G Model
ARTICLE IN PRESS
CHROMA-356426; No. of Pages 8
Journal of Chromatography A, xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A Dan Yu a , Xiaolei Hu a , Shoutai Wei a , Qiang Wang a , Chiyang He a,∗ , Shaorong Liu b a b
School of Chemistry and Chemical Engineering, Wuhan Textile University, 1 Textile Road, Wuhan 430073, China Department of Chemistry and Biochemistry, University of Oklahoma, 101 Stephenson Parkway, Norman, OK 73019, USA
a r t i c l e
i n f o
Article history: Received 10 January 2015 Received in revised form 1 April 2015 Accepted 2 April 2015 Available online xxx Keywords: Dummy molecular imprinting Hybrid imprinting Ordered mesoporous silica Solid-phase extraction Bisphenol A HPLC-UV
a b s t r a c t A novel hybrid dummy imprinting strategy was developed to prepare a mesoporous silica for the solidphase extraction (SPE) of bisphenol A (BPA). A new covalent template-monomer complex (BPAF-Si) was first synthesized with 2,2-bis(4-hydroxyphenyl)hexafluoropropane (BPAF) as the template. The imprinted silica was obtained through the gelation of BPAF-Si with tetraethoxysilane and the subsequent removal of template by thermal cleavage, and then it was characterized by FT-IR spectroscopy, scanning electron microscopy, transmission electron microscopy, and nitrogen adsorption–desorption isotherms. Results showed that the new silica had micron-level particle size and ordered mesoporous structure. The static binding test verified that the imprinted silica had much higher recognition ability for BPA than the non-imprinted silica. The imprinted silica also showed high extraction efficiencies and high enrichment factor for SPE of BPA. Using the imprinted silica, a SPE-HPLC-UV method was developed and successfully applied for detecting BPA in BPA-spiked tap water and lake water samples with a recovery of 99–105%, a RSD of 2.7–5.0% and a limit of detection (S/N = 3) of 0.3 ng/mL. The new imprinted silica avoided the interference of the residual template molecules and reduced the non-specific binding sites, and therefore it can be utilized as a good sorbent for SPE of BPA in environmental water samples. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Bisphenol A (BPA) is a xenoestrogenic chemical that is widely used in the production of polycarbonate plastic and epoxy resins [1–3]. BPA can cause cardiovascular disease, diabetes, neuro-behavioral disorders, carcinogenic hypersensitivity, and reproductive impairment [3–5]. BPA is commonly present in our environmental matrices, biological samples, food, and beverages [1,6,7]. However, it is challenging to analyze it because of its low concentration and matrix complexity in real-world samples [8]. Therefore, pretreatment and enrichment are crucial steps prior to its analysis. Solid-phase extraction (SPE) is by far the most frequently used pretreatment method for detecting BPA because SPE provides benefits of low cost, high enrichment factor, simple operation, and low consumption of toxic organic solvent [9,10]. A number of sorbents have been used for SPE of BPA such as divinylbenzene/N-vinylpyrrolidone copolymer [10], reversedphase silica [11], hydrophilic-hydrophobic balance sorbent [12],
∗ Corresponding author. Tel.: +86 27 59367685; fax: +86 27 59367336. E-mail address: [email protected] (C. He).
and mixed-mode cationic exchanger [13]. These traditional sorbents have high capacities, but they suffer from the low extraction selectivity towards BPA. Immuno-sorbents can provide high selectivity on the basis of molecular recognition. However, the main drawbacks of utilizing this sorbent are its high consumable costs (it consumes expensive antibodies) and limited column reusability [10]. Molecularly imprinted polymer (MIP) is a type of tail-made mimics of antibody with specific recognition ability, good stability, ease of preparation, and low cost [8,14–23]. MIPs are generally prepared via the co-polymerization of the imprint-monomer complex and cross-linking monomers, followed by removing the template molecules by solvent extraction and chemical cleavage [14]. The resulting imprint cavities, with the complementary size, shape and moieties to the template, can bind the target molecules specifically [15]. At present, there are three different strategies for synthesizing MIPs: non-covalent, covalent, and hybrid imprinting [18]. MIPs have been widely used as highly selective sorbents of SPE in recent years [24–33]. Some BPA-imprinted materials have also been applied to the determination of BPA [34–38]. Jiang et al. [34] prepared a BPA-imprinted amino-functionalized silica film coating on the surface of activated silica particles, Liu et al. [35] synthesized a BPA-imprinted polymer coated magnetic nanoparticle,
http://dx.doi.org/10.1016/j.chroma.2015.04.006 0021-9673/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: D. Yu, et al., Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.04.006
G Model CHROMA-356426; No. of Pages 8
ARTICLE IN PRESS D. Yu et al. / J. Chromatogr. A xxx (2015) xxx–xxx
2
Lofgreen et al. [36] developed a new BPA-imprinted mesoporous organosilica, Chen et al. [37] utilized a hierarchically BPA-imprinted mesoporous silica, and Yang et al. [38] utilized a sponge-like BPAimprinted silica for SPE of BPA. These BPA-imprinted sorbents, prepared by different non-covalent imprinting or hybrid imprinting techniques, improved the extraction selectivity to some extent, but the residual BPA leakage was a serious issue for determining trace amount of BPA. Dummy molecularly imprinted polymers (DMIPs) can effectively circumvent this problem. Several DMIPs have been prepared presently using different analogues of BPA as templates for recognition of BPA [6,39–41]. These include DMIP-coated magnetic nanoparticles prepared using bisphenol F as a template [39], a DMIP-coated stir bar synthesized using 4,4 -dihydroxybiphenyl as a template [40], and a hollow porous DMIP prepared using tetrabromobisphenol A as a template [41]. However, these materials were all synthesized by non-covalent imprinting techniques that required the use of excess functional monomers, creating a large number of nonspecific