Materials Science and Engineering C 49 (2015) 640–647
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
A bisphenol A sensor based on novel self-assembly of zinc phthalocyanine tetrasulfonic acid-functionalized graphene nanocomposites Keyu Hou a,b,c, Lei Huang b,c, Yongbo Qi b,c, Caixia Huang a,b,c, Haibo Pan a,b,c,⁎, Min Du a a b c
Fujian Key Lab of Medical Instrument & Pharmaceutical Technology, Yishan Campus, Fuzhou University, Fuzhou, Fujian 350002, China Institute of Research for Functional Materials, Yishan Campus, Fuzhou University, Fuzhou, Fujian 350002, China College of Chemistry, Qishan Campus, Fuzhou University, Fuzhou, Fujian 350116, China
a r t i c l e
i n f o
Article history: Received 21 September 2014 Received in revised form 2 December 2014 Accepted 17 January 2015 Available online 19 January 2015 Keywords: Graphene Self-assembly π–π stack Bisphenol A detection Metallophthalocyanine
a b s t r a c t In this work, a novel zinc phthalocyanine tetrasulfonic acid (ZnTsPc)-functionalized graphene nanocomposites (f-GN) was synthesized by a simple and efficient electrostatic self-assembly method, where the positive charged GN decorated by (3-aminopropyl) triethoxysilane (APTES) was self-assemblied with ZnTsPc, a two dimensional (2-D) molecules. It not only enhanced its stability for the hybrid structure, but also avoided the reaggregation of ZnTsPc or f-GN themselves. Based on layered ZnTsPc/f-GN nanocomposites modified glassy carbon electrode, a rapid and sensitive sensor was developed for the determination of bisphenol A (BPA). Under the optimal conditions, the oxidation peak current increased linearly with the concentration of BPA in the range of 5.0 × 10−8 to 4.0 × 10−6 M with correlation coefficient 0.998 and limits of detection 2.0 × 10−8 M. Due to high absorption nature for BPA and electron deficiency on ZnTsPc/f-GN, it presented the unique electron pathway arising from π–π stackable interaction during redox process for detecting BPA. The sensor exhibited remarkable long-term stability, good anti-interference and excellent electrocatalytic activity towards BPA detection. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Bisphenol A (BPA, 2,2-bis(4-hydroxyphenyl)propane) is a highly versatile hydrocarbon molecule that forms major feedstock in the production of epoxy resins and polycarbonate plastics [1–3]. BPA is readily enriched in the human body through environment and the food chain [4,5]. As an endocrine disrupting chemical, BPA is capable of disturbing the normal endocrine function, influencing sexual development and reproductive capability. In consideration of several laborious and timeconsuming steps in mass spectrometry, gas chromatography (GC) and liquid chromatography (LC) for detecting BPA, main attention is recently devoted to electroanalysis techniques [6–9], which are rapid, sensitive, inexpensive and suitable for on-site monitoring. As organic molecules including of a planar π-conjugated skeleton similarity in structure to the biological molecules (chlorophyll, hemoglobin) [10–13], metallophthalocyanine derivatives (MPcs) exhibit great thermal and chemical stability, and an efficient biomimetic catalysts for redox reactions. Zinc phthalocyanine tetrasulfonic acid
⁎ Corresponding author at: Fujian Key Lab of Medical Instrument & Pharmaceutical Technology, Yishan Campus, Fuzhou University, Fuzhou, Fujian 350002, China. E-mail address: [email protected] (H. Pan).
http://dx.doi.org/10.1016/j.msec.2015.01.064 0928-4931/© 2015 Elsevier B.V. All rights reserved.
(ZnTsPc, the molecule structure in Scheme 1), as electronic donor– acceptor coordination complexes, should contribute to electrocatalysis related with redox reaction. Generally, MPcs commonly suffer from hydrophobicity and aggregation (especially in aqueous solutions) owing to large π-conjugated systems. Although many water-soluble MPcs have been synthesized by introducing hydrophilic moieties and sulfonic groups, it is highly aggregated in water. Graphene (GN), 2-D monolayer comprising sp2-hybridized carbon atoms with the huge surface-to-volume ratio, fast charge carrier mobility, high thermal conductivity, and strong Young's modulus, has attracted wide attention for both fundamental science and application. For instance, some recent works related with the excellent ability of GN in electrochemical biosensors have come to our sight [14–16]. Recently, the GN modified with MPcs is under intensive exploitation, and few reports were published in synthesis and characterization of MPcs on GN [17–20], where MPcs combine with GN covalently or coordinatively. In present work, we developed a novel approach to combine the positively charged functionalized nanoscale graphene (f-GN) with negatively charged ZnTsPc by a simple and efficient electrostatic selfassembly method (Scheme 1), where GN was decorated by (3aminopropyl) triethoxysilane (APTES) at first. Note that they would form strong noncovalent π–π conjugated composite (ZnTsPc/f-GN)
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
641
Scheme 1. Schematic illustration for electrostatic self-assembly of ZnTsPc/f-GN nanocomposites.
through the π–π stackable interaction [21]. And also, it exhibited charming characteristics in electrocatalysis feature by their synergetic combination of two components. Owing to the above advantages, an alternative electrochemical sensor for the selective and sensitive detection of BPA was realized by ZnTsPc/f-GN modified glassy carbon electrode (GCE). The redox mechanism on ZnTsPc/f-GN/GCE was raised further in BPA. For confirming its practical application, the determination of BPA in real samples was also fulfilled by using differential pulse voltammetry (DPV). 2. Experimental 2.1. Materials and reagents Natural graphite (FP 99.95% pure) and APTES (C9H23NO3Si,) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). ZnTsPc (C32H16N8O12S4Zn) was obtained from J&K Scientific Ltd. (Beijing, China). 3,3′,5,5′-tetrabromobisphenol A (TBBPA) was obtained from Alfa Aesar Company (Tianjin, China). BPA was purchased from TCI (Shanghai) Chemical Industry Co. Ltd. and its stock solution (2 mM) was prepared with ethanol. The working solutions were prepared by diluting the stock solution with phosphate buffer saline (PBS, 0.1 M, pH 7.0). All other chemicals were of analytical grade and used as received without any further purification. Deionized (DI) water obtained from a Milli-Q plus water purification system (Millipore Co. Ltd., USA, 18 MΩ·cm) was used throughout the study. 2.2. Apparatus Zeta-potential value measurement was recorded on Zeta Sizer 3000 Laser Particle Size and Zeta Potential Tester (Malvern Corporation, UK),
and DI water was used as a dispersant. Fourier transformed infrared spectroscopy (FT-IR) were performed on a Nicolet Nexus 670 FT-IR spectrophotometer at a resolution of 4 cm−1. UV–vis spectra were measured with a UV–vis absorption spectrometer (PerkinElmer Lambda 900). The morphologies of samples were observed by AFM (Agilent 5500, USA). High-resolution transmission electron microscope (HRTEM) image was obtained using Tecnai G2 F20 S-TWIN, 200 kV (FEI Company, USA). All the electrochemical measurements were carried out on a CHI660D electrochemical workstation (Shanghai Chenhua Instrument Co. Ltd., China) with a conventional three-electrode system. Glass carbon electrode (3 mm diameter) loaded different materials was used as working electrode. The surface of GCE was polished with 0.3 and 0.05 μm alumina slurries, ultrasonicated in water and ethanol, sequentially. Then the bare GCE was dried under pure N2. Finally, these materials were dropped on the surface of bare GCE, and dried under room temperature. 2.3. Preparation of graphene nanosheets Graphene oxide (GO) was synthesized from natural graphite powder using modified Hummers' method. Typically, 1 g of graphite powder was put into a mixture of 6 ml of concentrated H2SO4, 1.25 g of K2S2O8, and 1.25 g of P2O5. The solution was heated to 80 °C in an oil-bath kept stirring for 24 h. It was carefully diluted with 300 ml of DI water, filtered, and washed until pH ~ 7. The pre-oxidized graphite powder was added to a mixture of 60 ml of concentrated H2SO4 and 30 ml HNO3 under vigorous stirring at 0 °C. Then, 8 g KMnO4 was added gradually under stirring and the temperature of mixture was kept to be below 20 °C by cooling. Successively, it was diluted with 500 ml of DI water in an ice bath to keep the temperature below 50 °C for 2 h. After further dilution with 500 ml of DI water, 20 ml of H2O2
642
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
