.0	2.0	2.1	2.2	2.2	—

---

3.4 Adsorption mechanism of La³⁺ ions onto calix8–APAN complex nanofibrous affinity membranes

To identify the interaction, the FTIR spectra of micro-nano structured APAN nanofibrous membrane, calix8–APAN complex nanofibrous membrane before and after La³⁺ adsorption were analyzed (shown in Fig. 10). Comparing to the spectra of APAN nanofibrous membrane (as shown in curve a) and virgin calix8–APAN membrane (as shown in curve b), the new peaks at 1159 and 1045 cm⁻¹ assigning to calix8 were observed in curve b, revealing that calix8 was adsorbed onto APAN nanofibrous membrane successfully.²² Furthermore, the peak at 3292 cm⁻¹ attributing to amine group (–NH₂) (shown in curve a) shifted to 3270 cm⁻¹, and the peaks at 1652 cm⁻¹ and 1567 cm⁻¹ assigning to amide group shifted to 1640 cm⁻¹ and 1557 cm⁻¹, respectively.²⁴ This is contributed to the electrostatic attraction between the positively charged –NH₂⁺– or –NH₃⁺ of APAN nanofibers and the negatively charged –SO₃⁻ of calix8. Besides, the peak at 3368 cm⁻¹ assigning to hydroxyl group shifted to 3365 cm⁻¹ due to the formation of intramolecular hydrogen bonding.¹⁴ The spectrum of calix8–APAN complex nanofibrous membrane after La³⁺ adsorption (as shown in curve c) showed that the two peaks assigning to –SO₃⁻ shifted to 1149 cm⁻¹ and 1040 cm⁻¹, respectively, indicating that interaction between La³⁺ ions and –SO₃⁻ occurred to form a complex via electrostatic interaction.²⁷ Moreover, there was a wide and strong peak in the range of 3200–3500 cm⁻¹ resulting in the peaks assigning to amine and phenolic hydroxyl being overlapped, owning to the binding water. However, there were no shifts for peaks assigning to amide group after La³⁺ loading, indicating that no interaction between La³⁺ and amide groups occurred.

Fig. 10 FTIR spectra of micro-nano structured APAN nanofibrous membranes (a), calix8–APAN membranes before (b) and after (c) La³⁺ adsorption, respectively.

The distribution of the electrons around the atoms of ligand and receptor will be impacted once the chemical bond formed. XPS is widely used to identify the interaction of an adsorbate with adsorbent.⁶ For better understand the adsorption mechanism, the XPS spectra before and after La(III) ions adsorption were investigated (shown in Fig. 11). For O 1s spectra of virgin calix8–APAN membrane (shown in Fig. 11A), there were two fitted peaks at binding energy of 532.4 eV and 531.9 eV assigning to the oxygen atom of OH and SO₃⁻, respectively.^6,28,29 The O 1s spectra were split into four individual component peaks after La³⁺ adsorption (shown in Fig. 11B). The peaks of binding energy at 531.9 eV, 532.6 eV, 533.4 eV and 534.8 eV were assigned to SO₃⁻, OH, coordination bond of O–La³⁰ and bound H₂O,³¹ respectively, indicating that there was a chemical bond between oxygen atom and La atom to form a complex. The binding energy of oxygen atom assigning to OH increased 0.2 eV after La³⁺ adsorption, the reason might be that the coordination bond of O–La reduced the charge density of oxygen atom of hydrogen bond in calix8 structure.¹⁴ Moreover, the relative content of oxygen atom of OH to S=O changed from 58.5 : 41.5 to 38.9 : 33.0, indicating that the oxygen atom involved in the coordination bond derived mostly from phenolic hydroxyl.

Fig. 11 XPS spectra of calix8–APAN membranes before (left) and after (right) La³⁺ ions adsorption: (A and B) O 1s; (C and D) N 1s; (E and F) S 2p; (G and H) survey of calix8–APAN membranes before and after La³⁺ adsorption, and La 3d (inset of H).

For the high resolution spectra of N 1s, the two splitting peaks of binding energy at 399.9 eV and 401.6 eV assigning to amine or imine (R–NH₂ or R–NH–) and ammonium (–NH₂⁺– or –NH₃⁺)^32,33 showed no shifts or new bond formation after La³⁺ adsorption (shown in Fig. 11C and D), indicating that there were no interaction between N atom and La atom. For the core-level spectra of S 2p, there are four fitted peaks at the binding energy of 171.0 eV, 170.8 eV, 170.2 and 169.2 eV assigning to the orbital of S 2p (shown in Fig. 11E and F).³⁴ It was found that there was no difference for S 2p spectra before and after La³⁺ adsorption, revealing that the sulfur atom didn't involve in the coordination with La atom. In the core-level spectra of La 3d (shown in inset of Fig. 11H), the peaks at 851.4 eV and 834.6 eV were ascribed to La 3d_3/2 and La 3d_5/2 respectively, indicating that La was adsorbed on the calix8–APAN membranes after adsorption experiment. The spin–orbit splitting of the 3d_3/2 and 3d_5/2 levels was 16.8 eV. The satellite lines on the high binding energy side of the 3d levels (854.9 and 838.1 eV for 3d_3/2 and 3d_5/2 respectively) can be interpreted in terms of the O(2p) → La(4f) charge-transfer excitation and suggested La–O bonding.³⁰