binding sites, and consequently affecting MIP’s selectivity for BPA [15,18]. The hybrid imprinting approach, a combination of covalent and non-covalent strategy, not only ensures the functional monomer residues to be present only in the imprint cavities (and thus greatly reduces the non-specific binding sites), but also makes the MIPs rapidly take up target molecules via the non-covalent interactions [18]. Herein, to avoid the interference of the residual template molecules and reduce the number of non-specific binding sites, we combine a dummy molecular imprinting with a hybrid strategy for the first time to prepare a mesoporous silica for SPE of BPA. A new covalent template-monomer complex (BPAF-Si) was first synthesized with 2,2-bis(4-hydroxyphenyl) hexafluoropropane (BPAF) as the dummy template. The imprinted mesoporous silica was obtained via gelation of BPAF-Si with tetraethoxysilane, followed by removing template by thermal cleavage of the urethane bonds. The mesoporous silica was characterized by FT-IR spectroscopy, scanning electron microscopy (SEM), transmission electron microscopy (TEM) and nitrogen adsorption-desorption isotherms. After that, the adsorption and SPE ability of the imprinted sorbent for both the template and BPA were investigated. Finally, coupled with a HPLC-UV method, the new imprinted silica was successfully applied to the SPE and detection of trace amount of BPA spiked in real water samples. 2. Experimental 2.1. Chemicals and materials Bisphenol A (BPA, 96%), 2,2-bis(4-hydroxyphenyl)hexafluoropropane (BPAF, 99%), dibutyltindilaurate (DBDU, 98%) and tetrahydrofuran (THF, superdry, 99.5%) were purchased from J&K Scientific LTD (Beijing, China). 3-(Triethyloxysily) propyl isocyanate (ICPTES, 95%) was obtained from TCI (Tokyo, Japan). Tetraethoxysilane (TEOS, 99%) was supplied by Alfa Aesar (Tianjin, China). Pluronic P123 (P123, Mn ∼ 5800 g/mol) was purchased from Aldrich (Steinheim, Germany). Methanol and acetonitrile were of HPLC grade supplied by Merck (Darmstadt, Germany). The ultrapure water was obtained from the KL-III-40 purification system (AK, Taiwan). All of other reagents were of analytical grade and used as supplied without further purification. 2.2. Preparation of BPAF-Si Under nitrogen gas, BPAF (2.0 g, 6 mmol) was dissolved in THF (60 mL). ICPTES (6.6 mL, 26.2 mmol) and DBDU (4 mL) were added to the solution at room temperature. The mixture was stirred at
75 ◦ C to allow the reaction to proceed for 24 h. After the solvent was evaporated, the product was isolated by column chromatography on silica gel using an eluent containing ethyl acetate and petroleum ether (ethyl acetate: petroleum ether = 1:2). The oily product, BPAF-Si, was obtained at a yield of 87%. FT-IR (at resolution of 4 cm−1 ): 2978/2931/2893 cm−1 (vC H ), 1728 cm−1 (vHNC O ), 1535 cm−1 (ıN H ), 1100–1000 cm−1 (vSi O Si ). 1 H NMR (300 MHz, CDCl3 ): ı (ppm) 7.27–7.30 (d, 4H), 7.03–7.12 (d, 4H), 6.71–6.74 (d, 2H), 3.70–3.80 (q, 12H), 3.15–3.22 (t, 4H), 1.49–1.57 (m, 4H), 1.11–1.18 (t, 18H), 0.52–70.63 (m, 4H). 13 C NMR (300 MHz, CDCl3 ): ı (ppm) 156.42, 153.32, 150.61, 130.29, 128.87, 120.27, 114.15, 57.47, 42.63, 22.29, 17.24, 6.64. For preparation of BPA-imprinted mesoporous silica, the covalent complex (BPA-Si) of BPA and ICPTES was also synthesized and characterized using the same methods as above. 2.3. Preparation of imprinted and non-imprinted silica A stock solution of porogen was prepared by mixing P123 (8.4 g, 1.5 mmol), NaCl (24.4 g, 0.418 mol), water (69.6 g, 3.86 mol), and 2 mol/L HCl (208.8 g, 11.6 mol H2 O, 0.42 mol HCl), and stirring the mixture until a homogeneous solution was obtained. Molecularly imprinted mesoporous silica (MIMO) was synthesized by mixing a pre-dissolved solution of BPAF-Si (0.347 g, 0.48 mmol) in TEOS (1.939 mL, 8.6 mmol) with 44 g of the stock porogen solution. The non-imprinted mesoporous silica (NIMO) was prepared in the almost same manner for the imprinted particles except that ICPTES (0.2365 g, 0.96 mmol) was used in place of BPAF-Si. Each mixture was stirred at room temperature for 24 h, then transferred to an oven at 80 ◦ C, and cured quiescently for 24 h. The resultant particulate material was collected by filtration, and the residual P123 was removed by washing with ethanol via Soxhlet extraction for 20 h. The removal of the template molecules from the imprinted silica was carried out by suspending MIMO (1 g) in the mixture of dimethylsulfoxide (DMSO) and water and subsequent heating at 160 ◦ C for 5 h with stirring. The imprint-removed silica (MIMO-ir) was isolated by centrifugation, rinsed three times with distilled water and ethanol alternately, and then oven-dried for 2 days. The same treatment for NIMO was achieved to obtain final nonimprinted silica (NIMO-ir). The removals of P123 and template were monitored by FT-IR. The obtained MIMO-ir particles were characterized by SEM, TEM, and nitrogen adsorption–desorption isotherms. To directly verify the leakage of residual BPA from BPAimprinted material, a BPA-imprinted mesoporous silica was prepared in the same manner as above except that BPAF-Si was replaced by BPA-Si. 2.4. Adsorption experiments For static adsorption tests, 20 mg of MIMO-ir particles were added in 1.0 mL of BPAF or BPA aqueous solution (containing 25% methanol) with a concentration range of 0.050–4.0 mmol/L, and then the mixtures were sonicated to disperse the particles completely. The suspensions were shaken continually at 25 ◦ C for 2 h and subsequently centrifuged. The amount of BPAF or BPA in the supernatant was measured by HPLC with a UV absorbance detector at 267 nm. The same procedure was performed for NIMO-ir particles. The adsorption capacity and the dissociation constant (Kd , mol/L) were calculated according to the Eqs. (1) and (2) [6]: Q =
(c0 − cf )v m
(1)
Please cite this article in press as: D. Yu, et al., Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.04.006
G Model CHROMA-356426; No. of Pages 8
ARTICLE IN PRESS D. Yu et al. / J. Chromatogr. A xxx (2015) xxx–xxx
3
a
b
Fig. 1. Schemes of (a) synthesis of BPAF-Si and (b) imprinting and recognition of the mesoporous silica.
