Highly-controllable imprinted polymer nanoshell at the surface of magnetic halloysite nanotubes for selective recognition and rapid adsorption of tetracycline

Abstract

Here, a general and effective method for preparing molecularly a imprinted polymer nanoshell on magnetic halloysite nanotubes (MHNTs) to make highly-controllable core–shell nanorods (MMINs) is described for the first time, and the as-obtained nanomaterials were then used for selective recognition and rapid adsorption of tetracycline (TC) from aqueous solution. Magnetic nanoparticles were uniformly loaded into the lumen of halloysite nanotubes (cheap, abundantly available and durable) using the impregnation and pyrolysis method. Vinyl groups were then anchored at the surface of MHNTs, subsequently directing the highly selective occurrence of imprinted polymerization at the surface, and the uniformly core–shell imprinted nanorods were easily produced via in situ precipitation polymerization, with tunable nanoshell thickness, by controlling the total amounts of monomers. The MMINs with a shell thickness of 35 nm exhibited the largest saturation adsorption capacity to TC, and the equilibrium data was well-described using the Langmuir isotherm model. The kinetic experiment showed the adsorption process reached equilibrium in about 10 min, and a pseudo-second-order kinetic model was used to fit the data well. The nanocomposites displayed selective recognition for TC and could be rapidly separated from solution by a magnet, with good stability and regeneration property, which provided practical applications for wastewater treatment, biological molecule purification and drug extraction.

---

1. Introduction

Tetracyclines (TCs), as a group of antimicrobial drugs, have been widely used in human and veterinary medicine, to prevent and treat a range of infectious diseases, due to broad-spectrum antibacterial activity and low cost.^1,2 However, a majority of administered TCs may be excreted as the original forms or transformation products, as well as the improper disposal, have resulted in the occurrence of residues in aquatic environment,^3,4 which are potentially toxic to aquatic organisms and human health through the food chain and drinking water,^5–7 causing a growing concern. It is thus of great necessity to develop efficient and reliable methods for the removal of TCs from aquatic environment.

Molecular imprinting,^8,9 as a versatile and potential technique, which involves creating artificial recognition sites with high affinity towards target species, has drawn extensive research interest to widespread applications, such as separation,^10,11 catalysis,¹² sensors,^13,14 and drug delivery,¹⁵ due to the low cost, stability and ease of preparation. To date, while various synthetic techniques were employed to prepare molecularly imprinted polymers (MIPs), most often exhibited high selectivity but many limitations including low rebinding capacity and poor site accessibility to template molecules, due to the high cross-linking nature.¹⁶ Thus, to resolve these problems, several research groups have explored convenient and controllable approaches for the development of nanostructured imprinted materials,^17–19 with controlling template molecules to be situated at the surface or in the proximity of the material's surface, which have a nanoscale size with extremely high specific surface area.

Fig. 1 shows the distribution of the effective binding sites in the imprinted bulk materials and imprinted nanomaterials by extraction of the template molecules. Here, assuming that the imprinted bulk materials are spherical with a diameter of 2R, the templates located within r nm from the surface can be extracted, and the resultant imprinted sites can capture the target species. By calculation, the effective volume of imprinted bulk materials with site accessibility is only 4[R³ − (R − r)³]π/3. Generally speaking, although various porogens or solvents were used in the imprinted process, the r value of bulk materials is extremely small. However, if the uniform imprinted materials were prepared in the form of nanostructures with a diameter of 2r nm, all templates can be completely extracted from the highly cross-linked polymer network, and the resultant sites are thus all effective for templates. As a result, molecular imprinting nanotechniques are expected to improve the adsorption capacity, kinetics, and easy-accessible recognition sites for imprinted materials. Typically, the MIP nanolayers were grafted onto the surface of various multiple functional supports to form core–shell composites, such as polystyrene microspheres,²⁰ silica spheres,²¹ fluorescent particles,²² and magnetic micro/nano particles,^23,24 providing more effective and easily-accessible recognition sites. However, the highly uniform core–shell imprinted materials, especially for fibrous natural minerals in the nanometer size, have been rarely reported,²⁵ due to the uncontrollable features of surface polymerization.

Fig. 1 Schematic illustration of the distribution of effective recognition sites in the imprinted bulk materials and the imprinted nanomaterials.

Halloysite is a kind of naturally occurring aluminosilicate clay, with a chemical formula of Al₂Si₂O₅(OH)₄·2H₂O, and structurally similar to kaolin, except for the intercalation of a monolayer of water molecules between the adjacent clay layers.²⁶ The lattice mismatch between neighboring silicone tetrahedron and aluminum octahedron layers creates a strain, which results in a halloysite sheet curving into tubes.²⁷ Halloysite nanotubes (HNTs) possess superior performance such as high porosity, large surface area and tunable surface chemistry, and therefore have attracted great interest in material science, especially catalysis,²⁸ drug delivery,²⁹ nano-template,³⁰ and separation.³¹ As compared to other nano-materials, for example carbon nanotubes, HNTs have many advantages in practical applications which are environmentally friendly and abundant in nature as a raw material.

