Highly efficient removal of ¹³⁷Cs in seawater by potassium titanium ferrocyanide functionalized magnetic microspheres with multilayer core–shell structure

Abstract

In this study, a novel kind of core–shell-structured magnetic microsphere functionalized with potassium titanium ferrocyanide (KTiFC) was developed for the highly efficient removal of radioactive cesium from seawater. During the synthesis, a compact silica protective interlayer was deliberately constructed to stabilize the nano-sized magnetite cores, while preventing erosion under harsh environmental conditions. Because of high ion exchange capacity of the KTiFC functional layer, the magnetic microspheres exhibited high removal efficiency (≥97.7%) of radiocesium from ¹³⁷Cs-spiked solutions (3000–35 000 Bq L⁻¹) and contaminated seawater. Batch experiments revealed that adsorption equilibrium was rapidly achieved within 30 min and the maximum adsorption capacity was up to 43.09 mg g⁻¹. Kinetic models and Langmuir/Freundlich adsorption isotherm equations were used to fit the experiment data for describing the adsorption process. Because of the favorable magnetic property, a facile separation and reclamation of the magnetic microspheres from aqueous solution was achieved under an external magnetic field. Moreover, from a practical viewpoint, the magnetic microspheres were proven to have good re-dispersion properties and long-term stability against strong HNO₃ solutions (1.0 mol L⁻¹). These magnetic microspheres are believed to hold great promise for the clean-up of radiocesium contaminated water around nuclear facilities and/or after nuclear accidents.

---

Introduction

Substantial amounts of radioactive contamination containing harmful fission products (FPs) are generated during the operation of nuclear facilities every year. Because of their long-term threat to the ecosystem and human health, the detection and separation of these hazards has been a world-wide concern. Among the FPs, radiocesium is a major component present in radioactive liquid waste and has been involved in nuclear accidents in the past years.¹ After the Fukushima Daiichi nuclear disaster in 2011, excessive levels of ¹³⁷Cs (half-life 30.2 years) were detected in seawater along the coastline of the Fukushima prefecture.^{2 137}Cs is a strong gamma emitter with high solubility, which enhances its migration through groundwater to the biosphere. Meanwhile, because of its chemical similarity to potassium, ¹³⁷Cs can be easily incorporated in terrestrial and aquatic organisms.³ Therefore, recently, numerous efforts have been made to seek out effective and eco-friendly methods for the removal of radiocesium from contaminated aqueous media.^4–6

Ion exchange⁷ is a well-established method for separating and recovering cesium from waste solutions. As a good class of inorganic ion exchanger, transition metal ferrocyanides have shown excellent adsorption ability towards cesium because of their high affinity to Cs(I) in solutions.^8,9 However, the inadequate strength and unsuitable mechanical properties are major disadvantages of these sorbent materials, which render them unsuitable for industrial applications.^10,11 For example, granular particles or slimes essential for liquid flow through columns are difficult to obtain by a simple precipitation reaction between soluble ferrocyanides and salts of divalent transition metals. To overcome these shortcomings, different types of support materials, such as polymers,¹² biopolymers,¹³ colloid,¹¹ anion exchanger resin,¹⁴ mineral oxide,¹⁵ porous silica gel particles,¹⁶ were employed for the binding/deposition of the ferrocyanides, which could then readily reduce the clogging effect and excessive pressure drop during the operation of fix-bed columns. However, the introduction of matrix materials inevitably sacrificed the operational simplicity and economic efficiency in column separation.

In recent years, magnetic nanoparticles (MNPs) have emerged as a fascinating functional material, which combine the advantages of nanomaterials with the facility of magnetic separation. Because of the unique magnetic responsivity, low cytotoxicity, and chemically-modifiable surface property,¹⁷ MNPs possess significant potential in bioseparation,^18,19 environmental remediation,²⁰ catalysis,²¹ enzyme immobilization,^22,23 magnetic resonance imaging,²⁴ etc. Lately, functionalization of the MNPs with transition metal ferrocyanides by deposition or coating has paved a way for the facile separation of cesium. Prussian blue and potassium nickel(II) hexacyanoferrate(II) modified MNPs have been prepared with favorable magnetism and good adsorption ability, which allowed the effective remediation of Cs-contaminated environments.^13,25–27 However, from a practical viewpoint, maintaining the stability of these magnetic composites is still quite challenging. Nano-sized particles tend to form agglomerates to minimize their surface energy.²⁸ Moreover, because the magnetic composites are mostly exposed to HNO₃⁻-containing radioactive liquid waste for Cs adsorption, particle degradation and magnetic susceptibility reduction are unavoidable due to the dissolution and leaching of the magnetite.²⁹ Fortunately, multilayer core–shell architecture, known as an appealing strategy for nanomaterial preparation, have offered substantial promise to solve this problem. Based upon this strategy, an impenetrable protective layer can be constructed in magnetic composites, which will not only stabilize the nano-sized particles and prevent damage to the external environment, but also provide many possibilities for further surface functionalization.^30–33 However, to our knowledge, the core–shell architecture strategy has never been employed for the synthesis of ferrocyanide modified MNPs.

