Selective separation of rare earth ions from aqueous solution using functionalized magnetite nanoparticles: kinetic and thermodynamic studies

Chemical Engineering Journal 327 (2017) 286–296

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Selective separation of rare earth ions from aqueous solution using functionalized magnetite nanoparticles: kinetic and thermodynamic studies Radwa M. Ashour a,b, Ramy El-sayed c, Ahmed F. Abdel-Magied b, Ahmed A. Abdel-khalek d, M.M. Ali b, Kerstin Forsberg e, A. Uheida a, Mamoun Muhammed f, Joydeep Dutta a,⇑ a

Functional Materials, Applied Physics Department, SCI School, Isafjordsgatan 22, SE-164 40 Kista Stockholm, Sweden Nuclear Materials Authority, P.O. Box 530, El Maadi, Cairo, Egypt c Experimental Cancer Medicine, KFC, Novum, Department of Laboratory Medicine, Karolinska Institute, 141 86 Stockholm, Sweden d Department of Chemistry, Faculty of Science, Beni-Suef University, Beni-Suef, Egypt e Department of Chemical Engineering, KTH Royal Institute of Technology, 100 44 Stockholm, Sweden f Institute of Graduate Studies and Research, Alexandria University, Alexandria, Egypt b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


Functionalized Fe3O4 NPs with citric

acid and L-cysteine were prepared. 3+

3+


Selective separation of La , Nd ,

Gd3+ and Y3+ by the functionalized Fe3O4 NPs. 3+
Cys@Fe3O4 NPs showed high La , Nd3+, Gd3+ and Y3+ adsorption capacity.
The adsorption process followed a pseudo-second order rate law.

a r t i c l e

i n f o

Article history: Received 11 May 2017 Received in revised form 18 June 2017 Accepted 19 June 2017 Available online 20 June 2017 Keywords: Magnetic nanoparticles Rare earths Functionalization Citric acid L-cysteine Adsorption

⇑ Corresponding author. E-mail address: [email protected] (J. Dutta). http://dx.doi.org/10.1016/j.cej.2017.06.101 1385-8947/Ó 2017 Elsevier B.V. All rights reserved.

a b s t r a c t Separation of rare earth ions (RE3+) from aqueous solution is a tricky problem due to their physicochemical similarities of properties. In this study, we investigate the influence of the functionalized ligands on the adsorption efficiency and selective adsorption of La3+, Nd3+, Gd3+ and Y3+ from aqueous solution using Magnetite (Fe3O4) nanoparticles (NPs) functionalized with citric acid (CA@Fe3O4 NPs) or L-cysteine (Cys@Fe3O4 NPs). The microstructure, thermal behavior and surface functionalization of the synthesized nanoparticles were studied. The general adsorption capacity of Cys@Fe3O4 NPs was found to be high (98 mg g1) in comparison to CA@Fe3O4 NPs (52 mg g1) at neutral pH 7.0. The adsorption kinetic studies revealed that the adsorption of RE3+ ions follows a pseudo second-order model and the adsorption equilibrium data fits well to the Langmuir isotherm. Thermodynamic studies imply that the adsorption process was endothermic and spontaneous in nature. Controlled desorption within 30 min of the adsorbed RE3+ ions from both Cys@Fe3O4 NPs and CA@Fe3O4 NPs was achieved with 0.5 M HNO3. Furthermore, Cys@Fe3O4 NPs exhibited a higher separation factor (SF) in the separation of Gd3+/ La3+, Gd3+/Nd3+, Gd3+/Y3+ ions compared to CA@Fe3O4 NPs. Ó 2017 Elsevier B.V. All rights reserved.

R.M. Ashour et al. / Chemical Engineering Journal 327 (2017) 286–296

1. Introduction Rare earth elements (REEs) have received attention due to their use in different applications such as in superconductors, telecommunications, nuclear and solar energy conversion, photocatalysis, and biosciences [1,2]. In recent years, RREs separation chemistry has gained a great attention not only for their industrial applications, but also their environmental mitigation. Gadolinium (Gd) is useful in ceramic industries, metallurgy, nuclear techniques in fuel element fabrications and heavy water reactors as neutron poison. Aqueous Gd3+ ion, even at trace level, considered as hazardous material [3,4]. In view of all the above-mentioned uses and its high toxicity, selective separation of Gd3+ from aqueous solutions is of extreme importance. Due to the similar chemical and physical properties of REEs, separating them from a mixture is still a challenging task [5,6]. The most commonly used techniques for the separation and recovery of REEs are precipitation [7], liquidliquid extraction [8], ion exchange [9] and adsorption [10]. Amongst these, solid-phase extraction (SPE) is proven to be an effective and convenient method for REEs recovery. SPE has been proven to be an easy operation offering high adsorption capacities, rapid phase separation and economical cost effectiveness [11]. Various types of adsorbents (organic, inorganic, carbon based material and bio-sorbents) have been developed for SPE recovery of REEs [12–17]. Magnetic nanoparticles have been found increased attention as nanoadsorbents for environmental decontamination due to the possibilities of using external magnetic field to guide the process of separation [18]. Furthermore, upon desorption of the adsorbed ions, the magnetic nanoparticles can be reused, making them promising sustainable adsorbents. Functionalized magnetic nanoparticles using inorganic materials have been demonstrated to be excellent candidates for selective adsorption of RE3+ ions and metal traces from aqueous solution [19–23]. Dupont et al. reported the functionalization of magnetite (Fe3O4) and nonmagnetic (silica and titanium dioxide) nanoparticles with N-[(3-trime thoxysilyl)propyl]ethylenediamine triacetic acid (TMS-EDTA) and their behavior towards the adsorption and separation of RE3+ ions in aqueous media [24]. Selective extraction of heavy samariumholmium (Sm–Ho) and Light lanthanum-niobium (La–Nd) lanthanides from aqueous solutions using diethylenetriaminepentaacetic acid (DTPA) functionalized Fe3O4 NPs, achieving high SF between heavy-lanthanides and light-lanthanides reaching the maximal value of 11.5 at low pH (2.0) in 30 min was reported by Qiang et al. recently [25]. The choice of appropriate functional chelating groups is a critical factor to obtain maximum adsorption efficiency and selectivity. Citric acid and L-cysteine are two major classes of chelating agents that have been studied as chelators for adsorption of metal ions [26]. Citric acid contains three carboxyl groups and a hydroxyl group, while L-cysteine is constituted of three different functional groups, amine, thiol and carboxylic acid (Fig. 1). Both citric acid and L-cysteine ligands lead to the availability of functional groups on Fe3O4 NPs for controlled adsorption but also affects the size, morphology and colloidal stability of the nanoparticles [21].

