Synergetic signal amplification of multi-walled carbon nanotubes-Fe3O4 hybrid and trimethyloctadecylammonium bromide as a highly sensitive detection platform for tetrabromobisphenol A

In this work, we fabricated an electrochemical sensor based on trimethyloctadecylammonium bromide and multi-walled carbon nanotubes-Fe3O4 hybrid (TOAB/MWCNTs-Fe3O4) for sensitive detection of tetrabromobisphenol A (TBBPA). The nanocomposite was characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM) and Fourier transform infrared spectroscopy (FT-IR) techniques. The electrochemical behaviors of TBBPA on TOAB/MWCNTs-Fe3O4 composite film modified glassy carbon electrode (GCE) were investigated by cyclic voltammetry (CV), differential pulse voltammetry (DPV) and electrochemical impedance spectroscopy (EIS) method. The experimental results indicated that the incorporation of MWCNTs-Fe3O4 with TOAB greatly enhanced the electrochemical response of TBBPA. This fabricated sensor displayed excellent analytical performance for TBBPA detection over a range from 3.0 nM to 1000.0 nM with a detection limit of 0.73 nM (S/N = 3). Moreover, the proposed electrochemical sensor exhibited good reproducibility and stability, and could be successfully applied to detect TBBPA in water samples with satisfactory results.


Results and Discussion
The characterization of the as-prepared MWCNTs-Fe 3 O 4 nannocomposite was investigated.
The Transmission electron microscopy (TEM) was used to characterize the microstructures of MWCNTs-COOH and the MWCNTs-Fe 3 O 4 nanocomposite (Fig. 1). The Fig. 1a showed the morphology of MWCNTs-COOH. And  The X-ray diffraction (XRD) patterns indicated that the crystal structures of the materials. As shown in Fig. 3a, the diffraction peaks at 2θ = 26.1° was identified as the (002) plane of the hexagonal graphite structure, which is consistent with previous reports 49 . The XRD pattern of MWCNTs-COOH (Fig. 3b) was similar to that of the MWCNTs. However, for MWCNTs-COOH, the diffraction peaks had higher intensities, and the crystallization peaks were more dominant. One possible reason is that the acid treatment of the MWCNTs advanced the removal of the amorphous carbon, carbon nanoparticles and metal particles. And as presented in Fig. 3c 50,51 . A diffraction peak at 2θ = 26.1° also appeared in the XRD pattern of MWCNTs-Fe 3 O 4 composite (Fig. 3d). The MWCNTs showed a lower intensity is probably ascribed to the reason that the Fe 3 O 4 grafted on the surface of MWCNTs weakened the diffraction peak at 26° of MWCNTs. The XRD results indicated that MWCNTs-Fe 3 O 4 nanocomposites were successfully synthesized by a facile hydrothermal method. No crystalline impurities were detected.
The comparative studies on the electrochemical behavior of different modified electrode were also investigated. The cyclic voltammetry method (CV) is an effective and convenient technique for probing the feature of the developed sensor. To investigate the electrochemical behaviors for each assembly step, different modified electrodes were studied with a 0.1 M KCl solution containing 1.0 mM [Fe (CN) 6 ] 4−/3− probe at scan rate of 100 mV s −1 . As seen in Fig. 4a where Ip is the anodic peak current, ν is the scan rate, n is counted as the number of electron transferred in the redox reaction, C is the concentration of [Fe (CN) 6  Electrochemical impedance spectroscopy (EIS) is an effective technique for probing the features of modified electrodes. The impedance spectra include a semicircle portion and a linear portion. The linear part at lower frequencies represents the diffusion process, whereas the semicircle portion at higher frequencies corresponds to the electron-transfer limit process, and the diameter of the semicircle indicates the electron transfer resistance (Ret) 53 . The stepwise construction process of the sensor was characterized by EIS, and the EIS patterns of different electrodes in a solution of 0.1 M KCl containing 5 mM [Fe (CN) 6 ] 4−/3− was recorded. As shown in Fig. 4c, a big well-defined semicircle at higher frequencies obtained for the bare GCE (curve a), which indicated a huge interface electron transfer resistance. When MWCNTs were modified on the GCE, a small semicircle in the Nyquist plot was obtained (curve b), which indicated that interface electron transfer resistance decreased obviously. For the MWCNTs-Fe 3 O 4 modified electrode (Fig. 4d, curve c), a little smaller semicircle in the Nyquist plot was obtained compared with curve b, which showed a smaller interface electron transfer resistance. And this phenomenon could be ascribed to cooperative action between MWCNTs and Fe 3 O 4 nanoparticles. After introduction of TOAB, the electrode modified with the TOAB/MWCNTs-Fe 3 O 4 showed the smallest semicircle, which indicated that the conductivity of the TOAB/MWCNTs-Fe 3 O 4 modified electrode was better than the other electrodes. This is good for the electrochemical sensing. These results agreed with the performance of the above-mentioned cyclic voltammograms.