1 Qmax Q =− Q+ cf Kd Kd
(2)
where c0 (mol/L) and cf (mol/L) are the initial and final concentrations of BPA, v (L) is the total volume of sample, m (g) is the mass of sorbent, Q and Qmax (mol/L) are the amount of BPA adsorbed at equilibrium and saturation, respectively. The kinetic adsorption study was performed using similar procedure with the static adsorption experiment except that 4.0 mmol/L BPA in 25% methanol aqueous solution and different equilibrium periods (7.5–60 min) were adopted. 2.5. Solid-phase extraction The SPE cartridge was prepared by packing 300 mg MIMO-ir or NIMO-ir particles into a 3-mL empty polyethylene syringe between two sieve plates. Each SPE cartridge was rinsed sequentially by methanol and deionized water for conditioning before loading sample aqueous solutions. Due to the poor solubility of BPAF and BPA in water, methanol-water mixtures were used as solvents to improve their solubility (25% methanol for 1.0 and 0.10 mmol/L BPA solution, 5% methanol for 0.010 and 0.0010 mmol/L BPA solution, 40% methanol for 1.0 and 0.10 mmol/L BPAF solution, 10% methanol for 0.010 and 0.0010 mmol/L BPAF solution). Each SPE cartridge was loaded with 1.0 mL sample solution, rinsed with 1.0 mL solvent, and eluted with 2.0 mL methanol in sequence. After that, the solvent in the effluent was evaporated to dryness by nitrogen. Finally, the dry residues were redissolved with 0.25 mL methanol and analyzed by HPLC. The extraction efficiency was calculated as the ratio of the mass of analyte after SPE to that before SPE. For the analysis of real water samples, tap water and lake water were centrifuged and then filtered with 0.45-m film prior to SPE. For each SPE test, 10.0 mL of the pretreated water sample was used. 2.6. HPLC analysis HPLC analysis was achieved on a Ultimate 3000 HPLC system (Thermo Fisher Scientific, USA), using a Dionex AcclaimTM 120 C18 column (250 mm × 4.6 mm, 5 m, Dionex Bonded Silica Products) and a UV detector (set at 276 nm for all compounds), with methanol and ultrapure water (70/30, v/v) as eluent at a rate of 1.0 mL/min for adsorption and SPE experiments. For analysis of real water samples, a gradient program was used at a flow rate of 1 mL/min, by combining solvent A (water) and solvent B (methanol) as follows: 35–100%
(25 min). The column temperature was kept at 25 ◦ C. The injection volume was 20 L for each analysis. 2.7. Characterization of prepared sorbents All infrared spectra were collected using a Nicolet Avatar 360 FT-IR spectrometer (Thermo Fisher Scientific, USA) within a scan range of 4000–500 cm−1 . SEM images were taken by a Hitachi SU8010 microscope. TEM images were obtained on a FEI Tecnai G2 F30 microscope. Nitrogen adsorption was carried out on a Gemini VII 2390 Series Surface Area Analyzers. 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. 3. Results and discussion 3.1. Preparation and characterization of MIMO-ir 3.1.1. Synthesis of BPAF-Si To eliminate the undesired leakage of residual BPA in the traditional BPA-imprinted materials, we used a dummy template molecule (BPAF, a structural analogue of BPA) for preparing the SPE sorbent. A new template-monomer complex (BPAF-Si) was first prepared; the schematic illustration of the preparation procedure for this complex is shown in Fig. 1a. Similar to what had been reported in the literatures [32,36,42], the reaction occurred between the isocyanate group of ICPTES and the phenol moieties of BPAF. The product contained two thermally cleavable urethane bonds that allowed convenient removal of the template molecules in the following step. The resultant product was characterized by FT-IR, 1 H NMR, and 13 C NMR. 3.1.2. Preparation of MIMO-ir The dummy molecularly imprinted mesoporous silica (MIMOir) was prepared via a triblock-copolymer-template sol-gel strategy similar to Lofgreen’s method [36]. The triblock-copolymer P123 was used as porogen to create ordered mesopores, which facilitated the removal of template molecules and recognition of BPA, and meanwhile it improved the silica absorption capability. As shown in Fig. 1b, the preparation included three major steps. The silica incorporating template and P123 (MIMO (P123)) was first synthesized by gelation of BPAF-Si and TEOS in an aqueous solution containing P123, NaCl, and HCl. After quiescent curing to improve the silica
Please cite this article in press as: D. Yu, et al., Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.04.006
G Model CHROMA-356426; No. of Pages 8 4
ARTICLE IN PRESS D. Yu et al. / J. Chromatogr. A xxx (2015) xxx–xxx
Fig. 4. Nitrogen adsorption-desorption isotherm of MIMO-ir. Fig. 2. FTIR spectra of the imprinted mesoporous silica. P123: the pure porogen; MIMO (P123): silica containing P123 and template; MIMO: silica containing template; MIMO-ir: silica after removing P123 and template.
rigidity and Soxhlet extraction with methanol to remove P123 in MIMO (P123), the MIMO was yielded. The final imprinted silica (MIMO-ir) was obtained by heating the MIMO particles to cleave the carbamate bonds and removing the template molecules. The removals of the porogen P123 and template BPAF were monitored by FT-IR spectra (Fig. 2). In the MIMO spectrum, the peak of OH group in P123 at 1641 cm−1 almost disappeared after Soxhlet extraction, indicating that the porogen was virtually eliminated. Furthermore, compared with MIMO spectrum, the strong peak of carbamate C O stretch at 1732 cm−1 disappeared in the MIMOir spectrum, demonstrating the complete removal of the template molecules. 3.1.3. Characterization of MIMO-ir The morphology of MIMO-ir was characterized by SEM, TEM, and nitrogen adsorption-desorption method. SEM image showed that the MIMO-ir particles have the irregular shape with size of 10–20 m, which were suitable to be used as the SPE packing material (Fig. 3a). Moreover, TEM image displayed the ordered mesopores in the imprinted silica (Fig. 3b). Fig. 4 presented the nitrogen gas adsorption–desorption isotherm of the imprinted silica. The curves exhibited a typical hysteresis loop for the mesoporous solids with ordered pores. The BET area of the imprinted silica was 606.70 m2 /g with a total pore volume of 0.4703 cm3 /g. The average pore size was 3.10 nm. The high BET area can greatly improve the adsorption capacity of the imprinted silica.