Over the past decades, magnetic nanoparticles (MNPs) have demonstrated great charm for a wide range of applications, including magnetic separation,³² catalysis,³³ magnetic resonance imaging,³⁴ and drug delivery.³⁵ Especially particles with a size below the critical value, less than 20 nm, often exhibit superparamagnetic behavior at room temperature, that is, fast response to applied magnetic fields with negligible remanence and coercivity. However, due to the high surface area to volume ratio and chemically high activation, superparamagnetic nanoparticles tend to agglomerate to reduce energy and are easily oxidized in air,³⁶ generally leading to loss of magnetism and redispersion. Thus, to overcome these limitations, the development of protection strategies is crucial to chemically stabilizing superparamagnetic nanoparticles against degradation, comprised of embedding into or coating with organic or inorganic shell, which can also be used for new multiple-functionality, depending on the desired applications.

Herein, we made a first attempt to prepare molecularly imprinted polymer nanoshells on the surface of magnetic halloysite nanotubes to form highly-controllable core–shell nanorods, and the as-synthesized nanomaterials were then used for selective recognition and rapid removal of the tetracycline (TC) template. Firstly, magnetic nanoparticles were uniformly loaded into the lumen of halloysite nanotubes by impregnation and pyrolysis method. Then 3-(trimethoxysilyl) propylmethacrylate (TMSPMA) was anchored at the surface of magnetic halloysite nanotubes (MHNTs) through a simple silanization reaction with the hydroxyl, to form polymerizable vinyl layers, subsequently directing the highly selective occurrence of imprinted polymerization to the surface of the supported materials in acetonitrile. The optimized reaction conditions can completely prohibit homogeneous copolymerization in solution phase, and as a result the uniformly core–shell imprinted nanorods were easily produced by precipitation polymerization, with tunable shell thickness on the nanometer scale, which could be achieved by controlling the amounts of precursors. The physical and chemical properties of the obtained nanomaterials were investigated using various characterizations, such as FT-IR, SEM, TEM, TGA and VSM, the adsorption dynamic, specific adsorption and selectivity capacity of which for TC were also studied in details.

2. Materials and methods

2.1. Materials

Halloysite nanotubes were purchased from Heinan, China and were purified by repeated sedimentation processes to remove the quartz impurities, followed by drying at 80 °C for 12 h and grinding. Ethylene glycol dimethacrylate (EGDMA, 98%), 2,2′-azobis (2-methyl-propionitrile) (AIBN, 99%), methacrylic acid (MAA, 98%), 3-(trimethoxysilyl) propylmethacrylate (TMSPMA, 98%), chlortetracycline (CTC, 98%), tetracycline (TC, 98%), ciprofloxacin (CIP, 98%) and cefalexin (CFX, 99%) were obtained from Aladdin Reagent Co., Ltd. (Shanghai, China) and used as received. Iron(III) nitrate nonahydrate (Fe(NO₃)₃·9H₂O, 98%), ethylene glycol (EG, 98%), dry toluene, methanol, ethanol and acetonitrile were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All reagents were of analytical grade. Double-distilled ultrapure water was purified with a Purelab ultra (Organo, Tokyo, Japan).

2.2. Preparation and surface modification of MHNTs

Incorporation of magnetic nanoparticles into the lumen of halloysite nanotubes was achieved according to a modified method.³⁷ Briefly, 1.0 g of HNTs was added into 20 mL of ethanol under ultrasound and vigorous stirring for complete dispersion, followed by 0.6 g of Fe (NO₃)₃·9H₂O being added to the above mixture. After stirring at room temperature overnight, the mixture was dried at 90 °C in a vacuum oven until ethanol was almost completely evaporated. Once the iron salt was deposited within the lumen of HNTs, the sample was impregnated with ethylene glycol up to incipient wetness (around 10.5 mmol of ethylene glycol per gram of HNTs). The impregnated sample was then collected and heated to 400 °C with a heating rate of 5.0 °C min⁻¹ and maintained at this temperature for 2.0 h under nitrogen. The resulting nanocomposites were denoted as MHNTs.

The as-made MHNTs were dispersed into 100 mL of dry toluene by ultrasonic vibration, and 2.0 mL of TMSPMA was slowly injected into the mixture under nitrogen at 90 °C, and the reaction then proceeded for 24 h under mechanical stirring. After cooling, the resulting products were collected and washed with fresh toluene several times. Vinyl-modified magnetic halloysite nanotubes (named as V-MHNTs) were dried in vacuum at 60 °C overnight.

2.3. Preparation of highly-controllable core–shell imprinted nanorods

The TC-imprinted polymer nanoshells coated on the surface of MHNTs were prepared by in situ precipitation polymerization, where methacrylic acid and EGDMA were employed as a functional monomer and cross-linking agent, respectively. Briefly, methacrylic acid (0.3 mmol), EGDMA (1.2 mmol) and TC (0.75 mmol) were dissolved in acetonitrile (50 mL) to self-assemble in the dark at room temperature. V-MHNTs (150 mg) were dispersed into the above solution by ultrasonication and then AIBN (10 mg) as the initiator was added. This mixture was purged with nitrogen for 30 min in an ice bath. To ensure homogeneous dispersion of V-MHNTs, the reaction was carried out in a water bath oscillator with a rate of 200 rpm. The slow pre-polymerization was first undertaken at 50 °C for 6 h, and the cross-linking polymerization was completed at 60 °C for 24 h. The resulting MMINs were separated from the mixed solution with the help of an external magnet, and were then washed with acetonitrile and ethanol several times. The imprinted core–shell nanorods with controlled thickness of imprinted shells were prepared by varying the total amount of polymeric monomers (MAA and EGDMA), with the mole ratio of MAA to EGDMA being kept at a constant value of 1/4. The template TC in the polymer nanoshell was extracted with a mixture of methanol and acetic acid (9 : 1, v/v). The corresponding non-imprinted core–shell nanorods (MNINs) were also synthesized by an identical procedure without the addition of TC template.