was added to remove excessive KMnO4. The brilliant yellow product called graphite oxide was purified by washing with 10% HCl and washed with warm water for several times. Exfoliation was accomplished by sonicating graphite oxide in aqueous solution for 120 min, and then it was centrifuged at 5000 rpm for 10 min to obtain GO solution. GO was reduced to GN by NaBH4 in water bath at 90 °C for 2 h. 2.4. Preparation of ZnTsPc/f-GN nanocomposites ZnTsPc/f-GN was fabricated by a simple and efficient electrostatic self-assembly method. In detail, APTES (400 μl) was added slowly into 40 ml GN aqueous solution (0.25 mg/ml) and the mixture was kept stirring for 10 h under room temperature. The decorated graphene nanosheets, f-GN, were collected by centrifugating at 8000 rpm for 15 min and washed with ethanol and water in succession until pH ~7. The final sediment was redispersed in DI water. Then, 40 mg ZnTsPc was dispersed in 1 ml of prepared f-GN solution by sonication for 30 min. The mixture was centrifuged and washed with DI water for several times after 24 hours' standing. Finally, well-distributed ZnTsPc/f-GN was obtained by redispersing the precipitates in DI water. 3. Results and discussion 3.1. Physicochemical characterization for ZnTsPc/f-GN nanocomposites After the decoration with APTES, the surface of GN nanosheets attached with amine functional groups was obtained. As displayed in Fig. S1, the zeta potential analysis of f-GN reveals a significant positively charged surface, in favor of avoiding the aggregation for 2-D GN layers. ZnTsPc/f-GN nanocomposites were also confirmed by the FT-IR spectra (Fig. 1). For GO (Fig. 1a), the stretching vibrations of O\H (3300 cm−1), C_O (1730 cm− 1), C_C (1625 cm−1), C\O\H (1350 cm− 1) and C\O\C (1050 cm−1) are observed, indicating the presence of various oxygen-containing functional groups. Compared with Fig. 1a, GN (Fig. 1b) shows survival of carboxyl groups and hydroxyl groups on the GN nanosheets even after reduction. As to f-GN (Fig. 1d), a sharp bend vibration of N\H at 1575 cm−1 presents through decorated GN with APTES, indicating that \NH2 groups attached onto the surface of GN. The characteristic peak around 1600 cm−1 is corresponding to the vibration of benzene ring. For ZnTsPc (Fig. 1c) and ZnTsPc/f-GN (Fig. 1e), these specific bands become much broader than that of other curves, implying that there are strong electronic interactions through the π–π stack between ZnTsPc and f-GN. A peak located at 1400 cm−1 is assigned to the stretching vibration of the C_N from ZnTsPc macrocycle which can also be found on ZnTsPc/f-GN. It is noted that a shift for coordination bond N → Zn from 912 cm−1 to 925 cm−1 occurs
Fig. 1. FT-IR spectra of (a) GO nanosheets, (b) GN nanosheets, (c) ZnTsPc, (d) f-GN and (e) ZnTsPc/f-GN nanocomposites.
(Fig. 1e), suggesting that the distribution of electrons around central metal Zn (II) prefers to f-GN. Such electron deficiency in the neighborhood of central metal Zn (II) is conducive to adsorption of phenolic hydroxyl groups in BPA, promoting electron transport from BPA to ZnTsPc/f-GN nanocomposites. Fig. 2 depicts UV–vis absorption spectra of f-GN, ZnTsPc and ZnTsPc/ f-GN dispersed in water and DMF for reference. ZnTsPc is known to readily form dimers in H2O due to the hydrophobic intermolecular affinity. Thus, Fig. 2b shows its partly aggregated form, and the inset of Fig. 2 represents its spectrum in monomeric form. A distinct blue shift of Q-band absorption of the dimer (λmax = 656 nm) as compared to the monomer (λmax = 677 nm) is associated with the overlap of π-electronic clouds of MPcs system in the dimer. The absorption peak of as-prepared f-GN at ~ 270 nm (Fig. 2a) is attributed to the n → π* transitions of C_O bonds, indicating that there is little influence after being decorated with APTES. It can be clearly seen that the spectrum of ZnTsPc/f-GN (Fig. 2c) displays four distinct absorption peaks. In addition, the color of ZnTsPc/f-GN dispersion exhibits blueblack that distinguishes from black f-GN and blue ZnTsPc (the inset in Fig. 2). Apparently, the light absorption below 300 nm is assigned to f-GN and a shoulder peak at 350 nm arises from the B-band absorption of ZnTsPc. As to a strong band at 705 nm, it represents that the red shift of Q-band of ZnTsPc in the composite occurred compared with that of pure ZnTsPc (Fig. 2b). It indicated that the π–π stackable interaction exists between ZnTsPc molecules and f-GN due to the benzene rings both in ZnTsPc and f-GN. At the same time, the interaction not only improves the stability of ZnTsPc/f-GN nanocomposites, but avoids the reaggregation of ZnTsPc or f-GN themselves. Nevertheless, these shifts in Q band of MPcs complexes are normally associated with electron transfer. In view of these shifts in FT-IR, we could conclude that electrons transfer from ZnTsPc to electron-accepting f-GN. According to the corresponding cross-sectional view, the typical AFM image (Fig. 3A) of f-GN indicates that the average thickness of nanosheets is about 1.021 nm, a little greater than the reported thickness of monolayer GN [22]. This can be also ascribed to APTES successfully decorated on the surface of f-GN. After being functionalized by ZnTsPc, ZnTsPc/f-GN became slightly rough with the increased mean thickness to 1.625 nm (Fig. 3B), declaring the formation of layered nanohybrids. It is known that the intensities, shapes and positions of the G band and 2D band peaks can be used to monitor the number of GN layers [23]. Raman spectra for f-GN, ZnTsPc, and ZnTsPc/f-GN were also measured (Fig. S2). According to the wider 2D band and ratio of the 2D/G intensity (f-GN: 0.409, ZnTsPc: 0.411), we estimated that the layers for GN diminish after ZnTsPc modifying on the f-GN.
Fig. 2. UV–vis absorption spectra of (a) f-GN, (b) ZnTsPc and (c) ZnTsPc/f-GN nanocomposites in H2O. Inset: The monomer spectrum of ZnTsPc in DMF.
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
643
Fig. 3. AFM images of (A) f-GN and (B) ZnTsPc/f-GN.
HRTEM (Fig. 4A) shows that the surface of f-GN is a thin wrinkling paper-like structure, while ZnTsPc/f-GN exhibits layered form after ZnTsPc molecules were loaded on f-GN (Fig. 4C). Moreover, in contrast with ZnTsPc/f-GN (Fig. 4D), based on the significant lattice fringes with typical fast Fourier transform (FFT) pattern, it verifies that the f-GN (Fig. 4B) is nearly single-crystalline.
3.2. Electrochemical properties of modified electrodes The electrochemical properties of the bare GCE and modified electrodes were performed by cyclic voltammograms (CVs) and electrochemical impedance spectroscopy (EIS), respectively. The CVs of different modified electrodes in 5.0 mM [Fe(CN)6]3−/4− solution containing
Fig. 4. HRTEM images of f-GN (A and B) and ZnTsPc/f-GN (C and D). Inset in (B) and (D): FFT pattern of lattice fringe images.
644
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
0.7
a b
10
c
0.5
I / µA
I / µA
5 0
-10
0.1
-15 -0.2
0.0
0.2
0.4
0.6
0.8
0.03 0.02
a
0.01
c
0.00 -0.01
0.3
0.4
0.3 0.2
0.5
0.6
E/V
0.7
0.8
b a 0.2
0.3
0.4
0.5
0.6
0.7
0.8
E/V
E/ V Fig. 5. Cyclic voltammograms of (a) bare GCE, (b) f-GN/GCE, (c) ZnTsPc/f-GN/GCE and (d) ZnTsPc in a mixture solution of 5.0 mmol·L−1 [Fe(CN)6]3−/4− and 0.1 mol·L−1 KCl with the scan rate at 0.1 V·s−1.
0.1 M KCl was collected in Fig. 5. As seen, a classic pair of well defined redox peaks was obtained at the bare GCE (curve a). After modification with f-GN (curve b), the redox peak currents of [Fe(CN)6]3 −/4 − decreased, representing that f-GN had been successfully covered on the electrode and hindered the electron transfer. However, the presence of ZnTsPc significantly diminished the redox of [Fe(CN)6]3−/4− on the electrode (curve d). In addition, the couple of redox peaks with the peak-to-peak (ΔEp) of 215 mV was observed, indicating the poor electrochemical reversibility. It was attributed to the higher steric hindrance effect resulted by macrocyclic compounds and electrostatic repulsion between negatively charged ZnTsPc and [Fe(CN)6]3−/4− probe. After assembly of ZnTsPc/f-GN on the electrode surface (curve c), the redox current was larger than that of ZnTsPc, which may be related to synergistic effect of the ZnTsPc and the f-GN arise from π–π stackable interaction. Fig. 6 shows the EIS of (a) bare GCE, (b) f-GN/GCE, (c) ZnTsPc/f-GN/ GCE and (d) ZnTsPc/GCE in 5.0 mM [Fe(CN)6]3−/4− containing 0.1 M KCl. Inset of Fig. 6 is the electrical equivalent circuits. Apparently, the bare GCE exhibits a very small semicircle domain (curve a), whereas the semicircle domain is much larger for the ZnTsPc modified electrode
Fig. 7. DPV of (a) bare GCE, (b) ZnTsPc/GCE, (c) f-GN/GCE and (d) ZnTsPc/f-GN/GCE in PBS (0.1 M, pH 7.0) containing 1.0 × 10−6 M BPA. Inset: an amplification of curve a.