3.2. Evaluation of MIMO-ir 3.2.1. Static and kinetic adsorption ability of MIMO-ir The static adsorption test was performed to evaluate the imprinting effect of the silica. The adsorption isotherms of BPAF and BPA on the mesoporous silica (MIMO-ir and NIMO-ir) were shown in Fig. 5a. It can be seen that the amounts of BPAF adsorbed on MIMO-ir are much higher than those on NIMO-ir in the initial concentration range of 0.050–4.0 mmol/L, indicating that MIMO-ir particles have obvious imprinting effect towards BPAF. The results demonstrated that the imprinting strategy used here was successful. Meanwhile, the MIMO-ir also showed higher adsorption ability towards BPA than the NIMO-ir in the initial concentration range of 0.030-4.0 mmol/L. The reason could be that the chemical structure of BPA was almost the same as that of BPAF except that the CF3 moieties are replaced by CH3 (Fig. 1). When mixed with the imprinted material, BPA molecules can easily enter the BPAFimprinted cavities and bind with the NH2 groups in the cavities via hydrogen bonding as shown in Fig. 1b, and therefore the imprinted silica presented high recognition ability to BPA. The binding data of the mesoporous silica for BPA were processed using Scatchard analysis. As shown in Fig. 5b, two straight lines were obtained in the plot region for MIMOir, indicating that there were two distinct binding sites: the high-affinity (specific) and the low-affinity (non-specific) sites. The linear regression equations for the two straight lines were Q/cf = 0.32007–0.03216Q (r = 0.9454) and Q/cf = 0.12423–0.00189Q (r = 0.9692). According to the slopes and intercepts of these equations, the corresponding dissociation constants (Kd ) and saturation
Fig. 3. (a) SEM and (b) TEM images of MIMO-ir.
Please cite this article in press as: D. Yu, et al., Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.04.006
G Model CHROMA-356426; No. of Pages 8
ARTICLE IN PRESS D. Yu et al. / J. Chromatogr. A xxx (2015) xxx–xxx
a
5
b
c
Fig. 5. (a) Adsorption isotherms of BPAF and BPA on MIMO-ir or NIMO-ir. The mean of three replicates was used as each adsorption value with error bar. (b) Scatchard plot analysis of the binding of BPA to MIMO-ir and NIMO-ir. (c) Kinetic binding profiles of BPA (4.0 mmol/L) on MIMO-ir and NIMO-ir.
capacity (Qmax ) values were calculated to be 31.09 mol/L and 9.952 mol/g for the high-affinity binding sites, and 529.10 mol/L and 65.73 mol/g for the low-affinity binding sites. The Kd and Qmax values for the high-affinity binding sites were much better than those on the non-covalently dummy MIPs [6]. The adsorption kinetics of MIMO-ir for BPA was also investigated. As shown in Fig. 5c, the adsorption equilibrium was reached within 30 min. The fast adsorption kinetics of the imprinted silica might be attributed to its mesoporous structure. 3.2.2. SPE ability of MIMO-ir In addition, the SPE ability of the synthesized silica particles (MIMO-ir and NIMO-ir) to BPAF and BPA were also investigated. A series of BPAF and BPA aqueous solutions with different concentrations (0.0010, 0.010, 0.10, and 1.0 mmol/L) were chosen for the SPE tests. As shown in Fig. 6a, the extraction efficiencies of BPAF on MIMO-ir are between 88-105%, which were much higher than those on NIMO-ir (35–61%), further verifying the success of the present imprinting technique. Meanwhile, the MIMO-ir also showed good recognition ability towards BPA with the extraction efficiencies of 94–100%, which were similarly higher than those on NIMO-ir (15–41%). These SPE results agreed well with those in the
a
static adsorption test. Since the MIMO-ir particles had high extraction efficiency towards BPA and avoided the interference of the residual BPA leaking from the BPA-imprinted materials, they could be potentially utilized for the SPE of trace BPA in real samples. To validate the new SPE method, its enrichment factors were measured with different volumes (5, 10, 15, and 20 mL) of 0.0010 mmol/L BPA aqueous solution percolating through the MIMO-ir cartridge. As depicted in Fig. 6b, the corresponding enrichment factors were 20, 40, 60, and 80 with extraction efficiencies of 94–105%. However, when 25 mL of 0.0010 mmol/L BPA solution percolated through the cartridge, the extraction efficiency decreased considerably. Therefore, a volume within 5.0–20 mL can be selected for the SPE test. It should be pointed out that when the volume of methanol to re-dissolve the SPE residues was less than 0.25 mL, a much higher enrichment factor could be obtained. For instance, if 0.10 mL of methanol was used for re-dissolution, the maximum enrichment factor reach ∼200. 3.2.3. Reusability of MIMO-ir The reusability of the imprinted sorbent was also investigated. After each SPE test, the cartridge was rinsed with methanol and water in sequence, and subsequently it was reused for the next SPE
b
Fig. 6. Extraction efficiencies of (a) BPA and BPAF aqueous solution on MIMO-ir or NIMO-ir cartridge with different concentrations and (b) 0.001 mmol/L BPA aqueous solution on MIMO-ir cartridge with different enrichment factors. The mean of three replicates was used as each extraction efficiency with error bar.
Please cite this article in press as: D. Yu, et al., Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.04.006
G Model
ARTICLE IN PRESS
CHROMA-356426; No. of Pages 8
D. Yu et al. / J. Chromatogr. A xxx (2015) xxx–xxx
6
Fig. 7. Reusability of MIMO-ir cartridge. Fig. 8. HPLC-UV chromatograms of 10 mL lake water spiked with 1 ng/mL BPA (a) after SPE and (b) before SPE using BPAF-imprinted silica as sorbent. HPLC-UV chromatograms of 10 mL ultrapure water after SPE using (c) BPAF- and (d) BPA-imprinted silica as sorbent.