In present study, we report a novel kind of core–shell structured magnetic microsphere functionalized with potassium titanium ferrocyanide (KTiFC) for the effective decontamination of radiocesium. The magnetic composite exhibited high removal efficiency (≥97.7%) of radiocesium from ¹³⁷Cs-spiked solutions (about 3000–35 000 Bq L⁻¹) and contaminated seawater, and could be easily separated from aqueous solution under an external magnetic field due to its favorable magnetic property. Moreover, because of the introduction of a silica interlayer, the material was endowed with good dispersion properties and excellent stability against strong HNO₃ solutions (1.0 mol L⁻¹). The structure, morphology, and magnetic properties were characterized by XRD, FT-IR, SEM and VSM. The adsorption behavior towards cesium including the adsorption kinetics and adsorption isotherm was fully investigated.

Experimental section

Chemicals

Ferric chloride hexahydrate (FeCl₃·6H₂O), anhydrous sodium acetate, potassium ferrocyanide (K₄Fe(CN)₆·3H₂O), ethylene glycol (EG), tetraethyl orthosilicate (TEOS), tetrabutyl orthotitanate (TBOT), and hydroxypropyl cellulose (HPC) were purchased from Sigma-Aldrich. Cocktail Ultima Gold AB (Perkin Elmer) and 20 mL polyethylene vials (Perkin Elmer) were used for the liquid scintillation counting (LSC) measurements. All other chemicals of analytical grade, including metal nitrates, nitric acid, ammonium hydroxide and other reagents, were commercially obtained and used without further purification. Deionized water was obtained from a Milli-Q water purification system. Cesium contaminated seawater was prepared using real seawater from both the shallow and deep districts of Bohai Sea in China. A calculated amount of cesium was added to evaluate the adsorption ability of the materials.

Characterization

The Morphology of the magnetic microspheres was investigated using a LEO 1530 scanning electron microscope (SEM) with an accelerating voltage of 20 kV, coupled with energy-dispersive X-ray spectroscopy (EDX) to determine the sample composition. Powder X-ray diffraction (XRD) patterns of the products were collected on a diffractometer with Cu Kα radiation, with a scan step of 0.02° and a scan range between 10° and 80°. The Fourier transform infrared (FT-IR) spectra were recorded on a Nicolet Nexus 470 spectrometer with KBr pellets in the range 4000–400 cm⁻¹ at room temperature. The magnetic properties (M–H curve) were measured at 300 K by cycling the external field between −10 and 10 kOe using 730T Vibrating Sample Magnetometer (VSM). The concentration of the metal ions was determined by inductively coupled plasma mass spectrometry (ICP-MS). Ultra-low background liquid scintillation counting (Quantulus 1220) from Perkin Elmer was used to measure the radioactivity.

Synthesis of Fe₃O₄ magnetic core

The magnetic particles Fe₃O₄ were prepared through a solvothermal reaction.¹⁷ Briefly, 2.70 g of FeCl₃·6H₂O and 7.2 g of sodium acetate were dissolved in 100 mL ethylene glycol under magnetic stirring. A homogeneous yellow solution was obtained and transferred to a Teflon-lined stainless-steel autoclave with capacity of 200 mL. The autoclave was sealed and heated at 200 °C for 8 h, followed by cooling to room temperature. The black magnetite particles were collected by magnet and washed several times with ethanol and deionized water. Finally, the product was dried in a vacuum at 60 °C for 12 h.

Synthesis of Fe₃O₄@SiO₂ microspheres

Magnetite microspheres were coated with a thin silica layer through a sol–gel approach, forming Fe₃O₄@SiO₂ microspheres with core–shell structure. Typically, the magnetic Fe₃O₄ particles (0.5 g) were treated with a 0.1 mol L⁻¹ HCl aqueous solution by ultrasonication for 20 min. The magnetite particles were then washed with deionized water and dispersed in a mixture of absolute ethanol (200 mL), deionized water (50 mL) and concentrated ammonia aqueous solution (2.5 mL, 28 wt%), followed by the injection of tetraethyl orthosilicate (TEOS, 0.5 mL). After stirring at room temperature for 6 h, the Fe₃O₄@SiO₂ microspheres were separated using a magnet, washed with ethanol and deionized water, and then dried in vacuum at 60 °C for 6 h.