(curve d). It suggests that the interfacial electron transfer was greatly hindered after modification with ZnTsPc on GCE. In the case of ZnTsPc/f-GN/GCE (curve c), the electron transfer resistance (Ret) was nearly the same as f-GN/GCE, indicating that the π–π stackable interaction between ZnTsPc and f-GN facilitated the electron transfer from the redox probe of [Fe(CN)6]3−/4− to the electrode surface. Moreover, by fitting the data using the electrical equivalent circuits, the sequence of corresponding Ret is also ZnTsPc/GCE N ZnTsPc/f-GN/GCE N f-GN/ GCE N bare GCE. This phenomenon corresponds to the performance of the cyclic voltammograms above. Fig. S3 presents an increase in oxidation peak currents and a positive shift in peak position as scan rate increases at ZnTsPc/f-GN/GCE for 1.0 × 10−6 M BPA. The anodic peak currents increase linearly with scan rate in the range from 50 to 500 mV s−1, indicating that the oxidation of BPA is an adsorption-controlled process. Therefore, accumulation time was necessary and the current response was obtained rapidly after 60 s. The recovery of electrode was achieved by scanning five times in a blank PBS solution. 3.3. Electrocatalytic behaviors of BPA at the modified electrodes Fig. S4 displays the CVs obtained at bare GCE (a), ZnTsPc/GCE (b), f-GN/GCE (c), and ZnTsPc/f-GN/GCE (d) in 0.1 M PBS (pH 7.0) containing 1 × 10−6 M BPA (υ = 50 mV s−1). The oxidation of BPA at those
5000
d
4000
-Z'' / ohm
0.4
-5
0.0
0.04
0.6
d
d
0.05
I / µA
15
3000
C Rs
2000
c 1000
a 0
0
Ret
b 2000
4000
Zw
6000
8000
10000
Z' / ohm Fig. 6. Nyquist diagrams of 5.0 mM [Fe(CN)6]3−/4− + 0.1 M KCl solution obtained at (a) bare GCE, (b) f-GN/GCE, (c) ZnTsPc/f-GN/GCE and (d) ZnTsPc/GCE. The inset is the electrical equivalent circuits. Rs: solution resistance, Ret: charge-transfer resistance, C: double-layer capacitance and Zw: Warbury resistance.
Scheme 2. Schematic diagram of the electrocatalytic mechanism towards BPA detection.
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
3
645
expressed as the following equations and Scheme 2, while f-GN acted as a media for charge transfer in ZnTsPc/f-GN nanocomposites.
I=-0.031+0.755c
ð1Þ
2
R =0.998 I / µA
2 4
I / µA
3
1
ð2Þ
2 1 0 0.3
0
0.4
0.5
0.6
0.7
0.8
E/V
0
1
2
3
C / µM
4
5 ð3Þ
Fig. 8. Calibration curve between BPA concentration and peak currents based on DPV. Inset: DPV of ZnTsPc/f-GN modified GCE with injection of different amounts of BPA (from bottom to top corresponding to BPA concentration from low to high) into PBS (0.1 M, pH = 7.0).
3.4. Optimization of analytical conditions electrodes is totally irreversible with only one oxidation peak under the selected experimental conditions. From the comparison of curves (a) and (b), it is very clear that ZnTsPc shows effective catalytic ability to oxidize BPA. While for f-GN/GCE (curve c), there is still a weak peak under large background currents. When ZnTsPc/f-GN was immobilized on GCE surface, there was no response in blank PBS (curve e) compared with a much higher response in the presence of 10−6 M BPA (curve d). Since DPV shows a much higher current sensitivity than CVs, it was used to evaluate the analytical performance of electrochemical sensor. As shown in Fig. 7 and inset, when 1.0 × 10−6 M BPA was added into pH 7.0 PBS, a very weak response at + 0.55 V was observed at the bare GCE within 0.20 to 0.80 V. While the peak (+ 0.54 V) current of the ZnTsPc/GCE (curve b) was much larger than that of bare GCE (curve a). Note that the peak potential of ZnTsPc/GCE (+ 0.54 V) displayed a slight negative than that of bare GCE (+ 0.55 V), which should arise from the catalytical nature of the central metal, Zn(II), in ZnTsPc (Eq. (1)). Compared to ZnTsPc/GCE, f-GN/GCE (curve c) exhibited a little higher current response with a peak potential at +0.53 V because BPA molecules were absorbed on the f-GN surface as described in Section 3.2 by the carboxyl and hydroxyl groups via the electrostatic attraction (Eq. (2)). Interestingly, the oxidation peak at around +0.53 V of BPA was further enhanced greatly on the ZnTsPc/fGN/GCE (curve d). Such excellent catalytic performance should be attributed to the improved accumulation efficiency for BPA through π–π interaction, as well as the synergistic effect between ZnTsPc and f-GN (Eq. (3)). In the light of our results, the proposed mechanism for electrocatalytic oxidation of BPA on the ZnTsPc/f-GN/GCE could be
To achieve a highly sensitive electrochemical sensor for BPA, the experimental conditions were optimized. The variation of peak current (Ipa) with pH and modified quantity of ZnTsPc/f-GN nanocomposites are exhibited in Figs. S5 and S6, respectively. As seen, the peak current increases and reaches a maximum at pH 7.0, and then decreases gradually, implying that protons have taken part in the electrochemical oxidation process [24]. According to modified quantity changing from 1.0 to 10.0 μl, we can easily get that 5 μl is the best choice for preparation of modified electrode. 3.5. Linear range, reproducibility and stability of BPA detection Under the optimum conditions, a series of BPA solutions with different concentrations were detected by DPV. As can be observed in Fig. 8, the oxidation peak currents are proportional to the concentration of BPA in a wider range of 5.0 × 10−8 to 4.0 × 10−6 M with a linear regression equation as Ipa (μA) = −0.031 + 0.755 c (μM) (R2 = 0.998, N = 9) and a detection limit as 2.0 × 10−8 M (S/N = 3). Moreover, compared with the reported analogous electrochemical sensors for BPA detection, this new sensing system exhibited comparable sensitivity, linear range and the detection limit in Table 1 [6,7,9,25–29]. According to the Chinese Health Standard (GB 13116-91, GB 14942-94), BPA content of carbonate resins and products is not greater than 2.2 × 10−7 M. Thus, ZnTsPc/fGN/GCE is an appropriate platform for the electrochemical sensing of BPA. The reproducibility and storage stability of the ZnTsPc/f-GN/GCE for BPA determination were investigated. The reproducibility was
Table 1 Performance comparison of electrochemical methods for BPA detection. Electrode
Sensitivity (μA/μM)
Linear range (M)
Detection limit (M)
References
GN nanofibers/gold/GCE Fe3O4-rGO/GCE Magnetic molecularly/CPEa Mesoporous SiO2/CPE N-doped graphene/GCE Cobalt phthalocyanine/CPE C60/GCE TA/N-Gb/Nafion/GCE ZnTsPc/f-GN/GCE
0.06515 0.0181 0.61 2.833 0.062 0.5672 42.584 1.4845 0.755
8.0 × 10−8–2.5 × 10−4 6.0 × 10−8–1.1 × 10−5 6.0 × 10−7–1.0 × 10−4 2.2 × 10−7–8.8 × 10−6 1.0 × 10−8–1.3 × 10−6 8.75 × 10−8–1.25 × 10−5 7.4 × 10−8–2.3 × 10−7 5.0 × 10−8–1.3 × 10−5 5.0 × 10−8–4.0 × 10−6
3.5 × 10−8 1.7 × 10−8 1.0 × 10−7 3.8 × 10−8 5.0 × 10−9 1.0 × 10−8 3.7 × 10−9 4.0 × 10−9 2.0 × 10−8
[7] [9] [25] [26] [6] [27] [28] [29] Present work
a b
CPE: carbon paste electrode. TA/N-G: tannic acid functionalized N-doped graphene.