287

Herein, we report the synthesis and characterization of magnetic nanoparticles functionalized with citric acid (CA@Fe3O4 NPs) and L-cysteine (Cys@Fe3O4 NPs) ligands, respectively. We investigate the influence of CA@Fe3O4 NPs and Cys@Fe3O4 NPs on the selective adsorption of RE3+ (= La3+, Nd3+, Gd3+ and Y3+) and the adsorption mechanism from aqueous solution. The dependence of adsorption parameters such as contact time, pH, temperature and initial concentration of metal ions on the adsorption efficiency were studied. In addition, we report the desorption characteristics of REEs from the loaded CA@Fe3O4 NPs and Cys@Fe3O4 NPs using different eluents such as NaOH and HNO3. Adsorption kinetics and adsorption isotherms were analyzed using a non-linear method and the thermodynamic parameters (DG°, DH° and DS°) were derived from the obtained experimental results. 2. Materials and methods 2.1. Chemicals Ferric chloride hexahydrate (FeCl36H2O,  99%), ferrous chloride tetrahydrate (FeCl24H2O,  98%), ammonium hydroxide (25%), citric acid monohydrate (98%), L-cysteine, nitric acid (65%), sodium hydroxide (98%), lanthanum(III) nitrate hexahydrate (99.9%) [La(NO3)36H2O], neodymium(III) nitrate hexahydrate (99.9%) [Nd(NO3)36H2O], gadolinium(III) nitrate hexahydrate (99.9%) [Gd(NO3)36H2O], yttrium(III) nitrate hexahydrate (99.9%) [Y(NO3)36H2O] and standard solutions (1000 mg/L) of Mg(NO3)2, Ca(NO3)2and Ni(NO3)2 were purchased from Sigma Aldrich. All chemicals were of analytical grade reagents and used as received without further purification. Deionized water with a resistivity of 18 MX cm was used in all the experiments. 2.2. Characterization High resolution transmission electron microscopy (HRTEM) (HR-FEG-TEM, JEOL JEM-2100, Tokyo, Japan) was used for the characterization of the morphology and size distribution of the synthesized nanoparticles. For structural evaluation X-Ray Powder Diffraction analysis (XRPD, PANalytical Empyrean) was carried out. Fourier transform infra–red (FT–IR) spectrometer (Nicolet Instruments model Avatar-100 equipped with diamond ATR, Madison, WI, USA) was used to study characteristic functional groups on the nanoparticle surfaces in the range of 500–4000 cm1. Specific surface area calculations were determined by the BrunauerEmmett-Teller (BET) method with N2 gas using Micromeritics ASAP 2000 surface area and porosity system (Quantachrome, USA) by first degassing the samples at 150 °C for 1 hour. Zeta potential of synthesized nanoparticles and surface charge of nanoparticles were determined by (Delsa Nano C, Beckman Coulter, CA, USA). The percentage weight loss of CA@Fe3O4 NPs and Cys@Fe3O4 NPs were studied in the temperature range of 30–600 °C at a heating rate of 10 °C min in nitrogen atmosphere, using Thermogravimetric analysis (TGA) (Q5000, TA instruments, DE, USA). The concentrations of metal ions were determined by Inductively Coupled Plasma-Optical Emission Spectroscopy (ICP-OES) iCAP 6500 ICP from Thermo Scientific (MA, USA). Adjustments of solutions pH were measured by a pH-meter (ORION 410). 2.3. Synthesis of precursor Fe3O4 NPs

Fig. 1. Chemical structure of the functionalization ligands used in this study: (a) Citric acid and (b) L-cysteine.

The detailed synthetic process of Fe3O4 NPs (10 nm) have been described elsewhere [27]. Briefly, 1.988 g (0.125 mol) of FeCl24H2O and 5.406 g (0.25 mol) of FeCl36H2O (1:2 molar ratio) were dissolved in 80 mL of water in a round bottomed flask at room temperature. Then the temperature was slowly increased to

288

R.M. Ashour et al. / Chemical Engineering Journal 327 (2017) 286–296

70 °C in nitrogen atmosphere with continuous mechanical stirring at 370 rpm for 30 min, then 20 mL of ammonium hydroxide solution (25%) was added to the reaction mixture and the mixture was then kept at 70 °C for another 30 min. The obtained black precipitate was then washed three times with deionized water and separated from the supernatant using a permanent magnet to obtain the Fe3O4 NPs (about 1.30 g was obtained typically). 2.4. Synthesis of CA@Fe3O4 NPs and Cys@Fe3O4 NPs In order to functionalize the synthesized Fe3O4 NPs with citrates, 4 mL of an aqueous solution of citric acid (0.3 g mL1) was added to the Fe3O4 NPs in deionized water and the temperature then raised slowly to 90 °C under refluxing conditions with continuous mechanical stirring for 1 h. The obtained precipitate was rinsed with water and separated by an external permanent magnet to obtain CA@Fe3O4 NPs. Surface functionalization of Fe3O4 NPs with L-cysteine was carried out by mixing a known quantity of Fe3O4 NPs with 50 mL (0.1 M) L-cysteine and the reaction mixture was kept under vigorous shaking overnight. The product was separated from the solution using magnetic decantation and was washed three times using deionized water to obtain Cys@Fe3O4 NPs. 2.5. Adsorption studies Stock solutions of lanthanum(III) nitrate, neodymium(III) nitrate, gadolinium(III) nitrate and yttrium(III) nitrate were prepared by dissolving an appropriate amount of the required salt in deionized water. Adsorption experiments were conducted by mixing 10 mL of RE3+ mixture solution containing lanthanum La3+, neodymium Nd3+, gadolinium Gd3+ and yttrium Y3+ with 2.5 mg of magnetic nanoadsorbents (CA@Fe3O4 NPs or Cys@Fe3O4 NPs) and the mixture was subjected to continuous shaking. Batch adsorption experiments were carried out to investigate RE3+ ions adsorption as a function of pH. The pH was adjusted in the range 3–8 by using nitric acid and sodium hydroxide, respectively. Initial RE3+ concentration was varied from 5 to 50 mg L1 to study the adsorption isotherms of RE3+ on surface functionalized Fe3O4 NPs. Contact time was changed from 1 to 120 min to study the adsorption kinetic efficiency. Temperature was varied from 278 to 318 K to study thermodynamic parameters. The magnetic nanoadsorbents CA@Fe3O4 NPs or Cys@Fe3O4 NPs were separated from the aqueous phase with an external magnet. The adsorption experiments were performed in duplicate and the average values of total adsorption are reported. The adsorption capacity qe (mg g1) of the adsorbents were determined by Eq. (1).

Adsorption capacityðqe Þ ¼

ðCi  Ce ÞV m

ð1Þ

where, Ci is the initial RE3+ concentration (mg L1), Ce is the equilibrium RE3+ concentration (mg L1) in aqueous solution, V is the total volume of water (mL) and m is the mass of CA@Fe3O4 NPs or Cys@Fe3O4 NPs (mg). The adsorption isotherms were analyzed with the nonlinear equations of Langmuir and Freundlich models using Eqs. (2) and (3).

qe ¼

qmax kCe 1 þ kCe

ð2Þ

qe ¼ K f C e1=n

ð3Þ 1

where, qmax (mg g ) is the maximum adsorption capacity, k (L mg1) is the Langmuir equilibrium constant, Kf (mg1n Ln g1) is Freundlich constant and 1/n is the heterogeneity factor.

