- Email: [email protected]
Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts
G Model ACA 233506 No. of Pages 9
Analytica Chimica Acta xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca
Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts Jiajia Yang a,b , Yun Li a , Jincheng Wang a , Xiaoli Sun a,b , Syed Mazhar Shah a , Rong Cao a,b , Jiping Chen a, * a Key Laboratory of Separation Sciences for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Dalian 116023, China b University of Chinese Academy of Sciences, Beijing 100049, China
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
Novel sponge-like molecularly imprinted mesoporous silica was synthesized. Extraordinarily large specific surface area and highly interconnected 3-D porous network. High specific adsorption capacity and fast adsorption kinetics for BPA. Good class-selectivity and clean-up effect for bisphenols in sediment under SPE mode. Good recoveries and sensitivity for bisphenols using the MISMS–SPE coupled with HPLC–DAD method.
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 June 2014 Received in revised form 29 August 2014 Accepted 29 September 2014 Available online xxx
Bisphenol A (BPA) imprinted sponge mesoporous silica was synthesized using a combination of semi-covalent molecular imprinting and simple self-assembly process. The molecularly imprinted sponge mesoporous silica (MISMS) material obtained was characterized by FT-IR, scanning electron microscopy, transmission electron microscopy, and nitrogen adsorption–desorption measurements. The results show that the MISMS possessed a large specific surface area (850.55 m2 g1) and a highly interconnected 3-D porous network. As a result, the MISMS demonstrated a superior specific adsorption capacity of 169.22 mmol g1 and fast adsorption kinetics (reaching equilibrium within 3 min) for BPA. Good class selectivity for BPA and its analogues (bisphenol F, bisphenol B, bisphenol E and bisphenol AF) was also demonstrated by the sorption experiment. The MISMS as solid-phase extraction (SPE) material was then evaluated for isolation and clean-up of these bisphenols (BPs) from sediment samples. An accurate and sensitive analytical method based on the MISMS–SPE coupled with HPLC–DAD has been successfully established for simultaneous determination of five BPs in river sediments with detection
Keywords: Molecularly imprinted Sponge mesoporous silica Solid-phase extraction Class-selectivity Bisphenols
Abbreviations: SMS, sponge mesoporous silica; MISMS, molecularly imprinted sponge mesoporous silica; NISMS, non-imprinted sponge mesoporous silica; TEOS, tetraethoxysilane; ICPTES, (3-isocyanatopropyl)triethoxysilane; APTES, 3-aminopropyltriethoxysilane; ACN, acetonitrile; DCM, dichloromethane; BET, Brunauer–Emmett– Teller; BJH, Barrett–Joyner–Halenda; TEM, transmission electron microscopy; BPs, bisphenols; BPA, bisphenol A; BPF, bisphenol F; BPB, bisphenol B; BPE, bisphenol E; BPAF, bisphenol AF; DDT, dichlorodiphenyltrichloroethane; E2, 17b-estradiol; DES, diethylstilbestrol; BPA-Si, BPA-ICPTES; dw, dry weight. * Corresponding author. Tel./fax: +86 411 8437 9562. E-mail address: [email protected] (J. Chen). http://dx.doi.org/10.1016/j.aca.2014.09.051 0003-2670/ ã 2014 Published by Elsevier B.V.
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
2
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
limits of 0.43–0.71 ng g1 dry weight (dw). The recoveries of BPs for lyophilizated sediment samples at two spiking levels (50 and 500 ng g1 dw for each BP) were in the range of 75.5–105.5% with RSD values below 7.5%. ã 2014 Published by Elsevier B.V.
1. Introduction Bisphenols (BPs) are a group of chemicals with two hydroxyphenyl functionalities and include several analogues such as bisphenol A (BPA), bisphenol F (BPF), bisphenol B (BPB), bisphenol E (BPE) and bisphenol AF (BPAF) [1]. In recent years, there has been a growing concern worldwide about the occurrence of BPs in the environment due to their potential adverse effects on human health [2–5]. BPA was known as an endocrine-disrupting chemical and its adverse effects on reproduction and development, neural networks, cardiovascular, metabolic, and immune systems in human have been well documented [3,4]. Its analogues (e.g., BPF, BPE and BPB) have shown moderate to slight acute toxicity and an estrogenic activity similar to BPA [5]. Therefore, the determination of BPs in environmental samples is essential for environmental health and human exposure assessments. Due to the complexity of environmental matrixes and the low concentrations of environmental pollutants, the development of highly specific or selective sample preparation techniques for BPs would be highly desirable. Molecular imprinting is known as a technique for generating tailor-made recognition sites complementary in shape, size and functional groups to the template molecules [6]. In general, three different strategies, namely covalent, non-covalent and semicovalent imprinting, have been employed to prepare molecularly imprinted materials, and non-covalent imprinting is the most used strategy owing to its experimental simplicity [7]. Due to some merits like good chemical stability and reusability, the majority of the research works for molecular imprinting, especially for solidphase extraction (SPE) applications [8–10], were focused on organic polymers. Nevertheless, molecularly imprinted polymers (MIPs) tend to swell or shrink when exposed to organic solvents [11]. The MIPs prepared with bulk polymerization or precipitation polymerization using non-covalent approach also exhibited many other drawbacks, such as low binding capacity, poor site accessibility, slow mass transfer and template leakage. These problems are mostly attributed to the fact that the recognition sites and residual template molecules are deeply embedded in the highly cross-linked polymer matrix [12]. Such shortcomings of MIPs have become one of the major bottlenecks in the development of molecular imprinting technology where extensive research of ideal materials is imperative.
With respect to organic polymer, silica materials are a good alternative for molecular imprinting due to its excellent mechanical resistance and non-swelling, rigid micro-structures [13]. Furthermore, mesoporous silicas such as MCM-41 [14,15], SBA15 [11,16–18], SBA-16 [19–21] and sponge mesoporous silica (SMS) [22–25] could guarantee high binding capacity and good site accessibility for target molecules by virtue of their large pore volumes and specific surface areas, well-defined pore sizes, and nanosized pore wall thicknesses. Particularly, the mesoporous silicas with highly interconnected 3-D pore structures such as SBA16 and SMS have been shown to be superior to those with 2-D channels (MCM-41 and SBA-15) for the acceleration of mass transfer [26,27]. SBA-16 possesses a higher degree of mesoscopic structural order than SMS, but it is much more difficult to prepare [25]. SMS has been fabricated through a self-assembly process between lecithin/dodecylamine mixed-micelles and tetraethoxysilane (TEOS) in an ethanolic/aqueous media [24]. A series of excellent features of SMS, such as very simple preparation procedure, extraordinarily high specific surface area and large pore volume, make it a promising material for molecular imprinting. Additionally, to overcome the limitations caused by template leakage existing in traditional non-covalent approach, a semicovalent approach was introduced. To date, the molecular imprinting of BPA in silica matrix based on non-covalent approach has been extensively studied [28–31], however, very few research works were focused on semi-covalent method. Currently, the commonly used semi-covalent approach in silica matrix is by forming the thermally reversible urethane bond between an isocyanate and a phenol, which is stable at room temperature but reversible cleavage occurs at elevated temperatures. Recently, this binding approach has been used for the imprinting of estrogen [32–34], testosterone [35], diethylstilbestrol (DES) [36], BPA [11,37,38] and DDT [39]. Lofgreen et al. [11] have used BPF, BPA and BPAF as the analytes for evaluating SPE performance of the BPA imprinted SBA-15, but no investigation about real samples was conducted. Xue et al. [38] determined only BPA in water samples using surface-imprinted nanoparticles, where BPAF and DES were just used as interferences to evaluate the selectivity for BPA. This study aims to fabricate a highly efficient BPA imprinted SMS and apply it for selective isolation of bisphenol A and its
Fig. 1. Molecular structures of BPA, BPF, BPB, BPE, BPAF, DDT, E2 and DES.
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
analogues from complex environmental sample. We chose spongelike mesoporous silica as the support matrix considering its easy synthetic procedure, high specific surface area and highly interconnected 3-D pore structure. Sediment is an important sink and reservoir of many contaminants and can be used to evaluate the applicability of optimized analysis procedures. Here, BPA, BPF, BPB, BPE, BPAF, dichlorodiphenyltrichloroethane (DDT), 17b-estradiol (E2) and DES (as shown in Fig. 1) were used to evaluate the binding properties and class selectivity of the prepared molecularly imprinted sponge mesoporous silica (MISMS) material. The MISMS–SPE purification coupled with HPLC–DAD detection method was successfully developed for selective isolation and determination of five BPs from sediment samples. 2. Experimental 2.1. Chemicals and materials Tetraethoxysilane (TEOS, 98%), (3-isocyanatopropyl) triethoxysilane (ICPTES, 96%), 3-aminopropyltriethoxysilane (APTES, 98%) and BPAF (99%) were purchased from J&K Scientific (Beijing, China). BPA (>99%), BPF (>99%), BPB (>98%), BPE (99%), E2 (>97%), lecithin from egg and dodecylamine (>97%) were all obtained from Tokyo Chemical Industry Co. (Tokyo, Japan). DDT was obtained from Wako Pure Chemical Industries. Ltd. (Osaka, Japan). DES (99%) was purchased from Dr. Ehrenstorfer GmbH (Augsburg, Germany). HPLC-grade methanol and acetonitrile (ACN) were supplied by Fisher Scientific (Fair Lawn, NJ, USA). Ethanol was obtained from Tianjin Hengxing Chemical Preparation Co. (Tianjin, China). Acetone was from Beijing Chemical Works (Beijing, China). Dimethyl sulfoxide (DMSO) was from Tianjin Fuchen Chemical Reagents Factory (Tianjin, China). Dichloromethane (DCM) and chloroform were obtained from Tianjin Tianhe Chemical Reagents Factory (Tianjin, China). HPLC-grade n-hexane was provided by J.T. Baker (Philipsburg, NJ, USA). HPLC-grade ethyl acetate, n-propanol and DMF were from Tianjin Kermel Chemical Reagent Co.
3
(Tianjin, China) and DMF was purified by distillation under reduced pressure after dried by calcium hydride. All solvents were analytical grade unless noted otherwise. The water used in all experiments was deionized water purified by a Milli-Q system (Millipore, Billerica, USA). The empty polypropylene SPE cartridges (3 mL) with polyethylene frits were from Dalian Sipore (Dalian, China). 2.2. Preparation of the MISMS and non-imprinted SMS (NISMS) 2.2.1. Synthesis of the BPA-ICPTES complex (BPA-Si) As shown in Scheme 1a, the template-monomer complex (BPA-Si) was synthesized as reported previously [35,40]. 2.28 g (10 mmol) of BPA was dissolved in 15 mL of purified DMF under nitrogen atmosphere. To this solution, 4.80 mL (20 mmol) of ICPTES was added gradually with stirring at room temperature. Then, the mixture was stirred for 72 h at 80 C under nitrogen atmosphere. After reaction, the product was isolated by a solvent evaporation step under reduced pressure below 90 C. The obtained transparent oily liquid was confirmed to be BPA-Si by FT-IR, 1H NMR (400 MHz, CDCl3) and 13C NMR (400 MHz, CDCl3), and the results were shown in the supporting information (Fig. S1). The yield was estimated to be 93% from 1H NMR. 2.2.2. Preparation of the MISMS and NISMS The BPA imprinted SMS was synthesized by a modified method based on the published literature [24]. In a typical procedure, 28 mL of water was added slowly under stirring to a solution containing 2.0 g of lecithin and 0.4 g of dodecylamine in 20.8 g of ethanol, until a homogeneous emulsion was formed. To this emulsion, 4.58 g (22 mmol) of TEOS predissolved with 1.45 g (2 mmol) of BPA-Si was added slowly under stirring for 15 min, and then the mixture was left under static conditions for 24 h at room temperature. The as-synthesized MISMS powder was filtrated, rinsed with ethanol, and then washed free of lecithin and dodecylamine by Soxhlet extraction with ethanol for 24 h. The obtained powder was washed with acetone, and dried overnight
Scheme 1. Schematic procedure for the preparation of BPA-Si (a) and MISMS (b).
