An Imprinted Organic−Inorganic Hybrid Sorbent for Selective

Sciences, Hebei University, Baoding 071002, China. A hierarchical double-imprinting concept was applied to the preparation of a new organic-inorganic ...
1 downloads 0 Views 194KB Size
Anal. Chem. 2004, 76, 453-457

An Imprinted Organic-Inorganic Hybrid Sorbent for Selective Separation of Cadmium from Aqueous Solution Yun-Kai Lu†,‡ and Xiu-Ping Yan*,†

State Key Laboratory of Functional Polymer Materials for Adsorption and Separation, Research Center for Analytical Sciences, College of Chemistry, Nankai University, Tianjin 300071, China, and College of Chemical and Environmental Sciences, Hebei University, Baoding 071002, China

A hierarchical double-imprinting concept was applied to the preparation of a new organic-inorganic hybrid sorbent for selective separation of Cd(II) from aqueous solution. In the prepared hierarchically imprinted sorbent, both Cd(II) and surfactant micelles (cetyltrimethylammonium bromide) were used as templates. The sorbent was prepared through self-hydrolysis, self-condensation, and co-condensation of the cross-linking agent (tetraethoxysilicate) and the functional precursor (3-(2-aminoethylamino)-propyltrimethoxysilane) in an alkaline media followed by gelation. The selectivity of the sorbent was investigated by a batch competitive ion-binding experiment using an aqueous Cd(II) and Zn(II) mixture. The largest selectivity coefficient for Cd(II) in the presence of Zn(II) was found to be over 100; the largest relative selectivity coefficient between Cd(II) and Zn(II), over 200. The uptake capacity of the prepared hierarchically imprinted sol-gel sorbent and the selectivity coefficient are much higher than those of the sorbent prepared in the absence of CTAB-template. The sorbent possesses a fast kinetics for the removal of Cd(II) from aqueous solution. Cadmium is one of the most toxic elements for animals and humans, even at low concentrations. Cd(II) is among the 13 toxic metal species on the priority pollutant list of the Environmental Protection Agency (EPA).1,2 Separation of toxic Cd(II) is of intense current interest in research and environmental cleanup. Selective removal of toxic metal ions from aqueous solutions is usually achieved by solvent extraction and solid-phase extraction with sorbents. Solid-phase extraction possesses many advantages over conventional solvent extraction, including a higher enrichment factor, absence emulsion; safety with respect to hazardous samples, minimal costs due to low consumption of reagents, and flexibility and incorporation into automated analytical techniques.3 * Corresponding author. Fax: (86)22-23503034. E-mail: [email protected]. † Nankai University. ‡ Hebei University. (1) Manahan, S. E. Environmental Chemistry, 6th ed.; Lewis Publishers: Boca Raton, 1994. (2) Watson, J. S. Separation Methods for Waste and Environmental Applications; Marcel Dekker: New York, 1999. (3) Thurman, E. M.; Mills, M. S. Solid-Phase Extraction: Principles and Practice; John Wiley & Sons: Chichester, 1998. 10.1021/ac0347718 CCC: $27.50 Published on Web 12/11/2003