2.4. Batch adsorption experiments

To investigate the effect of pH in the adsorption of TC in water, 5.0 mg of MMINs was dispersed into 10 mL of 200 μmol L⁻¹ solution with various initial pH values arranging from 3.0 to 7.0. After the reaction at 25 °C for 12 h, the supernatant was obtained by magnetic separation. The adsorption amount of TC could be calculated according to the difference between the initial and residual concentration in the solution, which was measured by a UV-vis spectrophotometer (Shimadzu UV-2450, Kyoto, Japan) at 275 nm. The equilibrium adsorption capacity towards the target molecule was investigated by placing 5.0 mg of MMINs in 10 mL aqueous solution with various TC concentrations, ranging from 10 to 800 μmol L⁻¹. Meanwhile, the adsorption kinetics was performed at different contact times with an initial concentration of 150 μmol L⁻¹. To study the selectivity properties of the TC-imprinted nanocomposites, 5.0 mg of MMINs was placed into each competitive antibiotic solution (CTC, CFX and CIP) with an initial concentration of 100 μmol L⁻¹. For comparison, the adsorption property of the MNINs was also investigated. All experiments were carried out in triplicate.

2.5. Characterization

The morphology and structure of the samples were observed by transmission electron microscopy (TEM, JEOL IEM-200CX) and field-emission scanning electron microscopy (FE-SEM, S-4800). Infrared spectra were recorded on a Nicolet NEXUS-470 spectrophotometer using KBr pellets (U.S.A.). The measurements of magnetic particles were carried out using a vibrating sample magnetometer (VSM, HH-15, China) under a magnetic field up to 10 kOe. The thermogravimetric analysis (TGA) of samples was measured using a Diamond TG/DTA instruments (STA 449C Jupiter, Netzsch, Germany) under an argon atmosphere up to 800 °C with a heating rate of 10 °C min⁻¹. Ultraviolet visible (UV-vis) absorption spectra of the samples were recorded by a UV-2401PC spectrometer.

3. Results and discussion

3.1. Method for highly-controllable core–shell imprinted nanorods

Fig. 2 illustrates the protocol for preparation of highly-controllable core–shell imprinted nanorods. Firstly, the Fe³⁺ irons were impregnated into the lumen of halloysite nanotubes using the solvent evaporation method, and were then reduced to magnetic nanoparticles by ethylene glycol at high temperature (400 °C) to obtain MHNTs, which were uniformly dispersed inside. Then, the MHNTs were functionalized with TMSPMA molecules through a simple silanization reaction with the hydroxyl at the surface, to form polymerizable vinyl layers. Subsequently, in the reaction system of functional monomers, cross-linking agents and initiator, the slow prepolymerization took place at 50 °C for 6 h between the vinyl groups of V-MHNTs and MAA/EGDMA monomers, leading to the formation of a thin layer of polymer oligomers at the surface. Finally, the resulting polymer preferentially nucleated and grew on the surface of MHNTs with polymer oligomers leading to the formation of uniform TC-imprinted nanoshells. The optimized reaction conditions could completely prohibit homogeneous copolymerization in the solution phase, and as a result, the uniformly core–shell imprinted nanorods were easily produced, with tunable shell thickness on the nanometer scale, which could be achieved by controlling the amounts of precursors. The strong hydrogen-bonding interactions occurred between the carboxyl group of MAA and the amine group of the hydroxyl, amine and amide groups of TC, which was thus imprinted within the polymer shells. After the removal of TC from the polymer shell, the resultant recognition sites could selectively rebind the template molecules due to perfect compatibility of shape, size and chemical interactions.

Fig. 2 Schematic route for highly-controllable core–shell imprinted nanorods.

3.2. Characterization of MMINs

Fig. 3 displayed the FT-IR spectra of MHNTs, TMSPMA, V-MHNTs, MMINs and MNINs. In the spectrum of MHNTs, the peaks at 1087 and 1031 cm⁻¹ were attributed to stretching vibrations of Si–O and Si–O–Si, respectively.³⁸ The deformation vibration of Al–O–Si band was clearly observed at 534 cm⁻¹. The peak intensity of Fe–O band was too weak so that could not be observed. The peaks at 3696 and 3623 cm⁻¹ were assigned to the stretching bands of the inner-surface –OH groups and the deformation vibration of the inner-surface –OH groups occurred at 912 cm⁻¹.³⁹ After the surface modification, as shown in Fig. 3A, a new peak could be observed at 1719 cm⁻¹ in the spectrum of V-MHNTs marked by an arrow, corresponding to the C=O stretching of TMSPMA. The spectrum of MMINs and MNINs in Fig. 3B showed almost the same curve. The C–H stretching vibration of the alkyl chain occurred at 2954 and 2981 cm⁻¹ for –CH₂ and –CH₃ groups, respectively. The characteristic peak at 3435 cm⁻¹ belonged to the –OH stretching of MAA molecules. The three significant peaks at 1731 (C=O stretching), 1249 and 1159 cm⁻¹ (C–O–C stretching) confirmed that the cross-linker agent EGDMA participated in the imprinted polymerization.⁴⁰ The above results successfully demonstrated the surface modification and imprinted polymerization at the surface of MHNTs.