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
4
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
under vacuum. For the synthesis of NISMS, the same procedure was adopted except that 4.58 g (22 mmol) of TEOS mixed with 0.89 g (4 mmol) of APTES was used as silica source. For template removal, 2.0 g of dry MISMS was first suspended in a mixture of 60 mL of DMSO and 12 mL of water, and then refluxed at 180 C for 5 h with stirring. The resulting MISMS was isolated by filtration, washed with water, ethanol and acetone in sequence, and dried overnight under vacuum. 2.3. Characterization of the MISMS The nitrogen adsorption–desorption isotherms were measured using a Quantachrome NOVA 4000 analyzer (Boynton Beach, FL, USA). Before the sorption measurements, the samples were outgassed for 8 h at 120 C. The specific surface area and pore volume were calculated using the standard Brunauer–Emmett– Teller (BET) method, and the pore size distribution was obtained using the Barrett–Joyner–Halenda (BJH) theory. The scanning electron microscopy (SEM) images were obtained using a Zeiss Supra 55 Sapphire microscope (Oberkochen, Germany). The transmission electron microscopy (TEM) images were taken by a JEOL JEM-2100 microscope (Tokyo, Japan) with 120 kV accelerating voltage. The FT-IR spectra were recorded on a PerkinElmer Spectrum GX Spectrometer (Waltham, MA, USA) in the 4000– 400 cm1 range with a resolution of 4 cm1. 2.4. Chromatographic conditions An Agilent 1200 series LC system (Santa Clara, CA, USA) consisting of a quaternary pump, an on-line degasser, an autosampler and a diode array detector (DAD), was used for the determination of BPs. Chromatographic separations were carried
out on an Agela Venusil MP-C18 column (5 mm, 250 mm 4.6 mm i.d., Tianjin, China). The mobile phase consisted of isocratic water/ methanol (30:70, v/v) for the analysis of single BPA. For the determination of five BPs, a gradient elution was adopted by combining solvent A (water) and solvent B (methanol) as follows: 35–100% B (25 min), 100% B (3 min), 100–35% B (1 min), 35% B (8 min). In both cases, the column temperature was kept at 25 C, the flow rate was 1 mL min1, and the detection wavelength was set at 225 nm. The injection volume was 5 mL for sorption experiments and 20 mL for real sample analysis. 2.5. Sorption experiments 20.0 mg of MISMS and NISMS particles were separately mixed with 2.0 mL of various concentrations of BPA (0.05–5.0 mmol L1) in n-hexane/n-propanol (95:5, v/v). The mixtures were shaken at 160 rpm for 24 h at 25 C. After binding, the mixtures were filtered through a 0.22 mm membrane, and 1.0 mL of the filtrates were dried under a moderate stream of nitrogen. After reconstituted in 1.0 mL of methanol, the free concentrations of BPA in the filtrates were determined by the HPLC method above. The adsorption capacity and the equilibrium dissociation constant (Kd, mmol L1) of MISMS were calculated according to the Eqs. (1) and (2) [6,9,41]: Q¼
ðC 0 C f Þv m
Q 1 Q ¼ Q þ max Cf Kd Kd
(1)
(2)
where C0 (mmol L1) and Cf (mmol L1) are the initial and final concentrations of BPA, v (L) is the total volume of the sample, m (g)
Fig. 2. SEM (a) and TEM ((b) inset: partial enlarged image) images of MISMS and schematic representation of the SMS pore structure (c).
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
is the mass of MISMS, Q and Qmax (mmol g1) are the amount of BPA adsorbed at equilibrium and saturation, respectively. The kinetic adsorption study was performed using similar procedures with the static adsorption experiment, except that 2.0 mmol L1 BPA in n-hexane/n-propanol (95:5, v/v) and different equilibrium periods (0.5–30 min) were adopted. The class selectivity was evaluated through similar procedures with the static adsorption experiment using the mixture standards of five BPs (BPA, BPF, BPB, BPE and BPAF) and three other analytes (DDT, E2 and DES) in n-hexane/n-propanol (95:5, v/v) (0.5 mmol L1 of each analyte). 2.6. Sample preparation and SPE procedure The sediment samples were collected from three different locations of Liaohe River (Liaoning, China). After lyophilization, 10 g dry weight (dw) of each sample (unspiked or spiked with 50 ng g1 dw and 500 ng g1 dw of each BP) were extracted by Soxhlet extraction with DCM for 24 h. The extracts were concentrated to around 10 mL by rotary evaporation. Triplicate experiments were conducted for each sample. The MISMS particles (100 mg) were packed into a 3 mL SPE cartridge using an upper frit and a lower frit to prevent sorbent loss. The cartridges were connected to a Supelco Visiprep SPE manifold (Bellefonte, PA, USA) equipped with a vacuum pump. Prior to extraction, the cartridges were equilibrated with 3 mL of DCM. Sample solutions were percolated at a constant flow rate of 1.0 mL min1. The cartridges were then rinsed with 4 mL of nhexane/n-propanol (95:5, v/v) and dried for 15 min under vacuum. Finally, the BPs retained on the sorbent were eluted with 3 mL of methanol. The eluates were evaporated to dryness under a moderate stream of nitrogen and the dry residues were reconstituted in 1.0 mL of methanol/water (35:65, v/v) for HPLC analysis. 3. Results and discussion 3.1. Preparation and characterization of the MISMS The MISMS for specifically recognizing BPA and its analogues was synthesized by combining semi-covalent molecular imprinting and a simple self-assembly process. As illustrated in Scheme 1b, BPA was trapped into the silica matrix via the urethane bond formed between an isocyanate and a phenol group. After heat treatment, BPA was removed due to a reversible cleavage of the urethane bond, and the dissociated isocyanato group was converted to amino group by its reaction with water. Thus, the delicate binding sites were formed in the silica matrix. The successful formation of imprinting sites was confirmed by FT-IR spectroscopy (see Fig. S2 in the supporting information). Before template removal, the carbonyl group band at 1712 cm1 is observed. After template removal, this band disappears and the N— H (1 amines) bending adsorption at 1560 cm1 appears, revealing the successful formation of imprinting sites. The two bands for N— H (1 amines) stretching vibration should also be observed at 3500–3200 cm1. Unfortunately, the signal in this range was overlapped with the broad band of O—H in silica matrix. The morphology of MISMS was investigated by SEM and TEM measurements. The SEM image shows the formation of irregular shaped particles with the mm-level particle sizes, which are suitable as SPE packing materials (see Fig. 2a). As shown in Fig. 2b (inset: partial enlarged image), the TEM image reveals an evident interconnected 3-D porous network, which was similar to the schematic representation of the SMS pore structure (Fig. 2c) [24]. The porosity of MISMS was also investigated by nitrogen adsorption–desorption measurement. The MISMS particles
5
showed typical “type IV” nitrogen adsorption–desorption isotherm, which is associated with capillary condensation taking place in mesoporous materials. The BET specific surface area obtained from the nitrogen isotherm was 850.55 m2 g1 and the total pore volume was 1.20 cm3 g1. The BJH pore size distribution calculated from the adsorption branch of the isotherm showed uniform mesopores with an average diameter of 3.53 nm. 3.2. Binding properties of the MISMS and NISMS 3.2.1. Effect of solvents on BPA binding performance To optimize the binding performance of MISMS, different types of organic solvents were investigated for BPA adsorption. In detail, 20.0 mg of MISMS and NISMS were separately equilibrated with 2.0 mL of BPA (2.0 mmol L1) in six different solvents, DCM, chloroform, n-hexane/n-propanol (95:5, v/v), n-propanol, ethyl acetate and ACN, and their adsorption amounts were determined. As shown in Fig. 3, both MISMS and NISMS show high binding affinities for BPA in nonpolar solvents (DCM and chloroform). In contrast, more polar solvents like n-propanol, ethyl acetate and ACN tend to weaken the interactions between the silica matrix and BPA, resulting in reduced binding affinities. However, the selectivity of MISMS for BPA can not be well demonstrated by using DCM and chloroform with respect to the slight distinction of adsorption amounts between MISMS and NISMS. Polar protic solvents with hydrogen bond forming ability (e.g., n-propanol) could reduce the binding affinity of NISMS more by competitively binding to the non-imprinting sites. After optimization, a mixed solvent of n-hexane/n-propanol (95:5, v/v) was chosen for the experiments to get a good compromise between the binding affinity and selectivity. 3.2.2. Static and kinetic adsorptions As can be seen from Fig. 4a, the adsorption amounts of MISMS and NISMS increase with increasing BPA initial concentration, and higher adsorption capacity of the MISMS is presented over the tested concentration range (0.05–5.0 mmol L1). This is usually indicative of the presence of recognition sites created by the molecular imprinting process. The binding data of MISMS were then processed using Scatchard analysis (Eq. (2)), and two straight lines were obtained in the plot region (Fig. 4b). The results indicate that the binding sites in MISMS are heterogeneous and can be classified into two distinct groups: the high-affinity (specific) and
Fig. 3. Effect of different solvents on BPA (2.0 mmol L1) adsorption amounts (Q) for MISMS and NISMS. Mixed solvent is n-hexane/n-propanol (95:5, v/v).