© 2004 American Chemical Society

Development of new solid sorbents for selective separation of toxic Cd(II) from aqueous solutions, therefore, is of great significance. Molecular imprinting is an attractive method for the preparation of selective sorbents.4-8 Recently, molecularly imprinted solgel materials (MISGMs) have been extensively studied.9-14 MISGMs are fabricated by a conventional sol-gel process and incorporation of the template molecules into rigid inorganic or inorganic-organic networks. After removal of the template, molecular cavities with distinct pore size, shape, or chemical functionality remain in the cross-linked host. These “molecularly designed cavities” show an affinity for the template molecule over other structurally related coumpounds. The most important type of the imprinted sol-gel material synthesis is surface imprinting on mesoporous material supports.13 Surface molecular imprinting of mesoporous silica was used in most of the work to increase metal loading capacities and selectivity, as compared to those sorbents using conventional silica. However, it has an inherent limitation in maximizing ligand densities of the functionalized silica sorbents because of the availability of hydroxyl groups on the mesoporous silica. This lack of hydroxyl groups can lead to small metal adsorption capacity. In the imprint coating process, the functional groups are introduced onto the pore surface of the mesoporous silica. The fine pores may not be available for the accommodation of functional ligands and may result in poor covalent attachment.15 A very promising type is direct formation of the imprinted mesoporous sorbent with functional ligands. Very recently, a novel (4) Martı´n-Esteban, A. Fresenius’ J. Anal. Chem. 2001, 370, 795-802. (5) Sellergren, B., Ed. Molecularly Imprinted Polymers: Man-made Mimics of Antibodies and Their Applications in Analytical Chemistry; Elsevier: Amsterdams, 2001;. (6) Andersson, L. I. Bioseparation 2001, 10, 353-364. (7) Lanza, F.; Sellergren, B. Chromatographia 2001, 53, 599-611. (8) Bartsch, R. A.; Maeda, M., Ed. Molecular and Ion Recognition with Imprinted Polymers; The American Chemical Society: Washington, DC, 1998. (9) Makote, R. D.; Dai, S. Anal. Chim. Acta 2001, 435, 169-175. (10) Dickert, F. L.; Hayden, O. Anal. Chem. 2002, 74, 1302-1306. (11) Makote, R.; Collinson, M. M. Chem. Mater. 1998, 10, 2440-2445. (12) Dai, S.; Burleigh, M. C.; Ju, Y. H.; Gao, H. J.; Lin, J. S.; Pennycook, S. J.; Barnes, C. E.; Xue, Z. L. J. Am. Chem. Soc. 2000, 122, 992-993. (13) Dai, S.; Burleigh, M. C.; Shin, Y.; Morrow, C. C.; Barnes, C. E.; Xue, Z. Angew. Chem., Int. Ed. Engl. 1999, 38, 1235-1239. (14) Hunnius, M.; Rufinska, A.; Maier, W. F. Micropor. Mesopor. Mater. 1999, 29, 389-403. (15) Lee, J. S.; Gomez-Salazar, S.; Tavlarides, L. L. React. Funct. Polym. 2001, 49, 159-172.

Analytical Chemistry, Vol. 76, No. 2, January 15, 2004 453

method for template-selective recognition sites on mesopororous sorbents through hierarchical double imprinting was proposed by Dai et al.12,16 The essence of this methodology is the combination of two powerful imprinting techniques at different scales: molecular imprinting and micelle templating synthesis. This method has been shown to be successful in creating Cu(II) ionselective recognition sites on mesoporous sorbent and precise control of stereochemical arrangement of ligands on the adsorption sites.12,16 Herein we apply this double imprinting concept to prepare a new hybrid sorbent with the selectivity of the cavities and the affinity of the functional ligands for selective separation of Cd(II) from aqueous solutions. The synthesis and characterization of this new sorbent, kinetics of Cd(II) uptake and removal (from the gels), and selectivity of the sorbent for Cd(II) are discussed. EXPERIMENTAL SECTION Instrumentation. A Hitachi 180-80 atomic absorption spectrometer (Hitachi, Japan) was used to measure the concentration of metal ions in aqueous solution for the study of the uptake and removal of Cd(II) and selectivity of the prepared new sorbent. FT-IR spectra (4000-400 cm-1) in KBr were recorded using a Magna-560 spectrometer (Nicolet, U.S.A.). Scanning electron microscopy (SEM) images were obtained at 20.0 KV on a Hitachi S-4100 field emission scanning electron microscope. Reagents. All reagents used were of the highest available purity and of at least analytical grade. Doubly deionized water (DDW) was used throughout this work. Tetraethoxysilicate (TEOS) and 3-(2-aminoethylamino)-propyltrimethoxysilane (AAPTS, Wuhan University Chemical Factory) were used in this study. The cetyltrimethylammonium bromide (CTAB), CdCl2‚2.5H2O and ZnCl2 (Tianjin Chemical Co., Tianjin, China) were used without further purification. The pH of the solutions was adjusted using the following buffers: sodium acetate/hydrochloric acid for pH 2-6 and potassium dihydrogen phosphate/sodium hydroxide for pH 6-7. Preparation of Hierarchically Imprinted Sol-Gel Sorbent. A 0.274-g portion of CdCl2‚2.5H2O and 0.730 g of CTAB were dissolved in 18.7 g of DDW. To the mixture, 0.556 g of AAPTS was added. After stirring the mixture for 1 h, TEOS (2.08 g of TEOS was dissolved in 2 mL of MeOH and stirred for 20 min.) and NaOH (1 mol L-1, 3.8 mL) were added. The solution was stirred for 2 day and then refluxed at 90 °C for 1 day. The solid product was recovered by filtration. This was then refluxed in ethanol/HCl to extract the surfactant templates with ethanol and to strip the cadmium ion templates by protonation of 4 mol L-1 HCl. The final materials were washed with copious amounts of 1 mol L-1 HCl to ensure the complete removal of the cadmium ion templates. The resulting gels were neutralized with 1 mol L-1 NaOH to pH 7.5, filtered, washed with deionized water, and dried under vacuum at 60 °C for 1 day. For comparison, the Cd(II)imprinted sol-gel sorbent was also prepared using an identical procedure, but without the addition of CTAB. The control blank samples were prepared in parallel without addition of Cd(II). Adsorption Test. Competitive loading of Cd(II) and Zn(II) by Cd(II)-imprinted and control blank sorbents was measured at pH 5.4 (acetic acid/sodium acetate buffer). The time-dependent (16) Dai, S. Chem. Eur. J. 2001, 7, 763-768.

