Benzene triamido-tetraphosphonic acid immobilized on mesoporous silica for adsorption of Nd3+ ions in aqueous solution

Microporous and Mesoporous Materials 258 (2018) 62e71

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso

Benzene triamido-tetraphosphonic acid immobilized on mesoporous silica for adsorption of Nd3þ ions in aqueous solution Seenu Ravi a, Yu-Ri Lee a, Kwangsun Yu a, b, Ji-Whan Ahn b, Wha-Seung Ahn a, * a b

Department of Chemistry and Chemical Engineering, Inha University, Incheon, South Korea Korea Institute of Geoscience and Mineral Resources (KIGAM), 124 Gwahang-no, Yuseong-gu, Daejeon, 305-350, South Korea

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 July 2017 Received in revised form 8 September 2017 Accepted 10 September 2017 Available online 12 September 2017

Mesoporous silica SBA-15 grafted with benzene-1,3,5-triamido-tetraphosphonic acid (SBA-15-BTATPA) was synthesized. After detailed characterization of the material using various analytical instrumentation, it was applied for the recovery of Nd3þ as a function of the contact time, ion concentration, and pH. The adsorption equilibria and kinetics were also examined. Under the optimal conditions, adsorption equilibrium was reached within 60 min of contact time, and the adsorption capacity of Nd3þ at room temperature and pH 6 was 129.8 mg g1, which was higher than that by other rare earth element (REE) ions tested for comparison (Y3þ, La3þ, and Ce3þ). The adsorbent showed excellent distribution coefficient for Nd3þ greater than 1.0  105 mL/g. The preferential order of adsorption in the ion mixture with equal ion concentrations was Y3þ>Nd3þ>Ce3þ>La3þ, which followed the sequence of their decreasing atomic size and increasing stability constant. The competing adsorption by transition metal ions, such as Cu2þ, Ni2þ, Co2þ, and Zn2þ, on the REEs recovery was negligible (<2%). The observed adsorption isotherms could be fitted well to a Langmuir model (correlation factor R2 > 0.99), whereas the adsorption kinetics could be described satisfactorily by a pseudo second order kinetic model. The material was reusable for up to 5 consecutive times with only a slight decrease in adsorption capacity. © 2017 Elsevier Inc. All rights reserved.

Keywords: Rare earth elements Neodymium Mesoporous silica Phosphonic acid Adsorption

1. Introduction The increasing use of rare earth elements (REEs) produces aqueous solutions contaminated with rare earth metal ions that become an environmental burden. The broad and rapidly growing applications of REEs in the energy industry and their limited deposits on earth have also highlighted the necessity of separating and recycling REEs from various sources including industrial and urban waste [1,2]. Therefore, effective recovery of REEs from water is desirable for both the economy and environment. The recovery of REEs have been conducted through various processes including solvent extraction [3,4], adsorption [5], biosorption [6], coprecipitation [7], hydrometallurgy [8], and ion exchange [9]. As adsorption processes employing an adsorbent with specific organic functionality have achieved selective metal separation on the small scale, a large number of studies have been reported recently for the adsorption of REEs in aqueous media using porous silica [10e14], carbons [15], metal oxides [16,17], metal

* Corresponding author. E-mail address: [email protected] (W.-S. Ahn). http://dx.doi.org/10.1016/j.micromeso.2017.09.006 1387-1811/© 2017 Elsevier Inc. All rights reserved.

organic frameworks [18,19], polymers [20], and biomaterials [6,21]. However, most of them showed poor adsorption capacity and/or reusability. The weak physicochemical stability of metal oxides and metal organic frameworks makes them less desirable for long term use in the separation of REEs. Carbon materials and resins, despite their stability, are difficult to functionalize with specific organic groups for the selective separation of REEs. Biomaterials show low chemical stability and poor mechanical properties. In contrast, porous silica materials tend to be more stable, eco-friendly, and possess a high population of surface hydroxyl groups that enable easy surface functionalization, and are widely used [22e27]. The high surface areas with uniform pore channel networks in mesoporous silica materials have great potential for REEs separation, in particular, for Nd3þ ions [28e31]. Roosen et al. reported a chitosansupported hybrid silica material functionalized with ethylenediaminetetraacetic acid (EDTA)- and diethylenetriaminepentaacetic acid (DTPA), which exhibited a Nd3þ adsorption capacity of ca. 60 and 106 mg/g, respectively [28]. Melnyk et al. reported SiPgrafted silica adsorbent for REEs separation that exhibited 45 and 46 mg/g adsorption of Nd3þ and Dy3þ, respectively [29]. Legaria et al. functionalized iminodiacetic acid over Fe3O4/SiO2 core shell

S. Ravi et al. / Microporous and Mesoporous Materials 258 (2018) 62e71

spherical particles that exhibited 27, 33 and 40 mg/g for La3þ, Nd3þ and Dy3þ ions, respectively [30]. These materials, despite their sufficient stability, showed rather low adsorption capacities for REE ions due most likely to the weakly binding functional groups on the silica surface. Thus, porous silica materials functionalized with stronger active sites, such as amide-based phosphonic acids, towards trivalent metal ions adsorption appear promising for effective adsorption of REE ions. Herein, a benzene-1,3,5-triamido-tetraphosphonic acid (BTATPA) grafted on an acid-stable SBA-15 mesoporous silica (SBA15-BTATPA) was proposed as a new adsorbent for the adsorption of Nd3þ ions. Adsorption of other REE ions such as La3þ, Ce3þ, and Y3þ were also examined for comparison. A systematic investigation on the adsorption equilibria and kinetics was made by examining the effect of the contact time, metal ion concentration, and pH of the system. In addition, REEs adsorption was carried out in the presence of potentially competing transition metal ions, and the recyclability of the adsorbent was examined.

2. Experimental

63

2.2.2. Preparation of amidotricarbonyl benzene-functionalized SBA15 (SBA-15-NHTCB) The prepared SBA-15-NH2 (1 g) was treated with 1,3,5benzenetricarbonyl trichloride (0.6 mmol) and triethylamine (0.06 mmol) in 100 mL of DMF. The solution was heated under reflux for 12 h, after which the material was filtered and washed with DMF and hot ethanol. The filtered material was vacuum-dried at 55  C for 6 h. The resulting yellow colored material was designated as SBA-15-NHTCB (Fig. 1).