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
6
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
Fig. 4. (a) BPA binding isotherms for MISMS and NISMS; (b) Scatchard plot for MISMS; (c) kinetic binding profiles of BPA (2.0 mmol L1) for MISMS and NISMS.
the low-affinity (non-specific) binding sites. The linear regression equations for the two straight lines are Q/Cf = 129.59 0.7658Q (r = 0.9946) and Q/Cf = 93.82 0.0692Q (r = 0.9540). The Kd and Qmax values can be calculated according to the slopes and intercepts of the two linear regression equations. The Kd and Qmax values are calculated to be 1.31 mmol L1 and 169.22 mmol g1 for the specific binding sites (upper line) and 14.45 mmol L1 and 1355.71 mmol g1 for the non-specific binding sites (lower line), respectively. The non-specific adsorption is mainly contributed by the Si—OH group and is difficult to be eliminated from the silica matrix. Although the Qmax (1355.71 mmol g1) for the non-specific adsorption of MISMS is much higher than or comparable to some traditional polymer and silica materials (0 mmol g1 [8], 700 mmol g1 [10] and 45.22 mmol g1 [41] for polymers; 805.87 mmol g1 [28] for silica), the Qmax value (169.22 mmol g1) for the specific binding sites created by the semi-covalent imprinting process is still superior to those materials. Furthermore, the specific adsorption is predominant in a wide concentration range of 0.05–1.0 mmol L1, which covers most sample concentrations for BPA analysis. Besides excellent binding capacity, fast adsorption kinetics for BPA (2.0 mmol L1) was also observed. As shown in Fig. 4c, the binding equilibrium for both MISMS and NISMS is reached within 3 min, which is a remarkable improvement compared to some traditional MIPs that required up to 10 min [8] and even 5 h [10] to reach equilibrium. The BPA binding amount at equilibrium for MISMS is almost 2-fold that of NISMS. The spongy mesoporous
silica prepared in this work features an isotropic 3-D pore structure. So the fast adsorption kinetics might be owing to the open diffusion channel and the highly accessible binding sites. Therefore, the superior binding capacity and fast adsorption kinetics make the prepared MISMS a promising and competitive sorbent for rapid and selective isolation of BPA from complex matrices. 3.2.3. Class selectivity The class selectivity of MISMS was assessed by evaluating the binding capacities for five BPs (BPA, BPF, BPB, BPE and BPAF) and three other analytes (DDT, E2 and DES) of MISMS and NISMS. Fig. 5 illustrates the data obtained from the selectivity experiment for both MISMS and NISMS, concerning the adsorption amounts and the ratios between QMISMS and QNISMS. The MISMS exhibits obviously higher adsorption capacity than NISMS for the five tested BPs due to the presence of the specific binding sites. Due to the lack of strong functional groups like the hydroxyl groups of BPs, the adsorption amounts of DDT for MISMS and NISMS are much lower than BPs. However, MISMS still demonstrates good selectivity for DDT with respect to the ratio between QMISMS and QNISMS. This might be owing to the very close molecular structure of DDT with the BPs (especially BPE). And BPE has been used as template on molecular imprinting for the selective detection of DDT in the published literature [39]. For E2 and DES, poor selectivities of MISMS were observed with respect to small distinctions and low ratios of the adsorption
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
7
Fig. 5. Adsorption amounts of MISMS and NISMS and ratios between QMISMS and QNISMS for eight analytes. Fig. 6. Effect of washing solvent (4 mL) on BPs recoveries for MISMS and NISMS.
amounts between MISMS and NISMS. This result shows that the MISMS has good class selectivity towards the five investigated BPs. 3.3. Optimization of SPE procedure The mass capacity of the MISMS cartridge was determined by percolating different amounts of BPs standards (5, 7.5, 10, 15 and 25 mg for each BP) in 10 mL of DCM. The BPs recoveries with values above 80% did not differ significantly over the tested range. Satisfactory recoveries were still obtained when the total loading amount of BPs reached a high value of 125 mg (25 mg for each BP), indicating that the MISMS sorbent had good mass capacity for the BPs. Considering the concentration levels of BPs in environmental samples, 50 mg (10 mg for each BP) of total loading amount was selected to optimize other factors. To realize the elimination of co-adsorbed matrix components without loss of target analytes in real sample analysis, an appropriate washing step was introduced. Since it exhibited appropriate polarity in the sorption experiments, the mixed nhexane/n-propanol (95:5, v/v) solvent was adopted as the washing solvent. The effect of washing volume on BPs recoveries was investigated by loading 50 mg of BPs mixture standards into the MISMS and NISMS SPE cartridges, followed by rinsing with different volumes of washing solvent. Without any washing step, the recoveries of BPs for both MISMS and NISMS were higher than 80% due to the existence of non-specific adsorption. After a 4 mL of n-hexane/n-propanol (95:5, v/v) washing step, the BPs recoveries for NISMS decreased obviously (especially for BPA and BPB) due to the reduced non-specific adsorption, as shown in Fig. 6. In comparison, the recoveries of BPs for MISMS kept almost constant due to the existence of significant specific adsorption. These results further confirm the good class selectivity of MISMS for the five BPs as discussed in Section 3.2.3. However, an obvious reduction of the BPs recoveries for MISMS was observed when the washing volume was higher than 4 mL. Thus, 4 mL was selected as the appropriate washing volume. Before elution, the SPE cartridge was dried with the aid of a vacuum pump for 15 min. The final elution of BPs was conducted by using 3 mL of methanol or 4 mL of ACN. Satisfactory recoveries (>80% for all BPs) were achieved when any of the two solvents was used. Considering that low boiling point and small volume were beneficial to reduce the drying time, 3 mL of methanol was finally adopted.
3.4. Analysis of spiked environmental sediment samples Twelve different concentrations of BPs (0.05–30 mg mL1) were tested to determine the linearity. Satisfactory results were obtained with a correlation coefficient higher than 0.9998 for all BPs, and the limits of detection (LOD) for BPs ranged from 0.43 to 0.71 ng g1 dw (4.3–7.1 ng mL1) (Table 1). The performance of the present SPE–HPLC–DAD method was compared with other methods for BPA determination in the literature (Table 2). Although the detection limit of this method is at the same level with or higher than other methods, the new sample type (sediment) and higher Qmax still make this method competitive and promising. Furthermore, sediment sample with enough mass could be a good indicator of pollution status and the high Qmax of this material could cover the wide linear range well for samples with a large mass. In the present work, a large mass of 10 g of sediment sample was adopted. As shown in Fig. 7a, the BPs standards (peak identifications: 1, BPF; 2, BPE; 3, BPA; 4, BPB; 5, BPAF) were well separated using the developed HPLC method. The existence of many matrix interferences in sediment samples was demonstrated in Fig. 7b. Both good selectivity towards BPs and clean-up effect of the MISMS were verified by the fact that the spiked BPs were well extracted from the sample matrix and most of interfering matrix components were removed (Fig. 7c). To demonstrate the reliability of the developed SPE procedure, the BPs recoveries were determined at two different spiking levels for the lyophilizated sediment samples without and with SPE. Before analysis, the three sediment samples were verified to be free of the five investigated BPs and could be used as blank samples. Due to the severe matrix interferences, unsatisfactory results (low BPs recoveries and high RSD values) were obtained when samples were analyzed without the developed SPE procedure (Table 1). Satisfactory recoveries of BPs were obtained when the SPE procedure was applied to the sample analysis. As can be seen from Table 1, the BPs recoveries after SPE were in the range from 75.5% to 105.5%, with RSD values below 7.5%. The results showed that a reliable and practical analytical procedure based on the MISMS–SPE purification coupled with HPLC–DAD detection was successfully established for simultaneous determination of five BPs from environmental sediment samples.
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
8
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
Table 1 Limits of detections (LODs), correlation coefficients (r) for BPs and average recoveries, relative standard deviations (RSDs, n = 3) of BPs from lyophilizated sediment samples at two different spiking levels without and with SPE procedure. Analyte LOD (ng g1 dw)
Correlation coefficient (r)
Spiked level (ng g1 dw)
SPE
Sample 1 Recovery (%)
BPF
0.43
0.9998
50 500
– BPE
0.45
0.9998
50 500
– BPA
0.46
0.9998
50 500
– BPB
0.43
0.9999
50 500
– BPAF
0.71
0.9999
50 500
Sample 2 RSD (%, n = 3)
Recovery (%)
Sample 3 RSD (%, n = 3)
Recovery (%)
RSD (%, n = 3)
Without With Without With
33.9 96.7 43.0 88.8
11.8 3.1 16.9 4.0
22.4 85.5 39.7 85.1
9.5 5.5 14.3 1.6
37.1 95.3 65.7 94.1
7.2 1.7 13.8 1.0
Without With Without With
48.3 95.9 46.8 79.9
15.4 2.7 27.1 3.6
26.8 83.1 43.2 79.5
10.0 3.4 25.1 3.3
37.5 90.0 67.7 85.6
14.2 2.8 15.4 1.4
Without With Without With
41.6 94.8 48.7 77.0
16.8 4.5 24.9 4.1
35.2 83.0 28.5 81.2
12.8 5.3 18.4 4.6
49.8 90.3 59.5 86.2
20.3 4.9 20.7 1.6
Without 53.3 With 103.1 Without 45.9 With 81.0
12.7 3.0 6.6 7.5
40.1 101.9 34.0 87.2
15.3 3.8 19.1 3.6
55.9 104.2 51.5 91.4
19.0 3.9 15.9 1.2
Without 89.5 With 104.6 Without 60.2 With 75.5
10.2 4.7 24.3 6.6
71.1 98.0 38.4 90.4
14.9 2.0 10.5 2.0
64.3 105.5 41.0 94.7
20.6 6.2 9.8 2.1
Table 2 Comparison of the proposed method with other analytical methods based on molecularly imprinted materials for BPA determination in various samples. Analytical methods
Material type
Matrix
Qmax
Linear range
LOD
References
SPE–HPLC–UV SPE–HPLC–FLDa SPE–UV SPE–HPLC–DAD MSPE–LC–FLDb SPE–HPLC–FLD SPE–HPLC–DAD
MI hollow porous polymer MI silica nanoparticlesc MI silica nanoparticles MI polymer MI magnetic polymerparticles MI silica nanoparticles MISMS
Tap water Chemical cleansingand cosmetics Tap water River water River water Fish River sediment
832.7 mmol g1) 822.3 mmol g1 30.26 mmol g1 57.95 mmol 1 3.55 mmol g1 39.4 mmol g1 1524.93 mmol g1
0.005–100 mg mL1 0.684–114 mg L1 – 0.02–2.0 mg L1 0.01–0.2 mg L1 0.7–114.1 mg L1 0.05–30 mg mL1
3 ng mL1 0.23 ng mL1 – 2.5 ng L1 10 ng L1 0.11 ng mL1 4.6 ng mL1
[9] [28] [31] [41] [42] [43] This work
a b c
FLD, fluorescence detector. MSPE, magnetic solid phase extraction. MI, molecularly imprinted.
4. Conclusions
Fig. 7. Chromatograms of the standards and real samples. (a) Standard mixtures of five BPs (5 mg mL1 for each BP); (b) extract of spiked sediment (500 ng g1 dw of each BP) before MISMS–SPE; (c) extract of spiked sediment (500 ng g1 dw of each BP) after MISMS–SPE. Peak identifications: 1, BPF; 2, BPE; 3, BPA; 4, BPB; 5, BPAF.
In this work, a sponge-like BPA imprinted mesoporous silica material was synthesized via the thermally reversible urethane bonds formed between isocyanates and phenol groups using lecithin/dodecylamine mixed-micelles as template and TEOS as cross-linker. When used as sorbent for BPA and its structural analogues, the synthesized MISMS showed high specific adsorption capacity and selectivity, as well as fast adsorption kinetics. The excellent performance of MISMS may be attributed to the combination of a sufficiently high specific surface area with an open 3-D pore architecture, which remains highly accessible. Furthermore, the developed SPE procedure using MISMS showed good selectivity and clean-up effect for isolation of BPs from sediment extracts. The developed MISMS–SPE coupled with HPLC–DAD method was accurate, sensitive and had high reliability for determination of BPs in environmental sediment samples. Further improvement of imprinting efficiency in terms of selectivity may be achieved by suppression of the nonspecific adsorption from silica matrix.