Synthesis of Fe₃O₄@SiO₂@KTiFC

In a typical procedure, 0.5 g Fe₃O₄@SiO₂ particles were mixed with deionized water (3 mL), HPC (1.25 g), and absolute ethanol (400 mL) under vigorous mechanical stirring. An ethanol solution (80 mL) of TBOT (5 mL) was added dropwise followed by heating the solution to 85 °C for 150 min under reflux. The obtained product was separated with magnet, washed with ethanol several times and dried in vacuum at 60 °C for 6 h.²¹ The particles were then soaked in 0.5 mol L⁻¹ K₄Fe(CN)₆ containing 1.5 mol L⁻¹ HCl. After mechanical stirring for 5 h, the black particles were separated with the help of a magnet and washed with deionized water until the washing liquor was colorless. Finally, the product Fe₃O₄@SiO₂@KTiFC was dried in vacuum at 60 °C for 10 h.

Adsorption kinetics

The adsorption kinetics were investigated by a batch operation in 10 mL plastic tube immersed in 25 °C constant temperature bath oscillator. The concentration of Cs(I) for the kinetic study was 0.3 g L⁻¹ 10 mg of Fe₃O₄@SiO₂@KTiFC particles were adding to 4 mL of the abovementioned cesium-containing solution with HNO₃ concentration of 1.0 mol L⁻¹. The mixed liquor was shaken at about 180 rpm for different times (5–240 min). The magnetic particles were then separated using an NdFeB magnet. The aqueous solution was filtered with syringe-type filters. The concentration of the residual cesium ions in the aqueous phase was measured by ICP-MS. The adsorption capacity, q_t, was used to evaluate the adsorption ability of the magnetic composite toward Cs(I). The q_t (mg g⁻¹) at time t was calculated by eqn (1):where C₀ and C_t represent the aqueous cesium concentration at initial time and at time t, respectively; V is the volume of the solution, and M is the mass of the adsorbent.

Adsorption isotherm

The adsorption isotherm was investigated based on batch experiments. The initial cesium concentration was varied from 0.15–1.0 g L⁻¹. The Fe₃O₄@SiO₂@KTiFC particles were placed in contact with 4 mL of solution containing various concentration of cesium with shaking on a swing bed for 2 h. After equilibrium, the particles were separated, filtered and the residual cesium concentration was analyzed by ICP-MS.

Decontamination of radioactive ¹³⁷Cs solution

Solutions containing radioactive ¹³⁷Cs were prepared by diluting a stock solution to the required concentrations. 20 mg of the Fe₃O₄@SiO₂@KTiFC particles was dispersed in 20 mL ¹³⁷Cs solution in a flask. After gently shaking the flask for 90 min, solid–liquid separation was accomplished by a magnet. An aliquot of the aqueous solution was filtered through syringe-type filter and added to a vial to determine the activity by liquid scintillation counting (LSC). The measurement method has been elaborated in our previous publication.³⁴ Here, we briefly describe it as follows. 10 mL of Cocktail Ultima AB was added to a vial for the background measurements in advance. The appropriate amount (e.g. 1 mL) of the ¹³⁷Cs solutions was then added to the vial, followed by shaking to obtain a homogeneous mixture. The radioactivity of the ¹³⁷Cs solutions was measured and determined by deducting the background signal.

The removal rate R (%) and decontamination factor (DF) were used to assess the adsorption performance of Fe₃O₄@SiO₂@KTiFC particles towards ¹³⁷Cs, which were defined by the following equations:

where A₀ and A_f were the aqueous cesium radioactivity in the initial solution and final solution after treatment with absorbent, respectively.

Cesium removal from seawater

5 mg of Fe₃O₄@SiO₂@KTiFC particles were placed in contact with 4 mL of cesium contaminated seawater. After shaking for 2 h, the aqueous solution was removed and filtered through syringe-type filters. The initial and residual cesium concentration were analyzed by ICP-MS. Except for the removal rate R (%), the distribution coefficient (K_d) was defined for evaluating the Cs removal ability of the particles from seawater:where C_e is the equilibrium concentration of Cs(I) in the seawater.

Results and discussions

Synthesis and characterization of Fe₃O₄@SiO₂@KTiFC

Potassium titanium ferrocyanides (KTiFC) have been recognized as the favorite inorganic ion exchanger for radiocesium clean-up. Here, by the covalent deposition of KTiFC to the silica encapsulated magnetite particles, we combined the advantages of such ion exchanger with the convenience of magnetic separation, which resulted in the facile and efficient removal of cesium in contaminated solutions. The synthesis scheme of the Fe₃O₄@SiO₂@KTiFC core–shell structured microspheres is illustrated in Fig. 1a. First, uniform magnetite microspheres with an average diameter of approximately 300 nm were obtained via a solvothermal synthesis. The micro-morphology of the product was examined by SEM, as shown in Fig. 1b. The well-formed magnetite core particles with good mechanical strength were then utilized for surface modification and/or functionalization. Through the typical Stöber method, a silica layer with a thickness of about 20 nm was generated as a protective coating, which can be clearly observed in the TEM inset of Fig. 1c.