2.2.3. Preparation of benzene-1,3,5-triamido-tetraphosphonic acid (BTATPA)-incorporated SBA-15 (SBA-15-BTATPA) The SBA-15-NHTCB (1 g) was treated with IDMP (1.2 mmol) in 100 mL DMF solvent, where 0.1 mmol of CDI was used to promote amide bond formation. The reaction mixture was heated refluxed for 20 h and filtered immediately under hot conditions, and washed with an ethanol and water mixture (1:1) to remove the unwanted residues from the material. The filtered material was dried at 55  C for 6 h in a vacuum oven. The obtained material was called SBA-15BTATPA (Fig. 1). The material was characterized and used as an adsorbent for the removal of REE ions.

2.1. Materials 2.3. Characterization Tetraethylorthosilicate (TEOS), poly(ethylene glycol)-blockpoly(propylene glycol)-block-poly(ethylene glycol) (Pluronic P123-EO/20:PO/70:EO/20), HCl, ethanol, N, N0 edimethylformamide (DMF), 1,3,5-benzenetricarbonyl trichloride, 3-aminopropyltriethoxysilane (APTES), iminodi(methylphosphonic acid) (IDMP), trimethylamine, and carbonyldiimidazole (CDI) were purchased from Sigma Aldrich. Aqueous rare-earth stock solutions for adsorption experiments were made from their corresponding REE salts such as Y(NO3)3.6H2O (99.9%), La(NO3)3.6H2O (99.9%), Ce(NO3)3.6H2O (99.9%), and Nd(NO3)3.6H2O (99.9%) supplied by Sigma-Aldrich. Transition metal nitrates Co(NO3)2.6H2O, Ni(NO3)2.6H2O, Cu(NO3)2.6H2O, and Zn(NO3)2.6H2O, were also supplied by Sigma Aldrich. These chemicals were used as received.

2.2. Material synthesis 2.2.1. Preparation of SBA-15 and APTES-functionalized SBA-15 (SBA-15-NH2) The synthesis procedures of SBA-15 and APTES-functionalized SBA-15 mesoporous silica were reported elsewhere [32,33]. In a typical synthesis, Pluronic P123 (0.017 mol) was added in deionized water (193 mol) at 35  C and stirred until fully dissolved, followed by addition of HCl (5.9 mol) solution. Once a clear solution was obtained, TEOS (1-x mole) was added to the reaction mixture and stirred for further 3 h. APTES (x mole, x ¼ 0 and 0.1) was then added dropwise to the gel solution, and stirred for 24 h at 35  C. The resulting gel solution was transferred to a sealed Teflon-lined stainless steel autoclave and heated to 100  C for 24e36 h. The products were filtered, washed with ethanol and water and dried at 60  C for 6 h in a vacuum oven. The as-synthesized SBA-15 was calcined at 550  C for 6 h with a heating rate of 1  C/min under air atmosphere. Besides, the APTES functionalized material i.e, SBA-15NH2 was solvent-extracted using an ethanol and HCl (36 wt %) (99:1) mixture, where 1 g of the as-synthesized material was dispersed in 300 mL of ethanol/HCl and heated under reflux for 12 h. The SBA-15-NH2 was filtered and washed three to four times with hot ethanol, and finally dried at 55  C for 4 h in a vacuum oven.

X-ray diffraction (XRD, Rigaku) patterns of the mesoporous silica products were recorded using CuKa (l ¼ 1.54 Å) radiation. The N2 isotherms were measured using a BELsorpMax (BEL, Japan) at 77 K. The samples were degassed at 373 K for 12 h under high vacuum before the isotherms were measured. The surface areas were estimated using the BrunauerEmmettTeller (BET) method over the relative pressure (P/P0) range of 0.0e0.2 The pore size distribution was calculated from the analysis of the isotherm desorption branch by the BJH (BarretteJoynereHalenda) method coupled with the apparatus software. The Fourier transform infrared (FT-IR, VERTEX 80 V FT-IR, Bruker, Germany) spectra were obtained at room temperature. KBr was used for preparing the pellet. The carbon, hydrogen, and nitrogen contents of the samples were measured using an elemental analyzer (EA, EA1112). The phosphorous content was measured by inductively coupled plasma optical emission spectroscopy (ICP-OES, Optima 7300DV) and Energy dispersive X-ray spectroscopy (EDX). X-ray photoelectron spectroscopy (XPS) was performed using a monochromatic Al Ka Xray source (Thermo Scientific, USA) and a hemispherical analyzer. The morphologies were examined by field emission-scanning electron microscopy (FE-SEM, Hitachi S-4300).

2.4. Adsorption experiments Batch adsorption experiments of REE ions over SBA-15-BTATPA were carried out and the adsorption under different conditions of the solution pH, contact time, and solution concentrations were investigated. The Nd3þ ion was used as model system for all rare earth ions (La3þ, Ce3þ, and Y3þ) in the optimization tests. The stock solutions of Nd3þ at specified concentrations were prepared. The pH of the solutions in the range of 3e9 was adjusted using either NaOH (0.01 M) or HNO3 (0.01 M). All the adsorption tests were performed in a 10 mL aliquot of a suitably diluted stock solution. The adsorbent (10 mg) was added to the adsorbate vials and the solutions were equilibrated in a thermostatic water-bath shaker operated at 25 C and 200 rpm for a preset time period of 6 h. The amount of Nd3þ ions adsorbed onto SBA-15-BTATPA was calculated using equation (1).

64

S. Ravi et al. / Microporous and Mesoporous Materials 258 (2018) 62e71

Fig. 1. Synthesis steps of the SBA-15-BTATPA.