Fig. 3 The FT-IR spectra of MHNTs, TMSPMA, V-MHNTs (A), MMINs and MNINs (B).

Scanning electron microscopy was used to observe the morphologies of HNTs, MHNTs and MMINs, as shown in Fig. 4, HNTs had a low tubular quality and high irregularity in diameter and morphology, with a length of 1.0–3.0 μm and an external diameter of 50–180 nm, which was also observed in the TEM images (in Fig. 5A). It is clearly evident that the surface of HNTs was rough and defective with many breakages, providing more surface hydroxyl groups than a perfect structure, which were potential reaction sites for surface modification.³⁸ In Fig. 5B, compared with HNTs, the external morphology of the MHNTs remained almost unchanged. As shown in Fig. 5B, most magnetic nanoparticles were uniformly dispersed into the inner lumen of HNTs, with an average size of about 12 nm, and a few were located at the external surface. Surprisingly, after the imprinted polymerization reaction, the products were still highly monodispersive nanorods with a smooth surface (see Fig. 4C). The MMINs showed highly core–shell nanorods in the TEM images of Fig. 6, in which the MHNTs core and polymer nanoshell could be clearly distinguished. Also, there were no pure polymer particles and bare MHNTs in the TEM images, further confirming the polymerization reaction selectively occurred at the surface of MHNTs with vinyl groups. The thickness of the imprinted nanoshell could be controlled by adjusting the total amount of monomers, which was demonstrated by the images in Fig. 6. When the total mass of the monomers changed from 0.18, 0.26–0.35 g, the nanoshell varied from 15, 27–35 nm, respectively (Fig. 6). As expected, in the presence of excessive monomers (larger than 0.4 g), homogeneous polymerization of the monomers unavoidably occurred in the solution phase, leading to forming of large bulk aggregates (in Fig. S1†).

Fig. 4 SEM images of HNTs (A), MHNTs (B) and MMINs (C).

Fig. 5 TEM images of HNTs (A) and MHNTs (B).

Fig. 6 TEM images of MMINs with different shell thickness: 15 (a and b), 27 (c and d) and 35 nm (e and f).

Thermal gravimetric analysis was used to quantify the content of imprinted polymers coated on the surface of MHNTs. Fig. 7 presented the TGA curves of MHNTs, V-MHNTs and MMINs. At temperature below 250 °C, only a low percentage of weight loss (about 2.0%) was observed for all samples, due to desorption of physically absorbed water. The weight loss for MHNTs (10.48%) was mainly assigned to the dehydroxylation of structural AlOH groups. The successful surface modification with TMSPMA resulted in an increase of weight loss to 15.73%. For MMINs, the rapid rate of weight loss in the temperature range between 250 and 600 °C was caused by thermal decomposition of the polymers (shown in Fig. 7c–e). With the increase of polymer shell thickness, the graft content of the polymer increased from 20.61, 34.39 to 40.19% through rough calculations, respectively. Also, the prepared uniform core–shell imprinted nanorods exhibited good thermal stability for practical applications.

Fig. 7 TGA curves of MHNTs (a), V-MHNTs (b) and MMINs with different shell thickness: 15 (c), 27 (d) and 35 nm (e).

Magnetization properties of MHNTs and MMINs were studied by VSM, as shown in Fig. 8A, the curves were symmetrical and passed through the origin with no hysteresis, regardless of the polymer shell thickness, suggesting that the MHNTs was superparamagnetic, which was in accordance with previous work.³⁷ The saturation magnetization value of MHNTs was 2.85 emu g⁻¹ measured at 298 °C, and the increase of polymer shell thickness resulted in a decrease of their response to the external magnetic field, the saturation magnetization values were 2.35, 1.88 and 1.59 emu g⁻¹ for MMINs with a shell thickness of 15, 27 and 35 nm, respectively. Although the saturation magnetization value was not very high, the MMINs could efficiently achieve magnetic collection, demonstrated by the photograph in Fig. 8B. The MMINs were rapidly drawn to the wall of the vial under an external magnetic field, obtaining a transparent and clear solution. MHNTs were used as a core to make the imprinted nanocomposites separate comfortably from the solution.

Fig. 8 (A) Magnetization curves of MHNTs (a) and MMINs with different shell thickness: 15 (b), 27 (c) and 35 nm (d); (B) photograph of the magnetic separation of MMINs (35 nm shell thickness) in the present of an external magnet.

3.3. Effect of solution pH

The value of pH is an important parameter for the adsorption of TC onto MMINs from aqueous solution. As shown in Fig. 9, the maximum adsorption amount for MMINs towards TC was 37.12 μmol L⁻¹, which was presented at pH value of 5.0. TC is predominantly neutral with internal zwitterion of dimethylamino group protonated and the hydroxyl group ionized in pH = 5.0 solution.⁴¹ Moreover, the pK_a of carboxylic acid in the MAA molecule is 5.65. Therefore, this condition is favorable for the formation of a strong interaction between TC and MAA. TC molecules existed either as a zwitterion or a cation in the pH 3.0 and 4.0 solutions. The adsorption amount of TC at higher pH was lower due to the repulsion interaction, where the MAA residues are presented in a negatively charged form. It was also obvious that the adsorption amounts of TC onto MMINs were larger than that of MNINs in the whole range of pH, due to the efficient imprinted recognition sites. Besides, the pH value of TC solution without any adjustment was approximately 5.0 in our experiments. Hence, pH = 5.0 was selected for the experiments.