The enrichment effect of different modified electrodes was also studied using chronocoulometry method. The curves of charge (Q)-time (t) in pH 7.0 phosphate buffer in the absence and presence of 100.0 nM TBBPA were individually recorded. Then, the data of Q-t curves were converted to the Q-t 1/2 plots, and two fitted curves were obtained (see Supplementary Fig. S1). According to the integrated Cottrell equation 54 : where Qdl is the double-layer charge, Qads is the Faradaic charge owing to the oxidation of adsorbed TBBPA, n is the electron transfer number, A is the electrode area, C is the concentration, D is the diffusion coefficient, Moreover, after the modification of the TOAB onto the electrode, the Qads value of TBBPA increased obviously, which indicated that the accumulation ability of TBBPA was improved obviously. This phenomenon may be attributed to the strong intermolecular force and hydrophobic interaction between the TOAB and TBBPA. It has been reported that the surfactants can enrich phenol compounds by its hydrophobic tails 55 . The TOAB, with three C18-alkyl chains, may play an important role in the enrichment of TBBPA. The long alkyl chains might provide a hydrophobic micro-environment on the surface of the electrode 33 , which may enhance the absorption of the lipophlic chemical TBBPA onto the electrode surface. Thus, the TOAB could facilitate the electron transfer, enlarge the effective surface area of sensor, and enrich the TBBPA with its long alkyl chains effectively. The results of chronocoulometry, along with the results of EIS and cyclic voltammograms, indicated that there may be a cooperative effect between each component in the stepwise construction process of the sensor.
The synergetic signal amplification of MWCNTs-Fe 3 O 4 nanocomposite hybrid and TOAB was studied. The oxidation behaviors of TBBPA on different modified electrodes were studied to deeply discuss their signal enhancement effects. The Fig. 5a records the detailed CV curves and Fig. 5b illustrates the variation of oxidation peak current (Ip) in the solution of pH 7.0 PBS containing 100.0 nM TBBPA. As shown in the Fig. 5a, only an oxidation peak was observed during the successive cyclic sweep from 0.2 to 0.9 V. Thus, the oxidation of TBBPA was totally irreversible on the surface of different modified electrodes. It could be seen in the Fig. 5, on the surface of bare GCE, a weak oxidation peak appeared at 0.59 V with the peak current of 0.12 μ A owing to the poor response to TBBPA. When using MWCNTs/GCE, the oxidation peak currents increased considerably and the peak current became 0.33 μ A, which suggested that the prepared MWCNTs displayed remarkable signal enhancement on TBBPA oxidation. The oxidation peak currents of TBBPA was further enhanced on the surface of MWCNTs-Fe 3 O 4 hybrid film, and the peak current further increased to 0.77 μ A, the value of oxidation peak currents are increased by 6.4-fold in comparison with those on bare GCE. And, the oxidation responses of TBBPA in the presence of TOAB were also examined. It should be noted here that the oxidation peak currents of TBBPA increased significantly after addition of TOAB. The peak current of TOAB/MWCNTs-Fe 3 O 4 modified GCE was 1.43 μ A, which was almost 12 times as large as the peak current of bare GCE. Such peak current enlargements testify the strong signal amplification of TOAB. From Fig. 5, we clearly found that the response signals of TBBPA are the highest on MWCNTs-Fe 3 O 4 hybrid film surface with the assistance of TOAB.