2.5. Adsorption kinetics

qe ¼ ðCi  Ce Þ  V=m

(1)

where, qe is the amount of adsorbed Nd3þ ions at equilibrium (mg/ g), Ci is the initial Nd3þ ion concentration in an aqueous solution (mg/L), Ce is the equilibrium Nd3þ ion concentration in an aqueous solution (mg/L), V is the volume of the solution (L) and m is the mass of the adsorbent (g). The adsorption efficiency of SBA-15-BTATPA was also determined from the distribution coefficient (Kd). The Kd value of adsorbent can be calculated using following equation (2),

Kd ¼





.  Cf  ðV=mÞ Ci  Cf

(2)

where, Ci is the initial Nd3þ ion concentration in aqueous solution (mg/L), Cf is the final Nd3þ ion concentration in aqueous solution (mg/L), V is the volume of the solution (mL), and m is the mass of the adsorbent (g). To determine Kd, the adsorbent (10 mg) was added to the 10 mg/L of Nd3þ solution, was shaken in a thermostatic water-bath operated at 25  C and 200 rpm for a preset time period of 2 h. Similar adsorption experimental protocols were followed for other metal ions including La3þ, Ce3þ and Y3þ. All the aliquots after filtration were measured by ICP-OES for metal ion detection.

The kinetics of REEs adsorption was examined using 10 mL of 50 mg/L metal ion solutions with 10 mg of adsorbents. The solutions were filtered at 5, 10, 15, 30, 60, 180, and 360 min of time intervals to measure the liquid phase concentrations. The adsorption data were analyzed using pseudo first order (PFO) and pseudo second order (PSO) models: The first order rate expression is given as [34],

logðqe  qÞ ¼ log qe  ðk1 =2:303Þt

(3)

where qe and q are the amount of adsorbate adsorbed on adsorbent (mg/g) at equilibrium and time t, respectively and k1 is the first order adsorption rate constant (min1). The slopes and intercepts of the log (qe-q) vs t plots were used to determine the first order rate constant k1. Similarly, the pseudo second order kinetic model is expressed as [34],

 . t=q ¼ ð1 k2 q2e þ ðt=qe Þ

(4)

where k2 (g/mg.min) is the rate constant of the second order adsorption and qe is the amount of adsorbate adsorbed at equilibrium (mg/g).

S. Ravi et al. / Microporous and Mesoporous Materials 258 (2018) 62e71

2.6. Desorption studies Desorption studies were carried out using SBA-15-BTATPA after the completed adsorption of REE ions. A 500 mg sample of the preadsorbed SBA-15-BTATPA was subjected to desorption in 50 mL of a 0.5 M HCl solution for 10 min at room temperature. The slurry was filtered off to recover the SBA-15-BTATPA and the aqueous phase was analyzed for the metal ion concentration. The percentage desorption (D) was calculated using the following equation,

Dð%Þ ¼ ðCe =C0 Þ  100

(5)

where, Ce is the aqueous metal concentration after desorption and Co is the initial metal ion concentration in the SBA-15-BTATPA phase. After the desorption experiment, the filtered SBA-15BTATPA was recovered and reused after drying at 50  C for 4 h. 3. Results and discussion 3.1. Characterization of the developed adsorbent Fig. 2 present the XRD patterns of SBA-15, SBA-15-NH2, SBA-15NHTCB, and SBA-15-BTATPA. The three peaks of the XRD pattern for SBA-15 could be indexed to the (100), (110) and (200) reflections at 0.8, 1.6 and 1.8 2q values, respectively, which are in line with the highly ordered hexagonally close-packed silica material with p6mm symmetry [33]. The gradual decrease in the peak intensities of the lattice patterns (100) of SBA-15-NH2, SBA-15eNHTCB and SBA-15-BTATPA were explained by functionalization of the respective organic groups onto the SBA-15 pore surface [32]. Fig. 3a presents the N2 adsorption-desorption isotherms of the corresponding samples, exhibiting type IV isotherms with a H1type hysteresis loop characteristic of mesoporous materials [35]. Fig. 3a showed sharp inflections at relative pressures (P/P0) between 0.4 and 0.9, which suggest the uniform mesopores of the pristine SBA-15 and functionalized SBA-15. The area of hysteresis loop gradually decreased upon successful functionalization of the organic ligands. Table 1 lists the textural parameters derived from N2 adsorption-desorption analysis. The BET surface area of SBA-15, SBA-15-NH2, SBA-15-NHTCB, and SBA-15-BTATPA were 857, 511, 394, and 276 m2/g, respectively. As shown in Table 1, the surface area and pore volumes decreased gradually from SBA-15 to SBA-15BTATPA as the stepwise functionalization steps were carried out.

Fig. 2. XRD patterns of SBA-15, SBA-15-NH2, SBA-15-NHTCB, and SBA-15-BTATPA.