Fig. 9 Chemistry structure of TC molecule (A). Effect of the solution pH on the adsorption of TC onto MMINs and MNINs with a shell thickness of 35 nm (B).

3.4. Effect of the shell thickness in adsorption property

To investigate the specific adsorption properties of MMINs with different shell thicknesses, the nanocomposites were suspended in the TC solution with various concentrations. As shown in Fig. 10, with the increase of TC concentration, the equilibrium adsorption amount of MMINs with different shell thickness gradually increased. Obviously, the equilibrium capacity of MMINs was dependent on the shell thickness. When the shell thickness was 35 nm, the MMINs exhibited the largest equilibrium adsorption capacity, which was the critical parameter to form the highest effective imprinted sites in the shell, achieving complete removal of templates and providing easily-accessible sites. The experiment result was in accordance with a previous report.⁴²

Fig. 10 The equilibrium adsorption capacity of TC onto MMINs with different shell thicknesses.

3.5. Adsorption isotherms

The adsorption capacities of MMINs and MNINs with a shell thickness of 35 nm toward target TCs are shown in Fig. 11, and the MMINs exhibited much larger adsorption capacity than the MNINs in the whole concentration range. The saturation adsorption capacity of TC molecules onto the MMINs was 48.37 μmol g⁻¹, which was much higher than that of other TC-imprinted materials previously reported,^43–45 owing to the more efficient recognition sites located at the surface of MMINs. Moreover, the value for MNINs was only 29.53 μmol g⁻¹, implying the existence of imprinted recognition sites.

Fig. 11 Equilibrium adsorption of target TC onto MMINs and MNINs with a shell thickness of 35 nm.

To obtain more insight into the adsorption characteristics of MMINs, the Langmuir and Freundlich isotherm models^46,47 in a linear form were tentatively used to simulate the equilibrium adsorption data, the linear forms of which were respectively expressed as follows:

where Q_e (μmol g⁻¹) is the equilibrium adsorption capacity of TC onto adsorbents, C_e (μmol L⁻¹) is the equilibrium concentration in the TC solution. Q_m (μmol g⁻¹) is the maximum adsorption capacity. K_L (L g⁻¹) is the Langmuir constant and K_F ((μmol g⁻¹) (L μmol⁻¹)^1/n) and n are the Freundlich constants.

The Langmuir and Freundlich plots of MMINs and MNINs are shown in Fig. 12 and the parameters of the two models were listed in Table 1. The correlation coefficients of the Freundlich isotherm were low, suggesting the Freundlich model could not describe well the experiment data. However, the use of the Langmuir model presented a good linear correlation with the values of R² close to 1, and the theoretical value (Q_m) by calculation was in agreement with the experimental value (Q_e,exp), indicating the surface of MMINs was homogeneous and a monolayer of TC covered the surface after adsorption.⁴⁸

Fig. 12 Langmuir (A) and Freundlich (B) plots of TC adsorption onto MMINs and MNINs with a shell thickness of 35 nm.

Table 1 Constants of the Langmuir and Freundlich isotherm models

		Langmuir model			Freundlich model
Adsorbents	Qe,exp (μmol g^−1)	Qm (μmol g^−1)	KL (L μmol^−1)	R^2	KF ((μmol g^−1) (L μmol^−1)^1/n)	1/n	R^2
MMINs	48.368	54.945	0.0114	0.9983	3.511	0.3910	0.9444
MNINs	29.532	34.129	0.0105	0.9955	2.090	0.4259	0.9233

---

3.6. Adsorption kinetics

Meanwhile, adsorption kinetics was investigated at various reaction times, and the initial concentration of TC solution was set at 100 μmol L⁻¹. Fig. 13A showed that the adsorption capacity of MMINs to TC molecules rapidly reached saturation in about 10 min, due to the ultrathin imprinted shell, suggesting an excellent application in ultrafast separation of the targets. In order to examine the mechanism and rate-controlling step in the overall adsorption process, pseudo-first-order and pseudo-second-order kinetic models are used to investigate the adsorption process, the linear forms of which are expressed as the following equations, respectively:⁴⁹

Ln(Q_e − Q_t) = Ln(Q_e) − k₁t (3)

where Q_e (μmol g⁻¹) and Q_t (μmol g⁻¹) are the adsorption amounts of TC at equilibrium and at time t, respectively. k₁ (min⁻¹) and k₂ (g μmol⁻¹ min⁻¹) are the constant of the pseudo-first-order and pseudo-second-order kinetic equation, respectively.

Fig. 13 Adsorption kinetics of TC onto MMINs and MNINs with a shell thickness of 35 nm (A); fitting curves of the pseudo-second-order kinetic equation for TC adsorption (B).

The kinetic parameters of the two models are given in Table 2. The values of R² for eqn (3) are low, and the calculated Q_e,cal were far from the experimental values of Q_e,exp. The results indicated that the pseudo-first-order kinetic model was not suitable for describing the adsorption process. Fig. 13B displays the fitting curves of MMINs and MNINs by the pseudo-second-order kinetic equation, which fitted the experimental data well (with R² > 0.99) and the experimental value were basically consistent with the calculated value, suggesting the adsorption process was well-described by the pseudo-second-order kinetic model. The results demonstrated the adsorption rate was mainly controlled by chemical adsorption, which might be caused by interaction between TC and carboxyl groups in the imprinted nanocomposites.