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
Acknowledgements Financial supports from the National High Technology Research and Development Program of China (863 program) (2013AA065203) and National Natural Science Foundation of China (Grant Nos. 21277139,21205117) are gratefully acknowledged. 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.aca.2014.09.051. References [1] C.Y. Liao, F. Liu, Y. Guo, H.B. Moon, H. Nakata, Q. Wu, K. Kannan, Occurrence of eight bisphenol analogues in indoor dust from the united states and several asian countries: implication for human exposure, Environ. Sci. Technol. 46 (2012) 9138–9145. [2] C.Y. Liao, F. Liu, H.B. Moon, N. Yamashita, S. Yun, K. Kannan, Bisphenol analogues in sediments from industrialized areas in the United States, Japan, and Korea: spatial and temporal distributions, Environ. Sci. Technol. 46 (2012) 11558–11565. [3] C.Y. Liao, F. Liu, H. Alomirah, V.D. Loi, M.A. Mohd, H.B. Moon, H. Nakata, K. Kannan, Bisphenol S in urine from the United States and seven Asian countries: occurrence and human exposures, Environ. Sci. Technol. 46 (2012) 6860–6866. [4] C.Y. Liao, F. Liu, K. Kannan, Bisphenol S, a new bisphenol analogue, in paper products and currency bills and its association with bisphenol A residues, Environ. Sci. Technol. 46 (2012) 6515–6522. [5] H. Gallart-Ayala, E. Moyano, M.T. Galceran, Analysis of bisphenols in soft drinks by on-line solid extraction fast liquid chromatography–tandem mass spectrometry, Anal. Chim. Acta 683 (2011) 227–233. [6] H. Zhou, Y.P. Xu, H.W. Tong, Y.X. Liu, F. Han, X.Y. Yan, S.M. Liu, Direct synthesis of surface molecularly imprinted polymers based on vinyl–SiO2 nanospheres for recognition of bisphenol A, J. Appl. Polym. Sci. 128 (2013) 3846–3852. [7] P.P. Qi, J.C. Wang, L.D. Wang, Y. Li, J. Jin, F. Su, Y.Z. Tian, J.P. Chen, Molecularly imprinted polymers synthesized via semi-covalent imprinting with sacrifical spacer for imprinting phenols, Polymer 51 (2010) 5417–5423. [8] T.H. Jiang, L.X. Zhao, B.L. Chu, Q.Z. Feng, W. Yan, J.M. Lin, Molecularly imprinted solid-phase extraction for the selective determination of 17b-estradiol in fishery samples with high performance liquid chromatography, Talanta 78 (2009) 442–447. [9] J. Li, X.B. Zhang, Y.X. Liu, H.W. Tong, Y.P. Xu, S.M. Liu, Preparation of a hollow porous molecularly imprinted polymer using tetrabromobisphenol A as a dummy template and its application as SPE sorbent for determination of bisphenol A in tap water, Talanta 117 (2013) 281–287. [10] X.L. Mou, W.P. Zhang, Z.L. Chen, Highly selective extraction of baicalin in natural herbs and medicinal preparations by molecularly imprinted polymer, J. Appl. Polym. Sci. 130 (2013) 1873–1878. [11] J.E. Lofgreen, I.L. Moudrakovski, G.A. Ozin, Molecularly imprinted mesoporous organo silica, ACS Nano 5 (2011) 2277–2287. [12] L.X. Chen, S.F. Xu, J.H. Li, Recent advances in molecular imprinting technology: current status, challenges and highlighted applications, Chem. Soc. Rev. 40 (2011) 2922–2942. [13] E.C. Morais, G.G. Correa, R. Brambilla, J.H.Z. dos Santos, A.G. Fisch, Selective silica-based sorbent materials synthesized by molecular imprinting for adsorption of pharmaceuticals in aqueous matrices, J. Sep. Sci. 36 (2013) 636–643. [14] D. Lewandowski, A. Olejnik, G. Schroeder, Adsorption studies and release of selected dyes from functionalized mesoporous MCM-41 silica, Cent. Eur. J. Chem. 12 (2014) 233–241. [15] Y. San, H.S. Li, M.H. Zhu, K. Ma, Q.Z. Jiao, Q. Wu, Catalytic performance of metal ion doped MCM-41 for methanol dehydration to dimethyl ether, J. Porous Mater. 20 (2013) 1509–1518. [16] A. Mehdinia, M. Ahmadifar, M.O. Aziz-Zanjani, A. Jabbari, M.S. Hashtroudi, Selective adsorption of 2,4-dinitrophenol on molecularly imprinted nanocomposites of mesoporous silica SBA-15/polyaniline, Analyst 137 (2012) 4368–4374. [17] A. Mehdinia, M.O. Aziz-Zanjani, M. Ahmadifar, A. Jabbari, Design and synthesis of molecularly imprinted polypyrrole based on nanoreactor SBA-15 for recognition of ascorbic acid, Biosens. Bioelectron. 39 (2013) 88–93. [18] Y. Liu, Z.C. Liu, J. Gao, J.D. Dai, J. Han, Y. Wang, J.M. Xie, Y.S. Yan, Selective adsorption behavior of Pb(II) by mesoporous silica SBA-15-supported Pb(II)imprinted polymer based on surface molecularly imprinting technique, J. Hazard. Mater. 186 (2011) 197–205.
9
[19] X.M. Xue, F.T. Li, Removal of Cu(II) from aqueous solution by adsorption onto functionalized SBA-16 mesoporous silica, Micropor. Mesopor. Mater. 116 (2008) 116–122. [20] L. Sierra, S. Valange, J.L. Guth, Formation mechanism and morphology of mesoporous SBA-16 type silica particles prepared with triblock copolymer surfactant PEO140PPO39PEO140, Micropor. Mesopor. Mater. 124 (2009) 100–109. [21] J. Hwang, H. Daiguji, Proton transport in mesoporous silica SBA-16 thin films with three-dimensional cubic structures, Langmuir 29 (2013) 2406–2411. [22] M. Mureseanu, A. Galarneau, G. Renard, F. Fajula, A new mesoporous micelletemplated silica route for enzyme encapsulation, Langmuir 21 (2005) 4648–4655. [23] A. Galarneau, M. Mureseanu, S. Atger, G. Renard, F. Fajula, Immobilization of lipase on silicas. Relevance of textural and interfacial properties on activity and selectivity, New J. Chem. 30 (2006) 562–571. [24] A. Galarneau, G. Renard, M. Mureseanu, A. Tourrette, C. Biolley, M. Choi, R. Ryoo, F.D. Renzo, F. Fajula, Synthesis of sponge mesoporous silicas from lecithin/dodecylamine mixed-micelles in ethanol/water media: a route towards efficient biocatalysts, Micropor. Mesopor. Mater. 104 (2007) 103–114. [25] A. Galarneau, F. Sartori, M. Cangiotti, T. Mineva, F.D. Renzo, M.F. Ottaviani, Sponge mesoporous silica formation using disordered phospholipid bilayers as template, J. Phys. Chem. B 114 (2010) 2140–2152. [26] F. Kleitz, T.W. Kim, R. Ryoo, Phase domain of the cubic Im m mesoporous silica in the EO106PO70EO106-butanol–H2O system, Langmuir 22 (2006) 440–445. [27] F. Kleitz, D. Liu, G.M. Anilkumar, I.S. Park, L.A. Solovyov, A.N. Shmakov, R. Ryoo, Large cage face-centered-cubic Fm3m mesoporous silica: synthesis and structure, J. Phys. Chem. B 107 (2003) 14296–14300. [28] R. Zhu, W.H. Zhao, M.J. Zhai, F.D. Wei, Z. Cai, N. Sheng, Q. Hu, Molecularly imprinted layer-coated silica nanoparticles for selective solid-phase extraction of bisphenol A from chemical cleansing and cosmetics samples, Anal. Chim. Acta 658 (2010) 209–216. [29] Y.Q. Wang, Y.Y. Yang, L. Xu, J. Zhang, Bisphenol A sensing based on surface molecularly imprinted, ordered mesoporous silica, Electrochim. Acta 56 (2011) 2105–2109. [30] X.M. Jiang, W. Tian, C.D. Zhao, H.X. Zhang, M.C. Liu, A novel sol–gel-material prepared by a surface imprinting technique for the selective solid-phase extraction of bisphenol A, Talanta 72 (2007) 119–125. [31] Y.M. Ren, W.Q. Ma, J. Ma, Q. Wen, J. Wang, F.B. Zhao, Synthesis and properties of bisphenol A molecular imprinted particle for selective recognition of BPA from water, J. Colloid Interface Sci. 367 (2012) 355–361. [32] C.D. Ki, C. Oh, S.G. Oh, J.Y. Chang, The use of a thermally reversible bond for molecular imprinting of silica spheres, J. Am. Chem. Soc. 124 (2002) 14838– 14839. [33] H.H. Yang, S.Q. Zhang, W. Yang, X.L. Chen, Z.X. Zhuang, J.G. Xu, X.R. Wang, Molecularly imprinted sol–gel nanotubes membrane for biochemical separations, J. Am. Chem. Soc. 126 (2004) 4054–4055. [34] X. Wang, L.Y. Wang, X.W. He, Y.K. Zhang, L.X. Chen, A molecularly imprinted polymer-coated nanocomposite of magnetic nanoparticles for estrone recognition, Talanta 78 (2009) 327–332. [35] C.Y. He, Y.Y. Long, J.L. Pan, K. Li, F. Liu, Molecularly imprinted silica prepared with immiscible ionic liquid as solvent and porogen for selective recognition of testosterone, Talanta 74 (2008) 1126–1131. [36] B.M. Jung, M.S. Kim, W.J. Kim, J.Y. Chang, Molecularly imprinted mesoporous silica particles showing a rapid kinetic binding, Chem. Commun. 46 (2010) 3699–3701. [37] Y.D. Kim, J.B. Jeon, J.Y. Chang, CdSe quantum dot-encapsulated molecularly imprinted mesoporous silica particles for fluorescent sensing of bisphenol A, J. Mater. Chem. 22 (2012) 24075–24080. [38] J.Q. Xue, D.W. Li, L.L. Qu, Y.T. Long, Surface-imprinted core–shell Au nanoparticles for selective detection of bisphenol A on surface-enhanced Raman scattering, Anal. Chim. Acta 777 (2013) 57–62. [39] A.L. Graham, C.A. Carlson, P.L. Edmiston, Development and characterization of molecularly imprinted sol–gel materials for the selective detection of DDT, Anal. Chem. 74 (2002) 458–467. [40] M.U. de la Orden, J. Arranz, V. Lorenzo, E. Pérez, J.M. Urreaga, Study of the effects of the reaction conditions on the modification of clays with polyelectrolytes and silanes, J. Colloid Interface Sci. 342 (2010) 185–191. [41] J.F. Yin, Z.H. Meng, Y.S. Zhu, M.Y. Song, H.L. Wang, Dummy molecularly imprinted polymer for selective screening of trace bisphenols in river water, Anal. Methods 3 (2011) 173–180. [42] Y. Hiratsuka, N. Funaya, H. Matsunaga, J. Haginaka, Preparation of magnetic molecularly imprinted polymers for bisphenol A and its analogues and their application to the assay of bisphenol A in river water, J. Pharm. Biomed. Anal. 75 (2013) 180–185. [43] F. Wei, X. Liu, M. Zhai, Z. Cai, G. Xu, J. Yang, S. Du, Q. Hu, Molecularly imprinted nanosilica solid-phase extraction for bisphenol A in fish samples, Food Anal. Methods 6 (2013) 415–420.