65

The average pore diameters (Fig. 3b) were also reduced with increasing chain length of the surface grafted organic groups. Fig. 4 shows the FT-IR spectra of the SBA-15-BTATPA and others. The intense bands at 802 and 1080 cm1 were assigned to Si-O-Si symmetric and asymmetric stretching modes, and the peak at 958 cm1 was due to the stretching vibration of the SieOH group [36]. APTES functionalization was accompanied by characteristic vibration at 2932 cm1, which could be assigned to the -CH stretching frequency from the CH2 groups of the propyl chain [37]. The reduced intensity of the broad peak in SBA-15-NH2 at 3457 cm1, which was ascribed for hydroxyl group stretching frequency, also signaled the successful functionalization of APTES. The FT-IR spectra of SBA-15-NHTCB exhibited peaks at 1665 cm1 corresponding to the amide carbonyl (-C¼O) stretching frequency, a small peak at 1720 cm1 corresponded to the acid chloride carbonyl stretching frequency, and a peak at 1549 cm1 is ascribed for C¼C stretching frequency from benzene moieties indicating the successful functionalization of tricarbonylbenzene trichloride. After functionalization with IDMP, the shifted peak with increased intensity at 1665 cm1 from the one at 1660 cm1 of pure IDMP and a small hump at 2318 cm1 in SBA-15-BTATPA were found corresponding to the -P¼O stretching frequencies. The presence of several different binding modes for residual P¼O and PeOeH sites in IDMP appears between 1300 cm1 and 900 cm1 in the IR region [38]. In agreement with this, a broad spectrum in the range of 950e1300 cm1 with increased intensity was detected after functionalization with IDMP. The broad peak at 1241 cm1 corresponded to the P-O and/or C-N stretching vibrations, which is slightly shifted from original C-N peak (1255 cm1) of IDMP. In addition, a peak at 571 cm1 could be assigned to C-P stretching frequency, which also was shifted from the IDMP peak (590 cm1). These IR assignments are in good agreement with the previously reported related materials [36e41], and indicated the successful functionalization of BTATPA over SBA-15. XPS was employed to determine the chemical state and presence of nitrogen, carbon and phosphorous in SBA-15-BTATPA. As shown in Fig. 5, APTES-functionalized SBA-15 showed the binding energy peak at 399.6 eV, which is related to the presence of primary amine group in SBA-15-NH2 (Fig. S1). The deconvolution peaks of the N 1s XPS binding energies of 399.6, 400.5, and 401.3 eV (Fig. 5a) in SBA-15-BTATPA are consistent with the presence of primary, secondary, and tertiary amine groups in BTATPA, respectively [42,43]. Similarly, Fig. 5b presents the XPS deconvolution peaks of the C(1s) core level of SBA-15-BTATPA, which showed an intense peak at ~285.2 eV, indicating the presence of a C¼C bond (sp2), and small peak for a C¼O bond at 287.5 eV [44]. The presence of phosphorous in SBA-15-BTATPA was confirmed, as shown in Fig. 5c with peaks at 132.5 eV and 133.5 eV corresponding to the 2p1/2 and 2p3/2 state regions [45]. The nitrogen, carbonyl, and phosphorous binding energies and their chemical states supported the successful functionalization of BTATPA on SBA-15. In addition, the amount of nitrogen was estimated using EA and the amount of phosphorous was determined by ICP-OES. Table 1 lists the corresponding elemental compositions of N and P in SBA-15-BTATPA. Fig. 6a, b and c show FE-SEM images and EDX profiles of all three functionalized SBA-15 samples. The slight decrease in N from 1.9 to 1.6 wt % in Fig. 6a1 and b1 could be due to the increased carbon content after the functionalization of benzenetricarbonyl chloride into SBA-15-NH2. The increase in N from 1.6 to 1.8 wt % in SBA-15BTATPA (Fig. 6b1 and c1) is due to the functionalization of IDMP. The particle sizes were in the range, ca. 7 mm, and the EDX quantified the P content of 3.1 wt % in SBA-15-BTATPA. In addition, the TEM image of SBA-15-BTATPA (Fig. 7) indicated that the material is in highly ordered hexagonal structure with well-defined pore channels.

66

S. Ravi et al. / Microporous and Mesoporous Materials 258 (2018) 62e71

Fig. 3. (a) N2 adsorption-desorption isotherms and (b) pore size distribution of SBA-15, SBA-15-NH2, SBA-15-NHTCB, and SBA-15-BTATPA.

Table 1 Physicochemical properties of the functionalized SBA-15 silica material. Materials

SBET (m2/g)

Pore volume (cm3/g)

Average pore diameter (nm)

Na (mmol/g)

Pb (mmol/g)

SBA-15 SBA-15-NH2 SBA-15-NHTCB SBA-15-BTATPA

857 521 394 276

1.18 0.48 0.40 0.34

6.9 4.6 4.1 3.9

e 1.18 1.06 1.23

e e e 1.04

a b

The amount of nitrogen was estimated by EA. The amount of phosphorous was estimated by ICP-OES.

varied from 1 to 9. The hydroxylation of REE ions reduces their solubility in alkaline pH medium. For example, the solubility product (Ksp) of Nd(OH)3 is ca. 1023 and precipitation can occur at pH > 7.3 [28]. In addition, the Si-O-Si bonds in SBA-15-BTATPA can rupture in basic medium. Therefore adsorption on alkaline pH is not advisable. From the figure, the pH value of 6 was selected as the optimal to obtain the maximized adsorption and it is used for the further adsorption experiment.

Fig. 4. FT-IR spectra of SBA-15, SBA-15-NH2, SBA-15-NHTCB, SBA-15-BTATPA and.

3.2. REEs adsorption studies 3.2.1. Effect of pH The pH effect on adsorption was investigated for aqueous solutions of REEs of La3þ, Ce3þ, Nd3þ, and Y3þ. Fig. 8 shows the effect of pH on the removal efficiency of SBA-15-BTATPA for these ions at 100 mg/L initial concentration. Varying pH of the solution could influence adsorption of a metal ion due to the protonation of active sites on the surface of the adsorbent. The pH of the system was

3.2.2. Adsorption equilibrium isotherms Fig. 9a presents the adsorption equilibria of REE ions at room temperature at pH 6. The amount of ions adsorbed at equilibrium increased with increasing solution concentration of REE ions, and Nd3þ and Y3þ ions were adsorbed in significantly higher amounts than those of La3þ and Ce3þ. The adsorption equilibrium values were then fitted to the widely used Langmuir and Freundlich models. This isotherm fitting not only generates more statistically meaningful adsorption capacities but also provides an understanding of the adsorbent surface properties by giving adsorbent/ adsorbate affinity constants. The Langmuir isotherm model assumes that adsorption takes place homogeneously over the surface of the adsorbent. The adsorption isotherm parameters by the Langmuir model were calculated using the following equation (6) [46],

Ce =qe ¼ ð1=bqm Þ þ ðCe =qm Þ

(6)

where Ce is the equilibrium concentration of REE ions in solution (mg/L); qe describes the amount of REE ions captured by the unit amount of the SBA-15-BTATPA at equilibrium concentration (mg/ g); qm defines the adsorption capacity (mg/g) of the SBA-15BTATPA; and b represents the affinity of the binding sites or the

S. Ravi et al. / Microporous and Mesoporous Materials 258 (2018) 62e71

67

Fig. 5. XPS of SBA-15-BTATPA (a) N 1s, (b) C 1s, and (c) P 2p.

Fig. 6. FE-SEM-EDX images of SBA-15-NH2, SBA-15-NHTCB, and SBA-15-BTATPA.