Table 2 Kinetic constants of the pseudo-first-order and pseudo-second-order kinetic models

		The pseudo-first-order model			The pseudo-second-order model
Adsorbents	Qe,exp (μmol g^−1)	Qe,c (μmol g^−1)	k1 (min^−1)	R^2	Qe,c (μmol g^−1)	k2 × 10^−3 (g μmol^−1 min^−1)	R^2
MMINs	31.336	154.563	0.3665	0.8670	37.736	4.973	0.9974
MNINs	19.010	24.073	0.1837	0.9549	22.883	7.700	0.9983

---

3.7. Selectivity property of MMINs

Here, three typical antibiotics including structurally analogous chlorotetracycline (CTC) and structurally different cefalexin (CFX) and ciprofloxacin (CIP) were used to examine the selectivity property of TC-imprinted sites in polymer nanoshells (in Fig. 14A). Fig. 14B shows the adsorption amounts of the analytes onto the MMINs and MNINs. Because the structures of TC and CTC were similar, the adsorption amount of MMINs to CTC was a little lower than to TC, which also exhibited a much larger adsorption amount to TC than to the CFX and CIP molecules, indicating the imprinted sites were matched with the TC template in the size, structure and chemical groups. Besides, the MNINs showed a similar amount to adsorb the CTC molecule and much smaller amounts to adsorb the CFX and CIP due to the larger steric hindrance. As a result, the imprinted nanoshells had good selectivity to tetracycline and its structurally analogous derivative.

Fig. 14 Chemical structures of CTC, CFX and CIP (A). Adsorption amounts for MMINs and MNINs with a shell thickness of 35 nm towards CTC, TC, CIP and CFX (B).

3.8. Regeneration performance of MMINs

The stability and reusable performance of the MMINs were further investigated by measuring the adsorption capacity to TC through the cyclic method, which was critical for the magnetic separation of the targets from aqueous solution. The MMINs (5.0 mg) were added into TC solution (0.5 mM) at 298 K for 0.5 h, and were then separated under a magnetic field. The adsorption and desorption procedures were repeated eight times using an identical batch of MMINs. The experiment data was intuitively presented in Fig. 15 and the MMINs exhibited high stability and continuous superior performance for adsorbing TC after eight cycles, providing a potential possibility for practical applications in wastewater treatment.

Fig. 15 Regeneration experiments for MMINs with a shell thickness of 35 nm to TC.

4. Conclusions

In summary, we developed for the first time an effective and general method for covering molecularly imprinted polymers onto MHNTs to obtain highly-controllable core–shell nanorods for selective recognition and rapid separation of target TC from water medium. Magnetic nanoparticles with an average size of about 10 nm were uniformly embedded into the lumen of HNTs. The modified vinyl end layers could direct the selective occurrence of imprinted polymerization at the surface of MHNTs. The imprinted shell thickness was controlled by varying the amount of polymerizable monomers. The MMINs with a shell thickness of 35 nm exhibited the largest saturation adsorption capacity to TC, and the equilibrium adsorption data were well described by the Langmuir isotherm model. Kinetic experiments showed the adsorption amount reaching equilibrium in about 10 min, due to the ultrathin imprinted polymer shell, and the pseudo-second-order kinetic model was used to fit the data well. Compared with competitive antibiotics, the adsorption capacity of the nanocomposites to TC antibiotic was larger than to others, suggesting the occurrence of selective recognition sites. The MMINs could be easily and rapidly separated from the solution phase by an external magnet and showed good stability and regeneration property, which could be possibly applied in wastewater treatment, biological molecule purification and drug extraction.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (no. 21077046, no. 21176107, no. 21174057, no. 21107037 and no. 21277063), the National Basic Research Program of China (973 Program, 2012CB821500), Natural Science Foundation of Jiangsu Province (SBK201122883) and Ph.D. Innovation Programs Foundation of Jiangsu Province (no. CXZZ13_0668).