Fig. 1 Schematic illustration of the fabrication procedure of Fe₃O₄@SiO₂@KTiFC (a), and SEM study of Fe₃O₄ (b), Fe₃O₄@SiO₂ microspheres (c), Fe₃O₄@SiO₂@TiO₂ microspheres (d) and Fe₃O₄@SiO₂@KTiFC (e). The insets show the TEM image (c) and EDS spectra (d & e) of the magnetic particles.

As mentioned above, a large amount of acid and salts co-exist in the cesium-containing radioactive liquid waste. The introduction of a silica protective layer is crucial for preventing the erosion of the magnetite cores. Meanwhile, the silica coating also provides a friendly surface for the subsequent deposition of the KTiFC functional layer. To this end, a thin layer of amorphous TiO₂ was first deposited on the Fe₃O₄@SiO₂ microspheres by the hydrolysis of TBOT in the ethanol solution with the help of hydroxypropyl cellulose (HPC).³⁵ The obtained Fe₃O₄@SiO₂@TiO₂ particles had a near-spherical appearance with a rough surface (Fig. 1d).

Ti signal peak was clearly observed in the EDS spectrum (Fig. 1d, inset). Finally, the Fe₃O₄@SiO₂@TiO₂ particles were immersed into a mixed solution containing potassium ferrocyanide and hydrochloric acid. Under the precipitation reaction between TiO₂ and Fe(CN)₆⁴⁻, potassium titanium ferrocyanides were formed on the surface of the magnetic microspheres. Fig. 1e shows the SEM image of the final Fe₃O₄@SiO₂@KTiFC particles. The presence of the Ti, K signal peaks in the EDS spectrum inset confirms the structure of the KTiFC functional layer.

The structure of the Fe₃O₄@SiO₂@KTiFC particles were also verified by powder X-ray diffraction (XRD) and Fourier transform infrared (FT-IR) spectroscopy. Fig. 2 shows the XRD patterns of Fe₃O₄@SiO₂ (a), Fe₃O₄@SiO₂@TiO₂ (b) and Fe₃O₄@SiO₂@KTiFC (c). The diffraction peaks in curve (a) and (b) are in agreement with a face center cubic Fe₃O₄ (JCPDS card 19-629). Obviously, after coating with the SiO₂ layer and TiO₂ layer, the diffraction patterns of the obtained materials showed no difference compared to pure magnetite, which indicates that both silica and titanium dioxide were amorphous. The XRD pattern of Fe₃O₄@SiO₂@KTiFC (curve c) reveals the superposition of the diffractions of a magnetite phase and a TiFe(CN)₆·2H₂O phase. The (200) (220) and (420) reflections of TiFe(CN)₆·2H₂O phase were clearly observed, which matches well with a previous study.³⁶ Meanwhile, an evident strong resonance at 2090 cm⁻¹ is observed in the FT-IR spectrum of Fe₃O₄@SiO₂@KTiFC (Fig. 3d), which is the characteristic stretching vibration of cyanide groups, confirming the successful deposition of potassium titanium ferrocyanide on the surface of the Fe₃O₄@SiO₂ microspheres. In addition, the typical adsorption bands at 582 cm⁻¹ contributed by the stretching vibration of Fe–O band were found in all the samples. Particles coated with a silica layer exhibited a strong Si–O–Si antisymmetric stretching vibration peak at 1089 cm⁻¹ (Fig. 3b–d). The signal of Si–O–Ti stretching vibration is around 950 cm⁻¹, which is partially overlapped by the broad Si–O peaks.²¹

Fig. 2 X-ray diffraction (XRD) patterns of Fe₃O₄@SiO₂ (a), Fe₃O₄@SiO₂@TiO₂ (b) and Fe₃O₄@SiO₂@KTiFC (c).

Fig. 3 FT-IR spectra of Fe₃O₄ particles (a), Fe₃O₄@SiO₂ (b), Fe₃O₄@SiO₂@TiO₂ (c), and Fe₃O₄@SiO₂@KTiFC (d).

Magnetic properties

A vibrating magnetometer was used to characterize the magnetic properties at 300 K. Fig. 4A displays the magnetic hysteresis loops of Fe₃O₄ (a), Fe₃O₄@SiO₂ (b), Fe₃O₄@SiO₂@TiO₂ (c), Fe₃O₄@SiO₂@KTiFC (d). The saturation magnetization values of the magnetic composites were 80.4, 61.0, 44.1, and 18.68 emu g⁻¹, respectively. The magnetite core and the silica encapsulated particles showed favorable magnetism, while the deposition of KTiFC resulted in a decrease in saturation magnetization. Under the applied external magnetic field, the Fe₃O₄@SiO₂@KTiFC particles showed a sensitive response and can be quickly separated from the aqueous medium (Fig. 4B). Moreover, negligible coercivity and remanence were observed from all the hysteresis loops, which confirmed the superparamagnetism of the magnetic particles. The superparamagnetic property will facilitate the stability and redispersion of the magnetic particles in solution, which is of great significance for practical magnetic separation applications.