Based on the results of electrochemical comparative studies and the chronocoulometry study, there are two main reasons may account for the mechanism of synergetic signal enhancement between each component. On the one hand, the effective surface area and the electron transfer rate of electrodes were improved by each modification step. Therefore, the electrochemical performance towards TBBPA may be improved in the construction process. On the other hand, the accumulation ability of different electrodes was also improved as demonstrated in the chronocoulometry study. The bare GCE showed a poor response to TBBPA, which may be attributed to the poor electrochemical performance and accumulation ability of bare GCE. When MWCNTs was modified onto the GCE, the electrochemical performance and accumulation ability were both improved. This may be ascribed to the higher electron transfer rate and good conductivity after the modification of MWCNTs. And the TBBPA could be absorbed on the nano-structure of MWCNTs. Consequently, the larger oxidation current peak of TBBPA was showed on the MWCNTs/GCE. After the Fe 3 O 4 nanoparticles were introduced onto the MWCNTs, the MWCNTs-Fe 3 O 4 /GCE showed a further improvement of electrochemical performance and accumulation ability. Fe 3 O 4 exhibits good electrical properties owing to the electron transfer between Fe 2+ and Fe 3+ , this could enhance the electrode conductivity and facilitate electron transfer 43 . So, the electrochemical performance of MWCNTs-Fe 3 O 4 /GCE was further improved compared with MWCNTs/GCE. At the same time, the three-dimensional structure of MWCNTs-Fe 3 O 4 hybrid could absorb more target molecule 56 . Thus, the signal enhancement effect could be observed. Finally, when TOAB was modified on the electrode, electrochemical performance and accumulation ability were greatly improved on the surface of the MWCNTs-Fe 3 O 4 /GCE. This could be explained by the reasons discussed above. Hence, the further signal enhancement was showed. To summarize, the system of MWCNTs-Fe 3 O 4 hybrid and TOAB exhibited very strong synergetic signal amplification on the oxidation of TBBPA, which confirmed our original hypothesis, and the detection sensitivity could be greatly improved.

The effects of different experimental conditions on the signal response were investigated.
he effect of TOAB concentration on the oxidation response signal of TBBPA was also discussed. The Supplementary Figure. S2 shows that the oxidation peak currents of TBBPA increased greatly with TOAB concentration from 0 to 0.19 mM (75 mg L −1 ), and then gradually decreased with the increasing TOAB concentration. The notable peak current improvements manifest that the prepared surfactant films are more active for the direct oxidation of TBBPA. Herein, the concentration of TOAB was fixed at 0.19 mM for optimum sensitivity.
The effect of pH value was also discussed. In most cases, the electrolyte pH is an important parameter in the electrochemical reaction. The influence of the solution pH on the electrocatalytic oxidation of TBBPA (500.0 nM) was investigated within the pH range of 5.5 to 8.5 phosphate buffer (Fig. 6a). As displayed in Fig. 6b, the oxidation peak currents of TBBPA on the TOAB/Fe 3 O 4 -MWNTs/GCE increased gradually when pH value improved from 5.5 to 7.0, and then decreased gradually with further improving pH value. Therefore, the pH 7.0 phosphate buffer solution was chosen for the subsequent analytical experiments to achieve higher response signal. The oxidation peak potential shifted negatively with the increase of pH value, which indicated that the protons were involved in the electrode reaction when pH increased from 5.5 to 8.5. Moreover, a good linear relationship can be established between anodic peak potential (Epa) and the solution pH. As shown in Fig. 6c, the regression equation can be expressed as Epa (V) = − 0.0559 pH + 1.004 (R 2 = 0.9979). The calculated slope value of 0.0559 V pH −1 is approximate close to the theoretical value of 0.0592 V pH −1 according to the Nernst equation, which indicates that the electron transfer is accompanied by an equal number of protons in the electrode reaction. The effect of scan rate on the electrochemical behavior of TBBPA (100.0 nM) at the TOAB/Fe 3 O 4 -MWNTs/ GCE was investigated by cyclic voltammetry. The Fig. 6d shows that, at scan rates from 10 to 250 mV s −1 , the oxidation peak currents (Ipa) of TBBPA increase linearly with the scan rates. The linear equation of TBBPA is Ipa (μ A) = 0.0104ν (mV s −1 ) + 0.0166 (R 2 = 0.9994) (Fig. 6e). The results indicate that the process was predominantly adsorption-controlled process. In addition, there was a linear relation between anodic peak potential (Epa) and logarithm of scan rate (logν ), which corresponds to the following equation: Epa (V) = 0.0555logν (mV s −1 ) + 0.4977 (R 2 = 0.9976) (Fig. 6f). In the case of an adsorption-controlled and irreversible electrode process, Laviron mentioned that the slope of the line is equal to 2.3RT/(α nF) 57 . Moreover, α was assumed to be 0.5 in totally irreversible electrode process 58 . Therefore, based on the value of the equation above, the electron transfer number (n) could be calculated as 2 in this study.