The adsorption kinetic experimental data were fitted to a nonlinear form of pseudo-first order and pseudo-second order models using Eqs. (4) and (5), respectively.

qt ¼ qe ð1  ek 1t Þ qt ¼

k2 q2e t 1 þ k 2 qe t

ð4Þ ð5Þ

where, k1 (min1) is the pseudo-first order rate constant of adsorption, k2 (g mg1 min) is the pseudo-second order rate constant of adsorption, qe (mg g1) is the adsorption capacity at equilibrium and qt (mg g1) is the adsorption capacity at a given time t. 2.6. Selective adsorption studies Experiments were performed to study the selectivity of CA@Fe3O4 NPs and Cys@Fe3O4 NPs towards RE3+ adsorption. The selective adsorption experiments were carried out by mixing a known amount of CA@Fe3O4 NPs or Cys@Fe3O4 NPs with La3+, Nd3+, Gd3+ and Y3+ solutions containing Ni2+, Ca2+ and Mg2+ at 1:1 molar ratio using a shaker at room temperature. The aqueous phase was separated from the solid phase by using an external magnet followed by ultra-centrifuging of the supernatant at 10,000 rpm and the concentration of metal ions in the supernatant was determined by ICP-OES. 2.7. Desorption studies To examine the feasibility of recycling the coated nanoparticles, desorption studies were performed using 0.1–1 mol L1 of eluting solution of NaOH or HNO3 for 30 min. The desorption efficiency of La3+, Nd3+, Gd3+ and Y3+ from the loaded CA@Fe3O4 NPs or Cys@Fe3O4 NPs was calculated using Eq. (6).

Desorption% ¼

Cdes x100 Cads

ð6Þ

where, Cdes is the amount of La3+, Nd3+, Gd3+ and Y3+ released into aqueous solution (mg L1) and Cads is the amount of La3+, Nd3+, Gd3+ and Y3+ adsorbed on the coated Fe3O4 NPs (mg L1). 3. Results and discussion 3.1. Characterization of CA@Fe3O4 NPs and Cys@Fe3O4 NPs The TEM images of CA@Fe3O4 NPs and Cys@Fe3O4 NPs (Fig. 2a and b, respectively), revealed that Fe3O4 NPs are well separated spherical particles. The particle size distribution determined from TEM micrographs, showed that the particles size of CA@Fe3O4 NPs and Cys@Fe3O4 NPs are 12 ± 3 nm and 10 ± 5 nm, respectively (Fig. 2a (inset) and b (inset)). The XRD patterns of the synthesized CA@Fe3O4 NPs and Cys@Fe3O4 NPs are presented in Fig. 2c and d, respectively. Five characteristic peaks for Fe3O4 NPs corresponding to (2 Theta) = 30.136, 35.554, 43.2, 53.529, 57.012 and 62.961 for (2 2 0), (3 1 1), (4 0 0), (4 2 2), (5 1 1) and (4 4 0) planes were observed. These peaks are well matched with the magnetite characteristic peaks confirming the functionalization process did not result in phase alteration of Fe3O4 NPs [28]. The TGA thermograms of pristine Fe3O4 NPs, CA@Fe3O4 NPs and Cys@Fe3O4 NPs are shown in Fig. 2e and f, respectively. About 1% weight loss for all NPs are observed at 50–200 °C due to the dehydration of NP surface. Above 200 °C, 8.8% weight loss of CA@Fe3O4 NPs and 2.8% for Cys@Fe3O4 NPs were observed due to the removal of the ligands. Weight losses above 450 °C is related to transformation of Fe3O4–Fe2O3 [28]. The obtained TGA results confirm that the

R.M. Ashour et al. / Chemical Engineering Journal 327 (2017) 286–296

289

Fig. 2. Characterization of the functionalized CA@Fe3O4 NPs and Cys@Fe3O4 NPs: TEM image of (a) CA@Fe3O4 NPs and (b) Cys@Fe3O4 NPs with particle size distribution (inset); XRD pattern of (c) CA@Fe3O4 NPs and (d) Cys@Fe3O4 NPs; TGA curves of (e) CA@Fe3O4 NPs, (f) Cys@Fe3O4 NPs and pristine Fe3O4 NPs; FT-IR analysis of (g) CA, Fe3O4 NPs and CA@Fe3O4 NPs and (h) Cys, Fe3O4 NPs and Cys@Fe3O4 NPs.

surface of Fe3O4 NPs was successfully functionalized with CA and Cys. FT-IR spectrum of CA@Fe3O4 NPs (Fig. 2g) showed that carboxylic functional group at 1680 cm1 assigned to C@O vibration of CA was shifted to lower wave number due to the covalent bonding on the surface of Fe3O4 nanoparticles as confirmed by the

appearance at 1582 cm1. The symmetric stretching of the COO group of CA at 1401 cm1 is also slightly shifted to lower frequencies with a strong but broad band appearing at 1389 cm1, confirming the successful functionalization of Fe3O4 nanoparticles with CA [29]. Similarly, the infrared spectrum of Cys@Fe3O4 NPs (Fig. 2h), reveal a strong broad peak at 3055 cm1, attributed to

290

R.M. Ashour et al. / Chemical Engineering Journal 327 (2017) 286–296

the ANH2 stretches, at 1586 cm1 due to the symmetric stretching of the C@O, at 1375 cm-1 due to the symmetric stretching of the COO The band at 2555 cm1 attributed to the –SH group of Cys, disappear in Cys@Fe3O4 NPs implying that the Cys molecules attaches to the surface of the Fe3O4 NPs via the formation of FeAS covalent bond [26,28]. The BET analysis for Fe3O4 NPs, CA@Fe3O4 NPs and Cys@Fe3O4 NPs are shown in Fig. S1a, b and c (ESIy), respectively. The specific surface area of Fe3O4 NPs, CA@Fe3O4 NPs and Cys@Fe3O4 NPs were 87.95, 94.65 and 53.95 m2 g1, respectively. The magnetic properties of the magnetic nanoparticles have been reported elsewhere [30]. The zeta potential of CA@Fe3O4 NPs and Cys@Fe3O4 NPs in the pH range of 2–10 are shown in Fig. 3. In case of CA@Fe3O4 NPs, high negative charges on the surface is observed and the isoelectric point (IEP) could not be found in the entire pH range studied, as also observed by others [29]. The amount of the negative charges increased with increasing pH due to the deprotonation of carboxylic groups in the case of functionalisation with citrates. The obtained results thus confirm that grafting of carboxylic groups to the surface of Fe3O4 NPs has occurred [31], as the IEP of Fe3O4 NPs is expected to be at pH 6.5 [32]. The zeta potential measurements of Cys@Fe3O4 NPs showed that the zeta potential values of Cys varies with pH, similar to other amino acids.[33] The IEP of Cys has been reported to be at pH 5.1 [34]. Below the IEP, a positive charge of Cys due to the protonation of amino and carboxylic groups on the surface of the Fe3O4 NPs, render the overall charge positive. Above IEP, the surface is negatively charged due to deprotonation of carboxylic and amino groups at higher pH values which causes the surface potential to cross from zero to the negatively charged region [35].

pH factor plays an important role in the adsorption of RE3+ ions on the surface of the NPs. To investigate the effect of pH on the adsorption of La3+, Nd3+, Gd3+ and Y3+ ions on CA@Fe3O4 NPs and Cys@Fe3O4 NPs, we prepared a series of mixed sample solutions containing 5 mg L1 of La3+, Nd3+, Gd3+ and Y3+ ions (Fig. 4c and d). The pH values of the solutions were adjusted by HNO3 or NaOH solution. This study is restricted up to a maximum pH 8.0 since beyond this pH, precipitation of RE3+ ions occur as observed from the speciation of REEs [38]. At pH > 8, REEs could be precipitated as insoluble hydroxides i.e. RE(OH)3(s) [38,39]. The obtained results revealed that the adsorption capacity of CA@Fe3O4 NPs and Cys@Fe3O4 NPs for La3+, Nd3+, Gd3+ and Y3+ ions increased with increasing the pH, so, pH = 7.0 was selected for subsequent studies.