References

1. S. Sczesny, H. Nau and G. Hamscher, Residue analysis of tetracyclines and their metabolites in eggs and in the environment by HPLC coupled with a microbiological assay and tandem mass spectrometry, J. Agric. Food Chem., 2003, 51, 697.
2. A. K. Sarmah, M. T. Meyer and A. B. A. Boxall, A global perspective on the use, sales, exposure pathways, occurrence, fate and effects of veterinary antibiotics (VAs) in the environment, Chemosphere, 2006, 65, 725–759.
3. B. Suárez, B. Santos, B. M. Simonet, S. Cárdenas and M. Valcárcel, Solid-phase extraction-capillary electrophoresis-mass spectrometry for the determination of tetracyclines residues in surface water by using carbon nanotubes as sorbent material, J. Chromatogr., A, 2007, 1175, 127.
4. J. Chico, S. Meca, R. Companyó, M. D. Prat and M. Granados, Restricted access materials for sample clean-up in the analysis of trace levels of tetracyclines by liquid chromatography: Application to food and environmental analysis, J. Chromatogr., A, 2008, 1181, 1.
5. A. B. A. Boxall, D. W. Kolpin, B. Halling-Sørensen and J. Tolls, Are veterinary medicines causing environmental risks?, Environ. Sci. Technol., 2003, 37, 286A.
6. V. F. Samanidou, K. I. Nikolaidou and I. N. Papadoyannis, Development and validation of an HPLC confirmatory method for the determination of tetracycline antibiotics residues in bovine muscle according to the European Union regulation 2002/657/EC, J. Sep. Sci., 2005, 28, 2247.
7. A. L. Cinquina, F. Longo, G. Anastasi, L. Giannetti and R. Cozzani, Validation of a high-performance liquid chromatography method for the determination of oxytetracycline, tetracycline, chlortetracycline and doxycycline in bovine milk and muscle, J. Chromatogr., A, 2003, 987, 227.
8. C. Alexander, H. S. Andersson, L. I. Andersson, R. J. Ansell, N. Kirsch, I. A. Nicholls, J. O'Mahony and M. J. Whitcombe, Molecular imprinting science and technology: a survey of the literature for the years up to and including 2003, J. Mol. Recognit., 2006, 19, 106–180.
9. G. Wulff, Enzyme-like catalysis by molecularly imprinted polymers, Chem. Rev., 2002, 102, 1.
10. C. Long, Z. Mai, Y. Yang, B. Zhu, X. Xu, L. Lu and X. Zou, Determination of multi-residue for malachite green, gentian violet and their metabolites in aquatic products by high-performance liquid chromatography coupled with molecularly imprinted solid-phase extraction, J. Chromatogr., A, 2009, 1216, 2275.
11. T. Jiang, L. Zhao, B. Chu, Q. Feng, W. Yan and J. Lin, Molecularly imprinted solid-phase extraction for the selective determination of 17β-estradiol in fishery samples with high performance liquid chromatography, Talanta, 2009, 78, 442.
12. S. J. Li, Y. Ge and A. P. F. Turner, A catalytic and positively thermosensitive molecularly imprinted polymer, Adv. Funct. Mater., 2011, 21, 1194.
13. R. N. Liang, D. A. Song, R. M. Zhang and W. Qin, Potentiometric Sensing of Neutral Species Based on a Uniform-Sized Molecularly Imprinted Polymer as a Receptor, Angew. Chem., Int. Ed., 2010, 49, 2556.
14. D. Lakshmi, A. Bossi, M. J. Whitcombe, I. Chianella, S. A. Fowler, S. Subrahmanyam, E. V. Piletska and S. A. Piletsky, Electrochemical sensor for catechol and dopamine based on a catalytic molecularly imprinted polymer-conducting polymer hybrid recognition element, Anal. Chem., 2009, 81, 3576–3584.
15. J. F. Yin, Y. Cui, G. L. Yang and H. L. Wang, Molecularly imprinted nanotubes for enantioselective drug delivery and controlled release, Chem. Commun., 2010, 46, 7688.
16. D. M. Gao, Z. P. Zhang, M. H. Wu, C. G. Xie, G. J. Guan and D. P. Wang, A surface functional monomer-directing strategy for highly dense imprinting of TNT at surface of silica nanoparticles, J. Am. Chem. Soc., 2007, 129, 7859.
17. C. G. Xie, Z. P. Zhang, D. P. Wang, G. J. Guan, D. M. Gao and J. Liu, Surface molecular self-assembly strategy for TNT imprinting of polymer nanowire/nanotube arrays, Anal. Chem., 2006, 78, 8339.
18. I. S. Chronakis, B. Milosevic, A. Frenot and L. Ye, Generation of molecular recognition sites in electrospun polymer nanofibers via molecular imprinting, Macromolecules, 2006, 39, 357.
19. H. H. Yang, S. Q. Zhang, F. Tan, Z. X. Zhuang and X. R. Wang, Surface molecularly imprinted nanowires for biorecognition, J. Am. Chem. Soc., 2005, 127, 1378.
20. G. J. Guan, R. Y. Liu, Q. S. Mei and Z. P. Zhang, Molecularly imprinted shells from polymer and xerogel matrices on polystyrene colloidal spheres, Chem.–Eur. J., 2012, 18, 4692.
21. T. H. Kim, C. D. Ki, H. Cho, T. Y. Chang and J. Y. Chang, Facile preparation of core–shell type molecularly imprinted particles: molecular imprinting into aromatic polyimide coated on silica spheres, Macromolecules, 2005, 38, 6423.
22. Y. Y. Zhao, Y. X. Ma, H. Li and L. Y. Wang, Composite QDs@MIP nanospheres for specific recognition and direct fluorescent quantification of pesticides in aqueous media, Anal. Chem., 2012, 84, 386.
23. C. Gonzato, M. Courty, P. Pasetto and K. Haupt, Magnetic molecularly imprinted polymer nanocomposites via surface-initiated RAFT polymerization, Adv. Funct. Mater., 2011, 21, 3947.