Table 1 Determination of BPA in real samples. Samples
Spiked (ng/mL)
Found (ng/mL)
Tap water
0 10 20 50 0 10 20 50
ARTICLE IN PRESS
CHROMA-356426; No. of Pages 8
Journal of Chromatography A, xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A Dan Yu a , Xiaolei Hu a , Shoutai Wei a , Qiang Wang a , Chiyang He a,∗ , Shaorong Liu b a b
School of Chemistry and Chemical Engineering, Wuhan Textile University, 1 Textile Road, Wuhan 430073, China Department of Chemistry and Biochemistry, University of Oklahoma, 101 Stephenson Parkway, Norman, OK 73019, USA
a r t i c l e
i n f o
Article history: Received 10 January 2015 Received in revised form 1 April 2015 Accepted 2 April 2015 Available online xxx Keywords: Dummy molecular imprinting Hybrid imprinting Ordered mesoporous silica Solid-phase extraction Bisphenol A HPLC-UV
a b s t r a c t A novel hybrid dummy imprinting strategy was developed to prepare a mesoporous silica for the solidphase extraction (SPE) of bisphenol A (BPA). A new covalent template-monomer complex (BPAF-Si) was first synthesized with 2,2-bis(4-hydroxyphenyl)hexafluoropropane (BPAF) as the template. The imprinted silica was obtained through the gelation of BPAF-Si with tetraethoxysilane and the subsequent removal of template by thermal cleavage, and then it was characterized by FT-IR spectroscopy, scanning electron microscopy, transmission electron microscopy, and nitrogen adsorption–desorption isotherms. Results showed that the new silica had micron-level particle size and ordered mesoporous structure. The static binding test verified that the imprinted silica had much higher recognition ability for BPA than the non-imprinted silica. The imprinted silica also showed high extraction efficiencies and high enrichment factor for SPE of BPA. Using the imprinted silica, a SPE-HPLC-UV method was developed and successfully applied for detecting BPA in BPA-spiked tap water and lake water samples with a recovery of 99–105%, a RSD of 2.7–5.0% and a limit of detection (S/N = 3) of 0.3 ng/mL. The new imprinted silica avoided the interference of the residual template molecules and reduced the non-specific binding sites, and therefore it can be utilized as a good sorbent for SPE of BPA in environmental water samples. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Bisphenol A (BPA) is a xenoestrogenic chemical that is widely used in the production of polycarbonate plastic and epoxy resins [1–3]. BPA can cause cardiovascular disease, diabetes, neuro-behavioral disorders, carcinogenic hypersensitivity, and reproductive impairment [3–5]. BPA is commonly present in our environmental matrices, biological samples, food, and beverages [1,6,7]. However, it is challenging to analyze it because of its low concentration and matrix complexity in real-world samples [8]. Therefore, pretreatment and enrichment are crucial steps prior to its analysis. Solid-phase extraction (SPE) is by far the most frequently used pretreatment method for detecting BPA because SPE provides benefits of low cost, high enrichment factor, simple operation, and low consumption of toxic organic solvent [9,10]. A number of sorbents have been used for SPE of BPA such as divinylbenzene/N-vinylpyrrolidone copolymer [10], reversedphase silica [11], hydrophilic-hydrophobic balance sorbent [12],
∗ Corresponding author. Tel.: +86 27 59367685; fax: +86 27 59367336. E-mail address: [email protected] (C. He).
and mixed-mode cationic exchanger [13]. These traditional sorbents have high capacities, but they suffer from the low extraction selectivity towards BPA. Immuno-sorbents can provide high selectivity on the basis of molecular recognition. However, the main drawbacks of utilizing this sorbent are its high consumable costs (it consumes expensive antibodies) and limited column reusability [10]. Molecularly imprinted polymer (MIP) is a type of tail-made mimics of antibody with specific recognition ability, good stability, ease of preparation, and low cost [8,14–23]. MIPs are generally prepared via the co-polymerization of the imprint-monomer complex and cross-linking monomers, followed by removing the template molecules by solvent extraction and chemical cleavage [14]. The resulting imprint cavities, with the complementary size, shape and moieties to the template, can bind the target molecules specifically [15]. At present, there are three different strategies for synthesizing MIPs: non-covalent, covalent, and hybrid imprinting [18]. MIPs have been widely used as highly selective sorbents of SPE in recent years [24–33]. Some BPA-imprinted materials have also been applied to the determination of BPA [34–38]. Jiang et al. [34] prepared a BPA-imprinted amino-functionalized silica film coating on the surface of activated silica particles, Liu et al. [35] synthesized a BPA-imprinted polymer coated magnetic nanoparticle,
http://dx.doi.org/10.1016/j.chroma.2015.04.006 0021-9673/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: D. Yu, et al., Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.04.006
G Model CHROMA-356426; No. of Pages 8
ARTICLE IN PRESS D. Yu et al. / J. Chromatogr. A xxx (2015) xxx–xxx
2
Lofgreen et al. [36] developed a new BPA-imprinted mesoporous organosilica, Chen et al. [37] utilized a hierarchically BPA-imprinted mesoporous silica, and Yang et al. [38] utilized a sponge-like BPAimprinted silica for SPE of BPA. These BPA-imprinted sorbents, prepared by different non-covalent imprinting or hybrid imprinting techniques, improved the extraction selectivity to some extent, but the residual BPA leakage was a serious issue for determining trace amount of BPA. Dummy molecularly imprinted polymers (DMIPs) can effectively circumvent this problem. Several DMIPs have been prepared presently using different analogues of BPA as templates for recognition of BPA [6,39–41]. These include DMIP-coated magnetic nanoparticles prepared using bisphenol F as a template [39], a DMIP-coated stir bar synthesized using 4,4 -dihydroxybiphenyl as a template [40], and a hollow porous DMIP prepared using tetrabromobisphenol A as a template [41]. However, these materials were all synthesized by non-covalent imprinting techniques that required the use of excess functional