3.2. Adsorption kinetics

3.4. Temperature dependence

The kinetic data for the adsorption of La3+, Nd3+, Gd3+ and Y3+ ions on CA@Fe3O4 NPs and Cys@Fe3O4 NPs are shown in Fig. 4a and b, respectively. It is clear that the adsorption of La3+, Nd3+, Gd3+ and Y3+ ions increase rapidly in the first 5 and 15 min for both the Cys@Fe3O4 NPs and CA@Fe3O4 NPs. and adsorption equilibrium was achieved within 15 and 30 min. The rapid adsorption of La3+, Nd3+, Gd3+ and Y3+ ions is attributed to the abundance of surface sites on the Fe3O4 surface-functionalized with Cys and CA [36]. The adsorption kinetics is an important aspect that can provide useful information of the adsorption mechanisms. Experimentally

The temperature dependence of La3+, Nd3+, Gd3+ and Y3+ adsorption on the surface of CA@Fe3O4 NPs and Cys@Fe3O4 NPs were studied at 278–318 K. From the data represented in Fig. S2 (ESIy), the adsorption capacity of CA@Fe3O4 NPs and Cys@Fe3O4 NPs for the adsorption of La3+, Nd3+, Gd3+ and Y3+ ions increased upon increasing the temperature due to the higher energy of the system facilitating the adsorption process, as well as due to the higher affinity of active binding sites on the surface of the coated Fe3O4 NPs [40].

determined kinetic data were analyzed using nonlinear pseudo first-order and pseudo second-order models. The obtained fit of the kinetic parameters of CA@Fe3O4 NPs and Cys@Fe3O4 NPs are shown in Table 1. The pseudo-second-order model fitted well with all experimental data (La3+: R2 = 0.999, Nd3+: R2 = 0.996, Gd3+: R2 = 0.995 and Y3+: R2 = 0.991 for CA@Fe3O4 NPs and La3+: R2 = 0.993, Nd3+: R2 = 0.995, Gd3+: R2 = 0.987 and Y3+: R2 = 0.998 for Cys@Fe3O4 NPs), and also the obtained qe,cal values from the pseudo-second-order model were found to agree well to the experimentally obtained qe,exp. The fitting results thus indicate that the adsorption rate was mainly determined by the chemical adsorption process through sharing or exchange of electrons between La3+, Nd3+, Gd3+ and Y3+ ions on the CA@Fe3O4 NPs and Cys@Fe3O4 NPs [37]. 3.3. Effect of pH

3.5. Thermodynamics parameters The standard enthalpy change (DH°), entropy change (DS°), and Gibb‘s energy change (DG°) were evaluated by applying Eqs. (7) and (8) [41].

lnKd ¼ 

DH  DS þ RT R

ð7Þ

DG ¼ DH  T DS

ð8Þ 1

1

where, R is the gas constant (8.314 J mol K ), T is the absolute temperature (Kelvin) and the distribution coefficients (Kd, mL g1) were defined by using Eq. (9) [42].

Kd ðRE3þ Þ ¼

Fig. 3. Zeta potential of CA@Fe3O4 NPs and Cys@Fe3O4 NPs at different pH.

ðCi  Cf Þ  V=m Cf

ð9Þ

where, Kd (RE3+) is the distribution coefficient, Ci = initial concentration, Cf = final concentration, V = volume of solution, and m = mass of CA@Fe3O4 NPs or Cys@Fe3O4 NPs, which is defined as the ratio of the concentration of rare earth ions in the solid phase to that in liquid phase.

291

R.M. Ashour et al. / Chemical Engineering Journal 327 (2017) 286–296

Fig. 4. Typical pseudo-second order fitting of the adsorption of La3+, Nd3+, Gd3+ and Y3+ ions using (a) CA@Fe3O4 NPs and (b) Cys@Fe3O4 NPs. Experimental conditions: Ci = 5 mg L1, pH = 7.0, m of adsorbents = 2.5 mg and v = 10 mL at room temperature. Effect of pH on adsorption of La3+, Nd3+, Gd3+ and Y3+ ions using (c) CA@Fe3O4 NPs, 1 time = 30 min and (d) Cys@Fe3O4 NPs, time = 15 min. Experimental conditions: C3+ , m of adsorbents = 2.5 mg and v = 10 mL at room temperature. )i(RE = 5 mg L

3+ (RE)

Table 1 Kinetic parameters of La3+, Nd3+, Gd3+ and Y3+ adsorption on CA@Fe3O4 NPs and Cys@Fe3O4 NPs. La3+

Nd3+

Gd3+

Y3+

qe,cal. (mg g1) qe,exp. (mg g1) k1 (min1) R2 qe,cal (mg g1) qe,exp. (mg g1) k2 (g mg1 min1) R2

9.1 ± 0.01 10 0.15 ± 0.005 0.961 10.1 ± 0.30 10 0.02 ± 0.001 0.999

12 ± 0.02 13.3 0.09 ± 0.005 0.982 14 ± 0.24 13.3 0.007 ± 0.005 0.996

16.9 ± 0.02 18 0.064 ± 0.003 0.982 19.6 ± 0.41 18 0.003 ± 0.003 0.995

11.1 ± 0.01 12 0.1 ± 0.001 0.984 12.4 ± 0.13 12 0.01 ± 0.001 0.991

qe,cal (mg g1) qe,exp.(mg g1) k1 (min1) R2 qe,cal. (mg g1) qe,exp.(mg g1) k2 (g mg1 min1) R2

18.4 ± 0.08 18.5 2.5 ± 0.15 0.891 18.5 ± 0.04 18.5 0.52 ± 0.03 0.993

20.1 ± 0.08 20.3 3.2 ± 0.13 0.863 20.32 ± 0.01 20.3 1.03 ± 0.002 0.995

21.3 ± 0.02 21.5 3.8 ± 0.16 0.833 21.45 ± 0.015 21.5 1.8 ± 0.15 0.987

20 ± 0.04 20.1 2.3 ± 0.14 0.892 20.1 ± 0.006 20.1 0.44 ± 0.008 0.998

Model CA@Fe3O4 NPs Pseudo-first order

Pseudo-second order

Cys@Fe3O4 NPs Pseudo-first order

Pseudo-second order

The standard enthalpy change (DH°) and the entropy change for CA@Fe3O4 NPs and Cys@Fe3O4 NPs can be calculated from the slope and intercept of the plot of ln Kd vs. 1/T using the van’t Hoff plot as shown in Fig. 5a and b, respectively. The obtained values of DG° (kJ mol1), DH° (kJ mol1) and DS° (J mol1) are summarized in Table 2. The values of enthalpy change (DH°) are positive, indicating that the adsorption of La3+, Nd3+, Gd3+ and Y3+ ions are endo-thermally driven and thus increasing the temperature is advantageous for higher adsorption. The positive values of entropy change (DS°) obtained in this study indicates an increase in the degree of freedom at the solid-liquid interface [27]. It favors

complexation and stability of the adsorption of RE3+ ions in both the systems studied in this work. In addition, the negative charge of Gibb’s free energy (DG°) values in the temperature range studied, suggest that the adsorption process is spontaneous and more favorable at higher temperatures indicating that the adsorption reactions are primarily driven towards the products. 3.6. Adsorption isotherms The effect of the concentration of RE3+ (La3+, Nd3+, Gd3+ and Y3+ ions) on the adsorption efficiency were investigated under the