454

Analytical Chemistry, Vol. 76, No. 2, January 15, 2004

uptake test indicated that 0.5 h is sufficient. In all batch experiments, 0.1 g of each sample was equilibrated with 10 mL of the buffer solution containing 2 mmol L-1 of Cd(II) and 2 mmol L-1 of Zn(II) in stoppered plastic vials, and these mixtures were stirred for 30 min at room temperature. The filtrates were measured for uptakes of Cd(II) and Zn(II) by a Hitachi 180-80 atomic absorption spectrometer. Regeneration of the Sol-Gel Derived Ion-Imprinted Sorbents. In a test of extraction/stripping cycles, once Cd(II) loading was complete; the metal ions were removed from the sorbents by washing with 5.0 mL of 4 mol L-1 HCl. The sorbents were rinsed several times with deionized water and then neutralized to pH 7.5 using 1 mol L-1 NaOH solution for up to 1 day to ensure complete H+ neutralization. The sorbents were washed again with deionized water and placed in a 60 °C oven to dry under vacuum before another extraction cycle. RESULTS AND DISCUSSION Factors Affecting Hierarchically Imprinted Sol-Gel Sorbent. The schematic representation of the synthesis route and ion-recognition are shown in Figure 1.15 In the synthesis process, first, the complexes were formed between cadmium ions and bifunctional ligands (dNH, -NH2), then AAPTS and TEOS were independently self-hydrolyzed and self-condensed, and both partially self-condensed silanes were added to surfactant micelle solution and co-condensed and gelled in alkaline media. The metal ion and the surfactant templates were removed from the sorbent by acid leaching and ethanol extraction, respectively. The effects of ion imprinting and the sol-gel process on the characteristics of the sorbent were investigated. The selectivity of imprinted sol-gel materials depends on various factors.14,15 These include the molar ratio of the functional precursor silane and the cross-linking silane; reaction time; reaction temperature; surfaction-templates; aging/drying conditions; types of acid/base catalysts; the specificity of the interaction of the ligand with the cation; the template ions have to be removed from the matrix; the adsorption site must not be blocked by any other molecule or ions; the matrix needs to be flexible or needs to have pores that are larger than the template; the cavity itself should be fairly flexible to allow the molecules (or ions) to enter and rebind, but the cavity must have kept its shape under adsorption; the sorbents offering reasonable kinetics for the adsorption; and the desorption of the cations. The molar ratio of AAPTS to TEOS is directly related to the moles of the active groups with respect to the weight of the silica matrix. For the ratio of TEOS/AAPTS 4 is reduced, possibly as a result of the decrease of the functional ligands. At a TEOS/AAPTS ratio of ∼4, obvious mechanical strength and high cadmium uptake capacity were observed. The scanning electron micrographs of these materials are shown in Figure 2. The material produced with a molar ratio of 2 exhibited a fractured surface (Figure 2a). The morphology of the material synthesized with a TEOS/AAPTS molar ratio of 4 was monoliths with large particles packed together and a smoother surface (Figure 2b). The reaction time and temperature affected the degree of functional group anchoring on the sorbent. Sol solution was stirred for 48 h at room temperature in order that gelation was completed