Langmuir constant (L/mg). As shown in Fig. 9b, the adsorption isotherms of trivalent REE ions over SBA-15-BTATPA displayed typical Langmuir behavior, which are similar to the reported findings on functionalized mesoporous materials, indicating monolayer adsorption on independent binding sites in SBA-15-BTATPA [26,31]. For comparison, the adsorption data were also fitted to the Freundlich model, as shown in Fig. S2. The Freundlich model given

by (6) can also describe the adsorption of REE ions over functionalized silica but are not restricted to monolayer adsorption [47],

 ln qe ¼ ln K þ

 =n ln Ce

1

(7)

68

S. Ravi et al. / Microporous and Mesoporous Materials 258 (2018) 62e71

Fig. 7. TEM image of SBA-15-BTATPA (Inset: Image of Fast Fourier transform).

Fig. 8. Effect of pH on the adsorption of La3þ, Ce3þ, Nd3þ and Y3þ ions over SBA-15BTATPA.

where K is a constant representing the adsorption capacity related to the bond strength and the slope, 1/n, is a measure of the adsorption intensity or surface heterogeneity. From the Langmuir isotherm, the maximum binding capacity of La3þ, Ce3þ, Nd3þ, and Y3þ over SBA-15-BTATPA were estimated to be 26.5, 34.2, 129.8, and 87.7 mg g1, respectively. Table 2 summarizes the calculated Langmuir constant, b, and the K and n parameters for the Freundlich models for all M3þ ions. The calculated correlation coefficient factor (R2) for all the Langmuir adsorption isotherms of REE ions investigated were >0.98 (Table 2). As shown in Table 2, the Langmuir isotherm model fits better than the Freundlich isotherm model (Fig. S2), suggesting that the binding sites of SBA-15-BTATPA have strong affinity for the adsorbate, leading to the formation of a monolayer of adsorbed molecules (saturation of the adsorbent). Generally, high Langmuir affinity constant implies high adsorption capacity [48]. As we can see from Table 2, the Langmuir affinity constant b for Nd3þ was higher when compare with other ions, and yielded high adsorption capacity. The adsorption capacities of the Nd3þ ions over SBA-15-BTATPA were then compared with those by the previously reported silica-based adsorbents in Table 3. The effectiveness of the SBA-15-BTATPA is higher than most reported articles. Only the silica monolith (Table 3, entry 1) appears to have exhibited higher adsorption capacity towards Nd3þ than SBA-15BTATPA, probably because this monolithic support has a larger average pore diameter (initial pore size: 12.7 nm, after functionalization:8.9 nm) than the rest, which is expected to offer enhanced diffusion of the metal ions to the adsorption sites. 3.2.3. Distribution coefficient Generally, adsorbents with Kd values greater than 104 are considered to be ideal for a given adsorption process. In some cases, high affinity of organic groups towards REE ions results in extremely low residual elements in the aqueous solution after the adsorption process affords high Kd values in the range of ~105 mL g1 [49]. The highly ordered SBA-15-BTATPA mesoporous materials with a well-developed porous structure and properly tethered amide and phosphonic groups provided resulted in Kd values in the order of >105 for Nd3þ and Y3þ ions, and also shown satisfactory Kd values for La3þ and Ce3þ (Table 2). The Kd values of SBA-15-BTATPA were far better than those by other reported adsorbents (Table 3, entries 4 and 7), and showed a similar value as with the benchmark adsorbent Ac-Phos (Table 3, entry 8).

Fig. 9. (a) Adsorption equilibrium capacity of REE ions over SBA-15-BTATPA at pH 6; (b) Adsorption isotherm data fitting to the Langmuir model for REE ions.

S. Ravi et al. / Microporous and Mesoporous Materials 258 (2018) 62e71

69

Table 2 Parameters of REE ions for the adsorption isotherms and kinetics models. Rare earth metals (M3þ ions)

La3þ

Ce3þ

Nd3þ

Y3þ

LAI

26.5 0.494 0.996 4.40 0.93  102 0.972 ~1  104 e e

34.2 0.512 0.995 4.14 1.09  102 0.994 ~1  104

129.8 0.875 0.997 3.11 4.37  102 0.850 ~1.5  105 0.061 0.934 0.086 48.65 0.999

87.71 0.637 0.995 4.95 3.14  102 0.954 ~1.04  105 0.122 0.765

qm (mg/g) b R2 n K (mg1-nLng1) R2

FAI

kd(mL/g) PFO

k1 (min1) R2 k2 (g mg1 min1) qe (mg/g) R2

PSO

e e

0.065 49.13 0.998

LAI-Langmuir adsorption isotherm, FAI-Freundlich adsorption isotherm, PFO- Pseudo first order kinetics and PSO-Pseudo second order kinetics.

Table 3 Comparison of REE ions removal by various silica-based adsorbents. Entries

Functionalized organic group type

1

OcTolPTA

2

EDTA

3 4

DGA DTPA

5

DPETES

6

H2IDA

7 8

N-DGA Ac-phos Prop-Phos BTATPA

9

Support

Silica monolith Chitosansilica Silica gel Chitosansilica Silica microspheres Core-shell g-Fe2O3@SiO2 KIT-6 silica Silica SAMMS SBA-15 Silica

Extracted Element

Extraction conditions

Ref

pH

Time (h)

Competitive elements

Kd (ml g1)

Capacity max (mg g1)

Nd, Eu, Yb

1e5.4

0e1.3

NA

162 (Nd)

[11]

Nd

1e7

0e6

Na, K, Ca, Mg, Al NA

NA

60 (Nd)

[28]

Lanthanides Nd

2 1e7

0e24 0e6

NA NA

NA 0e3.5  102

16.1 (Nd) 106 (Nd)

[31] [28]

Nd, Dy

5.8

1

NA

NA

45 (Nd)

[29]

La, Nd, Dy

NA.

0e50

NA

NA

33 (Nd)

[30]

Lanthanides La, Nd, Eu, Lu La,Ce,Nd, Y

4 1e6.5

0e0.5 2

Al, Fe, Th, U Fe, Ni, Cu, Zn, K, Ca

0e6 x 103 0e4 x 105

NA NA.

[12] [49]

1e6

0e6

Co, Ni, Cu, Zn

0e1.5  105

129.8 (Nd)

This work

diacetic acid, DTPA-diethylenetriaminepentaacetic acid, DPETESOcTolPTA(N-octylNtolyl-1,10phenanthroline-2carboxaide), H2IDA-imino diethylphosphatoethyltriethoxysilane and DGA-diglycolylamic acid. All of the organic group names mentioned here were from the corresponding literature. TW-this work.