646
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
investigated through repetitively recording at a fixed BPA concentration of 1.0 × 10− 6 M with eight different freshly prepared ZnTsPc/f-GN/ GCEs. All the electrodes exhibited similar electrochemical responses and an RSD of 2.9% was obtained, indicating excellent reproducibility of the modified electrode. Furthermore, the electrode retained 94% of its initial response after it was kept in refrigerator at 4 °C for one month, showing long-term stability of ZnTsPc/f-GN film on the GCE surface.
Table 2 Determination of BPA in real samples (n = 5).
3.6. Interference test and analytical application
(Zn(II)) in 2-D hierarchically ZnTsPc/f-GN architectures played as main active sites by accelerating electron transfer process towards the BPA detection. Results indicated that the ZnTsPc/f-GN/GCE not only exhibited excellent electrocatalytic activity towards the oxidation of BPA with broad linear range and low detection limit, but also provided remarkable long-term stability and strong anti-interference ability. The reasons can be attributed to the high adsorption capacity between ZnTsPc and f-GN, catalysis of the central mental (Zn(II)) in ZnTsPc and the synergistic effect between ZnTsPc and f-GN. Moreover, the ZnTsPc/f-GN/GCE has been proved to be an accurate and reliable method owing to the determination of BPA in real samples with satisfactory results. Therefore, the present method shows a potential application prospect for the detection of BPA with simple preparation and low cost.
The influence of some foreign substance on the determination of 1 μM BPA was investigated under the optimized experimental conditions. In addition to common inorganic ions, we gave more attention to several types of BPA analogs including phenol (PN), diethylstilbestrol (DES), bisphenol C (BPC) and 3,3′,5,5′-tetrabromobisphenol A (TBBPA). The results showed that no interference was observed in the presence of 100-fold concentrations of inorganic ions such as Na+, K+, Mg2+, Ca2+, 2− Al3+, NO− and Cl−. Likewise, as shown in Fig. 9, DPV response 3 , SO4 current of 1 μM BPA in coexistence with 20 μM BPA analogs was almost unaffected (signal change below ± 5%). These results indicate that ZnTsPc/f-GN/GCE has an excellent anti-interference ability for some foreign substance and good selectivity for BPA detection, which might be applied to determine BPA in real samples. In order to evaluate the practical performance of ZnTsPc/f-GN/GCE, and also BPA which is used generally in producing polymer, the determination of BPA in various drinking bottles (domestic products from supermarket in China) using standard addition method was carried out. In brief, 10 g dried plastics samples was immersed in 100 ml ethanol and sonicated for an hour, followed by standing still for 24 h. Then, the solution dissolving plastic samples was concentrated to 10 ml by using a 50 °C water bath after filtering. Furthermore, each sample solution underwent five parallel determinations. The concentrations measured were compared with the concentrations added, and the results are listed in Table 2. It can be summarized that the proposed electrochemical method for practical samples is accurate and reliable on the basis of the average recoveries varied from 96.8 to 103.6% with RSD below 5% in all cases. 4. Conclusions A simple and mild strategy has been successfully proposed for synthesizing novel ZnTsPc/f-GN nanocomposites. The central metal
Fig. 9. DPV response currents of ZnTsPc/f-GN/GCE towards 1.0 × 10−6 M BPA and the coexistence interference of 2.0 × 10−5 M PN, DES, BPC and TBBPA.
Drinking bottles
Quality (g)
Determination (μg/g)
Added (μg/g)
Found (μg/g)
Recovery (%)
RSD (%)
Sample 1 Sample 2 Sample 3
10.00 10.00 10.00
0 0 0
50 50 50
51.8 48.4 49.3
103.6 96.8 98.6
2.9 4.4 3.1
Acknowledgments The authors gratefully acknowledge the financial support from Project for International S & T Cooperation of China (2012DFM30040), National Natural Science Foundation of China (NSFC) (21201035, 61201397, J1103303(J2013-004)), and Fujian Department of Science and Technology (2012J01204). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.01.064. References [1] N.J. Cabaton, C. Canlet, P.R. Wadia, M. Tremblay-Franco, R. Gautier, J. Molina, C. Sonnenschein, J.P. Cravedi, B.S. Rubin, A.M. Soto, D. Zalko, Environ. Health Perspect. 121 (2013) 586–593. [2] T. Geens, D. Aerts, C. Berthot, J.-P. Bourguignon, L. Goeyens, P. Lecomte, G. MaghuinRogister, A.-M. Pironnet, L. Pussemier, M.-L. Scippo, Food Chem. Toxicol. 50 (2012) 3725–3740. [3] T.A. Patterson, N.C. Twaddle, C.S. Roegge, R.J. Callicott, J.W. Fisher, D.R. Doerge, Toxicol. Appl. Pharmacol. 267 (2013) 41–48. [4] A.W. Schwartz, P.J. Landrigan, Environ. Health Perspect. (2012) a14. [5] K. Ji, S. Hong, Y. Kho, K. Choi, Environ. Sci. Technol. 47 (2013) 8793–8800. [6] H. Fan, Y. Li, D. Wu, H. Ma, K. Mao, D. Fan, B. Du, H. Li, Q. Wei, Anal. Chim. Acta. 711 (2012) 24–28. [7] X. Niu, W. Yang, G. Wang, J. Ren, H. Guo, J. Gao, Electrochim. Acta 98 (2013) 167–175. [8] X. Tu, L. Yan, X. Luo, S. Luo, Q. Xie, Electroanalysis 21 (2009) 2491–2494. [9] Y. Zhang, Y. Cheng, Y. Zhou, B. Li, W. Gu, X. Shi, Y. Xian, Talanta 107 (2013) 211–218. [10] M.P. Siqueira-Moura, F.L. Primo, E.M. Espreafico, A.C. Tedesco, Mater. Sci. Eng. C 33 (2013) 1744–1752. [11] B. Wang, Y.Q. Wu, X.L. Wang, Z.M. Chen, C.Y. He, Sensors Actuators B 190 (2014) 157–164. [12] S.M. Yoon, S.J. Lou, S. Loser, J. Smith, L.X. Chen, A. Facchetti, T.J. Marks, Nano Lett. 12 (2012) 6315–6321. [13] B. Brozek-Pluska, A. Jarota, K. Kurczewski, H. Abramczyk, J. Mol. Struct. 924–26 (2009) 338–346. [14] O. Akhavan, E. Ghaderi, E. Hashemi, R. Rahighi, Nanoscale 6 (2014) 14810–14819. [15] O. Akhavan, E. Ghaderi, R. Rahighi, ACS Nano 6 (2012) 2904–2916. [16] O. Akhavan, E. Ghaderi, R. Rahighi, M. Abdolahad, Carbon 79 (2014) 654–663. [17] Y. Ogawa, T.C. Niu, S.L. Wong, M. Tsuji, A.T.S. Wee, W. Chen, H. Ago, J. Phys. Chem. C 117 (2013) 21849–21855. [18] M.E. Ragoussi, G. Katsukis, A. Roth, J. Malig, G. de la Torre, D.M. Guldi, T. Torres, J. Am. Chem. Soc. 136 (2014) 4593–4598. [19] J.P. Zhong, Y.J. Fan, H. Wang, R.X. Wang, L.L. Fan, X.C. Shen, Z.J. Shi, Electrochim. Acta 113 (2013) 653–660. [20] M. Mahyari, A. Shaabani, Appl. Catal. A Gen. 469 (2014) 524–531. [21] X.F. Zhang, X.N. Shao, J. Photochem. Photobiol. A 278 (2014) 69–74.
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647 [22] L. Wang, M. Xu, L. Han, M. Zhou, C. Zhu, S. Dong, Anal. Chem. 84 (2012) 7301–7307. [23] O. Akhavan, E. Ghaderi, Small 9 (2013) 3593–3601. [24] X. Chen, T. Ren, M. Ma, Z. Wang, G. Zhan, C. Li, Electrochim. Acta 111 (2013) 49–56. [25] L. Zhu, Y. Cao, G. Cao, Biosens. Bioelectron. 54 (2014) 258–261.
[26] [27] [28] [29]
F. Wang, J. Yang, K. Wu, Anal. Chim. Acta. 638 (2009) 23–28. H.S. Yin, Y.L. Zhou, S.Y. Ai, J. Electroanal. Chem. 626 (2009) 80–88. J.A. Rather, K. De Wael, Sensors Actuators B 176 (2013) 110–117. S. Jiao, J. Jin, L. Wang, Talanta 122 (2014) 140–144.