292

R.M. Ashour et al. / Chemical Engineering Journal 327 (2017) 286–296

Fig. 5. Linear plot of ln Kd versus 1/T for adsorption capacity of La3+, Nd3+, Gd3+ and Y3+ on (a) CA@Fe3O4 NPs, time = 30 min and (b) Cys@Fe3O4 NPs, time = 15 min. 1 Experimental condition: C3+ , pH = 7.0, m of adsorbents = 2.5 mg and v = 10 mL. )i(RE = 5.0 mg L

Table 2 Thermodynamic parameters for the adsorption of La3+, Nd3+, Gd3+ and Y3+ by CA@Fe3O4 NPs and Cys@Fe3O4 NPs at 278–318 K. Parameters

La3+

Nd3+

Gd3+

Y3+

CA@Fe3O4 NPs DH° (kJ mol1) DS° (J mol1) DG°(kJ mol1) R2

41.85 165 (–46 to –52) 0. 995

53 201 (–56 to –64) 0. 993

35 135 (–38 to–43) 0. 995

44 168 (–47 to –53) 0. 999

Cys@Fe3O4 NPs DH° (kJ mol1) DS° (J mol1) DG°(kJ mol1) R2

16.7 87 (–24 to –28) 0.996

14 89 (–25 to –28) 0. 991

15.5 92 (–25 to –29) 0.997

11 65 (–18 to –20) 0. 994

optimized conditions. The results demonstrate that the adsorbed amount of RE3+ ions (qe) on CA@Fe3O4 NPs and Cys@Fe3O4 NPs increased with increasing initial concentration of RE3+ and reached a plateau at higher concentrations due to saturation of the number of binding sites in a fixed amount of CA@Fe3O4 NPs and Cys@Fe3O4 NPs. The experimental data was interpreted by non-linear Langmuir and Freundlich sorption isotherms (Fig. 6a and b) and the corresponding parameters related to Langmuir and Freundlich models are summarized in Table 3. It was found that the equilibrium adsorption capacity was better described by the Langmuir isotherm model, indicating that the monolayer adsorption process can explain the metal adsorption. The obtained qmax values (Table 3) suggest that the adsorption affinity of both systems decreased in the order Gd3+ > Nd3+ > Y3+ > La3+ due to the competitive adsorption between La3+, Nd3+, Gd3+ and Y3+ ions on the same adsorption site on the surface of the NPs. The relative order is consistent with the chemical properties of RE3+ such as the sequence of the electronegativity and the size of the metal ions. The maximum adsorption capacity values of RE3+ ions for Cys@Fe3O4 NPs is almost double the values of CA@Fe3O4 NPs, which is attributed to the two different functional groups (amine and carboxylic groups in Cys) that leads to increased active sites on the Fe3O4 NPs surfaces. A comparison of the adsorption capacity with different functionalized nanoparticles are represented in Table 4. 3.7. Adsorption mechanism Based on the results obtained from zeta potential measurements (Fig. 3), FT-IR spectrum (Fig. S3, ESIy) and the trend of RE3+ ions adsorption at different pH values (Fig. 4c and d), we propose a schematic representation of the adsorption mechanisms as shown in scheme 1. Generally, The RE3+ ions can be estimated from the stability constants; it is clear that RE3+ ions in solution is

predominated by positively charged species such as RE3+ ions that are present in the form of RE3+, RE(OH)+, RE(OH)2 and RE(OH)3 at various pH values. At pH < 8, the main species of RE3+ ions are RE3+ and RE(OH)+ and the removal of RE3+ is mainly accomplished by surface complexation. For CA@Fe3O4 NPs, the negatively charged COO groups have strong coordinative affinity towards La3+, Nd3+, Gd3+ and Y3+ ions. These carboxylate ions capture the RE3+ ions by forming complexes and the ability of surface complexation increases with increasing pH of the solution. At higher pH, the chelating complexes of carboxylate and La3+, Nd3+, Gd3+ and Y3+ ions are expected more than at lower pH because the chelation sites occupied by H+ are released at higher pH leading to the desired chelation [47]. In case of Cys@Fe3O4 NPs, there are two different surface functional groups, carboxylate (COO) and amine (ANH2) At low pH, the Cys@Fe3O4 NPs surface has a net positive charge due to the protonation of amine (ANH+3). RE3+ ions adsorption occurred at pH < 5 in the range 5–10 mg g1. This could be attributed to the electrostatic repulsion between RE3+ and the surface of Fe3O4 NPs, which increase the conversion of ANH2 groups to ANH+3 and there were few of ANH2OH sites available on the Cys@Fe3O4 NPs surface for the adsorption of ions through surface complexation [48]. At higher pH the concentration of H+ increases as observed from the zeta potential measurements. The increase in the negative charge on the surface of Fe3O4 NPs are possibly due to the carboxylate sites on Fe3O4 NPs and the formation of NH2OH sites. Also, the increased adsorption capacity of metal ions through the electrostatic attraction mechanism between (NH2OH and RE3+) and surface complexation mechanism between (COO and RE3+) seems to be the major routes for the adsorption of RE3+ ions on the surface of Cys@Fe3O4 NPs [27]. FT-IR spectra of the RE3+ loaded (CA@Fe3O4 NPs-RE3+) is shown in Fig. S1a (ESIy). The stretching vibrational of C@O peak was longer, broad and shifted from 1582 to 1605 cm1. Additionally,

293

R.M. Ashour et al. / Chemical Engineering Journal 327 (2017) 286–296

Fig. 6. The non-linear Langmuir and Freundlich fitting of adsorption capacity of La3+, Nd3+, Gd3+ and Y3+ on (a) CA@Fe3O4 NPs, time = 30 min and (b) Cys@Fe3O4 NPs, time = 15 min. Experimental condition: pH = 7.0, m of adsorbents = 2.5 mg and v = 10 mL at room temperature. Table 3 Parameters of Langmuir and Freundlich adsorption isotherm models. La3+

Nd3+

Gd3+

Y3+

qe, (mg g1) kL (L mg1) R2 n kF (mg g1)(L g1)n R2

32.5 ± 0.71 0.7 ± 0.08 0.995 16 4.526 0.973

41 ± 0.11 0.3 ± 0.09 0.999 13.8 3.2 0.978

52 ± 0.20 0.3 ± 0.03 0.993 18 3.35 0.978

35.8 ± 0.72 0.67 ± 0.02 0.991 17.3 4.53 0.955

qe,(mg g1) kL(L mg1) R2 n kF (mg g1)(L g1)n R2

57.2 ± 0.31 1.73 ± 0.46 0.988 32.75 5.32 0.921

85.5 ± 0.39 1.812 ± 0.4 0. 987 46.73 4.0 0.979

98 ± 0.50 0.81 ± 0.11 0. 998 18 3.3 0.978

73 ± 0.34 0.47 ± 0.08 0.997 13.8 2.98 0.946

Model CA@Fe3O4 NPs Langmuir

Freundlich

Cys@Fe3O4 NPs Langmuir

Freundlich

Table 4 Reported literature review of La3+, Nd3+, Gd3+ and Y3+ ions adsorption capacity by different adsorbents. Adsorbents