Fig. 10. Kinetics studies of (a) Y3þ and (b) Nd3þadsorption over SBA-15-BTATPA. The inset figures a1 and b1 are for pseudo first order, whereas a2 and b2 are for the pseudo second order kinetics of Y3þ and Nd3þ adsorption, respectively.

3.2.4. Adsorption kinetics study For the adsorption kinetics study, measurements were taken at 5, 10, 15, 30, 60, 180, and 360 min whilst keeping the initial REE ion solution concentrations constant at 50 mg/L. The experimental

values were then fitted to the PFO and PSO kinetics models, as shown in Fig. 10. Table 2 lists the parameters of the PFO (k1) and PSO model (k2 and qe). The correlation coefficients (R2) of the PSO model for Y3þ and Nd3þ adsorption were ~1 (0.998 for Y3þ and

70

S. Ravi et al. / Microporous and Mesoporous Materials 258 (2018) 62e71

0.999 for Nd3þ), which were significantly higher than those of the PFO model (0.734 for Y3þ and 0.934 for Nd3þ), indicating that the adsorption of Y3þ and Nd3þ can be better described by the PSO model. The goodness of fit (R2) values of PSO and Langmuir model were higher than those of PFO and Freundlich model, which supported that chemisorption is the main adsorption controlling mechanism for REEs over SBA-15-BTATPA [50].

3.2.5. Selectivity The differences in the affinity of different metal ions with specific organic functionalities led to a selective adsorption process. Adsorption equilibrium at pH 6 was examined to quantify the selectivity of SBA-15-BTATPA towards a specific ion in a simple batch system of a La3þ, Ce3þ, Nd3þ, and Y3þsolution mixture at 50 mg/L each. The experimental runs were maintained for 6 h and aliquots after equilibrium was reached were analyzed by ICP-OES. Fig. 10a shows that the selectivity to Y3þ was >57%, which was higher than that of the other three REE ions. Next to the Y3þ, Nd3þ showed ca. 25% selectivity, followed by Ce3þ (10%) and La3þ (8%). Therefore, adsorption by SBA-15-BTATPA with amide and phosphonic acid functional group appears to proceed through a size selection process, in which a smaller cation (Y3þ) of REEs being adsorbed preferentially than the rest (Nd3þ, Ce3þ, and La3þ). The affinity of SBA-15-BTATPA increased from La3þ to Nd3þ probably due to lanthanide contraction. As shown earlier, the equilibrium adsorption capacity of Nd3þ was approximately four times higher than that of La3þ and Ce3þ ions. This is a consequence of the poor shielding of the nuclear charge by the 4f subshell of lanthanides, which causes the 5s and 5p orbital electrons to experience a larger effective nuclear charge. Therefore, the affinity of SBA-15-BTATPA enfolding the respective ion increased with decreasing ionic radius, resulting in stronger coordination. Among the lanthanides (La3þ, Ce3þ, and Nd3þ), the selectivity also increased with increasing stability constant of the respective ions (La3þ
Fig. 12. Reusability of SBA-15-BTATPA for Y3þ and Nd3þ ions.

BTATPA was negligible; SBA-15-BTATPA showed >98% selectivity towards REE ions under these experimental conditions.

3.2.6. Reusability SBA-15-BTATPA could be reused up to five times with only a slight loss of adsorption capacity, as shown in Fig. 12. The recovered SBA-15-BTATPA almost retained its previous adsorption capacities when reused, which showed that there was no loss of organic groups or silica during the acid treatment for desorption. The stability of SBA-15-BTATPA after 5 cycles was analyzed using XRD and FT-IR (Figs. S3 and S4), which showed the recovered adsorbent retained its structural integrity. In addition, the liquid solutions were tested for possible silicon, nitrogen, and phosphorous leaching by ICP-OES with no leaching detected. On the other hand, there was an adsorbent an approximate <1% weight loss in each cycle after desorption due to filtering and handling. 4. Conclusions In conclusion, the benzene-1,3,5-triamido-tetraphosphonic acid group (BTATPA) was introduced to the surface of SBA-15 (SBA-15BTATPA) via post-synthesis grafting. The SBA-15-BTATPA exhibited effective adsorption towards Nd3þ ions compare to other REE ions (La3þ, Ce3þ, and Y3þ). The adsorption equilibrium data correlated well with the Langmuir model and the maximum adsorption capacity for Nd3þ was estimated to be 129.8 at the optimal pH of 6. The selectivity of REE ions was higher than 98% in the presence of competitive transition metal ions, such as Co2þ, Ni2þ, Cu2þ, and Zn2þ. Adsorption equilibrium was reached within 60 min and could be well fitted by a pseudo second order kinetic model. Y3þ and Nd3þ showed higher selectivity compared to La3þ and Ce3þ in an ion mixture with equal concentrations, which reflects the affinity of these metal ions towards the BTATPA organic group. SBA-15BTATPA could be reused for a minimum of 5 cycles with only a slight loss of adsorption capacity.

Acknowledgements

Fig. 11. Adsorption selectivity of REE ions over SBA-15-BTATPA (a) Selectivity test among REEs; (b) Selectivity of individual REE in presence of other transition metal ions (Co2þ, Ni2þ, Cu2þ, and Zn2þ).

This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (Grant number: NRF2015R1A4A1042434) and also by a research funding supported by the Carbon mineralization flagship center in Korea (2017).