Based on the results from the FTIR and XPS studies, the adsorption of La³⁺ onto calix8–APAN self-assembly nanofibrous membrane occurred from the formation of coordination bond between O atom of phenolic hydroxyl and La atom, and the electrostatic interaction between La³⁺ ions and –SO₃⁻ groups. Sonoda et al.¹⁴ reported that the two phenolic hydroxyl groups in calix8 can be dissociated at pH about 5.0 and form two negatively charged phenoxide ions, which can form the intramolecular hydrogen bonds with neighboring in its macro cyclic ring, resulting in the considerably stabilized ionic species. Once the calix8 contacted with La³⁺ in aqueous solution, the complex would be formed via the coordination between La atom and eight oxygen atoms of phenolic hydroxyl, and the electrostatic interaction between La³⁺ ions and –SO₃⁻ groups also was involved.^14,18 As the coordination number for La³⁺ is 9 normally, one calix8 can provide eight oxygen atoms of phenolic hydroxyl to coordinate with La³⁺, resulting in the implant of La atom in the cavity of calix8. The remaining one coordination number may be provided by oxygen atom of H₂O molecule as shown in the split peak assigning to chemically bound H₂O in O 1s spectra.

3.5 Desorption and regeneration

Effective regeneration performance of an adsorbent is also very important for its potential practical application. In our previous work,⁶ 0.01 M H₃PO₄ aqueous solution was used to desorb La³⁺ ions from La³⁺ adsorbed nanofibrous polydopamine complex membranes efficiently. In this work, 0.01 M H₃PO₄ aqueous solution was chosen for desorption of La³⁺ ions from the calix8–APAN membranes. After washing with deionized water and dried in vacuum, the membrane was used in next adsorption–desorption cycle. The result of the desorption efficiency and the regeneration rate was summarized in Table 5. As can be seen, the desorption efficiency for La³⁺ ions reached 97.5% and the regeneration rate of the calix8–APAN membrane was still kept at above 95% without significant loss of their initial binding affinity after three adsorption–desorption cycles. This result indicated that the calix8–APAN nanofibrous affinity membrane could be recycled conveniently at least three times without significant loss of its initial properties.

Table 5 Desorption efficiency and regeneration rate after desorbed in 0.01 M H₃PO₄ solution of La³⁺ ions onto calix8–APAN nanofibrous membranes from three cycles of adsorption–desorption

Recycle times	Desorption efficiency (%)	Regeneration rate (%)
1	99.6	98.7
2	98.8	97.6
3	97.5	95.8

---

4 Conclusions

In conclusion, a novel calixarene affinity membrane for high efficient selective adsorption of La³⁺ ions was prepared via three-step procedure containing wet-electrospinning for porous PAN nanofiber mat, amination with DETA for micro-nano structured APAN nanofiber mat and then electrostatic adsorption of anionic calix8 on cationic APAN nanofiber mat. Two advantages of unique micro-nano structure and high amine loading made the APAN nanofiber membrane excellent adsorption capability for anionic calix8, and sequentially endowed calix8–APAN complex nanofiber membrane with micro-nano structure and high calix8 loading for highly efficient and selective adsorption of La(III) ions. The high adsorption capacity and selectivity toward La³⁺ were achieved attributing to the high content of functional phenolic hydroxyl in calix8 as confirmed by FTIR and XPS techniques. The adsorption of La³⁺ ions at equilibrium can be well fitted to Langmuir isotherm equation with the maximum uptake capacity of 155.1 mg g⁻¹. The kinetic study indicated that the adsorption of La³⁺ ions could be well fitted by pseudo-second-order kinetic equation, suggesting a rate-determining adsorption process. Moreover, the calix8–APAN affinity membrane showed slightly higher affinity toward La³⁺ than other lanthanide ions but significantly higher affinity than common metal ions (such as Fe³⁺, Al³⁺, Cu²⁺, Ca²⁺, Mg²⁺, K⁺). This special nanofibrous calix8 complex membrane with micro-nano structure could be regenerated successfully in 0.01 M H₃PO₄ aqueous solution without significantly loss of its adsorption capacity. The results of this study will be beneficial to the development of calixarene-containing adsorbents for the separation, recovery and enrichment of lanthanide ions from aqueous solution such as wastewater, ground water and surface water, and so on.

Acknowledgements

This work was supported by National Science Foundation of China (51273042 and 21174028), Program for New Century Excellent Talents in University, Program of Changjiang Scholars and Innovative Research Team in University (IRT1221), Innovation Program of Shanghai Municipal Education Commission and Fundamental Research Funds for the Central Universities.