Figure 1. Schematic synthesis route and mechanism of recognition of ion imprinted sol-gel sorbent.

and mesostructured products were obtained. The additional heat treatment greatly improved the structural order and mechanical strength of the sorbents. Surfactants are bifunctional molecules that form spherical micelles in water. Silicate species are deposited between surfactant tubules and interact with the micelle then condense to form the organic-inorganic network. The possible mechanism is similar to the formation of MCM-41.17 Surfactant templates improved the Cd(II)-uptake capacity, the selectivity, and surface areas. The results of an adsorption test (Table 1) clearly show the uptake capacity of the prepared hierarchically imprinted sol-gel sorbent (0.2 mmol g-1) and the selectivity coefficient (115) are much higher than those of the sorbent prepared in the absence of CTABtemplate. Characterization of the FTIR Spectra. To ascertain the presence of AAPTS and Cd(AAPTS)22+ in the silica matrix, FTIR spectra were obtained from SiO2, imprinted silica sorbent, and Cd(II)-loaded sorbent (Figure 3a-c). The presence of adsorption water was reflected by νOH vibration at 3446.4 and 1635.6 cm-1. The features around 1092.7 and 958.9 cm-1 indicate Si-O-Si and Si-O-H stretching vibrations, respectively. The bands around 799.3 and 469.6 cm-1 resulted from Si-O vibrations. A characteristic feature of the AAPTS-gel when compared with SiO2 is an aliphatic C-H bond and the stronger CH2-N bond, which therefore have a peak at 2957.7 cm-1 and a sharp peak at 1384.4 (17) Raman, N. K.; Anderson, M. T.; Brinker, C. J. Chem. Mater. 1996, 8, 16821701.

Figure 2. SEM image (20.0 kV) of ion-imprinted sol-gel materials prepared with molar ratios of TEOS/AAPTS of (a) 2 and (b) 4.

Analytical Chemistry, Vol. 76, No. 2, January 15, 2004

455

Table 1. Competitive Loading of Cd(II) and Zn(II) by Cd(II)-Imprinted and Control Blank Mesoporous Sorbents at pH 5.4 (Acetic Acid/Acetate) init soln (mmol L-1)

% uptake

capacity (mmol g-1)

Kd (mL g-1)a

sorbent

Cd(II)

Zn(II)

Cd(II)

Zn(II)

Cd(II)

Zn(II)

Cd(II)

nonimp-aapts (CTAB)d

8 5 2

8 5 2

63.4 99.2 99.6

70.8 99.7 99.8

0.507 0.492 0.199

0.566 0.499 0.200

173 12400 2500

242 33200 50000

imp-aapts (no surf.)e

2

2

48.5

21.0

0.097

0.042

94

27

8 5 2

8 5 2

60.6 95.0 99.8

11.3 20.4 81.3

0.485 0.475 0.200

0.091 0.102 0.163

153 1900 50000

imp-aapts

(CTAB)f

Zn(II)

13 25.6 435

kb

k′c

0.71 0.37 0.5 3.5 12 74 115

8 17 200 230

a K , distribution coefficient, the ratio of mole of Cd(II) absorbed per gram of sorbent to the molar concentration of Cd(II) in final solution. b k, d selectivity coefficient, Kd(Cd)/Kd(Zn). c k′, relative selectivity coefficient, k′ ) kimprinted/knonimprinted. d Nonimp-aapt (CTAB), nonimprinted sorbent synthesized with CTAB template. e Imp-aapts (no surf.), Cd(II) imprinted sorbent synthesized without surfactant template. f Imp-aapts (CTAB), imprinted sorbent synthesized with Cd(II) and CTAB templates.