monomers, creating a large number of nonspecific binding sites, and consequently affecting MIP’s selectivity for BPA [15,18]. The hybrid imprinting approach, a combination of covalent and non-covalent strategy, not only ensures the functional monomer residues to be present only in the imprint cavities (and thus greatly reduces the non-specific binding sites), but also makes the MIPs rapidly take up target molecules via the non-covalent interactions [18]. Herein, to avoid the interference of the residual template molecules and reduce the number of non-specific binding sites, we combine a dummy molecular imprinting with a hybrid strategy for the first time to prepare a mesoporous silica for SPE of BPA. A new covalent template-monomer complex (BPAF-Si) was first synthesized with 2,2-bis(4-hydroxyphenyl) hexafluoropropane (BPAF) as the dummy template. The imprinted mesoporous silica was obtained via gelation of BPAF-Si with tetraethoxysilane, followed by removing template by thermal cleavage of the urethane bonds. The mesoporous silica was characterized by FT-IR spectroscopy, scanning electron microscopy (SEM), transmission electron microscopy (TEM) and nitrogen adsorption-desorption isotherms. After that, the adsorption and SPE ability of the imprinted sorbent for both the template and BPA were investigated. Finally, coupled with a HPLC-UV method, the new imprinted silica was successfully applied to the SPE and detection of trace amount of BPA spiked in real water samples. 2. Experimental 2.1. Chemicals and materials Bisphenol A (BPA, 96%), 2,2-bis(4-hydroxyphenyl)hexafluoropropane (BPAF, 99%), dibutyltindilaurate (DBDU, 98%) and tetrahydrofuran (THF, superdry, 99.5%) were purchased from J&K Scientific LTD (Beijing, China). 3-(Triethyloxysily) propyl isocyanate (ICPTES, 95%) was obtained from TCI (Tokyo, Japan). Tetraethoxysilane (TEOS, 99%) was supplied by Alfa Aesar (Tianjin, China). Pluronic P123 (P123, Mn ∼ 5800 g/mol) was purchased from Aldrich (Steinheim, Germany). Methanol and acetonitrile were of HPLC grade supplied by Merck (Darmstadt, Germany). The ultrapure water was obtained from the KL-III-40 purification system (AK, Taiwan). All of other reagents were of analytical grade and used as supplied without further purification. 2.2. Preparation of BPAF-Si Under nitrogen gas, BPAF (2.0 g, 6 mmol) was dissolved in THF (60 mL). ICPTES (6.6 mL, 26.2 mmol) and DBDU (4 mL) were added to the solution at room temperature. The mixture was stirred at
75 ◦ C to allow the reaction to proceed for 24 h. After the solvent was evaporated, the product was isolated by column chromatography on silica gel using an eluent containing ethyl acetate and petroleum ether (ethyl acetate: petroleum ether = 1:2). The oily product, BPAF-Si, was obtained at a yield of 87%. FT-IR (at resolution of 4 cm−1 ): 2978/2931/2893 cm−1 (vC H ), 1728 cm−1 (vHNC O ), 1535 cm−1 (ıN H ), 1100–1000 cm−1 (vSi O Si ). 1 H NMR (300 MHz, CDCl3 ): ı (ppm) 7.27–7.30 (d, 4H), 7.03–7.12 (d, 4H), 6.71–6.74 (d, 2H), 3.70–3.80 (q, 12H), 3.15–3.22 (t, 4H), 1.49–1.57 (m, 4H), 1.11–1.18 (t, 18H), 0.52–70.63 (m, 4H). 13 C NMR (300 MHz, CDCl3 ): ı (ppm) 156.42, 153.32, 150.61, 130.29, 128.87, 120.27, 114.15, 57.47, 42.63, 22.29, 17.24, 6.64. For preparation of BPA-imprinted mesoporous silica, the covalent complex (BPA-Si) of BPA and ICPTES was also synthesized and characterized using the same methods as above. 2.3. Preparation of imprinted and non-imprinted silica A stock solution of porogen was prepared by mixing P123 (8.4 g, 1.5 mmol), NaCl (24.4 g, 0.418 mol), water (69.6 g, 3.86 mol), and 2 mol/L HCl (208.8 g, 11.6 mol H2 O, 0.42 mol HCl), and stirring the mixture until a homogeneous solution was obtained. Molecularly imprinted mesoporous silica (MIMO) was synthesized by mixing a pre-dissolved solution of BPAF-Si (0.347 g, 0.48 mmol) in TEOS (1.939 mL, 8.6 mmol) with 44 g of the stock porogen solution. The non-imprinted mesoporous silica (NIMO) was prepared in the almost same manner for the imprinted particles except that ICPTES (0.2365 g, 0.96 mmol) was used in place of BPAF-Si. Each mixture was stirred at room temperature for 24 h, then transferred to an oven at 80 ◦ C, and cured quiescently for 24 h. The resultant particulate material was collected by filtration, and the residual P123 was removed by washing with ethanol via Soxhlet extraction for 20 h. The removal of the template molecules from the imprinted silica was carried out by suspending MIMO (1 g) in the mixture of dimethylsulfoxide (DMSO) and water and subsequent heating at 160 ◦ C for 5 h with stirring. The imprint-removed silica (MIMO-ir) was isolated by centrifugation, rinsed three times with distilled water and ethanol alternately, and then oven-dried for 2 days. The same treatment for NIMO was achieved to obtain final nonimprinted silica (NIMO-ir). The removals of P123 and template were monitored by FT-IR. The obtained MIMO-ir particles were characterized by SEM, TEM, and nitrogen adsorption–desorption isotherms. To directly verify the leakage of residual BPA from BPAimprinted material, a BPA-imprinted mesoporous silica was prepared in the same manner as above except that BPAF-Si was replaced by BPA-Si. 2.4. Adsorption experiments For static adsorption tests, 20 mg of MIMO-ir particles were added in 1.0 mL of BPAF or BPA aqueous solution (containing 25% methanol) with a concentration range of 0.050–4.0 mmol/L, and then the mixtures were sonicated to disperse the particles completely. The suspensions were shaken continually at 25 ◦ C for 2 h and subsequently centrifuged. The amount of BPAF or BPA in the supernatant was measured by HPLC with a UV absorbance detector at 267 nm. The same procedure was performed for NIMO-ir particles. The adsorption capacity and the dissociation constant (Kd , mol/L) were calculated according to the Eqs. (1) and (2) [6]: Q =
(c0 − cf )v m
(1)
Please cite this article in press as: D. Yu, et al., Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.04.006
G Model CHROMA-356426; No. of Pages 8
ARTICLE IN PRESS D. Yu et al. / J. Chromatogr. A xxx (2015) xxx–xxx
3
a
b
Fig. 1. Schemes of (a) synthesis of BPAF-Si and (b) imprinting and recognition of the mesoporous silica.