S. Ravi et al. / Microporous and Mesoporous Materials 258 (2018) 62e71

Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.micromeso.2017.09.006. References [1] H.B. Kagan, Introduction: frontiers in lanthanide chemistry, Chem. Rev. 102 (2002) 1805e1806. [2] P. Gawryszewska, J. Sokolnicki, J. Legendziewicz, Photophysics and structure of selected lanthanide compounds, Coord. Chem. Rev. 249 (2005) 2489e2509. [3] T. Kakoi, T. Nishiyori, T. Oshima, F. Kubota, M. Goto, S. Shinkai, F. Nakashio, Extraction of rare-earth metals by liquid surfactant membranes containing a novel cyclic carrier, J. Membr. Sci. 136 (1997) 261e271. [4] A. Sengupta, P.K. Mohapatra, A.B. Patil, R.M. Kadam, W. Verboom, Radiation stability of diglycolamide functionalized calix[4]arenes in ionic liquid: solvent extraction, EPR and GCeMS studies, Sep. Purif. Technol. 162 (2016) 77e83. [5] S. Yang, P. Zong, X. Ren, Q. Wang, X. Wang, Rapid and highly efficient preconcentration of Eu(III) by coreeshell structured Fe3O4@Humic acid magnetic nanoparticles, ACS Appl. Mater. Interfaces 4 (2012) 6891e6900. [6] F. Wang, J. Zhao, F. Pan, H. Zhou, X. Yang, W. Li, H. Liu, Adsorption properties toward trivalent rare earths by alginate beads doping with silica, Industrial Eng. Chem. Res. 52 (2013) 3453e3461. €€ [7] F. Zhao, E. Repo, Y. Meng, X. Wang, D. Yin, M. Sillanpa a, An EDTA-b-cyclodextrin material for the adsorption of rare earth elements and its application in preconcentration of rare earth elements in seawater, J. Colloid Interface Sci. 465 (2016) 215e224. [8] X. Yang, J. Zhang, X. Fang, Rare earth element recycling from waste nickelmetal hydride batteries, J. Hazard. Mater. 279 (2014) 384e388. [9] F. Xie, T.A. Zhang, D. Dreisinger, F. Doyle, A critical review on solvent extraction of rare earths from aqueous solutions, Miner. Eng. 56 (2014) 10e28. [10] W. Yantasee, G.E. Fryxell, R.S. Addleman, R.J. Wiacek, V. Koonsiripaiboon, K. Pattamakomsan, V. Sukwarotwat, J. Xu, K.N. Raymond, Selective removal of lanthanides from natural waters, acidic streams and dialysate, J. Hazard. Mater. 168 (2009) 1233e1238. [11] M.R. Awual, T. Kobayashi, H. Shiwaku, Y. Miyazaki, R. Motokawa, S. Suzuki, Y. Okamoto, T. Yaita, Evaluation of lanthanide sorption and their coordination mechanism by EXAFS measurement using novel hybrid adsorbent, Chem. Eng. J. 225 (2013) 558e566. re, F. Kleitz, Nanostructured hybrid [12] J. Florek, F. Chalifour, F. Bilodeau, D. Larivie materials for the selective recovery and enrichment of rare earth elements, Adv. Funct. Mater. 24 (2014) 2668e2676. [13] S. Schneider, A. Caldas Garcez, M. Tremblay, F. Bilodeau, D. Lariviere, F. Kleitz, Nanoporous ammonium molybdophosphate-silica hybrids as regenerable ultra-selective extraction agents for radiocesium monitoring, New J. Chem. 37 (2013) 3877e3880. re, F. Kleitz, Large pore [14] P.J. Lebed, J.-D. Savoie, J. Florek, F. Bilodeau, D. Larivie mesostructured organosilica-phosphonate hybrids as highly efficient and regenerable sorbents for uranium sequestration, Chem. Mater. 24 (2012) 4166e4176. [15] A.N. Turanov, V.K. Karandashev, V.M. Masalov, A.A. Zhokhov, G.A. Emelchenko, Adsorption of lanthanides(III), uranium(VI) and thorium(IV) from nitric acid solutions by carbon inverse opals modified with tetraphenylmethylenediphospine dioxide, J. Colloid Interface Sci. 405 (2013) 183e188. [16] Y. Zhao, J. Li, S. Zhang, X. Wang, Amidoxime-functionalized magnetic mesoporous silica for selective sorption of U(vi), RSC Adv. 4 (2014) 32710e32717. [17] M.P. Moloney, J. Causse, C. Loubat, A. Grandjean, Sodium “activation” of silano-phosphonate modified mesoporous TiO2 leading to improved rareearth element extraction, Eur. J. Inorg. Chem. 2014 (2014) 2268e2277. [18] X. Zhao, M. Wong, C. Mao, T.X. Trieu, J. Zhang, P. Feng, X. Bu, Size-selective crystallization of homochiral camphorate metaleorganic frameworks for lanthanide separation, J. Am. Chem. Soc. 136 (2014) 12572e12575. [19] B.-C. Luo, L.-Y. Yuan, Z.-F. Chai, W.-Q. Shi, Q. Tang, U(VI) capture from aqueous solution by highly porous and stable MOFs: UiO-66 and its amine derivative, J. Radioanalytical Nucl. Chem. 307 (2016) 269e276. [20] M.-R. Huang, H.-J. Lu, X.-G. Li, Synthesis and strong heavy-metal ion sorption of copolymer microparticles from phenylenediamine and its sulfonate, J. Mater. Chem. 22 (2012) 17685e17699. [21] C. Han, L. Zhang, H. Li, Highly selective and sensitive colorimetric probes for Yb3þ ions based on supramolecular aggregates assembled from [small beta]cyclodextrin-4,4[prime or minute]-dipyridine inclusion complex modified silver nanoparticles, Chem. Commun. (2009) 3545e3547. [22] B. Chang, J. Guo, C. Liu, J. Qian, W. Yang, Surface functionalization of magnetic mesoporous silica nanoparticles for controlled drug release, J. Mater. Chem. 20 (2010) 9941e9947. [23] C. Perego, R. Millini, Porous materials in catalysis: challenges for mesoporous materials, Chem. Soc. Rev. 42 (2013) 3956e3976. [24] S.A. Idris, C.M. Davidson, C. McManamon, M.A. Morris, P. Anderson, L.T. Gibson, Large pore diameter MCM-41 and its application for lead removal from aqueous media, J. Hazard. Mater. 185 (2011) 898e904.