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
Analytica Chimica Acta xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca
Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts Jiajia Yang a,b , Yun Li a , Jincheng Wang a , Xiaoli Sun a,b , Syed Mazhar Shah a , Rong Cao a,b , Jiping Chen a, * a Key Laboratory of Separation Sciences for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Dalian 116023, China b University of Chinese Academy of Sciences, Beijing 100049, China
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
Novel sponge-like molecularly imprinted mesoporous silica was synthesized. Extraordinarily large specific surface area and highly interconnected 3-D porous network. High specific adsorption capacity and fast adsorption kinetics for BPA. Good class-selectivity and clean-up effect for bisphenols in sediment under SPE mode. Good recoveries and sensitivity for bisphenols using the MISMS–SPE coupled with HPLC–DAD method.
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 June 2014 Received in revised form 29 August 2014 Accepted 29 September 2014 Available online xxx
Bisphenol A (BPA) imprinted sponge mesoporous silica was synthesized using a combination of semi-covalent molecular imprinting and simple self-assembly process. The molecularly imprinted sponge mesoporous silica (MISMS) material obtained was characterized by FT-IR, scanning electron microscopy, transmission electron microscopy, and nitrogen adsorption–desorption measurements. The results show that the MISMS possessed a large specific surface area (850.55 m2 g1) and a highly interconnected 3-D porous network. As a result, the MISMS demonstrated a superior specific adsorption capacity of 169.22 mmol g1 and fast adsorption kinetics (reaching equilibrium within 3 min) for BPA. Good class selectivity for BPA and its analogues (bisphenol F, bisphenol B, bisphenol E and bisphenol AF) was also demonstrated by the sorption experiment. The MISMS as solid-phase extraction (SPE) material was then evaluated for isolation and clean-up of these bisphenols (BPs) from sediment samples. An accurate and sensitive analytical method based on the MISMS–SPE coupled with HPLC–DAD has been successfully established for simultaneous determination of five BPs in river sediments with detection
Keywords: Molecularly imprinted Sponge mesoporous silica Solid-phase extraction Class-selectivity Bisphenols
Abbreviations: SMS, sponge mesoporous silica; MISMS, molecularly imprinted sponge mesoporous silica; NISMS, non-imprinted sponge mesoporous silica; TEOS, tetraethoxysilane; ICPTES, (3-isocyanatopropyl)triethoxysilane; APTES, 3-aminopropyltriethoxysilane; ACN, acetonitrile; DCM, dichloromethane; BET, Brunauer–Emmett– Teller; BJH, Barrett–Joyner–Halenda; TEM, transmission electron microscopy; BPs, bisphenols; BPA, bisphenol A; BPF, bisphenol F; BPB, bisphenol B; BPE, bisphenol E; BPAF, bisphenol AF; DDT, dichlorodiphenyltrichloroethane; E2, 17b-estradiol; DES, diethylstilbestrol; BPA-Si, BPA-ICPTES; dw, dry weight. * Corresponding author. Tel./fax: +86 411 8437 9562. E-mail address: [email protected] (J. Chen). http://dx.doi.org/10.1016/j.aca.2014.09.051 0003-2670/ ã 2014 Published by Elsevier B.V.
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
2
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
limits of 0.43–0.71 ng g1 dry weight (dw). The recoveries of BPs for lyophilizated sediment samples at two spiking levels (50 and 500 ng g1 dw for each BP) were in the range of 75.5–105.5% with RSD values below 7.5%. ã 2014 Published by Elsevier B.V.
1. Introduction Bisphenols (BPs) are a group of chemicals with two hydroxyphenyl functionalities and include several analogues such as bisphenol A (BPA), bisphenol F (BPF), bisphenol B (BPB), bisphenol E (BPE) and bisphenol AF (BPAF) [1]. In recent years, there has been a growing concern worldwide about the occurrence of BPs in the environment due to their potential adverse effects on human health [2–5]. BPA was known as an endocrine-disrupting chemical and its adverse effects on reproduction and development, neural networks, cardiovascular, metabolic, and immune systems in human have been well documented [3,4]. Its analogues (e.g., BPF, BPE and BPB) have shown moderate to slight acute toxicity and an estrogenic activity similar to BPA [5]. Therefore, the determination of BPs in environmental samples is essential for environmental health and human exposure assessments. Due to the complexity of environmental matrixes and the low concentrations of environmental pollutants, the development of highly specific or selective sample preparation techniques for BPs would be highly desirable. Molecular imprinting is known as a technique for generating tailor-made recognition sites complementary in shape, size and functional groups to the template molecules [6]. In general, three different strategies, namely covalent, non-covalent and semicovalent imprinting, have been employed to prepare molecularly imprinted materials, and non-covalent imprinting is the most used strategy owing to its experimental simplicity [7]. Due to some merits like good chemical stability and reusability, the majority of the research works for molecular imprinting, especially for solidphase extraction (SPE) applications [8–10], were focused on organic polymers. Nevertheless, molecularly imprinted polymers (MIPs) tend to swell or shrink when exposed to organic solvents [11]. The MIPs prepared with bulk polymerization or precipitation polymerization using non-covalent approach also exhibited many other drawbacks, such as low binding capacity, poor site accessibility, slow mass transfer and template leakage. These problems are mostly attributed to the fact that the recognition sites and residual template molecules are deeply embedded in the highly cross-linked polymer matrix [12]. Such shortcomings of MIPs have become one of the major bottlenecks in the development of molecular imprinting technology where extensive research of ideal materials is imperative.
With respect to organic polymer, silica materials are a good alternative for molecular imprinting due to its excellent mechanical resistance and non-swelling, rigid micro-structures [13]. Furthermore, mesoporous silicas such as MCM-41 [14,15], SBA15 [11,16–18], SBA-16 [19–21] and sponge mesoporous silica (SMS) [22–25] could guarantee high binding capacity and good site accessibility for target molecules by virtue of their large pore volumes and specific surface areas, well-defined pore sizes, and nanosized pore wall thicknesses. Particularly, the mesoporous silicas with highly interconnected 3-D pore structures such as SBA16 and SMS have been shown to be superior to those with 2-D channels (MCM-41 and SBA-15) for the acceleration of mass transfer [26,27]. SBA-16 possesses a higher degree of mesoscopic structural order than SMS, but it is much more difficult to prepare [25]. SMS has been fabricated through a self-assembly process between lecithin/dodecylamine mixed-micelles and tetraethoxysilane (TEOS) in an ethanolic/aqueous media [24]. A series of excellent features of SMS, such as very simple preparation procedure, extraordinarily high specific surface area and large pore volume, make it a promising material for molecular imprinting. Additionally, to overcome the limitations caused by template leakage existing in traditional non-covalent approach, a semicovalent approach was introduced. To date, the molecular imprinting of BPA in silica matrix based on non-covalent approach has been extensively studied [28–31], however, very few research works were focused on semi-covalent method. Currently, the commonly used semi-covalent approach in silica matrix is by forming the thermally reversible urethane bond between an isocyanate and a phenol, which is stable at room temperature but reversible cleavage occurs at elevated temperatures. Recently, this binding approach has been used for the imprinting of estrogen [32–34], testosterone [35], diethylstilbestrol (DES) [36], BPA [11,37,38] and DDT [39]. Lofgreen et al. [11] have used BPF, BPA and BPAF as the analytes for evaluating SPE performance of the BPA imprinted SBA-15, but no investigation about real samples was conducted. Xue et al. [38] determined only BPA in water samples using surface-imprinted nanoparticles, where BPAF and DES were just used as interferences to evaluate the selectivity for BPA. This study aims to fabricate a highly efficient BPA imprinted SMS and apply it for selective isolation of bisphenol A and its
Fig. 1. Molecular structures of BPA, BPF, BPB, BPE, BPAF, DDT, E2 and DES.
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
analogues from complex environmental sample. We chose spongelike mesoporous silica as the support matrix considering its easy synthetic procedure, high specific surface area and highly interconnected 3-D pore structure. Sediment is an important sink and reservoir of many contaminants and can be used to evaluate the applicability of optimized analysis procedures. Here, BPA, BPF, BPB, BPE, BPAF, dichlorodiphenyltrichloroethane (DDT), 17b-estradiol (E2) and DES (as shown in Fig. 1) were used to evaluate the binding properties and class selectivity of the prepared molecularly imprinted sponge mesoporous silica (MISMS) material. The MISMS–SPE purification coupled with HPLC–DAD detection method was successfully developed for selective isolation and determination of five BPs from sediment samples. 2. Experimental 2.1. Chemicals and materials Tetraethoxysilane (TEOS, 98%), (3-isocyanatopropyl) triethoxysilane (ICPTES, 96%), 3-aminopropyltriethoxysilane (APTES, 98%) and BPAF (99%) were purchased from J&K Scientific (Beijing, China). BPA (>99%), BPF (>99%), BPB (>98%), BPE (99%), E2 (>97%), lecithin from egg and dodecylamine (>97%) were all obtained from Tokyo Chemical Industry Co. (Tokyo, Japan). DDT was obtained from Wako Pure Chemical Industries. Ltd. (Osaka, Japan). DES (99%) was purchased from Dr. Ehrenstorfer GmbH (Augsburg, Germany). HPLC-grade methanol and acetonitrile (ACN) were supplied by Fisher Scientific (Fair Lawn, NJ, USA). Ethanol was obtained from Tianjin Hengxing Chemical Preparation Co. (Tianjin, China). Acetone was from Beijing Chemical Works (Beijing, China). Dimethyl sulfoxide (DMSO) was from Tianjin Fuchen Chemical Reagents Factory (Tianjin, China). Dichloromethane (DCM) and chloroform were obtained from Tianjin Tianhe Chemical Reagents Factory (Tianjin, China). HPLC-grade n-hexane was provided by J.T. Baker (Philipsburg, NJ, USA). HPLC-grade ethyl acetate, n-propanol and DMF were from Tianjin Kermel Chemical Reagent Co.
3
(Tianjin, China) and DMF was purified by distillation under reduced pressure after dried by calcium hydride. All solvents were analytical grade unless noted otherwise. The water used in all experiments was deionized water purified by a Milli-Q system (Millipore, Billerica, USA). The empty polypropylene SPE cartridges (3 mL) with polyethylene frits were from Dalian Sipore (Dalian, China). 2.2. Preparation of the MISMS and non-imprinted SMS (NISMS) 2.2.1. Synthesis of the BPA-ICPTES complex (BPA-Si) As shown in Scheme 1a, the template-monomer complex (BPA-Si) was synthesized as reported previously [35,40]. 2.28 g (10 mmol) of BPA was dissolved in 15 mL of purified DMF under nitrogen atmosphere. To this solution, 4.80 mL (20 mmol) of ICPTES was added gradually with stirring at room temperature. Then, the mixture was stirred for 72 h at 80 C under nitrogen atmosphere. After reaction, the product was isolated by a solvent evaporation step under reduced pressure below 90 C. The obtained transparent oily liquid was confirmed to be BPA-Si by FT-IR, 1H NMR (400 MHz, CDCl3) and 13C NMR (400 MHz, CDCl3), and the results were shown in the supporting information (Fig. S1). The yield was estimated to be 93% from 1H NMR. 2.2.2. Preparation of the MISMS and NISMS The BPA imprinted SMS was synthesized by a modified method based on the published literature [24]. In a typical procedure, 28 mL of water was added slowly under stirring to a solution containing 2.0 g of lecithin and 0.4 g of dodecylamine in 20.8 g of ethanol, until a homogeneous emulsion was formed. To this emulsion, 4.58 g (22 mmol) of TEOS predissolved with 1.45 g (2 mmol) of BPA-Si was added slowly under stirring for 15 min, and then the mixture was left under static conditions for 24 h at room temperature. The as-synthesized MISMS powder was filtrated, rinsed with ethanol, and then washed free of lecithin and dodecylamine by Soxhlet extraction with ethanol for 24 h. The obtained powder was washed with acetone, and dried overnight
Scheme 1. Schematic procedure for the preparation of BPA-Si (a) and MISMS (b).