1 Qmax Q =− Q+ cf Kd Kd
(2)
where c0 (mol/L) and cf (mol/L) are the initial and final concentrations of BPA, v (L) is the total volume of sample, m (g) is the mass of sorbent, Q and Qmax (mol/L) are the amount of BPA adsorbed at equilibrium and saturation, respectively. The kinetic adsorption study was performed using similar procedure with the static adsorption experiment except that 4.0 mmol/L BPA in 25% methanol aqueous solution and different equilibrium periods (7.5–60 min) were adopted. 2.5. Solid-phase extraction The SPE cartridge was prepared by packing 300 mg MIMO-ir or NIMO-ir particles into a 3-mL empty polyethylene syringe between two sieve plates. Each SPE cartridge was rinsed sequentially by methanol and deionized water for conditioning before loading sample aqueous solutions. Due to the poor solubility of BPAF and BPA in water, methanol-water mixtures were used as solvents to improve their solubility (25% methanol for 1.0 and 0.10 mmol/L BPA solution, 5% methanol for 0.010 and 0.0010 mmol/L BPA solution, 40% methanol for 1.0 and 0.10 mmol/L BPAF solution, 10% methanol for 0.010 and 0.0010 mmol/L BPAF solution). Each SPE cartridge was loaded with 1.0 mL sample solution, rinsed with 1.0 mL solvent, and eluted with 2.0 mL methanol in sequence. After that, the solvent in the effluent was evaporated to dryness by nitrogen. Finally, the dry residues were redissolved with 0.25 mL methanol and analyzed by HPLC. The extraction efficiency was calculated as the ratio of the mass of analyte after SPE to that before SPE. For the analysis of real water samples, tap water and lake water were centrifuged and then filtered with 0.45-m film prior to SPE. For each SPE test, 10.0 mL of the pretreated water sample was used. 2.6. HPLC analysis HPLC analysis was achieved on a Ultimate 3000 HPLC system (Thermo Fisher Scientific, USA), using a Dionex AcclaimTM 120 C18 column (250 mm × 4.6 mm, 5 m, Dionex Bonded Silica Products) and a UV detector (set at 276 nm for all compounds), with methanol and ultrapure water (70/30, v/v) as eluent at a rate of 1.0 mL/min for adsorption and SPE experiments. For analysis of real water samples, a gradient program was used at a flow rate of 1 mL/min, by combining solvent A (water) and solvent B (methanol) as follows: 35–100%
(25 min). The column temperature was kept at 25 ◦ C. The injection volume was 20 L for each analysis. 2.7. Characterization of prepared sorbents All infrared spectra were collected using a Nicolet Avatar 360 FT-IR spectrometer (Thermo Fisher Scientific, USA) within a scan range of 4000–500 cm−1 . SEM images were taken by a Hitachi SU8010 microscope. TEM images were obtained on a FEI Tecnai G2 F30 microscope. Nitrogen adsorption was carried out on a Gemini VII 2390 Series Surface Area Analyzers. 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. 3. Results and discussion 3.1. Preparation and characterization of MIMO-ir 3.1.1. Synthesis of BPAF-Si To eliminate the undesired leakage of residual BPA in the traditional BPA-imprinted materials, we used a dummy template molecule (BPAF, a structural analogue of BPA) for preparing the SPE sorbent. A new template-monomer complex (BPAF-Si) was first prepared; the schematic illustration of the preparation procedure for this complex is shown in Fig. 1a. Similar to what had been reported in the literatures [32,36,42], the reaction occurred between the isocyanate group of ICPTES and the phenol moieties of BPAF. The product contained two thermally cleavable urethane bonds that allowed convenient removal of the template molecules in the following step. The resultant product was characterized by FT-IR, 1 H NMR, and 13 C NMR. 3.1.2. Preparation of MIMO-ir The dummy molecularly imprinted mesoporous silica (MIMOir) was prepared via a triblock-copolymer-template sol-gel strategy similar to Lofgreen’s method [36]. The triblock-copolymer P123 was used as porogen to create ordered mesopores, which facilitated the removal of template molecules and recognition of BPA, and meanwhile it improved the silica absorption capability. As shown in Fig. 1b, the preparation included three major steps. The silica incorporating template and P123 (MIMO (P123)) was first synthesized by gelation of BPAF-Si and TEOS in an aqueous solution containing P123, NaCl, and HCl. After quiescent curing to improve the silica
Please cite this article in press as: D. Yu, et al., Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.04.006
G Model CHROMA-356426; No. of Pages 8 4
ARTICLE IN PRESS D. Yu et al. / J. Chromatogr. A xxx (2015) xxx–xxx
Fig. 4. Nitrogen adsorption-desorption isotherm of MIMO-ir. Fig. 2. FTIR spectra of the imprinted mesoporous silica. P123: the pure porogen; MIMO (P123): silica containing P123 and template; MIMO: silica containing template; MIMO-ir: silica after removing P123 and template.
rigidity and Soxhlet extraction with methanol to remove P123 in MIMO (P123), the MIMO was yielded. The final imprinted silica (MIMO-ir) was obtained by heating the MIMO particles to cleave the carbamate bonds and removing the template molecules. The removals of the porogen P123 and template BPAF were monitored by FT-IR spectra (Fig. 2). In the MIMO spectrum, the peak of OH group in P123 at 1641 cm−1 almost disappeared after Soxhlet extraction, indicating that the porogen was virtually eliminated. Furthermore, compared with MIMO spectrum, the strong peak of carbamate C O stretch at 1732 cm−1 disappeared in the MIMOir spectrum, demonstrating the complete removal of the template molecules. 3.1.3. Characterization of MIMO-ir The morphology of MIMO-ir was characterized by SEM, TEM, and nitrogen adsorption-desorption method. SEM image showed that the MIMO-ir particles have the irregular shape with size of 10–20 m, which were suitable to be used as the SPE packing material (Fig. 3a). Moreover, TEM image displayed the ordered mesopores in the imprinted silica (Fig. 3b). Fig. 4 presented the nitrogen gas adsorption–desorption isotherm of the imprinted silica. The curves exhibited a typical hysteresis loop for the mesoporous solids with ordered pores. The BET area of the imprinted silica was 606.70 m2 /g with a total pore volume of 0.4703 cm3 /g. The average pore size was 3.10 nm. The high BET area can greatly improve the adsorption capacity of the imprinted silica.