The electrochemical oxidation of TBBPA on TOAB/MWCNTs-Fe 3 O 4 /GCE was suggested to be a two-electron and two-proton process given that the number of electron and proton involved in the oxidation process of TBBPA was equal as demonstrated in the pH-dependent electrochemical response. The electro-oxidation mechanism of TBBPA could be illustrated in Supplementary Fig. S3.
Fixing the accumulation potential and accumulation time was significant when adsorption studies were intended. Both conditions could affect the amount of adsorption of TBBPA at the electrode. Thus, the effect of accumulation potential and time on the CV signal was studied. The concentration of TBBPA was 100.0 nM. A little change in the peak current was observed when accumulation potential varied from − 0.30 to + 0.30 V. However, comparison of the obtained peak current for the open-circuit accumulation with those for polarization accumulation did not show significant difference. Therefore, the open circuit accumulation was applied. The influence of accumulation time ranges from 0 to 10 min on the oxidation of TBBPA at TOAB/MWCNTs-Fe 3 O 4 / GCE was also investigated (see Supplementary Fig. S4). The peak current increased gradually as the accumulation time increased from 0 to 6 min. However, the peak current tends to be almost stable with the further increase of accumulation time beyond 6 min. Therefore, the optimal accumulation time of 6 min was chosen in further investigations.
The electrochemical Determination of TBBPA was studied. Differential pulse voltammetry (DPV) technology was used for quantitative detection of TBBPA because of the sensitivity of DPV was higher than the CV. Under the optimal conditions, DPV was performed to investigate the relationship between the peak current and the concentration of TBBPA. The pulse amplitude was 50 mV, the pulse width was 50 ms, and the scan rate was 40 mV s −1 . As shown in Fig. 7a and b, the oxidation peak current of TBPPA increased linearly with the concentration in the range from 3.0 to 1000.0 nM. The linear regression equation was calculated as: Ipa (μ A) = 0.0136 C (nM) − 0.022 (R 2 = 0.9988), and the limit of detection was estimated to be 0.73 nM based on three signal-to-noise ratio (S/N).
The repeatability of one TOAB/MWCNTs-Fe 3 O 4 /GCE was examined by the detection of 100.0 nM TBBPA for five successive determinations. The relative standard deviation (RSD) is 2.85%, which shows that the excellent repeatability of the modified electrode for TBBPA detection was obtained. Fabrication reproducibility was also estimated with five different electrodes, which were fabricated independently by the same procedure. The RSD is 4.67% for the peak current measured in 100.0 nM TBBPA, which demonstrates the reliability of the fabrication procedure. In addition, the fabricated electrode retained 93.04% of its initial response after stored in the refrigerator at 4 °C for one week, which indicates that the good long-term stability of the TOAB/MWCNTs-Fe 3 O 4 /GCE can be achieved.