Fig. 4 (A) Magnetic hysteresis loops of Fe₃O₄ particles (a), Fe₃O₄@SiO₂ (b), Fe₃O₄@SiO₂@TiO₂ (c), and Fe₃O₄@SiO₂@KTiFC (d); (B) magnetic separation–redispersion process of Fe₃O₄@SiO₂@KTiFC.

Adsorption kinetics behavior

The effect of the contact time on the adsorption capacity, q_t, of Fe₃O₄@SiO₂@KTiFC towards cesium was investigated in a 1.0 mol L⁻¹ HNO₃ solution containing 0.3 g L⁻¹ of Cs(I). After different contact time intervals between 5 min and 4 h, the mixture was separated and the aqueous cesium concentration was measured. The results are displayed in Fig. 5. The curve shows that cesium were adsorbed by KTiFC deposited magnetic materials fleetly and that the equilibrium was established after 30 min with q_t ≈ 33.56 mg g⁻¹.

Fig. 5 Influence of contact time on the adsorption capacity of cesium.

To better describe the adsorption of cesium by the Fe₃O₄@SiO₂@KTiFC magnetic composite, two representative adsorption kinetic models, including the pseudo-first-order model³⁷ and pseudo-second-order model,³⁸ were utilized to fit the experimental data. The related equations are given as eqn (5) and (6):

where q_t and q_e are the adsorption capacity at equilibrium and at the time t (mg g⁻¹), respectively. k₁ is the rate constant of pseudo-first-order model (min⁻¹) and k₂ is the rate constant of the pseudo-second-order model (g μg⁻¹ min⁻¹). Through fitting the experimental data by linear regression, the kinetic parameters k₁, k₂, q_e and the correlation coefficient R² were obtained and are summarized in Table 1. Obviously, the results show that the Cs(I) adsorption process by Fe₃O₄@SiO₂@KTiFC magnetic composite was better described by a pseudo-second-order model. The fitted curve with R² = 0.999 is plotted in Fig. 6. It can be inferred that the adsorption capacity is proportional to the number of active sites on the magnetic composite, and the rate-limiting step may be chemical sorption or chemisorption involving valency forces through sharing or exchange of electrons between sorbent and sorbate.^38–40

Table 1 Kinetic parameters fitted by the pseudo-first-order model and pseudo-second-order model for cesium adsorption by the Fe₃O₄@SiO₂@KTiFC magnetic composite

Pseudo-first-order model			Pseudo-second-order model
k1 (min^−1)	qe,1 (mg g^−1)	R^2	k2 (g mg^−1 min^−1)	qe,2 (mg g^−1)	R^2
0.368	33.08	0.989	0.024	33.98	0.999

---

Fig. 6 Pseudo-second-order model for the adsorption of Cs(I) by Fe₃O₄@SiO₂@KTiFC.

Adsorption isotherm

The adsorption capacity of the Fe₃O₄@SiO₂@KTiFC magnetic microspheres as a function of cesium concentration was studied in batch experiments by altering the initial concentration from 0.15 to 1.0 g L⁻¹. The adsorption capacity, q_e (mg L⁻¹), versus concentration of cesium after equilibrium C_e (g L⁻¹) is plotted in Fig. 7. It is observed that the adsorption capacity increased rapidly with increasing Cs(I) ion concentration when the initial concentration of Cs(I) was less than 0.3 g L⁻¹, which may be attributed to the fact that sufficient active sites were available.⁴¹ While at a higher concentration, competition for available adsorption sites resulted in a gradual slowing of the increase in adsorption capacity.

Fig. 7 Experimental data and the fitted curves by the Langmuir and Freundlich adsorption isotherm models.

The Langmuir and Freundlich adsorption isotherm models⁴² were applied to fit the adsorption data at equilibrium. The Langmuir model is valid for single-layer adsorption with the hypothesis that all the binding sites are free. The nonlinear form of the equation is written as:

where q_e and q_max are the equilibrium adsorption capacity and the monolayer maximum adsorption capacity (mg g⁻¹), respectively. K is a constant related to the affinity between the sorbent and sorbate.

In comparison, the Freundlich adsorption isotherm model is considered to be an empirical equation that is used to describe multi-layer adsorption with several kinds of adsorption sites on the surface of sorbent. The model was in the following form:

q_e = K_FC_e^1/n (8)

where K_F and n are the Freundlich constants relative to the multilayer adsorption capacity and adsorption intensity, respectively. Fig. 7 shows the experimental data and the curves fitted by the Langmuir and Freundlich models. It can be seen that the Langmuir model is more suitable for the description of the Cs(I) adsorption by Fe₃O₄@SiO₂@KTiFC. The fitted parameters and correlation coefficient for both the models are listed in Table 2. The correlation coefficient R² for the Langmuir model is up to 0.989. The maximum adsorption capacity of the Fe₃O₄@SiO₂@KTiFC magnetic microspheres toward Cs(I) is determined to be 43.09 mg g⁻¹.