647
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
A bisphenol A sensor based on novel self-assembly of zinc phthalocyanine tetrasulfonic acid-functionalized graphene nanocomposites Keyu Hou a,b,c, Lei Huang b,c, Yongbo Qi b,c, Caixia Huang a,b,c, Haibo Pan a,b,c,⁎, Min Du a a b c
Fujian Key Lab of Medical Instrument & Pharmaceutical Technology, Yishan Campus, Fuzhou University, Fuzhou, Fujian 350002, China Institute of Research for Functional Materials, Yishan Campus, Fuzhou University, Fuzhou, Fujian 350002, China College of Chemistry, Qishan Campus, Fuzhou University, Fuzhou, Fujian 350116, China
a r t i c l e
i n f o
Article history: Received 21 September 2014 Received in revised form 2 December 2014 Accepted 17 January 2015 Available online 19 January 2015 Keywords: Graphene Self-assembly π–π stack Bisphenol A detection Metallophthalocyanine
a b s t r a c t In this work, a novel zinc phthalocyanine tetrasulfonic acid (ZnTsPc)-functionalized graphene nanocomposites (f-GN) was synthesized by a simple and efficient electrostatic self-assembly method, where the positive charged GN decorated by (3-aminopropyl) triethoxysilane (APTES) was self-assemblied with ZnTsPc, a two dimensional (2-D) molecules. It not only enhanced its stability for the hybrid structure, but also avoided the reaggregation of ZnTsPc or f-GN themselves. Based on layered ZnTsPc/f-GN nanocomposites modified glassy carbon electrode, a rapid and sensitive sensor was developed for the determination of bisphenol A (BPA). Under the optimal conditions, the oxidation peak current increased linearly with the concentration of BPA in the range of 5.0 × 10−8 to 4.0 × 10−6 M with correlation coefficient 0.998 and limits of detection 2.0 × 10−8 M. Due to high absorption nature for BPA and electron deficiency on ZnTsPc/f-GN, it presented the unique electron pathway arising from π–π stackable interaction during redox process for detecting BPA. The sensor exhibited remarkable long-term stability, good anti-interference and excellent electrocatalytic activity towards BPA detection. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Bisphenol A (BPA, 2,2-bis(4-hydroxyphenyl)propane) is a highly versatile hydrocarbon molecule that forms major feedstock in the production of epoxy resins and polycarbonate plastics [1–3]. BPA is readily enriched in the human body through environment and the food chain [4,5]. As an endocrine disrupting chemical, BPA is capable of disturbing the normal endocrine function, influencing sexual development and reproductive capability. In consideration of several laborious and timeconsuming steps in mass spectrometry, gas chromatography (GC) and liquid chromatography (LC) for detecting BPA, main attention is recently devoted to electroanalysis techniques [6–9], which are rapid, sensitive, inexpensive and suitable for on-site monitoring. As organic molecules including of a planar π-conjugated skeleton similarity in structure to the biological molecules (chlorophyll, hemoglobin) [10–13], metallophthalocyanine derivatives (MPcs) exhibit great thermal and chemical stability, and an efficient biomimetic catalysts for redox reactions. Zinc phthalocyanine tetrasulfonic acid
⁎ Corresponding author at: Fujian Key Lab of Medical Instrument & Pharmaceutical Technology, Yishan Campus, Fuzhou University, Fuzhou, Fujian 350002, China. E-mail address: [email protected] (H. Pan).
http://dx.doi.org/10.1016/j.msec.2015.01.064 0928-4931/© 2015 Elsevier B.V. All rights reserved.
(ZnTsPc, the molecule structure in Scheme 1), as electronic donor– acceptor coordination complexes, should contribute to electrocatalysis related with redox reaction. Generally, MPcs commonly suffer from hydrophobicity and aggregation (especially in aqueous solutions) owing to large π-conjugated systems. Although many water-soluble MPcs have been synthesized by introducing hydrophilic moieties and sulfonic groups, it is highly aggregated in water. Graphene (GN), 2-D monolayer comprising sp2-hybridized carbon atoms with the huge surface-to-volume ratio, fast charge carrier mobility, high thermal conductivity, and strong Young's modulus, has attracted wide attention for both fundamental science and application. For instance, some recent works related with the excellent ability of GN in electrochemical biosensors have come to our sight [14–16]. Recently, the GN modified with MPcs is under intensive exploitation, and few reports were published in synthesis and characterization of MPcs on GN [17–20], where MPcs combine with GN covalently or coordinatively. In present work, we developed a novel approach to combine the positively charged functionalized nanoscale graphene (f-GN) with negatively charged ZnTsPc by a simple and efficient electrostatic selfassembly method (Scheme 1), where GN was decorated by (3aminopropyl) triethoxysilane (APTES) at first. Note that they would form strong noncovalent π–π conjugated composite (ZnTsPc/f-GN)
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
641
Scheme 1. Schematic illustration for electrostatic self-assembly of ZnTsPc/f-GN nanocomposites.
through the π–π stackable interaction [21]. And also, it exhibited charming characteristics in electrocatalysis feature by their synergetic combination of two components. Owing to the above advantages, an alternative electrochemical sensor for the selective and sensitive detection of BPA was realized by ZnTsPc/f-GN modified glassy carbon electrode (GCE). The redox mechanism on ZnTsPc/f-GN/GCE was raised further in BPA. For confirming its practical application, the determination of BPA in real samples was also fulfilled by using differential pulse voltammetry (DPV). 2. Experimental 2.1. Materials and reagents Natural graphite (FP 99.95% pure) and APTES (C9H23NO3Si,) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). ZnTsPc (C32H16N8O12S4Zn) was obtained from J&K Scientific Ltd. (Beijing, China). 3,3′,5,5′-tetrabromobisphenol A (TBBPA) was obtained from Alfa Aesar Company (Tianjin, China). BPA was purchased from TCI (Shanghai) Chemical Industry Co. Ltd. and its stock solution (2 mM) was prepared with ethanol. The working solutions were prepared by diluting the stock solution with phosphate buffer saline (PBS, 0.1 M, pH 7.0). All other chemicals were of analytical grade and used as received without any further purification. Deionized (DI) water obtained from a Milli-Q plus water purification system (Millipore Co. Ltd., USA, 18 MΩ·cm) was used throughout the study. 2.2. Apparatus Zeta-potential value measurement was recorded on Zeta Sizer 3000 Laser Particle Size and Zeta Potential Tester (Malvern Corporation, UK),
and DI water was used as a dispersant. Fourier transformed infrared spectroscopy (FT-IR) were performed on a Nicolet Nexus 670 FT-IR spectrophotometer at a resolution of 4 cm−1. UV–vis spectra were measured with a UV–vis absorption spectrometer (PerkinElmer Lambda 900). The morphologies of samples were observed by AFM (Agilent 5500, USA). High-resolution transmission electron microscope (HRTEM) image was obtained using Tecnai G2 F20 S-TWIN, 200 kV (FEI Company, USA). All the electrochemical measurements were carried out on a CHI660D electrochemical workstation (Shanghai Chenhua Instrument Co. Ltd., China) with a conventional three-electrode system. Glass carbon electrode (3 mm diameter) loaded different materials was used as working electrode. The surface of GCE was polished with 0.3 and 0.05 μm alumina slurries, ultrasonicated in water and ethanol, sequentially. Then the bare GCE was dried under pure N2. Finally, these materials were dropped on the surface of bare GCE, and dried under room temperature. 2.3. Preparation of graphene nanosheets Graphene oxide (GO) was synthesized from natural graphite powder using modified Hummers' method. Typically, 1 g of graphite powder was put into a mixture of 6 ml of concentrated H2SO4, 1.25 g of K2S2O8, and 1.25 g of P2O5. The solution was heated to 80 °C in an oil-bath kept stirring for 24 h. It was carefully diluted with 300 ml of DI water, filtered, and washed until pH ~ 7. The pre-oxidized graphite powder was added to a mixture of 60 ml of concentrated H2SO4 and 30 ml HNO3 under vigorous stirring at 0 °C. Then, 8 g KMnO4 was added gradually under stirring and the temperature of mixture was kept to be below 20 °C by cooling. Successively, it was diluted with 500 ml of DI water in an ice bath to keep the temperature below 50 °C for 2 h. After further dilution with 500 ml of DI water, 20 ml of H2O2
642
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