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
4
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
under vacuum. For the synthesis of NISMS, the same procedure was adopted except that 4.58 g (22 mmol) of TEOS mixed with 0.89 g (4 mmol) of APTES was used as silica source. For template removal, 2.0 g of dry MISMS was first suspended in a mixture of 60 mL of DMSO and 12 mL of water, and then refluxed at 180 C for 5 h with stirring. The resulting MISMS was isolated by filtration, washed with water, ethanol and acetone in sequence, and dried overnight under vacuum. 2.3. Characterization of the MISMS The nitrogen adsorption–desorption isotherms were measured using a Quantachrome NOVA 4000 analyzer (Boynton Beach, FL, USA). Before the sorption measurements, the samples were outgassed for 8 h at 120 C. The specific surface area and pore volume were calculated using the standard Brunauer–Emmett– Teller (BET) method, and the pore size distribution was obtained using the Barrett–Joyner–Halenda (BJH) theory. The scanning electron microscopy (SEM) images were obtained using a Zeiss Supra 55 Sapphire microscope (Oberkochen, Germany). The transmission electron microscopy (TEM) images were taken by a JEOL JEM-2100 microscope (Tokyo, Japan) with 120 kV accelerating voltage. The FT-IR spectra were recorded on a PerkinElmer Spectrum GX Spectrometer (Waltham, MA, USA) in the 4000– 400 cm1 range with a resolution of 4 cm1. 2.4. Chromatographic conditions An Agilent 1200 series LC system (Santa Clara, CA, USA) consisting of a quaternary pump, an on-line degasser, an autosampler and a diode array detector (DAD), was used for the determination of BPs. Chromatographic separations were carried
out on an Agela Venusil MP-C18 column (5 mm, 250 mm 4.6 mm i.d., Tianjin, China). The mobile phase consisted of isocratic water/ methanol (30:70, v/v) for the analysis of single BPA. For the determination of five BPs, a gradient elution was adopted by combining solvent A (water) and solvent B (methanol) as follows: 35–100% B (25 min), 100% B (3 min), 100–35% B (1 min), 35% B (8 min). In both cases, the column temperature was kept at 25 C, the flow rate was 1 mL min1, and the detection wavelength was set at 225 nm. The injection volume was 5 mL for sorption experiments and 20 mL for real sample analysis. 2.5. Sorption experiments 20.0 mg of MISMS and NISMS particles were separately mixed with 2.0 mL of various concentrations of BPA (0.05–5.0 mmol L1) in n-hexane/n-propanol (95:5, v/v). The mixtures were shaken at 160 rpm for 24 h at 25 C. After binding, the mixtures were filtered through a 0.22 mm membrane, and 1.0 mL of the filtrates were dried under a moderate stream of nitrogen. After reconstituted in 1.0 mL of methanol, the free concentrations of BPA in the filtrates were determined by the HPLC method above. The adsorption capacity and the equilibrium dissociation constant (Kd, mmol L1) of MISMS were calculated according to the Eqs. (1) and (2) [6,9,41]: Q¼
ðC 0 C f Þv m
Q 1 Q ¼ Q þ max Cf Kd Kd
(1)
(2)
where C0 (mmol L1) and Cf (mmol L1) are the initial and final concentrations of BPA, v (L) is the total volume of the sample, m (g)
Fig. 2. SEM (a) and TEM ((b) inset: partial enlarged image) images of MISMS and schematic representation of the SMS pore structure (c).
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
is the mass of MISMS, Q and Qmax (mmol g1) are the amount of BPA adsorbed at equilibrium and saturation, respectively. The kinetic adsorption study was performed using similar procedures with the static adsorption experiment, except that 2.0 mmol L1 BPA in n-hexane/n-propanol (95:5, v/v) and different equilibrium periods (0.5–30 min) were adopted. The class selectivity was evaluated through similar procedures with the static adsorption experiment using the mixture standards of five BPs (BPA, BPF, BPB, BPE and BPAF) and three other analytes (DDT, E2 and DES) in n-hexane/n-propanol (95:5, v/v) (0.5 mmol L1 of each analyte). 2.6. Sample preparation and SPE procedure The sediment samples were collected from three different locations of Liaohe River (Liaoning, China). After lyophilization, 10 g dry weight (dw) of each sample (unspiked or spiked with 50 ng g1 dw and 500 ng g1 dw of each BP) were extracted by Soxhlet extraction with DCM for 24 h. The extracts were concentrated to around 10 mL by rotary evaporation. Triplicate experiments were conducted for each sample. The MISMS particles (100 mg) were packed into a 3 mL SPE cartridge using an upper frit and a lower frit to prevent sorbent loss. The cartridges were connected to a Supelco Visiprep SPE manifold (Bellefonte, PA, USA) equipped with a vacuum pump. Prior to extraction, the cartridges were equilibrated with 3 mL of DCM. Sample solutions were percolated at a constant flow rate of 1.0 mL min1. The cartridges were then rinsed with 4 mL of nhexane/n-propanol (95:5, v/v) and dried for 15 min under vacuum. Finally, the BPs retained on the sorbent were eluted with 3 mL of methanol. The eluates were evaporated to dryness under a moderate stream of nitrogen and the dry residues were reconstituted in 1.0 mL of methanol/water (35:65, v/v) for HPLC analysis. 3. Results and discussion 3.1. Preparation and characterization of the MISMS The MISMS for specifically recognizing BPA and its analogues was synthesized by combining semi-covalent molecular imprinting and a simple self-assembly process. As illustrated in Scheme 1b, BPA was trapped into the silica matrix via the urethane bond formed between an isocyanate and a phenol group. After heat treatment, BPA was removed due to a reversible cleavage of the urethane bond, and the dissociated isocyanato group was converted to amino group by its reaction with water. Thus, the delicate binding sites were formed in the silica matrix. The successful formation of imprinting sites was confirmed by FT-IR spectroscopy (see Fig. S2 in the supporting information). Before template removal, the carbonyl group band at 1712 cm1 is observed. After template removal, this band disappears and the N— H (1 amines) bending adsorption at 1560 cm1 appears, revealing the successful formation of imprinting sites. The two bands for N— H (1 amines) stretching vibration should also be observed at 3500–3200 cm1. Unfortunately, the signal in this range was overlapped with the broad band of O—H in silica matrix. The morphology of MISMS was investigated by SEM and TEM measurements. The SEM image shows the formation of irregular shaped particles with the mm-level particle sizes, which are suitable as SPE packing materials (see Fig. 2a). As shown in Fig. 2b (inset: partial enlarged image), the TEM image reveals an evident interconnected 3-D porous network, which was similar to the schematic representation of the SMS pore structure (Fig. 2c) [24]. The porosity of MISMS was also investigated by nitrogen adsorption–desorption measurement. The MISMS particles
5
showed typical “type IV” nitrogen adsorption–desorption isotherm, which is associated with capillary condensation taking place in mesoporous materials. The BET specific surface area obtained from the nitrogen isotherm was 850.55 m2 g1 and the total pore volume was 1.20 cm3 g1. The BJH pore size distribution calculated from the adsorption branch of the isotherm showed uniform mesopores with an average diameter of 3.53 nm. 3.2. Binding properties of the MISMS and NISMS 3.2.1. Effect of solvents on BPA binding performance To optimize the binding performance of MISMS, different types of organic solvents were investigated for BPA adsorption. In detail, 20.0 mg of MISMS and NISMS were separately equilibrated with 2.0 mL of BPA (2.0 mmol L1) in six different solvents, DCM, chloroform, n-hexane/n-propanol (95:5, v/v), n-propanol, ethyl acetate and ACN, and their adsorption amounts were determined. As shown in Fig. 3, both MISMS and NISMS show high binding affinities for BPA in nonpolar solvents (DCM and chloroform). In contrast, more polar solvents like n-propanol, ethyl acetate and ACN tend to weaken the interactions between the silica matrix and BPA, resulting in reduced binding affinities. However, the selectivity of MISMS for BPA can not be well demonstrated by using DCM and chloroform with respect to the slight distinction of adsorption amounts between MISMS and NISMS. Polar protic solvents with hydrogen bond forming ability (e.g., n-propanol) could reduce the binding affinity of NISMS more by competitively binding to the non-imprinting sites. After optimization, a mixed solvent of n-hexane/n-propanol (95:5, v/v) was chosen for the experiments to get a good compromise between the binding affinity and selectivity. 3.2.2. Static and kinetic adsorptions As can be seen from Fig. 4a, the adsorption amounts of MISMS and NISMS increase with increasing BPA initial concentration, and higher adsorption capacity of the MISMS is presented over the tested concentration range (0.05–5.0 mmol L1). This is usually indicative of the presence of recognition sites created by the molecular imprinting process. The binding data of MISMS were then processed using Scatchard analysis (Eq. (2)), and two straight lines were obtained in the plot region (Fig. 4b). The results indicate that the binding sites in MISMS are heterogeneous and can be classified into two distinct groups: the high-affinity (specific) and
Fig. 3. Effect of different solvents on BPA (2.0 mmol L1) adsorption amounts (Q) for MISMS and NISMS. Mixed solvent is n-hexane/n-propanol (95:5, v/v).