3.2. Evaluation of MIMO-ir 3.2.1. Static and kinetic adsorption ability of MIMO-ir The static adsorption test was performed to evaluate the imprinting effect of the silica. The adsorption isotherms of BPAF and BPA on the mesoporous silica (MIMO-ir and NIMO-ir) were shown in Fig. 5a. It can be seen that the amounts of BPAF adsorbed on MIMO-ir are much higher than those on NIMO-ir in the initial concentration range of 0.050–4.0 mmol/L, indicating that MIMO-ir particles have obvious imprinting effect towards BPAF. The results demonstrated that the imprinting strategy used here was successful. Meanwhile, the MIMO-ir also showed higher adsorption ability towards BPA than the NIMO-ir in the initial concentration range of 0.030-4.0 mmol/L. The reason could be that the chemical structure of BPA was almost the same as that of BPAF except that the CF3 moieties are replaced by CH3 (Fig. 1). When mixed with the imprinted material, BPA molecules can easily enter the BPAFimprinted cavities and bind with the NH2 groups in the cavities via hydrogen bonding as shown in Fig. 1b, and therefore the imprinted silica presented high recognition ability to BPA. The binding data of the mesoporous silica for BPA were processed using Scatchard analysis. As shown in Fig. 5b, two straight lines were obtained in the plot region for MIMOir, indicating that there were two distinct binding sites: the high-affinity (specific) and the low-affinity (non-specific) sites. The linear regression equations for the two straight lines were Q/cf = 0.32007–0.03216Q (r = 0.9454) and Q/cf = 0.12423–0.00189Q (r = 0.9692). According to the slopes and intercepts of these equations, the corresponding dissociation constants (Kd ) and saturation
Fig. 3. (a) SEM and (b) TEM images of MIMO-ir.
Please cite this article in press as: D. Yu, et al., Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.04.006
G Model CHROMA-356426; No. of Pages 8
ARTICLE IN PRESS D. Yu et al. / J. Chromatogr. A xxx (2015) xxx–xxx
a
5
b
c
Fig. 5. (a) Adsorption isotherms of BPAF and BPA on MIMO-ir or NIMO-ir. The mean of three replicates was used as each adsorption value with error bar. (b) Scatchard plot analysis of the binding of BPA to MIMO-ir and NIMO-ir. (c) Kinetic binding profiles of BPA (4.0 mmol/L) on MIMO-ir and NIMO-ir.
capacity (Qmax ) values were calculated to be 31.09 mol/L and 9.952 mol/g for the high-affinity binding sites, and 529.10 mol/L and 65.73 mol/g for the low-affinity binding sites. The Kd and Qmax values for the high-affinity binding sites were much better than those on the non-covalently dummy MIPs [6]. The adsorption kinetics of MIMO-ir for BPA was also investigated. As shown in Fig. 5c, the adsorption equilibrium was reached within 30 min. The fast adsorption kinetics of the imprinted silica might be attributed to its mesoporous structure. 3.2.2. SPE ability of MIMO-ir In addition, the SPE ability of the synthesized silica particles (MIMO-ir and NIMO-ir) to BPAF and BPA were also investigated. A series of BPAF and BPA aqueous solutions with different concentrations (0.0010, 0.010, 0.10, and 1.0 mmol/L) were chosen for the SPE tests. As shown in Fig. 6a, the extraction efficiencies of BPAF on MIMO-ir are between 88-105%, which were much higher than those on NIMO-ir (35–61%), further verifying the success of the present imprinting technique. Meanwhile, the MIMO-ir also showed good recognition ability towards BPA with the extraction efficiencies of 94–100%, which were similarly higher than those on NIMO-ir (15–41%). These SPE results agreed well with those in the
a
static adsorption test. Since the MIMO-ir particles had high extraction efficiency towards BPA and avoided the interference of the residual BPA leaking from the BPA-imprinted materials, they could be potentially utilized for the SPE of trace BPA in real samples. To validate the new SPE method, its enrichment factors were measured with different volumes (5, 10, 15, and 20 mL) of 0.0010 mmol/L BPA aqueous solution percolating through the MIMO-ir cartridge. As depicted in Fig. 6b, the corresponding enrichment factors were 20, 40, 60, and 80 with extraction efficiencies of 94–105%. However, when 25 mL of 0.0010 mmol/L BPA solution percolated through the cartridge, the extraction efficiency decreased considerably. Therefore, a volume within 5.0–20 mL can be selected for the SPE test. It should be pointed out that when the volume of methanol to re-dissolve the SPE residues was less than 0.25 mL, a much higher enrichment factor could be obtained. For instance, if 0.10 mL of methanol was used for re-dissolution, the maximum enrichment factor reach ∼200. 3.2.3. Reusability of MIMO-ir The reusability of the imprinted sorbent was also investigated. After each SPE test, the cartridge was rinsed with methanol and water in sequence, and subsequently it was reused for the next SPE
b
Fig. 6. Extraction efficiencies of (a) BPA and BPAF aqueous solution on MIMO-ir or NIMO-ir cartridge with different concentrations and (b) 0.001 mmol/L BPA aqueous solution on MIMO-ir cartridge with different enrichment factors. The mean of three replicates was used as each extraction efficiency with error bar.
Please cite this article in press as: D. Yu, et al., Dummy molecularly imprinted mesoporous silica prepared by hybrid imprinting method for solid-phase extraction of bisphenol A, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.04.006
G Model
ARTICLE IN PRESS
CHROMA-356426; No. of Pages 8
D. Yu et al. / J. Chromatogr. A xxx (2015) xxx–xxx
6
Fig. 7. Reusability of MIMO-ir cartridge. Fig. 8. HPLC-UV chromatograms of 10 mL lake water spiked with 1 ng/mL BPA (a) after SPE and (b) before SPE using BPAF-imprinted silica as sorbent. HPLC-UV chromatograms of 10 mL ultrapure water after SPE using (c) BPAF- and (d) BPA-imprinted silica as sorbent.
Table 1 Determination of BPA in real samples. Samples
Spiked (ng/mL)
Found (ng/mL)
Tap water
0 10 20 50 0 10 20 50