Notes and references

1. C. Morais and V. Ciminelli, Hydrometallurgy, 2004, 73, 237–244.
2. P. Maestro and D. Huguenin, J. Alloys Compd., 1995, 225, 520–528.
3. X. Du and T. E. Graedel, Environ. Sci. Technol., 2011, 45, 4096–4101.
4. N. Das and D. Das, J. Rare Earths, 2013, 31, 933–943.
5. D. Wu, Y. Sun and Q. Wang, J. Hazard. Mater., 2013, 260, 409–419.
6. G. Hong, L. Shen, M. Wang, Y. Yang, X. Wang, M. Zhu and B. S. Hsiao, Chem. Eng. J., 2014, 244, 307–316.
7. B. Zhu, D. Wu, Y. Yang, Y. Chen, W. Li, J. Guo and Q. Wang, J. Chem. Eng. Data, 2012, 57, 553–560.
8. A. Toutianoush, J. Schnepf, A. El Hashani and B. Tieke, Adv. Funct. Mater., 2005, 15, 700–708.
9. D.-S. Guo, K. Wang and Y. Liu, J. Inclusion Phenom. Macrocyclic Chem., 2008, 62, 1–21.
10. Y. Liu, H. Wang, L. H. Wang and H. Y. Zhang, Thermochim. Acta, 2004, 414, 65–70.
11. C. B. Smith, L. J. Barbour, M. Makha, C. L. Raston and A. N. Sobolev, New J. Chem., 2006, 30, 991–996.
12. J. L. Atwood, L. J. Barbour, S. Dalgarno, C. L. Raston and H. R. Webb, J. Chem. Soc., Dalton Trans., 2002, 4351–4356.
13. Y. Israëli, C. Bonal, C. Detellier, J.-P. Morel and N. Morel-Desrosiers, Can. J. Chem., 2002, 80, 163–168.
14. M. Sonoda, K. Hayashi, M. Nishida, D. Ishii and I. Yoshida, Anal. Sci., 1998, 14, 493–500.
15. P. T. Hammond, Adv. Mater., 2004, 16, 1271–1293.
16. A. El-Hashani and B. Tieke, J. Nanosci. Nanotechnol., 2006, 6, 1710–1717.
17. B. Tieke, A. Toutianoush and W. Jin, Adv. Colloid Interface Sci., 2005, 116, 121–131.
18. A. Toutianoush, A. El-Hashani, J. Schnepf and B. Tieke, Appl. Surf. Sci., 2005, 246, 430–436.
19. M. Min, L. Shen, G. Hong, M. Zhu, Y. Zhang, X. Wang, Y. Chen and B. S. Hsiao, Chem. Eng. J., 2012, 197, 88–100.
20. C. D. Gutsche, B. Dhawan, K. H. No and R. Muthukrishnan, J. Am. Chem. Soc., 1981, 103, 3782–3792.
21. C. D. Gutsche and L. G. Lin, Tetrahedron, 1986, 42, 1633–1640.
22. S. Shinkai, K. Araki, T. Tsubaki, T. Arimura and O. Manabe, J. Chem. Soc., Perkin Trans. 1, 1987, 2297–2299.
23. G. Hong, X. Li, Y. Yang, L. Shen and M. Wang, Appl. Mech. Mater., 2014, 556–562, 60–63.
24. P. K. Neghlani, M. Rafizadeh and F. A. Taromi, J. Hazard. Mater., 2011, 186, 182–189.
25. W. E. Teo and S. Ramakrishna, Compos. Sci. Technol., 2009, 69, 1804–1817.
26. D. Wu, L. Zhang, L. Wang, B. Zhu and L. Fan, J. Chem. Technol. Biotechnol., 2011, 86, 345–352.
27. J. L. Atwood, L. J. Barbour, M. J. Hardie and C. L. Raston, Coord. Chem. Rev., 2001, 222, 3–32.
28. G. Han, S. Zhang, X. Li, N. Widjojo and T.-S. Chung, Chem. Eng. Sci., 2012, 80, 219–231.
29. L. Ruangchuay, J. Schwank and A. Sirivat, Appl. Surf. Sci., 2002, 199, 128–137.
30. J. M. Ouyang, W. J. Zheng, N. X. Huang and Z. H. Tai, Thin Solid Films, 1999, 340, 257–261.
31. M. Schindler, F. C. Hawthorne, M. S. Freund and P. C. Burns, Geochim. Cosmochim. Acta, 2009, 73, 2488–2509.
32. S. Deng, G. Yu, S. Xie, Q. Yu, J. Huang, Y. Kuwaki and M. Iseki, Langmuir, 2008, 24, 10961–10967.
33. Y. Lei, X. Qian, J. Shen and X. An, Ind. Eng. Chem. Res., 2012, 51, 10408–10415.
34. M. M. Nasef, H. Saidi, H. M. Nor and M. A. Yarmo, J. Appl. Polym. Sci., 2000, 76, 336–349.

---