was added to remove excessive KMnO4. The brilliant yellow product called graphite oxide was purified by washing with 10% HCl and washed with warm water for several times. Exfoliation was accomplished by sonicating graphite oxide in aqueous solution for 120 min, and then it was centrifuged at 5000 rpm for 10 min to obtain GO solution. GO was reduced to GN by NaBH4 in water bath at 90 °C for 2 h. 2.4. Preparation of ZnTsPc/f-GN nanocomposites ZnTsPc/f-GN was fabricated by a simple and efficient electrostatic self-assembly method. In detail, APTES (400 μl) was added slowly into 40 ml GN aqueous solution (0.25 mg/ml) and the mixture was kept stirring for 10 h under room temperature. The decorated graphene nanosheets, f-GN, were collected by centrifugating at 8000 rpm for 15 min and washed with ethanol and water in succession until pH ~7. The final sediment was redispersed in DI water. Then, 40 mg ZnTsPc was dispersed in 1 ml of prepared f-GN solution by sonication for 30 min. The mixture was centrifuged and washed with DI water for several times after 24 hours' standing. Finally, well-distributed ZnTsPc/f-GN was obtained by redispersing the precipitates in DI water. 3. Results and discussion 3.1. Physicochemical characterization for ZnTsPc/f-GN nanocomposites After the decoration with APTES, the surface of GN nanosheets attached with amine functional groups was obtained. As displayed in Fig. S1, the zeta potential analysis of f-GN reveals a significant positively charged surface, in favor of avoiding the aggregation for 2-D GN layers. ZnTsPc/f-GN nanocomposites were also confirmed by the FT-IR spectra (Fig. 1). For GO (Fig. 1a), the stretching vibrations of O\H (3300 cm−1), C_O (1730 cm− 1), C_C (1625 cm−1), C\O\H (1350 cm− 1) and C\O\C (1050 cm−1) are observed, indicating the presence of various oxygen-containing functional groups. Compared with Fig. 1a, GN (Fig. 1b) shows survival of carboxyl groups and hydroxyl groups on the GN nanosheets even after reduction. As to f-GN (Fig. 1d), a sharp bend vibration of N\H at 1575 cm−1 presents through decorated GN with APTES, indicating that \NH2 groups attached onto the surface of GN. The characteristic peak around 1600 cm−1 is corresponding to the vibration of benzene ring. For ZnTsPc (Fig. 1c) and ZnTsPc/f-GN (Fig. 1e), these specific bands become much broader than that of other curves, implying that there are strong electronic interactions through the π–π stack between ZnTsPc and f-GN. A peak located at 1400 cm−1 is assigned to the stretching vibration of the C_N from ZnTsPc macrocycle which can also be found on ZnTsPc/f-GN. It is noted that a shift for coordination bond N → Zn from 912 cm−1 to 925 cm−1 occurs
Fig. 1. FT-IR spectra of (a) GO nanosheets, (b) GN nanosheets, (c) ZnTsPc, (d) f-GN and (e) ZnTsPc/f-GN nanocomposites.
(Fig. 1e), suggesting that the distribution of electrons around central metal Zn (II) prefers to f-GN. Such electron deficiency in the neighborhood of central metal Zn (II) is conducive to adsorption of phenolic hydroxyl groups in BPA, promoting electron transport from BPA to ZnTsPc/f-GN nanocomposites. Fig. 2 depicts UV–vis absorption spectra of f-GN, ZnTsPc and ZnTsPc/ f-GN dispersed in water and DMF for reference. ZnTsPc is known to readily form dimers in H2O due to the hydrophobic intermolecular affinity. Thus, Fig. 2b shows its partly aggregated form, and the inset of Fig. 2 represents its spectrum in monomeric form. A distinct blue shift of Q-band absorption of the dimer (λmax = 656 nm) as compared to the monomer (λmax = 677 nm) is associated with the overlap of π-electronic clouds of MPcs system in the dimer. The absorption peak of as-prepared f-GN at ~ 270 nm (Fig. 2a) is attributed to the n → π* transitions of C_O bonds, indicating that there is little influence after being decorated with APTES. It can be clearly seen that the spectrum of ZnTsPc/f-GN (Fig. 2c) displays four distinct absorption peaks. In addition, the color of ZnTsPc/f-GN dispersion exhibits blueblack that distinguishes from black f-GN and blue ZnTsPc (the inset in Fig. 2). Apparently, the light absorption below 300 nm is assigned to f-GN and a shoulder peak at 350 nm arises from the B-band absorption of ZnTsPc. As to a strong band at 705 nm, it represents that the red shift of Q-band of ZnTsPc in the composite occurred compared with that of pure ZnTsPc (Fig. 2b). It indicated that the π–π stackable interaction exists between ZnTsPc molecules and f-GN due to the benzene rings both in ZnTsPc and f-GN. At the same time, the interaction not only improves the stability of ZnTsPc/f-GN nanocomposites, but avoids the reaggregation of ZnTsPc or f-GN themselves. Nevertheless, these shifts in Q band of MPcs complexes are normally associated with electron transfer. In view of these shifts in FT-IR, we could conclude that electrons transfer from ZnTsPc to electron-accepting f-GN. According to the corresponding cross-sectional view, the typical AFM image (Fig. 3A) of f-GN indicates that the average thickness of nanosheets is about 1.021 nm, a little greater than the reported thickness of monolayer GN [22]. This can be also ascribed to APTES successfully decorated on the surface of f-GN. After being functionalized by ZnTsPc, ZnTsPc/f-GN became slightly rough with the increased mean thickness to 1.625 nm (Fig. 3B), declaring the formation of layered nanohybrids. It is known that the intensities, shapes and positions of the G band and 2D band peaks can be used to monitor the number of GN layers [23]. Raman spectra for f-GN, ZnTsPc, and ZnTsPc/f-GN were also measured (Fig. S2). According to the wider 2D band and ratio of the 2D/G intensity (f-GN: 0.409, ZnTsPc: 0.411), we estimated that the layers for GN diminish after ZnTsPc modifying on the f-GN.
Fig. 2. UV–vis absorption spectra of (a) f-GN, (b) ZnTsPc and (c) ZnTsPc/f-GN nanocomposites in H2O. Inset: The monomer spectrum of ZnTsPc in DMF.
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
643
Fig. 3. AFM images of (A) f-GN and (B) ZnTsPc/f-GN.
HRTEM (Fig. 4A) shows that the surface of f-GN is a thin wrinkling paper-like structure, while ZnTsPc/f-GN exhibits layered form after ZnTsPc molecules were loaded on f-GN (Fig. 4C). Moreover, in contrast with ZnTsPc/f-GN (Fig. 4D), based on the significant lattice fringes with typical fast Fourier transform (FFT) pattern, it verifies that the f-GN (Fig. 4B) is nearly single-crystalline.
3.2. Electrochemical properties of modified electrodes The electrochemical properties of the bare GCE and modified electrodes were performed by cyclic voltammograms (CVs) and electrochemical impedance spectroscopy (EIS), respectively. The CVs of different modified electrodes in 5.0 mM [Fe(CN)6]3−/4− solution containing
Fig. 4. HRTEM images of f-GN (A and B) and ZnTsPc/f-GN (C and D). Inset in (B) and (D): FFT pattern of lattice fringe images.
644
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
0.7
a b
10
c
0.5
I / µA
I / µA
5 0
-10
0.1
-15 -0.2
0.0
0.2
0.4
0.6
0.8
0.03 0.02
a
0.01
c
0.00 -0.01
0.3
0.4
0.3 0.2
0.5
0.6
E/V
0.7
0.8
b a 0.2
0.3
0.4
0.5
0.6
0.7
0.8
E/V
E/ V Fig. 5. Cyclic voltammograms of (a) bare GCE, (b) f-GN/GCE, (c) ZnTsPc/f-GN/GCE and (d) ZnTsPc in a mixture solution of 5.0 mmol·L−1 [Fe(CN)6]3−/4− and 0.1 mol·L−1 KCl with the scan rate at 0.1 V·s−1.
0.1 M KCl was collected in Fig. 5. As seen, a classic pair of well defined redox peaks was obtained at the bare GCE (curve a). After modification with f-GN (curve b), the redox peak currents of [Fe(CN)6]3 −/4 − decreased, representing that f-GN had been successfully covered on the electrode and hindered the electron transfer. However, the presence of ZnTsPc significantly diminished the redox of [Fe(CN)6]3−/4− on the electrode (curve d). In addition, the couple of redox peaks with the peak-to-peak (ΔEp) of 215 mV was observed, indicating the poor electrochemical reversibility. It was attributed to the higher steric hindrance effect resulted by macrocyclic compounds and electrostatic repulsion between negatively charged ZnTsPc and [Fe(CN)6]3−/4− probe. After assembly of ZnTsPc/f-GN on the electrode surface (curve c), the redox current was larger than that of ZnTsPc, which may be related to synergistic effect of the ZnTsPc and the f-GN arise from π–π stackable interaction. Fig. 6 shows the EIS of (a) bare GCE, (b) f-GN/GCE, (c) ZnTsPc/f-GN/ GCE and (d) ZnTsPc/GCE in 5.0 mM [Fe(CN)6]3−/4− containing 0.1 M KCl. Inset of Fig. 6 is the electrical equivalent circuits. Apparently, the bare GCE exhibits a very small semicircle domain (curve a), whereas the semicircle domain is much larger for the ZnTsPc modified electrode
Fig. 7. DPV of (a) bare GCE, (b) ZnTsPc/GCE, (c) f-GN/GCE and (d) ZnTsPc/f-GN/GCE in PBS (0.1 M, pH 7.0) containing 1.0 × 10−6 M BPA. Inset: an amplification of curve a.