Ions

Adsorption capacity (mg g1)

Magnetic nano-hydroxyapatite [40] Magnetic silica nanocomposite (P507) [43] Magnetic alginate (P507) microcapsules [44] Fe3O4 (humic acid) [45] Dibenzo- 18-crown-6 ether onto mesoporous silica monoliths [46] CA@Fe3O4 NPs (this work) Cys@Fe3O4 NPs (this work)

Nd and Sm La Nd Eu Cs

323 and 370 55.9 149.3 10.6 50.32

La, Nd, Gd, Y La, Nd, Gd, Y

32.5, 41, 52 and 35.8 57.2, 85.5, 98 and 73

the symmetric stretching of COO group shifted to 1361 cm1, indicating that the oxygen in COO and C@O are involved in chelating adsorption of RE3+. The FTIR spectrum for Cys@Fe3O4 NPs after adsorption of RE3+ ions are presented in Fig. S1b (ESIy). It is clear

that the v(C@O) band at 1586 cm1 shifted to 1628 cm1. The broad and strong bands of tertiary amine at 3055 cm1 were shifted to 3378 cm1 due to the interaction between carboxylate and amine groups in Cys and RE3+ ions [49].

294

R.M. Ashour et al. / Chemical Engineering Journal 327 (2017) 286–296

Scheme 1. Schematic representations of proposed mechanisms for adsorption of RE3+ ions by (a) CA@Fe3O4 NPs and (b) Cys@Fe3O4 NPs.

3.8. Desorption studies The recovery of adsorbed RE3+ ions from the loaded CA@Fe3O4 NPs and Cys@Fe3O4 NPs was tested using different eluting solutions. The desorption experiments were carried out in an aqueous solution of sodium hydroxide (0–0.72%), whereby it was found that the recovered amounts of La3+, Nd3+, Gd3+ and Y3+ were negligible. The low recovery of all metals may be attributed to the strong binding between CA@Fe3O4 NPs/Cys@Fe3O4 NPs with the RE3+ ions in alkaline conditions which is in conformation to the observations made from pH studies and zeta potential measurements as well as the observations from the FT-IR measurements. Different concentrations of nitric acid were tested for the recovery of RE3+ species from the loaded coated Fe3O4 NPs. as shown in Fig. 7a and b. It is clear that recovery of all metals can be accomplished with 0.5 M HNO3 in which 85% and 97% recovery can be achieved for CA@Fe3O4 NPs and Cys@Fe3O4 NPs, respectively. 3.9. Selective separation The selective separation performance of La3+, Nd3+, Gd3+ and Y3+ ions from a mixture of a base metal solution containing RE3+ and

Mg2+, Ca2+ and Ni2+ ions with initial concentration of 0.01 mmol L1 and 0.05 mmol L1 at pH 7.0 was investigated. From the obtained results (Fig. 8), CA@Fe3O4 NPs and Cys@Fe3O4 NPs adsorbed much more RE3+ ions than other base metal ions. This can be explained in terms of strong binding of La3+, Nd3+, Gd3+ and Y3+ with O and N donor atoms than with the common interfering ions. The RE3+ ions are hard Lewis acid cations and have high tendency to complex with hard Lewis base atoms [50]. Here, the CA@Fe3O4 NPs and Cys@Fe3O4 NPs containing O and N donor atoms strongly bind to La3+, Nd3+, Gd3+ and Y3+ than the other competing cations according to Pearson’s hard and soft acid and base theory [51]. The results indicate that the CA@Fe3O4 NPs and Cys@Fe3O4 NPs have higher adsorption affinity of heavy rare earth ions such as (Gd3+, Y3+) ions than light rare earth ions such as (Nd3+, La3+). Ionic strength does not significantly seem to affect adsorption process. The selectivity was reported using separation factor (SF), which compares the molar ratio of two element ions. The separation of Gd3+ ions was chosen because of the difference in the ionic radius. SF was defined as the ratio of distribution coefficients; SF(Gd/La) = Kd(Gd)/Kd(La), SF(Gd/Nd) = Kd(Gd)/Kd(Nd) and SF(Gd/Y) = Kd(Gd)/Kd(Y) (see Eq. (9) for more details).

Fig. 7. Recovery efficiency% of La3+, Nd3+, Gd3+ and Y3+ from the loaded (a) CA@Fe3O4 NPs and (b) Cys@Fe3O4 NPs using different concentration of HNO3 as an eluting solution.

295

R.M. Ashour et al. / Chemical Engineering Journal 327 (2017) 286–296

Fig. 8. Separation of La3+, Nd3+, Gd3+ and Y3+ from a mixture of (RE3+ ions and Mg2+, Ca2+ and Ni2+) of 0.01 and 0.05 mmol L1 concentrations by (a) CA@Fe3O4 NPs, time = 30 min and (b) Cys@Fe3O4 NPs time = 15 min and pH = 7.0 at room temperature.

Table 5 Selective SF of Gd3+ from the mixed Gd3+/La3+, Gd3+/Nd3+ and Gd3+/Y3+ solution using CA@Fe3O4 NPs and Cys@Fe3O4 NPs. CA@Fe3O4 NPs

0.01 M

0.05 M

Cys@Fe3O4 NPs

0.01 M

0.05 M

SF(Gd3+/La3+) SF(Gd3+/Nd3+) SF(Gd3+/Y3+)

3.7 1.4 1.8

1.98 1.04 1.66

SF(Gd3+/La3+) SF(Gd3+/Nd3+) SF(Gd3+/Y3+)

4.8 2.3 2.23

2.3 1.5 1.8

One can notice that SF increases rapidly at the beginning of the Gd3+/La3+, Gd3+/Nd3+and Gd3+/Y3+ separation series and stagnates for the heaviest lanthanides. The obtained results in Table 5, demonstrate that the separation of Gd3+ from mixed Gd3+/La3+, Gd3+/Nd3+ and Gd3+/Y3+ solution was achieved by using CA@Fe3O4 NPs and Cys@Fe3O4 NPs. The Cys@Fe3O4 NPs have a higher selectivity separation toward heavier rare earth ions than CA@Fe3O4 NPs, which can be explained by the fact that it has amine and carboxylic functional groups on the surface.

Acknowledgments Radwa M. Ashour is grateful to Erasmus Mundus program (EMWELCOME) for financial support and KTH for hosting and giving the opportunity to work in Functional Materials Laboratory. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cej.2017.06.101.