The influence of potentially interfering substances on the determination of analyte compound was investigated to evaluate the selectivity of the method for the determination of TBBPA. The tolerance limit for interfering species was considered as the maximum concentration that gave a relative error less than ± 5.0% at a concentration of 100.0 nM of TBBPA. Table S1 presents that the inorganic species K + , Na + , NH 4+ , Ca 2+ , Mn 2+ , Cd 2+ , Zn 2+ , Cu 2+ , Al 3+ , Fe 2+ , Fe 3+ , Cl − , Ac − , SO4 2− , and PO4 3− in a 500-fold concentrations almost had no influence on the detection of 100.0 nmol L −1 TBPPA with deviations below 5% (see in the Supplementary Table S1). Moreover, 100-fold concentration of some organic substances and organic phenolic compounds such as glucose, p-aminophenol, 4-nitrophenol, 3-aminophenol, o-nitrophenol, m-nitrophenol, p-nitrophenol, catechol, hydroquinone, phenol, nonyl phenol, showed no influence on the electrochemical signal with deviations below 5%.
Although, bisphenol A ( BPA ), 4, 4′ -Dihydroxydiphenylmethane ( BPF ), 4,4-Dihydroxydiphenyl sulphone (BPS) hexafluorobisphenol A ( BPAF ), 2, 2′ ,6, 6′ -tetrabromobisphenol A diallyl ether ( TBBDE ), tetrabromobisphenol A bis ( dibromopropyl ether ) ( TBBME ), 4, 4′ -Sulphonylbis ( 2, 6-dibromophenol ) ( TBBPS ), tetrachlorobisphenol A (TCBPA), showed interference effect on the determination of TBBPA in different levels owing to the similar structure to TBBPA, the signal changes were all less than 5%, thereby indicating the selectivity of the prepared sensor was acceptable. The comparison with other reported methods shown in Supplementary Table S2 reveals that the newly developed TOAB/MWCNTs-Fe 3 O 4 /GCE based sensor is sensitive and competitive. For the further evaluation of the applicability of the method, the proposed sensor was applied to detect TBBPA in water samples collected from the environment. And, to verify the performance of the proposed method, HPLC method was used to determine TBBPA in real samples at the same time. The standard addition method was used for calculating the TBBPA concentrations given that TBBPA had not been detected from original water samples. The results obtained are shown in Table 1. The recovery ranged from 95.5% to 106.5%, and relative standard derivation (RSD, n = 5) was below 5.0%. These results indicated that the relative standard derivations (RSDs) and the recovery rates were acceptable. Furthermore, the results obtained by the proposed electrochemical method were in a good agreement with the HPLC method. Thus, the prepared sensor could be applied reliably and effectively for the determination of TBBPA in water samples.

Conclusions
In summary, a novel electrochemical sensor was constructed to determine TBBPA in the water samples. After hybridization with Fe 3 O 4 nanoparticles, the obtained MWCNTs-Fe 3 O 4 hybrid exhibited stronger signal enhancement on TBBPA oxidation, compared with MWCNTs. The addition of TOAB further enhanced the oxidation signal of TBBPA and displayed remarkable synergetic signal amplification effects. This new sensing method exhibited excellent sensitivity and stability with the detection limit as low as 0.73 nM. Finally, the new sensing method could be successfully applied to detect TBBPA in water samples with satisfactory results.

Methods
Reagents and apparatus. Tetrabromobisphenol A (TBBPA, 99%) and Trimethyloctadecylammonium bromide (TOAB) were purchased from Sigma-Aldrich. The 0.01 mol L −1 of stock solution of TBBPA was prepared with methanol and stored at 4 °C. The multi-walled carbon nanotubes (MWNTs, purity > 95%) was obtained from the Nanjing XFNano Advanced Materials Supplier (Nanjing, China). All other reagents and solvents were of analytical grade and used without further purification. Ultrapure water was produced from a Milli-Q purification system (Millipore, Germany) and used throughout the experiments. The phosphate buffer solutions (PBS) were prepared in the pH range 5.5 to 8.5 and were used as supporting electrolytes.