(curve d). It suggests that the interfacial electron transfer was greatly hindered after modification with ZnTsPc on GCE. In the case of ZnTsPc/f-GN/GCE (curve c), the electron transfer resistance (Ret) was nearly the same as f-GN/GCE, indicating that the π–π stackable interaction between ZnTsPc and f-GN facilitated the electron transfer from the redox probe of [Fe(CN)6]3−/4− to the electrode surface. Moreover, by fitting the data using the electrical equivalent circuits, the sequence of corresponding Ret is also ZnTsPc/GCE N ZnTsPc/f-GN/GCE N f-GN/ GCE N bare GCE. This phenomenon corresponds to the performance of the cyclic voltammograms above. Fig. S3 presents an increase in oxidation peak currents and a positive shift in peak position as scan rate increases at ZnTsPc/f-GN/GCE for 1.0 × 10−6 M BPA. The anodic peak currents increase linearly with scan rate in the range from 50 to 500 mV s−1, indicating that the oxidation of BPA is an adsorption-controlled process. Therefore, accumulation time was necessary and the current response was obtained rapidly after 60 s. The recovery of electrode was achieved by scanning five times in a blank PBS solution. 3.3. Electrocatalytic behaviors of BPA at the modified electrodes Fig. S4 displays the CVs obtained at bare GCE (a), ZnTsPc/GCE (b), f-GN/GCE (c), and ZnTsPc/f-GN/GCE (d) in 0.1 M PBS (pH 7.0) containing 1 × 10−6 M BPA (υ = 50 mV s−1). The oxidation of BPA at those
5000
d
4000
-Z'' / ohm
0.4
-5
0.0
0.04
0.6
d
d
0.05
I / µA
15
3000
C Rs
2000
c 1000
a 0
0
Ret
b 2000
4000
Zw
6000
8000
10000
Z' / ohm Fig. 6. Nyquist diagrams of 5.0 mM [Fe(CN)6]3−/4− + 0.1 M KCl solution obtained at (a) bare GCE, (b) f-GN/GCE, (c) ZnTsPc/f-GN/GCE and (d) ZnTsPc/GCE. The inset is the electrical equivalent circuits. Rs: solution resistance, Ret: charge-transfer resistance, C: double-layer capacitance and Zw: Warbury resistance.
Scheme 2. Schematic diagram of the electrocatalytic mechanism towards BPA detection.
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
3
645
expressed as the following equations and Scheme 2, while f-GN acted as a media for charge transfer in ZnTsPc/f-GN nanocomposites.
I=-0.031+0.755c
ð1Þ
2
R =0.998 I / µA
2 4
I / µA
3
1
ð2Þ
2 1 0 0.3
0
0.4
0.5
0.6
0.7
0.8
E/V
0
1
2
3
C / µM
4
5 ð3Þ
Fig. 8. Calibration curve between BPA concentration and peak currents based on DPV. Inset: DPV of ZnTsPc/f-GN modified GCE with injection of different amounts of BPA (from bottom to top corresponding to BPA concentration from low to high) into PBS (0.1 M, pH = 7.0).
3.4. Optimization of analytical conditions electrodes is totally irreversible with only one oxidation peak under the selected experimental conditions. From the comparison of curves (a) and (b), it is very clear that ZnTsPc shows effective catalytic ability to oxidize BPA. While for f-GN/GCE (curve c), there is still a weak peak under large background currents. When ZnTsPc/f-GN was immobilized on GCE surface, there was no response in blank PBS (curve e) compared with a much higher response in the presence of 10−6 M BPA (curve d). Since DPV shows a much higher current sensitivity than CVs, it was used to evaluate the analytical performance of electrochemical sensor. As shown in Fig. 7 and inset, when 1.0 × 10−6 M BPA was added into pH 7.0 PBS, a very weak response at + 0.55 V was observed at the bare GCE within 0.20 to 0.80 V. While the peak (+ 0.54 V) current of the ZnTsPc/GCE (curve b) was much larger than that of bare GCE (curve a). Note that the peak potential of ZnTsPc/GCE (+ 0.54 V) displayed a slight negative than that of bare GCE (+ 0.55 V), which should arise from the catalytical nature of the central metal, Zn(II), in ZnTsPc (Eq. (1)). Compared to ZnTsPc/GCE, f-GN/GCE (curve c) exhibited a little higher current response with a peak potential at +0.53 V because BPA molecules were absorbed on the f-GN surface as described in Section 3.2 by the carboxyl and hydroxyl groups via the electrostatic attraction (Eq. (2)). Interestingly, the oxidation peak at around +0.53 V of BPA was further enhanced greatly on the ZnTsPc/fGN/GCE (curve d). Such excellent catalytic performance should be attributed to the improved accumulation efficiency for BPA through π–π interaction, as well as the synergistic effect between ZnTsPc and f-GN (Eq. (3)). In the light of our results, the proposed mechanism for electrocatalytic oxidation of BPA on the ZnTsPc/f-GN/GCE could be
To achieve a highly sensitive electrochemical sensor for BPA, the experimental conditions were optimized. The variation of peak current (Ipa) with pH and modified quantity of ZnTsPc/f-GN nanocomposites are exhibited in Figs. S5 and S6, respectively. As seen, the peak current increases and reaches a maximum at pH 7.0, and then decreases gradually, implying that protons have taken part in the electrochemical oxidation process [24]. According to modified quantity changing from 1.0 to 10.0 μl, we can easily get that 5 μl is the best choice for preparation of modified electrode. 3.5. Linear range, reproducibility and stability of BPA detection Under the optimum conditions, a series of BPA solutions with different concentrations were detected by DPV. As can be observed in Fig. 8, the oxidation peak currents are proportional to the concentration of BPA in a wider range of 5.0 × 10−8 to 4.0 × 10−6 M with a linear regression equation as Ipa (μA) = −0.031 + 0.755 c (μM) (R2 = 0.998, N = 9) and a detection limit as 2.0 × 10−8 M (S/N = 3). Moreover, compared with the reported analogous electrochemical sensors for BPA detection, this new sensing system exhibited comparable sensitivity, linear range and the detection limit in Table 1 [6,7,9,25–29]. According to the Chinese Health Standard (GB 13116-91, GB 14942-94), BPA content of carbonate resins and products is not greater than 2.2 × 10−7 M. Thus, ZnTsPc/fGN/GCE is an appropriate platform for the electrochemical sensing of BPA. The reproducibility and storage stability of the ZnTsPc/f-GN/GCE for BPA determination were investigated. The reproducibility was
Table 1 Performance comparison of electrochemical methods for BPA detection. Electrode
Sensitivity (μA/μM)
Linear range (M)
Detection limit (M)
References
GN nanofibers/gold/GCE Fe3O4-rGO/GCE Magnetic molecularly/CPEa Mesoporous SiO2/CPE N-doped graphene/GCE Cobalt phthalocyanine/CPE C60/GCE TA/N-Gb/Nafion/GCE ZnTsPc/f-GN/GCE
0.06515 0.0181 0.61 2.833 0.062 0.5672 42.584 1.4845 0.755
8.0 × 10−8–2.5 × 10−4 6.0 × 10−8–1.1 × 10−5 6.0 × 10−7–1.0 × 10−4 2.2 × 10−7–8.8 × 10−6 1.0 × 10−8–1.3 × 10−6 8.75 × 10−8–1.25 × 10−5 7.4 × 10−8–2.3 × 10−7 5.0 × 10−8–1.3 × 10−5 5.0 × 10−8–4.0 × 10−6
3.5 × 10−8 1.7 × 10−8 1.0 × 10−7 3.8 × 10−8 5.0 × 10−9 1.0 × 10−8 3.7 × 10−9 4.0 × 10−9 2.0 × 10−8
[7] [9] [25] [26] [6] [27] [28] [29] Present work
a b
CPE: carbon paste electrode. TA/N-G: tannic acid functionalized N-doped graphene.