Figure 3. FTIR spectra of (a) SiO2 prepared using TEOS, (b) the imprinted sorbent without loaded Cd(II), and (c) the imprinted sorbent with loaded Cd(II). The presence of adsorption water was reflected by νOH vibration at 3446.4 and 1635.6 cm-1. The features around 1092.7 and 958.9 cm-1 indicate Si-O-Si and Si-O-H stretching vibrations, respectively. The bands around 799.3 and 469.6 cm-1 resulted from Si-O vibrations. A characteristic feature of the AAPTSgel when compared with SiO2 is an aliphatic C-H bond at 2957.7 cm-1 and the stronger CH2-N bond at 1384.4 cm-1.

cm-1. Compared with the AAPTS gel, the FTIR features of CdAAPTSgel showed very similar location and appearance of the major bands, but relatively strong and sharp peak at 1384.4 cm-1 likely due to the coordination effect. Selectivity for Cd(II) in Competitive Adsorption on the Imprinted Sol-Gel Sorbents. Adsorption and competitive ionrecognition studies were performed with Cd(II) and Zn(II) ions in order to measure the selectivity of the imprinted materials. The Zn(II) ion was chosen as the competitor species because it has the same charge and similar chemical properties and also binds well with the diamine ligand. The stability of the metal-diamine complex decreases in the following order: Cu(II) > Ni(II) > Co(II) > Zn(II) > Cd(II) > Pb(II). The metal-diamine complex ability can result in the difference of selectivity. The Zn(II) ion with higher complex stability was used in the competitive binding experiments between Zn(II) and Cd(II) ions. Table 1 summarizes the data for the percentage metal ion adsorbed, uptake capacity, distribution coefficient (Kd), selectivity coefficient of the sorbent toward Cd(II) (k), and the relative selectivity coefficient (k′) obtained in these competitive ion-binding 456 Analytical Chemistry, Vol. 76, No. 2, January 15, 2004

experiments between Zn(II) and Cd(II) ions. Comparison of the k values for the Cd(II)-imprinted sorbents with the corresponding control blank samples reveals a significant increase in k for Cd(II) through imprinting, with the largest k over100 and the largest k′ over 200. The Kd and k values of doubly imprinted sorbents (i.e., in the presence of surfactant temperates) are significantly larger in comparison with the singly imprinted sorbent (i.e., without surfactant temperates). The doubly imprinted sorbents offer much higher uptake capacity for Cd(II) (0.2 mmol g-1) at 2 mmol L-1 than the singly imprinted sorbent (0.097 mmol g-1). In contrast, imprint coating on the surfaces of commercial, amorphous silica gel showed very little evidence of imprinting effects in selectivity experiments, and mesoporous silica support improved the value of k′ (40).13 The imprinted sol-gel sorbents prepared by means of the hierarchically double imprinting approach have the largest k′, over 200. This may be attributed to the flexibility of the cavity and which specific binding sites contained functional groups in a predetermined orientation. Although Zn(II) has the same coordination configuration as Cd(II), the Cd(II)-imprinted sol-gel sorbents exhibit much higher selectivity for Cd(II) than Zn(II) due to the imprinting effect because of the ion radius of Zn(II) being much smaller than that of Cd(II). Effect of pH in Absorption Test. The pH dependence of the percentage of Cd(II) extracted is shown in Figure 4. The percentage of Cd(II) extracted increased as the pH of the aqueous solution increased from 2 to 5.4, then it remained constant with further increase in pH. Below pH 3, only 6, precipitation of the metal hydroxide is expected.18 As seen in Figure 4, the optimum pH value for the extraction of Cd(II) from aqueous solution ranged from 5 to 6. In this pH range, neither precipitation of the metal hydroxide nor the protonation of the amine with the active chelating groups is expected. Uptake Kinetics of Cd(II) by the Doubly Imprinted Sorbent. In a typical uptake kinetics test, 100 mg of the sorbent was added to 10 mL of 0.01 mol L-1 Cd(II) aqueous solution buffered (18) Mahmoud, M. E. Anal. Lett. 1996, 29, 1791-1804.