Electrochemical measurements were performed with a CHI660C electrochemical workstation (CH Instruments, Shanghai, China). A conventional three-electrode system was used, containing a bare or modified glassy carbon electrode as the working electrode (Φ = 3 mm), a saturated calomel electrode (SCE) as the reference electrode and a platinum wire as the counter electrode. X-ray diffraction (XRD) pat-terns were measured using a X'Pert PRO diffractometer (Panalytical Company, Netherlands). Transmission electron microscopy (TEM) images were recorded with a Tecnai G220U-Twin microscope (FEI Com-pany, Netherlands). The Fourier transform infrared (FTIR) spectra were recorded with a VERTEX 70 spectrometer (Bruker, Germany). The high-performance liquid chromatograph (HPLC) was performed on Agilent 1200 series with a UV-vis detector (Agilent Technologies, United States). The analytical column used was Waters Sunfire C18 column (150 × 4.6 mm, 5 mm). The mobile phase was methanol-water (85/15, v/v) and the flow rate was 1.0 mL min −1 . The injection volume was 20 μ L and the UV spectra was recorded at 212 nm. All measurements were performed at ambient temperature.

Synthesis of Fe 3 O 4 /MWNTs-COOH nanocomposite.
Initially, chemical pretreatment is a simple and effective way to introduce oxygen groups (COOH, C= O and OH) onto the MWCNTs surface to improve its properties 59 . The MWCNTs-COOH was prepared according to a method reported by Zhang 60 . An amount of 500 mg of MWCNTs was added to a 60 mL solution of H 2 SO 4 -HNO 3 (3 : 1, v/v raio) and the mixture was ultrasonicated for 15 min. Then, the mixture was stirred at 85 °C for 12 h. After cooling to room temperature, the product was isolated by filtration through a 0.22 mm polycarbonate membrane and washed with water several times until the pH of the filtrate was neutral. Finally, the resulting MWCNTs-COOH was dried under vacuum at 40 °C for 12 h. NaAc (3.6 g) and a suitable amount of MWCNTs were then added into the homogeneous solution of FeCl 3 (0.81 g) and dissolved in ethylene glycol (40 mL). Ultrasonic for 30 min was employed to disperse the mixture and the mixture was then transferred to a teflon-lined stainless-steel autoclave. The autoclave was heated to and   Table 1. Determination of TBBPA in real samples by the developed electrochemical method (n = 5) and HPLC method (n = 5).
maintained at 200 °C for 8 h, and allowed to cool to room temperature. Finally, the black products were obtained and washed several times with water and ethanol and then dried at 60 °C.
Preparation of working electrodes. First, in order to obtain the stable MWCNTs-Fe 3 O 4 suspension, 10.0 mg of MWCNTs-Fe 3 O 4 composite was exactly weighed and added into 10.0 mL dimethylformamide (DMF). After an ultrasonic treatment for 60 min, a stable and black suspension with concentration of 1 mg mL −1 was achieved. The solutions of TOAB with concentration of 1.0 mg mL −1 was prepared using ethanol, and then diluted to different concentrations for the subsequent modification.
Prior to the preparation of the modified electrode, the bare glassy carbon electrode (GCE) was polished to obtain a mirror-like surface with 0.3 and 0.05 μ m alumina slurry, respectively, and then sonicated in alcohol and ultrapure water successively to remove any residue. After drying in the air at room temperature, the clean GCE was modified with a certain amount of the MWCNTs-Fe 3 O 4 suspension and dried under an infrared lamp. Afterwards, 2 μ L of the diluted TOAB solution was dropped onto GCE surface, and kept at room temperature to evaporate ethanol completely. The prepared electrode was denoted as TOAB/MWCNTs-Fe 3 O 4 /GCE.
Water sample preparation. Water sample was collected from the East Lake and Yangtze River (Wuhan, China) and detected using the presented method and HPLC method, respectively. The sample was filtered by 0.45 μ m membrane prior to the electrochemical detection. The filtered water sample was then mixed with NaHPO 4 and Na 2 PO 4 for preparing 0.1 M phosphate buffer. The pH value was adjusted to 7.0. For HPLC method, the water sample was filtered by 0.45 μ m and 0.22 μ m membranes successively and then the pH value was adjusted to 7.0. The sample was then mixed with methanol (1: 1, v/v). All prepared samples were stored in refrigerator at 4 °C before analysis.