646
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647
investigated through repetitively recording at a fixed BPA concentration of 1.0 × 10− 6 M with eight different freshly prepared ZnTsPc/f-GN/ GCEs. All the electrodes exhibited similar electrochemical responses and an RSD of 2.9% was obtained, indicating excellent reproducibility of the modified electrode. Furthermore, the electrode retained 94% of its initial response after it was kept in refrigerator at 4 °C for one month, showing long-term stability of ZnTsPc/f-GN film on the GCE surface.
Table 2 Determination of BPA in real samples (n = 5).
3.6. Interference test and analytical application
(Zn(II)) in 2-D hierarchically ZnTsPc/f-GN architectures played as main active sites by accelerating electron transfer process towards the BPA detection. Results indicated that the ZnTsPc/f-GN/GCE not only exhibited excellent electrocatalytic activity towards the oxidation of BPA with broad linear range and low detection limit, but also provided remarkable long-term stability and strong anti-interference ability. The reasons can be attributed to the high adsorption capacity between ZnTsPc and f-GN, catalysis of the central mental (Zn(II)) in ZnTsPc and the synergistic effect between ZnTsPc and f-GN. Moreover, the ZnTsPc/f-GN/GCE has been proved to be an accurate and reliable method owing to the determination of BPA in real samples with satisfactory results. Therefore, the present method shows a potential application prospect for the detection of BPA with simple preparation and low cost.
The influence of some foreign substance on the determination of 1 μM BPA was investigated under the optimized experimental conditions. In addition to common inorganic ions, we gave more attention to several types of BPA analogs including phenol (PN), diethylstilbestrol (DES), bisphenol C (BPC) and 3,3′,5,5′-tetrabromobisphenol A (TBBPA). The results showed that no interference was observed in the presence of 100-fold concentrations of inorganic ions such as Na+, K+, Mg2+, Ca2+, 2− Al3+, NO− and Cl−. Likewise, as shown in Fig. 9, DPV response 3 , SO4 current of 1 μM BPA in coexistence with 20 μM BPA analogs was almost unaffected (signal change below ± 5%). These results indicate that ZnTsPc/f-GN/GCE has an excellent anti-interference ability for some foreign substance and good selectivity for BPA detection, which might be applied to determine BPA in real samples. In order to evaluate the practical performance of ZnTsPc/f-GN/GCE, and also BPA which is used generally in producing polymer, the determination of BPA in various drinking bottles (domestic products from supermarket in China) using standard addition method was carried out. In brief, 10 g dried plastics samples was immersed in 100 ml ethanol and sonicated for an hour, followed by standing still for 24 h. Then, the solution dissolving plastic samples was concentrated to 10 ml by using a 50 °C water bath after filtering. Furthermore, each sample solution underwent five parallel determinations. The concentrations measured were compared with the concentrations added, and the results are listed in Table 2. It can be summarized that the proposed electrochemical method for practical samples is accurate and reliable on the basis of the average recoveries varied from 96.8 to 103.6% with RSD below 5% in all cases. 4. Conclusions A simple and mild strategy has been successfully proposed for synthesizing novel ZnTsPc/f-GN nanocomposites. The central metal
Fig. 9. DPV response currents of ZnTsPc/f-GN/GCE towards 1.0 × 10−6 M BPA and the coexistence interference of 2.0 × 10−5 M PN, DES, BPC and TBBPA.
Drinking bottles
Quality (g)
Determination (μg/g)
Added (μg/g)
Found (μg/g)
Recovery (%)
RSD (%)
Sample 1 Sample 2 Sample 3
10.00 10.00 10.00
0 0 0
50 50 50
51.8 48.4 49.3
103.6 96.8 98.6
2.9 4.4 3.1
Acknowledgments The authors gratefully acknowledge the financial support from Project for International S & T Cooperation of China (2012DFM30040), National Natural Science Foundation of China (NSFC) (21201035, 61201397, J1103303(J2013-004)), and Fujian Department of Science and Technology (2012J01204). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.01.064. References [1] N.J. Cabaton, C. Canlet, P.R. Wadia, M. Tremblay-Franco, R. Gautier, J. Molina, C. Sonnenschein, J.P. Cravedi, B.S. Rubin, A.M. Soto, D. Zalko, Environ. Health Perspect. 121 (2013) 586–593. [2] T. Geens, D. Aerts, C. Berthot, J.-P. Bourguignon, L. Goeyens, P. Lecomte, G. MaghuinRogister, A.-M. Pironnet, L. Pussemier, M.-L. Scippo, Food Chem. Toxicol. 50 (2012) 3725–3740. [3] T.A. Patterson, N.C. Twaddle, C.S. Roegge, R.J. Callicott, J.W. Fisher, D.R. Doerge, Toxicol. Appl. Pharmacol. 267 (2013) 41–48. [4] A.W. Schwartz, P.J. Landrigan, Environ. Health Perspect. (2012) a14. [5] K. Ji, S. Hong, Y. Kho, K. Choi, Environ. Sci. Technol. 47 (2013) 8793–8800. [6] H. Fan, Y. Li, D. Wu, H. Ma, K. Mao, D. Fan, B. Du, H. Li, Q. Wei, Anal. Chim. Acta. 711 (2012) 24–28. [7] X. Niu, W. Yang, G. Wang, J. Ren, H. Guo, J. Gao, Electrochim. Acta 98 (2013) 167–175. [8] X. Tu, L. Yan, X. Luo, S. Luo, Q. Xie, Electroanalysis 21 (2009) 2491–2494. [9] Y. Zhang, Y. Cheng, Y. Zhou, B. Li, W. Gu, X. Shi, Y. Xian, Talanta 107 (2013) 211–218. [10] M.P. Siqueira-Moura, F.L. Primo, E.M. Espreafico, A.C. Tedesco, Mater. Sci. Eng. C 33 (2013) 1744–1752. [11] B. Wang, Y.Q. Wu, X.L. Wang, Z.M. Chen, C.Y. He, Sensors Actuators B 190 (2014) 157–164. [12] S.M. Yoon, S.J. Lou, S. Loser, J. Smith, L.X. Chen, A. Facchetti, T.J. Marks, Nano Lett. 12 (2012) 6315–6321. [13] B. Brozek-Pluska, A. Jarota, K. Kurczewski, H. Abramczyk, J. Mol. Struct. 924–26 (2009) 338–346. [14] O. Akhavan, E. Ghaderi, E. Hashemi, R. Rahighi, Nanoscale 6 (2014) 14810–14819. [15] O. Akhavan, E. Ghaderi, R. Rahighi, ACS Nano 6 (2012) 2904–2916. [16] O. Akhavan, E. Ghaderi, R. Rahighi, M. Abdolahad, Carbon 79 (2014) 654–663. [17] Y. Ogawa, T.C. Niu, S.L. Wong, M. Tsuji, A.T.S. Wee, W. Chen, H. Ago, J. Phys. Chem. C 117 (2013) 21849–21855. [18] M.E. Ragoussi, G. Katsukis, A. Roth, J. Malig, G. de la Torre, D.M. Guldi, T. Torres, J. Am. Chem. Soc. 136 (2014) 4593–4598. [19] J.P. Zhong, Y.J. Fan, H. Wang, R.X. Wang, L.L. Fan, X.C. Shen, Z.J. Shi, Electrochim. Acta 113 (2013) 653–660. [20] M. Mahyari, A. Shaabani, Appl. Catal. A Gen. 469 (2014) 524–531. [21] X.F. Zhang, X.N. Shao, J. Photochem. Photobiol. A 278 (2014) 69–74.
K. Hou et al. / Materials Science and Engineering C 49 (2015) 640–647 [22] L. Wang, M. Xu, L. Han, M. Zhou, C. Zhu, S. Dong, Anal. Chem. 84 (2012) 7301–7307. [23] O. Akhavan, E. Ghaderi, Small 9 (2013) 3593–3601. [24] X. Chen, T. Ren, M. Ma, Z. Wang, G. Zhan, C. Li, Electrochim. Acta 111 (2013) 49–56. [25] L. Zhu, Y. Cao, G. Cao, Biosens. Bioelectron. 54 (2014) 258–261.
[26] [27] [28] [29]
F. Wang, J. Yang, K. Wu, Anal. Chim. Acta. 638 (2009) 23–28. H.S. Yin, Y.L. Zhou, S.Y. Ai, J. Electroanal. Chem. 626 (2009) 80–88. J.A. Rather, K. De Wael, Sensors Actuators B 176 (2013) 110–117. S. Jiao, J. Jin, L. Wang, Talanta 122 (2014) 140–144.
647