4. Conclusions In this work, different chelating functional groups with magnetic nanoparticles (CA@Fe3O4 NPs and Cys@Fe3O4 NPs) were successfully synthesized and characterized. The role of the functional groups on the adsorption of rare earth ion and selective separation of Gd3+ from aqueous solutions was investigated. The adsorption behaviour was found to be highly dependent on the surface functionality of Fe3O4 NPs. The time needed to reach maximum adsorption was attained in less than 30 min and high loading capacity was obtained (98 mg g1 and 52 mg g1 for Cys@Fe3O4 NPs and CA@Fe3O4 NPs, respectively). Maximum adsorption capacity qmax of the RE3+ was better descried by Langmuir isotherm model, and the relative order of qmax values Gd3+ > Nd3+ > Y3+ > La3+ was consistent with the sequence of electronegativity and ionic radial sizes of REEs. The adsorption process is of endothermic nature in both the systems are more favourable at higher temperatures. Adsorption kinetic studies follow a pseudo-second order model suggesting that the rate-limiting step is chemisorption. The adsorption of La3+, Nd3+, Gd3+ and Y3+ was not affected by commonly coexisting ions such as Ca2+, Mg2+ and Ni2+, which illustrates the selective adsorption of La3+, Nd3+, Gd3+ and Y3+ from diluted aqueous solutions. High desorption efficiency was obtained in which more than 90% RE3+ ions recovery was achieved in both systems under acidic conditions. The Cys@Fe3O4 NPs exhibits highly selective separation factor for pairs of Gd3+ /La3+, Gd3+/Nd3+, Gd3+/Y3+ ions more than CA@Fe3O4 NPs. These results therefore demonstrate that Cys@Fe3O4 NPs possesses high specificity for the separation of RE3+ ions from aqueous solution and could be applied in the field of recovery and separation of REEs from mineral sources.

References [1] Y. Li, B. Hu, Cloud point extraction with/without chelating agent on-line coupled with inductively coupled plasma optical emission spectrometry for the determination of trace rare earth elements in biological samples, J. Hazard. Mater. 174 (2010) 534–540. [2] S.V. Eliseeva, J.-C.G. Bunzli, Rare earths: jewels for functional materials of the future, New J. Chem. 35 (2011) 1165–1176. [3] P. Sasikumar, S.V. Narasimhan, S. Velmurugan, Development of a Modified Ion Exchange Resin Column for Removal of Gadolinium From the Moderator System of PHWRs, Sep. Sci. Technol. 48 (2013) 1220–1225. [4] W. Yongxing, W. Xiaorong, H. Zichun, Genotoxicity of lanthanum (III) and gadolinium (III) in human peripheral blood lymphocytes, Bull. Environ. Contam. Toxicol. 64 (2000) 611–616. [5] K. Binnemans, P.T. Jones, B. Blanpain, T. Van Gerven, Y.X. Yang, A. Walton, M. Buchert, Recycling of rare earths: a critical review, J Clean. Prod. 51 (2013) 1– 22. [6] N. Das, D. Das, Recovery of rare earth metals through biosorption: an overview, J. Rare Earths 31 (2013) 933–943. [7] E. Fujimori, T. Hayashi, K. Inagaki, H. Haraguchi, Determination of lanthanum and rare earth elements in bovine whole blood reference material by ICP-MS after coprecipitation preconcentration with heme-iron as coprecipitant, Fresenius J. Anal. Chem. 363 (1999) 277–282. [8] S. Tong, N. Song, Q. Jia, W. Zhou, W. Liao, Solvent extraction of rare earths from chloride medium with mixtures of 1-phenyl-3-methyl-4-benzoyl-pyrazalone5 and sec-octylphenoxyacetic acid, Sep. Purif. Technol. 69 (2009) 97–101. [9] Q. Jia, X. Kong, W. Zhou, L. Bi, Flow injection on-line preconcentration with an ion-exchange resin coupled with microwave plasma torch-atomic emission spectrometry for the determination of trace rare earth elements, Microchem. J. 89 (2008) 82–87. [10] X. Sun, B. Peng, Y. Ji, J. Chen, D. Li, The solid–liquid extraction of yttrium from rare earths by solvent (ionic liquid) impreganated resin coupled with complexing method, Sep. Purif. Technol. 63 (2008) 61–68. [11] P. Liang, Y. Qin, B. Hu, T. Peng, Z. Jiang, Nanometer-size titanium dioxide microcolumn on-line preconcentration of trace metals and their determination by inductively coupled plasma atomic emission spectrometry in water, Anal. Chim. Acta 440 (2001) 207–213.

296

R.M. Ashour et al. / Chemical Engineering Journal 327 (2017) 286–296

[12] S. Radhika, V. Nagaraju, B. Nagaphani Kumar, M.L. Kantam, B.R. Reddy, Solidliquid extraction of Gd(III) and separation possibilities of rare earths from phosphoric acid solutions using Tulsion CH-93 and Tulsion CH-90 resins, J. Rare Earths 30 (2012) 1270–1275. [13] P. Liang, J. Cao, R. Liu, Y. Liu, Determination of trace rare earth elements by inductively coupled plasma optical emission spectrometry after preconcentration with immobilized nanometer titanium dioxide, Microchim. Acta 159 (2007) 35–40. [14] S. Tong, S. Zhao, W. Zhou, R. Li, Q. Jia, Modification of multi-walled carbon nanotubes with tannic acid for the adsorption of La, Tb and Lu ions, Microchim. Acta 174 (2011) 257–264. [15] Y. Yao, S. Miao, S. Liu, L.P. Ma, H. Sun, S. Wang, Synthesis, characterization, and adsorption properties of magnetic Fe3O4@graphene nanocomposite, Chem. Eng. J. 184 (2012) 326–332. [16] V. Diniz, B. Volesky, Biosorption of La Eu and Yb using Sargassum biomass, Water Res. 39 (2005) 239–247. [17] R.M. Ashour, H.N. Abdelhamid, A.F. Abdel-Magied, A.A. Abdel-Khalek, M.M. Ali, A. Uheida, M. Muhammed, X. Zou, J. Dutta, Rare Earth Ions Adsorption onto Graphene Oxide Nanosheets, Solvent Extr. Ion Exch. (2017) 1–13. [18] R. Hao, R. Xing, Z. Xu, Y. Hou, S. Gao, S. Sun, Synthesis, functionalization, and biomedical applications of multifunctional magnetic nanoparticles, Adv. Mater. 22 (2010) 2729–2742. [19] B. Gao, Y. Gao, Y. Li, Preparation and chelation adsorption property of composite chelating material poly(amidoxime)/SiO2 towards heavy metal ions, Chem. Eng. J. 158 (2010) 542–549. [20] L. Li, K.Y. Mak, C.W. Leung, K.Y. Chan, W.K. Chan, W. Zhong, P.W.T. Pong, Effect of synthesis conditions on the properties of citric-acid coated iron oxide nanoparticles, Microelectron. Eng. 110 (2013) 329–334. [21] Y. Su, A.S. Adeleye, A.A. Keller, Y. Huang, C. Dai, X. Zhou, Y. Zhang, Magnetic sulfide-modified nanoscale zerovalent iron (S-nZVI) for dissolved metal ion removal, Water Res. 74 (2015) 47–57. [22] E.-J. Kim, C.-S. Lee, Y.-Y. Chang, Y.-S. Chang, Hierarchically structured manganese oxide-coated magnetic nanocomposites for the efficient removal of heavy metal ions from aqueous systems, ACS Appl. Mater. Interfaces 5 (2013) 9628–9634. [23] S. Shin, J. Jang, Thiol containing polymer encapsulated magnetic nanoparticles as reusable and efficiently separable adsorbent for heavy metal ions, Chem. Commun. 4230–4232 (2007). [24] D. Dupont, W. Brullot, M. Bloemen, T. Verbiest, K. Binnemans, Selective uptake of rare earths from aqueous solutions by edta-functionalized magnetic and nonmagnetic nanoparticles, ACS Appl. Mater. Interfaces 6 (2014) 4980–4988. [25] H. Zhang, R.G. McDowell, L.R. Martin, Y. Qiang, Selective extraction of heavy and light lanthanides from aqueous solution by advanced magnetic nanosorbents, ACS Appl. Mater. Interfaces 8 (2016) 9523–9531. [26] X. Shen, Q. Wang, W. Chen, Y. Pang, One-step synthesis of water-dispersible cysteine functionalized magnetic Fe3O4 nanoparticles for mercury(II) removal from aqueous solutions, Appl. Surf. Sci. 317 (2014) 1028–1034. [27] S. Singh, K.C. Barick, D. Bahadur, Surface engineered magnetic nanoparticles for removal of toxic metal ions and bacterial pathogens, J. Hazard. Mater. 192 (2011) 1539–1547. [28] J. Sangeetha, J. Philip, Synthesis, characterization and antimicrobial property of Fe3O4-Cys-HNQ nanocomplex, with l-cysteine molecule as a linker, RSC Adv. 3 (2013) 8047–8057. [29] S. Nigam, K.C. Barick, D. Bahadur, Development of citrate-stabilized Fe3O4 nanoparticles: conjugation and release of doxorubicin for therapeutic applications, J. Magn. Magn. Mater. 323 (2011) 237–243. [30] G. Salazar-Alvarez, J. Sort, A. Uheida, M. Muhammed, S. Surinach, M.D. Baro, J. Nogues, Reversible post-synthesis tuning of the superparamagnetic blocking temperature of [gamma]-Fe2O3 nanoparticles by adsorption and desorption of Co(ii) ions, J. Mater. Chem. 17 (2007) 322–328. [31] C. Kotsmar, K.Y. Yoon, H. Yu, S.Y. Ryoo, J. Barth, S. Shao, M. Prodanovic´, T.E. Milner, S.L. Bryant, C. Huh, K.P. Johnston, Stable citrate-coated iron oxide