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
6
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
Fig. 4. (a) BPA binding isotherms for MISMS and NISMS; (b) Scatchard plot for MISMS; (c) kinetic binding profiles of BPA (2.0 mmol L1) for MISMS and NISMS.
the low-affinity (non-specific) binding sites. The linear regression equations for the two straight lines are Q/Cf = 129.59 0.7658Q (r = 0.9946) and Q/Cf = 93.82 0.0692Q (r = 0.9540). The Kd and Qmax values can be calculated according to the slopes and intercepts of the two linear regression equations. The Kd and Qmax values are calculated to be 1.31 mmol L1 and 169.22 mmol g1 for the specific binding sites (upper line) and 14.45 mmol L1 and 1355.71 mmol g1 for the non-specific binding sites (lower line), respectively. The non-specific adsorption is mainly contributed by the Si—OH group and is difficult to be eliminated from the silica matrix. Although the Qmax (1355.71 mmol g1) for the non-specific adsorption of MISMS is much higher than or comparable to some traditional polymer and silica materials (0 mmol g1 [8], 700 mmol g1 [10] and 45.22 mmol g1 [41] for polymers; 805.87 mmol g1 [28] for silica), the Qmax value (169.22 mmol g1) for the specific binding sites created by the semi-covalent imprinting process is still superior to those materials. Furthermore, the specific adsorption is predominant in a wide concentration range of 0.05–1.0 mmol L1, which covers most sample concentrations for BPA analysis. Besides excellent binding capacity, fast adsorption kinetics for BPA (2.0 mmol L1) was also observed. As shown in Fig. 4c, the binding equilibrium for both MISMS and NISMS is reached within 3 min, which is a remarkable improvement compared to some traditional MIPs that required up to 10 min [8] and even 5 h [10] to reach equilibrium. The BPA binding amount at equilibrium for MISMS is almost 2-fold that of NISMS. The spongy mesoporous
silica prepared in this work features an isotropic 3-D pore structure. So the fast adsorption kinetics might be owing to the open diffusion channel and the highly accessible binding sites. Therefore, the superior binding capacity and fast adsorption kinetics make the prepared MISMS a promising and competitive sorbent for rapid and selective isolation of BPA from complex matrices. 3.2.3. Class selectivity The class selectivity of MISMS was assessed by evaluating the binding capacities for five BPs (BPA, BPF, BPB, BPE and BPAF) and three other analytes (DDT, E2 and DES) of MISMS and NISMS. Fig. 5 illustrates the data obtained from the selectivity experiment for both MISMS and NISMS, concerning the adsorption amounts and the ratios between QMISMS and QNISMS. The MISMS exhibits obviously higher adsorption capacity than NISMS for the five tested BPs due to the presence of the specific binding sites. Due to the lack of strong functional groups like the hydroxyl groups of BPs, the adsorption amounts of DDT for MISMS and NISMS are much lower than BPs. However, MISMS still demonstrates good selectivity for DDT with respect to the ratio between QMISMS and QNISMS. This might be owing to the very close molecular structure of DDT with the BPs (especially BPE). And BPE has been used as template on molecular imprinting for the selective detection of DDT in the published literature [39]. For E2 and DES, poor selectivities of MISMS were observed with respect to small distinctions and low ratios of the adsorption
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
7
Fig. 5. Adsorption amounts of MISMS and NISMS and ratios between QMISMS and QNISMS for eight analytes. Fig. 6. Effect of washing solvent (4 mL) on BPs recoveries for MISMS and NISMS.
amounts between MISMS and NISMS. This result shows that the MISMS has good class selectivity towards the five investigated BPs. 3.3. Optimization of SPE procedure The mass capacity of the MISMS cartridge was determined by percolating different amounts of BPs standards (5, 7.5, 10, 15 and 25 mg for each BP) in 10 mL of DCM. The BPs recoveries with values above 80% did not differ significantly over the tested range. Satisfactory recoveries were still obtained when the total loading amount of BPs reached a high value of 125 mg (25 mg for each BP), indicating that the MISMS sorbent had good mass capacity for the BPs. Considering the concentration levels of BPs in environmental samples, 50 mg (10 mg for each BP) of total loading amount was selected to optimize other factors. To realize the elimination of co-adsorbed matrix components without loss of target analytes in real sample analysis, an appropriate washing step was introduced. Since it exhibited appropriate polarity in the sorption experiments, the mixed nhexane/n-propanol (95:5, v/v) solvent was adopted as the washing solvent. The effect of washing volume on BPs recoveries was investigated by loading 50 mg of BPs mixture standards into the MISMS and NISMS SPE cartridges, followed by rinsing with different volumes of washing solvent. Without any washing step, the recoveries of BPs for both MISMS and NISMS were higher than 80% due to the existence of non-specific adsorption. After a 4 mL of n-hexane/n-propanol (95:5, v/v) washing step, the BPs recoveries for NISMS decreased obviously (especially for BPA and BPB) due to the reduced non-specific adsorption, as shown in Fig. 6. In comparison, the recoveries of BPs for MISMS kept almost constant due to the existence of significant specific adsorption. These results further confirm the good class selectivity of MISMS for the five BPs as discussed in Section 3.2.3. However, an obvious reduction of the BPs recoveries for MISMS was observed when the washing volume was higher than 4 mL. Thus, 4 mL was selected as the appropriate washing volume. Before elution, the SPE cartridge was dried with the aid of a vacuum pump for 15 min. The final elution of BPs was conducted by using 3 mL of methanol or 4 mL of ACN. Satisfactory recoveries (>80% for all BPs) were achieved when any of the two solvents was used. Considering that low boiling point and small volume were beneficial to reduce the drying time, 3 mL of methanol was finally adopted.
3.4. Analysis of spiked environmental sediment samples Twelve different concentrations of BPs (0.05–30 mg mL1) were tested to determine the linearity. Satisfactory results were obtained with a correlation coefficient higher than 0.9998 for all BPs, and the limits of detection (LOD) for BPs ranged from 0.43 to 0.71 ng g1 dw (4.3–7.1 ng mL1) (Table 1). The performance of the present SPE–HPLC–DAD method was compared with other methods for BPA determination in the literature (Table 2). Although the detection limit of this method is at the same level with or higher than other methods, the new sample type (sediment) and higher Qmax still make this method competitive and promising. Furthermore, sediment sample with enough mass could be a good indicator of pollution status and the high Qmax of this material could cover the wide linear range well for samples with a large mass. In the present work, a large mass of 10 g of sediment sample was adopted. As shown in Fig. 7a, the BPs standards (peak identifications: 1, BPF; 2, BPE; 3, BPA; 4, BPB; 5, BPAF) were well separated using the developed HPLC method. The existence of many matrix interferences in sediment samples was demonstrated in Fig. 7b. Both good selectivity towards BPs and clean-up effect of the MISMS were verified by the fact that the spiked BPs were well extracted from the sample matrix and most of interfering matrix components were removed (Fig. 7c). To demonstrate the reliability of the developed SPE procedure, the BPs recoveries were determined at two different spiking levels for the lyophilizated sediment samples without and with SPE. Before analysis, the three sediment samples were verified to be free of the five investigated BPs and could be used as blank samples. Due to the severe matrix interferences, unsatisfactory results (low BPs recoveries and high RSD values) were obtained when samples were analyzed without the developed SPE procedure (Table 1). Satisfactory recoveries of BPs were obtained when the SPE procedure was applied to the sample analysis. As can be seen from Table 1, the BPs recoveries after SPE were in the range from 75.5% to 105.5%, with RSD values below 7.5%. The results showed that a reliable and practical analytical procedure based on the MISMS–SPE purification coupled with HPLC–DAD detection was successfully established for simultaneous determination of five BPs from environmental sediment samples.
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
8
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
Table 1 Limits of detections (LODs), correlation coefficients (r) for BPs and average recoveries, relative standard deviations (RSDs, n = 3) of BPs from lyophilizated sediment samples at two different spiking levels without and with SPE procedure. Analyte LOD (ng g1 dw)
Correlation coefficient (r)
Spiked level (ng g1 dw)
SPE
Sample 1 Recovery (%)
BPF
0.43
0.9998
50 500
– BPE
0.45
0.9998
50 500
– BPA
0.46
0.9998
50 500
– BPB
0.43
0.9999
50 500
– BPAF
0.71
0.9999
50 500
Sample 2 RSD (%, n = 3)
Recovery (%)
Sample 3 RSD (%, n = 3)
Recovery (%)
RSD (%, n = 3)
Without With Without With
33.9 96.7 43.0 88.8
11.8 3.1 16.9 4.0
22.4 85.5 39.7 85.1
9.5 5.5 14.3 1.6
37.1 95.3 65.7 94.1
7.2 1.7 13.8 1.0
Without With Without With
48.3 95.9 46.8 79.9
15.4 2.7 27.1 3.6
26.8 83.1 43.2 79.5
10.0 3.4 25.1 3.3
37.5 90.0 67.7 85.6
14.2 2.8 15.4 1.4
Without With Without With
41.6 94.8 48.7 77.0
16.8 4.5 24.9 4.1
35.2 83.0 28.5 81.2
12.8 5.3 18.4 4.6
49.8 90.3 59.5 86.2
20.3 4.9 20.7 1.6
Without 53.3 With 103.1 Without 45.9 With 81.0
12.7 3.0 6.6 7.5
40.1 101.9 34.0 87.2
15.3 3.8 19.1 3.6
55.9 104.2 51.5 91.4
19.0 3.9 15.9 1.2
Without 89.5 With 104.6 Without 60.2 With 75.5
10.2 4.7 24.3 6.6
71.1 98.0 38.4 90.4
14.9 2.0 10.5 2.0
64.3 105.5 41.0 94.7
20.6 6.2 9.8 2.1
Table 2 Comparison of the proposed method with other analytical methods based on molecularly imprinted materials for BPA determination in various samples. Analytical methods
Material type
Matrix
Qmax
Linear range
LOD
References
SPE–HPLC–UV SPE–HPLC–FLDa SPE–UV SPE–HPLC–DAD MSPE–LC–FLDb SPE–HPLC–FLD SPE–HPLC–DAD
MI hollow porous polymer MI silica nanoparticlesc MI silica nanoparticles MI polymer MI magnetic polymerparticles MI silica nanoparticles MISMS
Tap water Chemical cleansingand cosmetics Tap water River water River water Fish River sediment
832.7 mmol g1) 822.3 mmol g1 30.26 mmol g1 57.95 mmol 1 3.55 mmol g1 39.4 mmol g1 1524.93 mmol g1
0.005–100 mg mL1 0.684–114 mg L1 – 0.02–2.0 mg L1 0.01–0.2 mg L1 0.7–114.1 mg L1 0.05–30 mg mL1
3 ng mL1 0.23 ng mL1 – 2.5 ng L1 10 ng L1 0.11 ng mL1 4.6 ng mL1
[9] [28] [31] [41] [42] [43] This work
a b c
FLD, fluorescence detector. MSPE, magnetic solid phase extraction. MI, molecularly imprinted.
4. Conclusions
Fig. 7. Chromatograms of the standards and real samples. (a) Standard mixtures of five BPs (5 mg mL1 for each BP); (b) extract of spiked sediment (500 ng g1 dw of each BP) before MISMS–SPE; (c) extract of spiked sediment (500 ng g1 dw of each BP) after MISMS–SPE. Peak identifications: 1, BPF; 2, BPE; 3, BPA; 4, BPB; 5, BPAF.
In this work, a sponge-like BPA imprinted mesoporous silica material was synthesized via the thermally reversible urethane bonds formed between isocyanates and phenol groups using lecithin/dodecylamine mixed-micelles as template and TEOS as cross-linker. When used as sorbent for BPA and its structural analogues, the synthesized MISMS showed high specific adsorption capacity and selectivity, as well as fast adsorption kinetics. The excellent performance of MISMS may be attributed to the combination of a sufficiently high specific surface area with an open 3-D pore architecture, which remains highly accessible. Furthermore, the developed SPE procedure using MISMS showed good selectivity and clean-up effect for isolation of BPs from sediment extracts. The developed MISMS–SPE coupled with HPLC–DAD method was accurate, sensitive and had high reliability for determination of BPs in environmental sediment samples. Further improvement of imprinting efficiency in terms of selectivity may be achieved by suppression of the nonspecific adsorption from silica matrix.