Table 2 Parameters of the adsorption isotherms fitted by Langmuir and Freundlich models

Langmuir model			Freundlich model
KL (L mg^−1)	qmax (mg g^−1)	R^2	KF	n	R^2
43.09	42.23	0.989	43.99	8.24	0.848

---

Selectivity

The selectivity of Fe₃O₄@SiO₂@KTiFC toward cesium was investigated in solution containing Cs(I) and other typical interference ions such as Na(I), Ni(II), Fe(III), Sr(II), Mo(VI), Zr(IV), Ba(II), and Nd(III). The solution was prepared to simulate a real acidic radioactive liquid waste with different initial concentrations for each ion.⁴³ The contact time was 2 h, and the phase ratio was 0.02 g mL⁻¹. After the adsorption, the residual metal ions in the solution were measured. The distribution coefficient (K_d) was calculated and are summarized in Table 3. Obviously, the Fe₃O₄@SiO₂@KTiFC particles showed selective adsorption toward Cs(I) in the simulated radioactive liquid waste with the highest K_d (7849.8 mL g⁻¹). In comparison, the K_d values for all the interference ions were less than 10 mL g⁻¹. It is worth noting that the concentration of K(I) showed a significant increase after treatment with Fe₃O₄@SiO₂@KTiFC. This suggested that the ion exchange mechanism between K(I) and Cs(I) was responsible for the adsorption ability of potassium titanium ferrocyanide.⁸ Cs(I) replaced K(I) in the crystalline structures of cyanometallates, while the latter was released into the solution. Due to the size mismatch and charge unbalance, most of the interference ions could not be inserted into the crystalline structure.⁹

Table 3 Selectivity of Fe₃O₄@SiO₂@KTiFC toward Cs(I)

Ions	Concentration (g L^−1)		Kd (mL g^−1)
	Initial	Treatment
Cs^+	0.405	0.003	7849.8
Ni^2+	0.281	0.254	5.5
Fe^3+	0.240	0.239	0.1
Sr^2+	0.157	0.157	0.0
Mo^6+	1.805	1.717	2.6
Zr^4+	0.129	0.124	2.0
Ba^2+	0.405	0.379	3.4
Nd^4+	2.170	2.104	1.6
Na^+	1.138	1.112	2.5
K^+	0.002	1.457	—

---

Radioactive ¹³⁷Cs decontamination

The adsorption ability of the Fe₃O₄@SiO₂@KTiFC magnetic composite in radioactive ¹³⁷Cs-spiked solutions with different initial activity was examined. The performance parameters including the decontamination factor (DF) and removal efficiency (R%) were obtained, as summarized in Table 4. It can be seen that the ¹³⁷Cs activity in the solutions shows a significant decrease after contact with the particles for 90 min. The removal efficiency, which exceeded 97% under all the experimental conditions, indicates the high adsorption competence of Fe₃O₄@SiO₂@KTiFC magnetic composite for ¹³⁷Cs decontamination. It is notable that the residual activity of the ¹³⁷Cs solution after treatment with Fe₃O₄@SiO₂@KTiFC was 38 Bq L⁻¹ at the initial concentration of 2993 Bq L⁻¹, which was slightly higher than the background activity. In addition, high DF (>40) further confirms the application potential of the Fe₃O₄@SiO₂@KTiFC magnetic composite for the decontamination of ¹³⁷Cs-containing radioactive liquid wastes.

Table 4 Adsorption performance of Fe₃O₄@SiO₂@KTiFC in ¹³⁷Cs-spiked solutions^a

Activity of ^137Cs		Performance parameters
A0 (Bq L^−1)	Af (Bq L^−1)	DF A0/Af	R (%)
a A0: Initial activity; Af: final activity after treatment.
2993	38	79	98.7
5347	48	112	99.1
16 901	389	43	97.7
34 244	437	78	98.7

---

Cesium removal from seawater

As mentioned above, radionuclide-contaminated seawater in the Fukushima accident is a serious environmental concern. Because of the extremely low amount as well as the co-existence of a large number of competing metal ions, the removal of cesium in seawater remains a challenging task. In our study, real seawater from different districts of the Bohai Sea in China was employed to prepare cesium contaminated water for batch adsorption. The removal efficiency (R%) and distribution coefficient (K_d) is listed in Table 5. For the shallow-ocean seawaters with the addition of 15- or 150-fold cesium, the Fe₃O₄@SiO₂@KTiFC magnetic composite showed a removal efficiency higher than 98%, and the K_d was higher than 104 mL g⁻¹. After the treatment, the amount of cesium in the contaminated seawater was reduced to less than the background level. Similar results were obtained for the deep-ocean seawater, even though it had a lower concentration of cesium.