Table 2. Effects of Stripping Agents on Cd(II) Recoverya stripping agent

time (h)

Cd(II) recovery (%)

0.1 mol L-1 EDTA 0.1 mol L-1 EDTA 1 mol L-1 HNO3 2 mol L-1 HNO3 1 mol L-1 HCl 4 mol L-1 HCl 4 mol L-1HCl

0.5 24 1 12 1 1 0.5

9.4 31.2 14.3 56.5 41.6 94.7 75.6

a

Figure 4. Effect of pH on the uptake of 0.002 mol L-1 Cd(II) onto 0.1 g of the imprinted sorbent for 30 min.

5 mL of the stripping agent, 0.1 g of the imprinted sorbent.

Table 3. Extraction Recyclability through Six Extraction/Stripping Cycles with 0.1 g of the Cd(II)-Imprinted Sol-Gel Sorbent in 20 mL of 0.001 mol L-1 Cd(II) Solutions at pH 5.4 (Acetic Acid/ Sodium Acetate Buffer) extraction cycle

% uptake

Kd (mL g-1)

capacity (mmol g-1)

1 2 3 4 5 6

99.4 94.5 91.4 89.7 88.6 90.2

33000 3436 2126 1742 1554 1841

0.199 0.189 0.183 0.179 0.177 0.180

through six extraction/stripping cycles. The stripping agent used in this experiment was 4 mol L-1 HCl. The results of the recyclability studies are shown in Table 3. The same sample of the sorbent was found to remove 92 ( 4% of the Cd(II) from solution through six extraction cycles. The capacity of the sorbent in these five recycles was 91 ( 2% of the fresh sorbent.

Figure 5. Study of the uptake kinetics for 0.002 mol L-1 Cd(II) onto 0.1 g of the imprinted sorbent at pH 5.4 (acetic acid/sodium acetate buffer).

to pH 5.4 in a 25-mL flask. The mixture was mechanically shaken for 2-30 min at room temperature. The mixture was then filtered off and washed with doubly distilled water, and the excess unextracted Cd(II) in the filtrate was determined by atomic absorption spectrometry. Figure 5 shows the uptake kinetics of cadmium ions. It is clear that the solid extraction process of the sorbent is fairly rapid; the 95% uptake of Cd(II) was achieved within 5 min. Removal of Cd(II) from the Sorbent. Several stripping agents, EDTA, HNO3, and HCl, were investigated to remove the adsorbed Cd(II) from the sorbent. The results are shown in Table 2. Among the stripping reagents studied, 4 mol L-1 of HCl solution was found to be quite effective for removing the absorbed Cd(II) from the sorbent. With a single wash of 4 mol L-1 of HCl solution, 95% of Cd(II) of the absorbed Cd(II) were recovered. Recycling of the Sorbent. In a test of sorbent recyclability, the ion-imprinted sol-gel sorbent was used to extract Cd(II)

CONCLUSIONS The successful preparation of ion-imprinted sol-gel materials using metal ion and surfactant templates demonstrated the feasibility of the direct formation of the imprinted mesoporous sorbent with functional ligands for Cd(II) ion. This method provides the mild imprinting reactive condition, which may incorporate various metal ions into inorganic and organicinorganic hybrid materials, and improves ion recognition capabilities. This methodological study and application in the preconcentration and separation of toxic metal ions will be an important field. ACKNOWLEDGMENT This work was supported by the National Natural Science Foundation of China for Distinguished Young Scholars (No. 20025516) and by the Research Foundation for Excellent Young Teachers, State Education Ministry.

Received for review July 9, 2003. Accepted November 6, 2003. AC0347718

Analytical Chemistry, Vol. 76, No. 2, January 15, 2004

457