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051
G Model ACA 233506 No. of Pages 9
J. Yang et al. / Analytica Chimica Acta xxx (2014) xxx–xxx
Acknowledgements Financial supports from the National High Technology Research and Development Program of China (863 program) (2013AA065203) and National Natural Science Foundation of China (Grant Nos. 21277139,21205117) are gratefully acknowledged. 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.aca.2014.09.051. References [1] C.Y. Liao, F. Liu, Y. Guo, H.B. Moon, H. Nakata, Q. Wu, K. Kannan, Occurrence of eight bisphenol analogues in indoor dust from the united states and several asian countries: implication for human exposure, Environ. Sci. Technol. 46 (2012) 9138–9145. [2] C.Y. Liao, F. Liu, H.B. Moon, N. Yamashita, S. Yun, K. Kannan, Bisphenol analogues in sediments from industrialized areas in the United States, Japan, and Korea: spatial and temporal distributions, Environ. Sci. Technol. 46 (2012) 11558–11565. [3] C.Y. Liao, F. Liu, H. Alomirah, V.D. Loi, M.A. Mohd, H.B. Moon, H. Nakata, K. Kannan, Bisphenol S in urine from the United States and seven Asian countries: occurrence and human exposures, Environ. Sci. Technol. 46 (2012) 6860–6866. [4] C.Y. Liao, F. Liu, K. Kannan, Bisphenol S, a new bisphenol analogue, in paper products and currency bills and its association with bisphenol A residues, Environ. Sci. Technol. 46 (2012) 6515–6522. [5] H. Gallart-Ayala, E. Moyano, M.T. Galceran, Analysis of bisphenols in soft drinks by on-line solid extraction fast liquid chromatography–tandem mass spectrometry, Anal. Chim. Acta 683 (2011) 227–233. [6] H. Zhou, Y.P. Xu, H.W. Tong, Y.X. Liu, F. Han, X.Y. Yan, S.M. Liu, Direct synthesis of surface molecularly imprinted polymers based on vinyl–SiO2 nanospheres for recognition of bisphenol A, J. Appl. Polym. Sci. 128 (2013) 3846–3852. [7] P.P. Qi, J.C. Wang, L.D. Wang, Y. Li, J. Jin, F. Su, Y.Z. Tian, J.P. Chen, Molecularly imprinted polymers synthesized via semi-covalent imprinting with sacrifical spacer for imprinting phenols, Polymer 51 (2010) 5417–5423. [8] T.H. Jiang, L.X. Zhao, B.L. Chu, Q.Z. Feng, W. Yan, J.M. Lin, Molecularly imprinted solid-phase extraction for the selective determination of 17b-estradiol in fishery samples with high performance liquid chromatography, Talanta 78 (2009) 442–447. [9] J. Li, X.B. Zhang, Y.X. Liu, H.W. Tong, Y.P. Xu, S.M. Liu, Preparation of a hollow porous molecularly imprinted polymer using tetrabromobisphenol A as a dummy template and its application as SPE sorbent for determination of bisphenol A in tap water, Talanta 117 (2013) 281–287. [10] X.L. Mou, W.P. Zhang, Z.L. Chen, Highly selective extraction of baicalin in natural herbs and medicinal preparations by molecularly imprinted polymer, J. Appl. Polym. Sci. 130 (2013) 1873–1878. [11] J.E. Lofgreen, I.L. Moudrakovski, G.A. Ozin, Molecularly imprinted mesoporous organo silica, ACS Nano 5 (2011) 2277–2287. [12] L.X. Chen, S.F. Xu, J.H. Li, Recent advances in molecular imprinting technology: current status, challenges and highlighted applications, Chem. Soc. Rev. 40 (2011) 2922–2942. [13] E.C. Morais, G.G. Correa, R. Brambilla, J.H.Z. dos Santos, A.G. Fisch, Selective silica-based sorbent materials synthesized by molecular imprinting for adsorption of pharmaceuticals in aqueous matrices, J. Sep. Sci. 36 (2013) 636–643. [14] D. Lewandowski, A. Olejnik, G. Schroeder, Adsorption studies and release of selected dyes from functionalized mesoporous MCM-41 silica, Cent. Eur. J. Chem. 12 (2014) 233–241. [15] Y. San, H.S. Li, M.H. Zhu, K. Ma, Q.Z. Jiao, Q. Wu, Catalytic performance of metal ion doped MCM-41 for methanol dehydration to dimethyl ether, J. Porous Mater. 20 (2013) 1509–1518. [16] A. Mehdinia, M. Ahmadifar, M.O. Aziz-Zanjani, A. Jabbari, M.S. Hashtroudi, Selective adsorption of 2,4-dinitrophenol on molecularly imprinted nanocomposites of mesoporous silica SBA-15/polyaniline, Analyst 137 (2012) 4368–4374. [17] A. Mehdinia, M.O. Aziz-Zanjani, M. Ahmadifar, A. Jabbari, Design and synthesis of molecularly imprinted polypyrrole based on nanoreactor SBA-15 for recognition of ascorbic acid, Biosens. Bioelectron. 39 (2013) 88–93. [18] Y. Liu, Z.C. Liu, J. Gao, J.D. Dai, J. Han, Y. Wang, J.M. Xie, Y.S. Yan, Selective adsorption behavior of Pb(II) by mesoporous silica SBA-15-supported Pb(II)imprinted polymer based on surface molecularly imprinting technique, J. Hazard. Mater. 186 (2011) 197–205.
9
[19] X.M. Xue, F.T. Li, Removal of Cu(II) from aqueous solution by adsorption onto functionalized SBA-16 mesoporous silica, Micropor. Mesopor. Mater. 116 (2008) 116–122. [20] L. Sierra, S. Valange, J.L. Guth, Formation mechanism and morphology of mesoporous SBA-16 type silica particles prepared with triblock copolymer surfactant PEO140PPO39PEO140, Micropor. Mesopor. Mater. 124 (2009) 100–109. [21] J. Hwang, H. Daiguji, Proton transport in mesoporous silica SBA-16 thin films with three-dimensional cubic structures, Langmuir 29 (2013) 2406–2411. [22] M. Mureseanu, A. Galarneau, G. Renard, F. Fajula, A new mesoporous micelletemplated silica route for enzyme encapsulation, Langmuir 21 (2005) 4648–4655. [23] A. Galarneau, M. Mureseanu, S. Atger, G. Renard, F. Fajula, Immobilization of lipase on silicas. Relevance of textural and interfacial properties on activity and selectivity, New J. Chem. 30 (2006) 562–571. [24] A. Galarneau, G. Renard, M. Mureseanu, A. Tourrette, C. Biolley, M. Choi, R. Ryoo, F.D. Renzo, F. Fajula, Synthesis of sponge mesoporous silicas from lecithin/dodecylamine mixed-micelles in ethanol/water media: a route towards efficient biocatalysts, Micropor. Mesopor. Mater. 104 (2007) 103–114. [25] A. Galarneau, F. Sartori, M. Cangiotti, T. Mineva, F.D. Renzo, M.F. Ottaviani, Sponge mesoporous silica formation using disordered phospholipid bilayers as template, J. Phys. Chem. B 114 (2010) 2140–2152. [26] F. Kleitz, T.W. Kim, R. Ryoo, Phase domain of the cubic Im m mesoporous silica in the EO106PO70EO106-butanol–H2O system, Langmuir 22 (2006) 440–445. [27] F. Kleitz, D. Liu, G.M. Anilkumar, I.S. Park, L.A. Solovyov, A.N. Shmakov, R. Ryoo, Large cage face-centered-cubic Fm3m mesoporous silica: synthesis and structure, J. Phys. Chem. B 107 (2003) 14296–14300. [28] R. Zhu, W.H. Zhao, M.J. Zhai, F.D. Wei, Z. Cai, N. Sheng, Q. Hu, Molecularly imprinted layer-coated silica nanoparticles for selective solid-phase extraction of bisphenol A from chemical cleansing and cosmetics samples, Anal. Chim. Acta 658 (2010) 209–216. [29] Y.Q. Wang, Y.Y. Yang, L. Xu, J. Zhang, Bisphenol A sensing based on surface molecularly imprinted, ordered mesoporous silica, Electrochim. Acta 56 (2011) 2105–2109. [30] X.M. Jiang, W. Tian, C.D. Zhao, H.X. Zhang, M.C. Liu, A novel sol–gel-material prepared by a surface imprinting technique for the selective solid-phase extraction of bisphenol A, Talanta 72 (2007) 119–125. [31] Y.M. Ren, W.Q. Ma, J. Ma, Q. Wen, J. Wang, F.B. Zhao, Synthesis and properties of bisphenol A molecular imprinted particle for selective recognition of BPA from water, J. Colloid Interface Sci. 367 (2012) 355–361. [32] C.D. Ki, C. Oh, S.G. Oh, J.Y. Chang, The use of a thermally reversible bond for molecular imprinting of silica spheres, J. Am. Chem. Soc. 124 (2002) 14838– 14839. [33] H.H. Yang, S.Q. Zhang, W. Yang, X.L. Chen, Z.X. Zhuang, J.G. Xu, X.R. Wang, Molecularly imprinted sol–gel nanotubes membrane for biochemical separations, J. Am. Chem. Soc. 126 (2004) 4054–4055. [34] X. Wang, L.Y. Wang, X.W. He, Y.K. Zhang, L.X. Chen, A molecularly imprinted polymer-coated nanocomposite of magnetic nanoparticles for estrone recognition, Talanta 78 (2009) 327–332. [35] C.Y. He, Y.Y. Long, J.L. Pan, K. Li, F. Liu, Molecularly imprinted silica prepared with immiscible ionic liquid as solvent and porogen for selective recognition of testosterone, Talanta 74 (2008) 1126–1131. [36] B.M. Jung, M.S. Kim, W.J. Kim, J.Y. Chang, Molecularly imprinted mesoporous silica particles showing a rapid kinetic binding, Chem. Commun. 46 (2010) 3699–3701. [37] Y.D. Kim, J.B. Jeon, J.Y. Chang, CdSe quantum dot-encapsulated molecularly imprinted mesoporous silica particles for fluorescent sensing of bisphenol A, J. Mater. Chem. 22 (2012) 24075–24080. [38] J.Q. Xue, D.W. Li, L.L. Qu, Y.T. Long, Surface-imprinted core–shell Au nanoparticles for selective detection of bisphenol A on surface-enhanced Raman scattering, Anal. Chim. Acta 777 (2013) 57–62. [39] A.L. Graham, C.A. Carlson, P.L. Edmiston, Development and characterization of molecularly imprinted sol–gel materials for the selective detection of DDT, Anal. Chem. 74 (2002) 458–467. [40] M.U. de la Orden, J. Arranz, V. Lorenzo, E. Pérez, J.M. Urreaga, Study of the effects of the reaction conditions on the modification of clays with polyelectrolytes and silanes, J. Colloid Interface Sci. 342 (2010) 185–191. [41] J.F. Yin, Z.H. Meng, Y.S. Zhu, M.Y. Song, H.L. Wang, Dummy molecularly imprinted polymer for selective screening of trace bisphenols in river water, Anal. Methods 3 (2011) 173–180. [42] Y. Hiratsuka, N. Funaya, H. Matsunaga, J. Haginaka, Preparation of magnetic molecularly imprinted polymers for bisphenol A and its analogues and their application to the assay of bisphenol A in river water, J. Pharm. Biomed. Anal. 75 (2013) 180–185. [43] F. Wei, X. Liu, M. Zhai, Z. Cai, G. Xu, J. Yang, S. Du, Q. Hu, Molecularly imprinted nanosilica solid-phase extraction for bisphenol A in fish samples, Food Anal. Methods 6 (2013) 415–420.
Please cite this article in press as: J. Yang, et al., Novel sponge-like molecularly imprinted mesoporous silica material for selective isolation of bisphenol A and its analogues from sediment extracts, Anal. Chim. Acta (2014), http://dx.doi.org/10.1016/j.aca.2014.09.051