71

[25] S. Ravi, D.H. Kang, R. Roshan, J. Tharun, A.C. Kathalikkattil, D.W. Park, Organic sulphonate salts tethered to mesoporous silicas as catalysts for CO2 fixation into cyclic carbonates, Catal. Sci. Technol. 5 (2015) 1580e1587. [26] S. Ravi, M. Selvaraj, Incessant formation of chain-like mesoporous silica with a superior binding capacity for mercury, Dalton Trans. 43 (2014) 5299e5308. [27] S. Ravi, M. Selvaraj, H. Park, H.-H. Chun, C.-S. Ha, Novel hierarchically dispersed mesoporous silica spheres: effective adsorbents for mercury from wastewater and a thermodynamic study, New J. Chem. 38 (2014) 3899e3906. [28] J. Roosen, J. Spooren, K. Binnemans, Adsorption performance of functionalized chitosan-silica hybrid materials toward rare earths, J. Mater. Chem. A 2 (2014) 19415e19426. [29] I.V. Melnyk, V.P. Goncharyk, L.I. Kozhara, G.R. Yurchenko, A.K. Matkovsky, Y.L. Zub, В. Alonso, Sorption properties of porous spray-dried microspheres functionalized by phosphonic acid groups, Microporous Mesoporous Mater. 153 (2012) 171e177. [30] E. Polido Legaria, S.D. Topel, V.G. Kessler, G.A. Seisenbaeva, Molecular insights into the selective action of a magnetically removable complexone-grafted adsorbent, Dalton Trans. 44 (2015) 1273e1282. [31] T. Ogata, H. Narita, M. Tanaka, Adsorption behavior of rare earth elements on silica gel modified with diglycol amic acid, Hydrometallurgy 152 (2015) 178e182. [32] A.S. Maria Chong, X.S. Zhao, Functionalization of SBA-15 with APTES and characterization of functionalized materials, J. Phys. Chem. B 107 (2003) 12650e12657. [33] D. Zhao, Q. Huo, J. Feng, B.F. Chmelka, G.D. Stucky, Nonionic triblock and star diblock copolymer and oligomeric surfactant syntheses of highly ordered, hydrothermally stable, mesoporous silica structures, J. Am. Chem. Soc. 120 (1998) 6024e6036. [34] Y.S. Ho, G. McKay, Comparative sorption kinetic studies of dye and aromatic compounds onto fly ash, J. Environ. Sci. Health, Part A 34 (1999) 1179e1204. [35] L. Sarkisov, P.A. Monson, Modeling of adsorption and desorption in pores of simple geometry using molecular dynamics, Langmuir 17 (2001) 7600e7604. [36] J. Tu, N. Li, W. Geng, R. Wang, X. Lai, Y. Cao, T. Zhang, X. Li, S. Qiu, Study on a type of mesoporous silica humidity sensing material, Sensors Actuators B Chem. 166e167 (2012) 658e664. [37] R.K. Sharma, S. Sharma, S. Gulati, A. Pandey, Fabrication of a novel nanocomposite carbon paste sensor based on silica-nanospheres functionalized with isatin thiosemicarbazone for potentiometric monitoring of Cu2þ ions in real samples, Anal. Methods 5 (2013) 1414e1426. [38] R. Luschtinetz, G. Seifert, E. Jaehne, H.-J.P. Adler, Infrared spectra of alkylphosphonic acid bound to aluminium surfaces, Macromol. Symp. 254 (2007) 248e253. [39] L. Paul, B. Banerjee, A. Bhaumik, M. Ali, Catecholase activity of a manganese Schiff base complex functionalized over SBA-15 in aqueous heterogeneous medium, Microporous Mesoporous Mater. 249 (2017) 78e87. [40] J. Lu, Y. Li, C. Deng, Facile synthesis of zirconium phosphonate-functionalized magnetic mesoporous silica microspheres designed for highly selective enrichment of phosphopeptides, Nanoscale 3 (2011) 1225e1233. [41] C. Qi, Y.-J. Zhu, G.-J. Ding, J. Wu, F. Chen, Solvothermal synthesis of hydroxyapatite nanostructures with various morphologies using adenosine 5 [prime or minute]-monophosphate sodium salt as an organic phosphorus source, RSC Adv. 5 (2015) 3792e3798. [42] S. Ravi, R. Roshan, J. Tharun, D.-W. Park, H.-H. Chun, H. Park, M. Selvaraj, Mesoporous silica-giant particle with slit pore arrangement as an adsorbent for heavy metal oxyanions from aqueous medium, RSC Adv. 5 (2015) 10260e10266. [43] D. Chaudhuri, D.D. Sarma, Blue emitting polyaniline, Chem. Commun. (2006) 2681e2683. [44] B. Song, C. Sizemore, L. Li, X. Huang, Z. Lin, K.-s. Moon, C.-P. Wong, Triethanolamine functionalized graphene-based composites for high performance supercapacitors, J. Mater. Chem. A 3 (2015) 21789e21796. [45] L.M. Cornaglia, E.A. Lombardo, XPS studies of the surface oxidation states on vanadium-phosphorus-oxygen (VPO) equilibrated catalysts, Appl. Catal. A General 127 (1995) 125e138. [46] I. Langmuir, The adsorption of gases on plane surfaces of glass, mica and platinum, J. Am. Chem. Soc. 40 (1918) 1361e1403. [47] S. Aytas, M. Yurtlu, R. Donat, Adsorption characteristic of U(VI) ion onto thermally activated bentonite, J. Hazard. Mater. 172 (2009) 667e674. [48] J.-H. Park, Y.S. Ok, S.-H. Kim, J.-S. Cho, J.-S. Heo, R.D. Delaune, D.-C. Seo, Competitive adsorption of heavy metals onto sesame straw biochar in aqueous solutions, Chemosphere 142 (2016) 77e83. [49] G.E. Fryxell, H. Wu, Y. Lin, W.J. Shaw, J.C. Birnbaum, J.C. Linehan, Z. Nie, K. Kemner, S. Kelly, Lanthanide selective sorbents: self-assembled monolayers on mesoporous supports (SAMMS), J. Mater. Chem. 14 (2004) 3356e3363. [50] Y.S. Ho, G. McKay, Pseudo-second order model for sorption processes, Process Biochem. 34 (1999) 451e465. [51] V.V. Nikonorov, Determination of the stability constants of lanthanide complexes with oxyacids using capillary electrophoresis, J. Anal. Chem. 65 (2010) 359e365.