Table 5 Adsorption performance of Fe₃O₄@SiO₂@KTiFC in cesium contaminated seawater

Solution	Cs(I) concentration (ppb)		R (%)	Kd (mL g^−1)
	Initial	Treatment
Shallow-ocean water	10.01	2.61	73.9	2100.2
Shallow-ocean water + 0.15 ppm Cs(I)	161.35	2.52	98.4	48580.8
Shallow-ocean water + 1.5 ppm Cs(I)	1603.5	8.28	99.5	151197.2
Deep-ocean water	2.58	1.23	52.5	834.8
Deep-ocean water + 0.15 ppm Cs(I)	162	5.46	96.6	22507.9
Deep-ocean water + 1.5 ppm Cs(I)	1660.5	11.25	99.3	112769.2

---

Stability evaluation in HNO₃ medium

As is known, a large amount of acid co-exists with radiocesium in radioactive liquid waste. For practical application, it is essential to evaluate the acid-resistance performance of Fe₃O₄@SiO₂@KTiFC magnetic composites in harsh acidic conditions. For this purpose, 30 mg of the magnetic composite was soaked into 20 mL of 1.0 mol L⁻¹ nitric acid solution. After shaking for different interval times, aliquots of liquid solution were removed for the determination of the Fe³⁺ concentration in the aqueous phase through spectrophotometry. Fig. 8 gives the variation of the leached Fe³⁺ in the solution during 168 hours treatment. It can be seen that the naked Fe₃O₄ particles without a silica protective layer showed a fast degradation under the experimental condition. On the contrary, the Fe₃O₄@SiO₂@KTiFC magnetic composite with a silica protective layer was stable enough under strong acid conditions. No detectable Fe³⁺ ions were found in the leaching solution. Therefore, it is believed that the Fe₃O₄@SiO₂@KTiFC magnetic microspheres should be sufficiently stable for practical applications.

Fig. 8 Acid-resistance test of the magnetite microspheres and Fe₃O₄@SiO₂@KTiFC particles.

Conclusions

A novel kind of potassium titanium ferrocyanide (KTiFC) functionalized magnetic microsphere was synthesized through a core–shell architecture strategy for the removal of radiocesium. The structure, morphology, and adsorption behavior of the Fe₃O₄@SiO₂@KTiFC magnetic composite were comprehensively investigated. The high ion exchange efficiency of KTiFC in combination of the ease of the magnetic separation provided a facile way for the cesium decontamination. The magnetic composite showed high removal efficiency even in seawater containing extremely low concentrations of cesium. Moreover, for practical application, the material has been proven to have excellent stability in strong HNO₃ solutions due to the existence of silica coating as a protection layer. This magnetic composite has promising applications for the cleanup of radioactive cesium in contaminated environments.

Acknowledgements

The study was supported by Program for Changjiang Scholars and Innovative Research Team in University (IRT13026), and National Natural Science Foundation of China under Project 51103079 and 9122611.