24. C. H. Lu, Y. Wang, Y. Li, H. H. Yang, X. Chen and X. R. Wang, Bifunctional superparamagnetic surface molecularly imprinted polymer core–shell nanoparticles, J. Mater. Chem., 2009, 19, 1077.
25. L. G. Chen, J. Liu, Q. L. Zeng, H. Wang, A. M. Yu, H. Q. Zhang and L. Ding, Preparation of magnetic molecularly imprinted polymer for the separation of tetracycline antibiotics from egg and tissue samples, J. Chromatogr., A, 2009, 1216, 3710.
26. E. Joussein, S. Petit, J. Churchman, B. Theng, D. Righi and B. Delvaux, Halloysite clay minerals: a review, Clay Miner., 2005, 40, 383.
27. G. Tari, I. Bobos, C. Gomes and J. Ferreira, Modification of surface charge properties during kaolinate to halloysite-7A transformation, J. Colloid Interface Sci., 1999, 210, 360.
28. D. Papoulis, S. Komarneni, A. Nikolopoulou, P. Tsolis-Katagas, D. Panagiotaras, H. G. Kacandes, P. Zhang, S. Yin, T. Sato and H. Katsuki, Palygorskite- and halloysite-TiO₂ nanocomposites: synthesis and photocatalytic activity, Appl. Clay Sci., 2010, 50, 118.
29. E. Abdullayev, K. Sakakibara, K. Okamoto, W. Wei, K. Ariga and Y. Lvov, Natural tubule clay template synthesis of silver nanorods for antibacterialcomposite coating, ACS Appl. Mater. Interfaces, 2011, 3, 4040.
30. S. R. Levis and P. B. Deasy, Characterisation of halloysite for use as a microtubular drug delivery system, Int. J. Pharm., 2002, 243, 125.
31. M. F. Zhao and P. Liu, Adsorption behavior of methylene blue on halloysite nanotubes, Microporous Mesoporous Mater., 2008, 112, 419.
32. T. Sen, A. Sebastianelli and I. J. Bruce, Mesoporous silica-magnetite nanocomposite: fabrication and applications in magnetic bioseparations, J. Am. Chem. Soc., 2006, 128, 7130.
33. A. H. Lu, W. Schmidt, N. Matoussevitch, H. Bpnnermann, B. Spliethoff, B. Tesche, E. Bill, W. Kiefer and F. Schvth, Nanoengineering of a magnetically separable hydrogenation catalyst, Angew. Chem., 2004, 116, 4403.
34. Z. Li, L. Wei, M. Y. Gao and H. Lei, One-pot reaction to synthesize biocompatible magnetite nanoparticles, Adv. Mater., 2005, 17, 1001.
35. C. J. Sunderland, M. Steiert, J. E. Talmadge, A. M. Derfus and S. E. Barry, Targeted nanoparticles for detecting and treating cancer, Drug Dev. Res., 2006, 67, 70.
36. A. H. Lu, E. L. Salabas and F. Schüth, Magnetic nanoparticles: synthesis, protection, functionalization and application, Angew. Chem., Int. Ed., 2007, 46, 1222.
37. A. B. Fuertes and P. Tartaj, A facile route for the preparation of superparamagnetic porous carbons, Chem. Mater., 2006, 18, 1675.
38. P. Yuan, P. D. Southon, Z. W. Liu, M. E. R. Green, J. M. Hook, S. J. Antill and C. J. Kepert, Functionalization of halloysite clay nanotubes by grafting with γ-aminopropyltriethoxysilane, J. Phys. Chem. C, 2008, 112, 15742.
39. E. Abdullayev, A. Joshi, W. B. Wei, Y. F. Zhao and Y. Lvov, Enlargement of halloysite clay nanotube lumen by selective etching of aluminum oxide, ACS Nano, 2012, 28, 7216.
40. K. Yoshimatsu, K. Reimhult, A. Krozer, K. Mosbach, K. Sode and L. Ye, Uniform molecularly imprinted microspheres and nanoparticles prepared by precipitation polymerization: the control of particle size suitable for different analytical applications, Anal. Chim. Acta, 2007, 584, 112.
41. P. Wang, Y. L. He and C. H. Huang, Reactions of tetracycline antibiotics with chlorine dioxide and free chlorine, Water Res., 2011, 45, 1838.
42. B. H. Liu, M. Y. Han, G. J. Guan, S. H. Wang, R. Y. Liu and Z. P. Zhang, Highly-controllable molecular imprinting at superparamagnetic iron oxide nanoparticles for ultrafast enrichment and separation, J. Phys. Chem. C, 2011, 115, 17320.
43. X. G. Hu, J. L. Pan, Y. L. Hu, Y. Huo and G. K. Li, Preparation and evaluation of solid-phase microextraction fiber based on molecularly imprinted polymers for trace analysis of tetracyclines in complicated samples, J. Chromatogr., A, 2008, 1188, 97.
44. W. S. Cai and R. B. Gupta, Molecularly-imprinted polymers selective for tetracycline binding, Sep. Purif. Technol., 2004, 35, 215–222.
45. R. Suedee, T. Srichana, T. Chuchome and U. Kongmark, Use of molecularly imprinted polymers from a mixture of tetracycline and its degradation products to produce affinity membranes for the removal of tetracycline from water, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2004, 811, 191.
46. I. Langmuir, The adsorption of gases on plane surfaces of glass, mica, and platinum, J. Am. Chem. Soc., 1918, 40, 1361.
47. H. M. F. Freundlich, Over the adsorption in solution, J. Phys. Chem., 1906, 57, 385.
48. P. Luo, Y. F. Zhao, B. Zhang, J. D. Liu, Y. Yang and J. F. Liu, Study on the adsorption of neutral red from aqueous solution onto halloysite nanotubes, Water Res., 2010, 44, 1489.
49. A. Lara and E. Houssam, Adsorption kinetics and thermodynamics of azo-dye Orange II onto highly porous titania aerogel, Chem. Eng. J., 2009, 150, 403.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3ra45779f

---