[32] [33]

[34]

[35]

[36]

[37]

[38]

[39]

[40] [41] [42]

[43]

[44]

[45]

[46]

[47]

[48] [49]

[50]

[51]

superparamagnetic nanoclusters at high salinity, Ind. Eng. Chem. Res. 49 (2010) 12435–12443. M. Kosmulski, Chemical Properties of Material Surfaces, CRC press, 2001. R.A. Bini, R.F.C. Marques, F.J. Santos, J.A. Chaker, M. Jafelicci Jr., Synthesis and functionalization of magnetite nanoparticles with different amino-functional alkoxysilanes, J. Magn. Magn. Mater. 324 (2012) 534–539. Z. Durmus, H. Kavas, M.S. Toprak, A. Baykal, T.G. Altınçekiç, A. Aslan, A. Bozkurt, S. Cosßgun, l-lysine coated iron oxide nanoparticles: synthesis, structural and conductivity characterization, J. Alloys Compd. 484 (2009) 371–376. A. Mocanu, I. Cernica, G. Tomoaia, L.-D. Bobos, O. Horovitz, M. Tomoaia-Cotisel, Self-assembly characteristics of gold nanoparticles in the presence of cysteine, Colloids Surfaces A 338 (2009) 93–101. M.R. Awual, T. Kobayashi, Y. Miyazaki, R. Motokawa, H. Shiwaku, S. Suzuki, Y. Okamoto, T. Yaita, Selective lanthanide sorption and mechanism using novel hybrid Lewis base (N-methyl-N-phenyl-1,10-phenanthroline-2-carboxamide) ligand modified adsorbent, J. Hazard. Mater. 252–253 (2013) 313–320. S. Yusan, C. Gok, S. Erenturk, S. Aytas, Adsorptive removal of thorium (IV) using calcined and flux calcined diatomite from Turkey: Evaluation of equilibrium, kinetic and thermodynamic data, Appl. Clay Sci. 67–68 (2012) 106–116. E. Kim, K. Osseo-Asare, Aqueous stability of thorium and rare earth metals in monazite hydrometallurgy: Eh–pH diagrams for the systems Th–, Ce–, La–, Nd– (PO4)–(SO4)–H2O at 25 °C, Hydrometallurgy 113–114 (2012) 67–78. M. Vaca, R. Mier López Callejas, R. Gehr, B.E. Jiménez Cisneros, P.J.J. Alvarez, Heavy metal removal with mexican clinoptilolite: multi-component ionic exchange, Water Res. 35 (2001) 373–378. C. Gok, Neodymium and samarium recovery by magnetic nanohydroxyapatite, J. Radioanal. Nucl. Chem. 301 (2014) 641–651. C.-H. Weng, C.P. Huang, Adsorption characteristics of Zn(II) from dilute aqueous solution by fly ash, Colloids Surfaces A 247 (2004) 137–143. R. Rahal, F. Annani, S. Pellet-Rostaing, G. Arrachart, S. Daniele, Surface modification of titanium oxide nanoparticles with chelating molecules: New recognition devices for controlling the selectivity towards lanthanides ionic separation, Sep. Purif. Technol. 147 (2015) 220–226. D. Wu, Y. Sun, Q. Wang, Adsorption of lanthanum (III) from aqueous solution using 2-ethylhexyl phosphonic acid mono-2-ethylhexyl ester-grafted magnetic silica nanocomposites, J. Hazard. Mater. 260 (2013) 409–419. L. Zhang, D. Wu, B. Zhu, Y. Yang, L. Wang, Adsorption and selective separation of neodymium with magnetic alginate microcapsules containing the extractant 2-ethylhexyl phosphonic acid mono-2-ethylhexyl ester, J. Chem. Eng. Data 56 (2011) 2280–2289. S. Yang, P. Zong, X. Ren, Q. Wang, X. Wang, Rapid and highly efficient preconcentration of Eu(III) by core-shell structured Fe3O4@Humic acid magnetic nanoparticles, ACS Appl. Mater. Interfaces 4 (2012) 6891–6900. M.R. Awual, S. Suzuki, T. Taguchi, H. Shiwaku, Y. Okamoto, T. Yaita, Radioactive cesium removal from nuclear wastewater by novel inorganic and conjugate adsorbents, Chem. Eng. J. 242 (2014) 127–135. W. Brullot, N.K. Reddy, J. Wouters, V.K. Valev, B. Goderis, J. Vermant, T. Verbiest, Versatile ferrofluids based on polyethylene glycol coated iron oxide nanoparticles, J. Magn. Magn. Mater. 324 (2012) 1919–1925. S.S. Banerjee, D.-H. Chen, Fast removal of copper ions by gum arabic modified magnetic nano-adsorbent, J. Hazard. Mater. 147 (2007) 792–799. F. Zhao, E. Repo, M. Sillanpää, Y. Meng, D. Yin, W.Z. Tang, Green synthesis of magnetic EDTA- and/or DTPA-cross-linked chitosan adsorbents for highly efficient removal of metals, Ind. Eng. Chem. Res. 54 (2015) 1271–1281. Y. Hasegawa, S. Tamaki, H. Yajima, B. Hashimoto, T. Yaita, Selective separation of samarium (III) by synergistic extraction with b-diketone and methylphenylphenanthroline carboxamide, Talanta 85 (2011) 1543–1548. R.G. Pearson, Hard and soft acids and bases, HSAB, part 1: fundamental principles, J Chem Educ 45 (1968) 581.