Notes and references

1. C. Liu, Y. Huang, N. Naismith, J. Economy and J. Talbott, Environ. Sci. Technol., 2003, 37, 4261.
2. H. Kawamura, T. Kobayashi, A. Furuno, T. In, Y. Ishikawa, T. Nakayama, S. Shima and T. Awaji, J. Nucl. Sci. Technol., 2011, 48, 1349.
3. R. R. Sheha, J. Colloid Interface Sci., 2012, 388, 21.
4. P. K. Mohapatra, D. S. Lakshmi, A. Bhattacharyya and V. K. Manchanda, J. Hazard. Mater., 2009, 169, 472.
5. A. Y. Zhang and Z. F. Chai, Ind. Eng. Chem. Res., 2012, 51, 6196.
6. Y. X. Leng, G. Ye, J. Xu, J. Wei, J. C. Wang and J. Chen, J. Sol–Gel Sci. Technol., 2013, 66, 413.
7. T. Sangvanich, V. Sukwarotwat, R. J. Wiacek, R. M. Grudzien, G. E. Fryxell, R. S. Addleman, C. Timchalk and W. Yantasee, J. Hazard. Mater., 2010, 182, 225.
8. P. A. Haas, Sep. Sci. Technol., 1993, 28, 2479.
9. C. Delchet, A. Tokarev, X. Dumail, G. Toquer, Y. Barre, Y. Guari, C. Guerin, J. Larionova and A. Grandjean, RSC Adv., 2012, 2, 5707.
10. A. Nilchi, A. Khanchi, H. Atashi, A. Bagheri and L. Nematollahi, J. Hazard. Mater., 2006, 137, 1271.
11. V. Avramenko, S. Bratskaya, V. Zheleznov, I. Sheveleva, O. Voitenko and V. Sergienko, J. Hazard. Mater., 2011, 186, 1343.
12. H. H. Someda, A. A. ElZahhar, M. K. Shehata and H. A. El-Naggar, Sep. Purif. Technol., 2002, 29, 53.
13. C. Dwivedi, S. K. Pathak, M. Kumar, S. C. Tripathi and P. N. Bajaj, RSC Adv., 2013, 3, 22102.
14. T. P. Valsala, S. C. Roy, J. G. Shah., J. Gabriel, K. Raj and V. Venugopal, J. Hazard. Mater., 2009, 166, 1148.
15. A. V. Voronina, V. S. Semenishchev, E. V. Nogovitsyna and N. D. Betenekov, J. Radioanal. Nucl. Chem., 2013, 298, 67.
16. T. Sangvanich, V. Sukwarotwat, R. J. Wiacek, R. M. Grudzien, G. E. Fryxell, R. S. Addleman, C. Timchalk and W. Yantasee, J. Hazard. Mater., 2010, 182, 225.
17. A. Lu, E. L. Salabas and F. Schüth, Angew. Chem., Int. Ed., 2007, 46, 1222.
18. H. Chen, C. Deng and X. Zhang, Angew. Chem., Int. Ed., 2010, 49, 607.
19. S. Wu, N. Duan, X. Ma, Y. Xia, Y. Yu, Z. Wang and H. Wang, Chem. Commun., 2012, 48, 4866.
20. Y. Zhao, J. Li, S. Zhang, H. Chen and D. Shao, RSC Adv., 2013, 3, 18952.
21. M. Ye, Q. Zhang, Y. Hu, J. Ge, Z. Lu, L. He, Z. Chen and Y. Yin, Chem.–Eur. J., 2010, 16, 6243.
22. J. Huang, R. Zhao, H. Wang, W. Zhao and L. Ding, Biotechnol. Lett., 2010, 32, 817.
23. S. Xuan, F. Wang, X. Gong, S. Kong, J. C. Yu and K. C. Leung, Chem. Commun., 2011, 47, 2514.
24. H. Tan, J. M. Xue, B. Shuter, X. Li and J. Wang, Adv. Funct. Mater., 2010, 20, 722.
25. R. D. Ambashta, P. K. Wattal, S. Singh and D. Bahadur, J. Magn. Magn. Mater., 2003, 267, 335.
26. C. Thammawong, P. Opaprakasit, P. Tangboriboonrat and P. Sreearunothai, J. Nanopart. Res., 2013, 15, 1689.
27. H. Yang, L. Sun, J. Zhai, H. Li, Y. Zhao and H. Yu, J. Mater. Chem. A, 2014, 2, 326.
28. S. Laurent, D. Forge, M. Port, A. Roch, C. Robic, L. V. Elst and R. N. Muller, Chem. Rev., 2008, 108, 2064.
29. M. D. Kaminski and L. Nuñez, Sep. Sci. Technol., 2002, 37, 3703.
30. M. F. Shao, F. Y. Ning, J. W. Zhao, M. Wei, D. G. Evans and X. Duan, J. Am. Chem. Soc., 2012, 134, 1071.
31. Y. Deng, D. Qi, C. Deng, X. Zhang and D. Zhao, J. Am. Chem. Soc., 2007, 130, 28.
32. R. Yi, G. Ye, D. Pan, F. Wu, M. Wen and J. Chen, J. Mater. Chem. A, 2014, 2, 6840.
33. M. Kaur, A. Johnson, G. X. Tian, W. L. Jiang, L. F. Rao, A. Paszczynski and Y. Qiang, Nano Energy, 2013, 2, 124.
34. X. Feng, Q. He, Z. Chen, X. Han and J. Guo, Radiat. Meas., 2011, 46, 533.
35. Q. Zhang, D. Q. Lima, I. Lee, F. Zaera, M. F. Chi and Y. D. Yin, Angew. Chem., Int. Ed., 2011, 123, 7226.
36. X. Feng, S. Jing, Q. Wu, J. Chen and C. Song, Chin. J. Chem. Eng., 2007, 15, 184.
37. Y. S. Ho and G. McKay, Process Saf. Environ. Prot., 1998, 76, 332.
38. Y. S. Ho and G. McKay, Process Biochem., 1999, 34, 451.
39. Y. S. Ho, J. Hazard. Mater., 2006, 136, 681.
40. Y. S. Ho, Water Res., 2006, 40, 119.
41. C. Dwivedi, A. Kumar, K. K. Singh, A. K. Juby, M. Kumar, P. K. Wattal and P. N. Bajaj, J. Appl. Polym. Sci., 2013, 129, 152.
42. E. Njikam and S. Schiewer, J. Hazard. Mater., 2012, 213–214, 242.
43. F. F. Bai, G. Ye, G. J. Chen, J. C. Wei, J. C. Wang and J. Chen, React. Funct. Polym., 2013, 73, 228.

---